nsrd(1m) nsrd(1m) · option has no additional effect. on all other platforms, the normal behavior...

NSRD(1m) NSRD(1m)

NAMEnsrd − daemon providing the NetWorker service

SYNOPSISnsrd [ −k virtual-service-name]

ansrd [ commentary ]

DESCRIPTIONThe nsrd daemon provides an RPC-based save and recover service. This service allows users to save,query for, and recover their files across a network. The RPC program number provided bynsrd is 390103.

Normally nsrd is invoked from a startup shell script (for examplerc.local, rc.boot) at boot-time, andshould never need to be started directly by a user. After it is started,nsrd starts up the other daemons itneeds to provide the NetWorker service.

Thensrd command must be run on a machine with appropriate resources. These resources include devices(for example, tape drives) which are under the control of the media multiplexor software (seensr-mmd(1m)), and sufficient disk space for the index daemons, (seensrindexd(1m) andnsrmmdbd(1m)) tomaintain the index of sav ed user files and volumes with corresponding files.

Each time a backup, recover, or another session begins,nsrd starts the program,ansrd, to process therequested session. Theansrd program is called anagent. The agent is in charge of monitoring thatbackup, recover, or another session, and automatically exits when a session completes. Usingps(1) oranother process monitoring tool, you can inspect the subsequent parameters ofansrd to see what kind ofsession it is monitoring. If necessary, agents can be forcibly terminated to abort a backup or recover ses-sion. Agents cannot be run directly; they can only be started bynsrd.

Whennsrd is started with the−k option, it checks to see whether it has been installed as a cluster serviceand that the virtual host which owns/nsr/res matchesvirtual-service-name.If either of these validationsteps fails,nsrd exits immediately. (To check whether NetWorker has been installed as a cluster service,nsrd checks for a file calledNetWorker.clustersvr in the directory containing thensrd binary. To checkthat/nsr/res is owned byvirtual-service-name, nsrd queries the cluster management software.)

If the −k option is not used when starting NetWorker in a cluster, the server assumes the identity of the vir-tual host which owns/nsr/res. If no virtual host owns/nsr/res, thennsrd will not start.

OPTIONS−k virtual-service-name

Instructs nsrd to start up in cluster failover mode usingvirtual-service-nameas its host-name/identity. This option is used by the NetWorker cluster control script which starts NetWorker.

FILES/nsr/logs/daemon.log

The file to whichnsrd and other NetWorker daemons send information about various errorconditions that cannot otherwise be logged using the NetWorker event mechanism.

/nsr/res/nsr.resAttributes describing the NetWorker service and its resources (Seensr_service(5)).

/nsr/res/nsrjb.resAttributes describing the jukebox resources of the NetWorker service.

NetWorker.clustersvrIf this file exists in the directory containing NetWorker’s daemons, it indicates that the Net-Worker server has been installed as a cluster service.

SEE ALSOnsr(1m),nsr_service(5), nsrmmd(1m),nsrmmdbd(1m),nsrindexd(1m),ps(1), rc(1m).

NetWorker 6.1.1 October 15, 2001 1

CHANGERS(1m) CHANGERS(1m)

NAMEchangers− list SCSI autochangers attached to the system

SYNOPSISchangers[ -dv ] [ -a b.t.l ] [ -l ]

DESCRIPTIONThechangersprogram lists all of the SCSI autochangers (jukeboxes) connected to the current system.

OPTIONS−d Determines the names and addresses of the autochanger’s media elements (for example, tape

drives).

-v Lists more detailed information about each autochanger. The details provided may indicate howmany media transports (MT, for example, robot arm), storage transports (ST, for example, slot),import/export elements (IE, for example, mail slot), of data transport (DT) elements theautochanger contains. The-v option also provides information about the element movement matrixsupported by the autochanger.

−a b.t.l Selects a specific ordinal SCSI address, whereb is the logical SCSI bus,t is the SCSI target, andlis the SCSI logical unit number (LUN) on that target. Seelibscsi(1m).

−l Performs a complete LUN search for all SCSI adapters in the system. This argument is acceptedon all systems, but does not have any effect on HP-UX systems. Due to the method used to scanfor available devices on HP-UX systems, all accessible devices are always shown, and the−loption has no additional effect. On all other platforms, the normal behavior is to start checking atLUN 0 for SCSI deivces. The first empty LUN found will end the search for a given target ID.With the −l option, all LUNS present on all target IDs for all SCSI busses in the system will bechecked for devices. This can take avery long time and should therefore only be used when nec-essary. For example, a Fibre Channel adapter can support 126 target IDs, each of which may have80 or more LUNs. Checking all LUNs on this single adapter may take over 10 minutes.

EXAMPLESample output is shown below: hal$ changers -dv -a 0.2.0

[email protected]:Vendor <SPECTRA>, Product <4000>Data Transfer Element at address 80 is [email protected]

Device:Vendor <HP>, Product <C1533A>Type:Tape

System Name: /dev/rmt2.1Data Transfer Element at address 81 is [email protected]

Device:Vendor <HP>, Product <C1533A>Type:Tape

System Name: /dev/rmt3.1

1 MT Element starting at address 7960 ST Elements starting at address 11 IE Element starting at address 02 DT Elements starting at address 80

Element Movement Matrix

->DT, ->IE, ->ST, ->MTMT->DT,MT->IE,MT->ST,______ST->DT,ST->IE,ST->ST,ST->MTIE->DT,______,IE->ST,IE->MTDT->DT,DT->IE,DT->ST,DT->MT


CHANGERS(1m) CHANGERS(1m)

______,______,______,____________,______,______,____________,______,______,____________,______,______,______

SEE ALSOlibscsi(1m)


EMASS_silo_files(8) EMASS_silo_files(8)

NAMEdasadmin − ADIC/EMASS/Grau silo administrative utilitylibstlemass − shared library for communication to

ADIC/EMASS/Grau silo

SYNOPSISdasadmincommand [options] [parameters]

dasadmin.exe command [options] [parameters] (NT only)libstlemass.so (Solaris)libstlemass.so.a(AIX)libstlemass.sl (HPUX)libstlemass.so.1(SGI)libstlemass.so (DECAXP)libstlemass.dll (NT i386)

DESCRIPTIONdasadmin

This is not a complete listing of all possible dasadmin commands, but does include those commands thatare of use with NetWorker. For a complete discussion, please see the DAS Installation and Administrationguide provided by ADIC, EMASS or Grau.

mo[unt] [ −t type] volser[ drive-name]mounts the tape with the barcode label ofvolserinto either the first available drive (ifdrive-nameis not specified) or into the drive specicified bydrive-nameIf the tape is not the type defined byDAS_MEDIUM or ACI_MEDIA_TYPE , you can use the−t type option to get the tapemounted. If the type of the tape and the defined type for the drive do not match, the silo will notload the tape. Note that the drive you are attempting to use must be allocated for your use beforeyou can mount or dismount tapes. Seelistd andallocd below.

dism[ount] [ −t type] volser | −d drive-namedismounts the tape that is either specified byvolseror whatever is in the drive specified bydrive-name. If the tape or drive are of a different type than your default, use the−t typeparameter. Aswith mount, you must have the drive allocated to you to use this command.

ej[ect] [ −c ] [ −t type] volser-range area-nameejects one or more tapes to the specified eject area. As with other commands, if the type of thetape you are ejecting is different from that defined byDAS_MEDIUM orACI_MEDIA_TYPE ,you will need the−t typeoption. The−c specifies a ’complete’ ejection for the specified volsers.A complete ejection removes the entry for that volser from the silo controller’s internal database.A NON-complete ejection will eject the tape, but the volser’s entry in the database will remain,and the volser’s state will be set to ’ejected’. This is useful if you anticipate replacing the tapeinto the silo soon.

in[sert] area-nameMoves all tapes that are currently in the specified insertarea-namefrom the insert area to thenormal storage locations for tapes.

inventoryStarts a full inventory of the silo.USE WITH CAUTION! An inventory of this sort can take avery long time! An inventory of our silo with 180 slots takes over 20 minutes.

view [ −t type] volserWill display the current status ofvolser. This will show the volser, type, attribute and coordinate.

all[ocd] drive-nameUP|DOWN clientnameallocd is used to allocate and deallocate drives for different clients. Before you can use a tapedrive, the drive must beallocd’ed UP for your system. If it is currentlyallocd’ed UP for adifferent client, if must first beallocd’ed DOWN for that client before beingallocd’ed UP foryour system. You cannotallocd DOWN a drive that has a tape in it. The tape must bedismounted first.

l[ist]dlistdorld will show the current state of all the tape drives defined in the silo. The informationpresented will include thedrive-name,the amu drive (the location in the silo), status (UP or



DOWN), type, client the drive is allocated to, and the volser of any loaded tape.show −op| −acclient-name

Shows the operational or access parameters for the specifiedclient-name. You must includeeither−ac if you wish to see access parameters, or−op if you wish to see operational parametersfor the client-name. Access parameters include volser ranges and drive ranges that theclient-nameis allowed to use. Operational parameters include whether theclient-namehas completeaccess, dismount priviledges along with the IP address entered forclient-name

list client-nameLists any outstanding requests that have been made byclient-name. If there are any, they areshown, along with the request number and type.

can[cel] request-idAllows you to cancel an outstanding request, assuming that you have the necessary priviledges.Use therequest-idthat was shown by thelist command.

qversionThis shows the version of the DAS server that you are connected to and the version of the ACIprotocol you are using to talk to DAS over.

qvolsrangebeginvolser endvolser count[ clientname]qvolsrangeis the way to obtain a list of the volsers that are available in the silo.beginvolserandendvolserare volsers of the form "123456". To use the first available or the last available, youcan use "".countspecifies the maimum number of volsers you wish to see.

ENVIRONMENT VARIABLESThese environment variables affect the operation of the silo, and since the processes that are using theminclude both commands the user will enter and processes that are spawned fromnsrd, they need to beset in a location where they will be in place whennsrd is started. The three DAS_ variables are used bylibstlemass,while dasadminusesACI_MEDIA_TYPE instead ofDAS_MEDIUM

For Solaris, the definitions should be placed in/etc/rc.2/S95networker.

For AIX, the definitions should be placed in/etc/rc.nsr.

For HPUX, the definitions should be placed in/sbin/rc2.d/S900networker.

DAS_SERVERThis is either the network name of the IP address of the system that is running DAS. For a single silo,this will usually be the silo controller system. In larger installations, there will probably be only oneDAS server for the whole network. It is case sensitive.

DAS_CLIENTThis is the network name of the system that NetWorker is running on. It is case-sensitive.

DAS_MEDIUMThis variable is used by libstlemass. It should be the same as ACI_MEDIA_TYPE.This is the type of tape drive you are connected to. If this is not specified, the default value of DLT willbe used. Acceptable values are:

3480, 3590, OPTICAL_THICK, OPTICAL_THIN, DECDLT, DLT , 8MM, 4MM, D2, VHS, CD,TRAVAN and DTF

ACI_MEDIA_TYPEThis variable is used by dasadmin. It should be the same as DAS_MEDIUMThis is the type of tape drive you are connected to. If this is not specified, the default value of DLT willbe used. Acceptable values are the same as those listed underDAS_MEDIUM.

EXAMPLESNOTE on ranges:

dasadmin will accept volser ranges for some commands. There are three acceptable variations for theseranges:single volser: "000635"multiple volsers: "000635, 000789, 098732"true range: "000610 - 000745"



NOTE onarea-nameanddrive-name:area-namesusually consist of a letter and 2 digits. The letter corresponds to whether you are referring toan insert area ("I") or an eject area ("E"). You will need to get the correct values from your siloadministrator before using them.drive-namesare essentially free-form labels concocted by whomever installed the silo. They may or maynot have any relevance to physical reality, so you will need to see the silo admin to get the correct names.If the silo admin is not available, you can get the same information usingdasadmin listd along withdasadmin show −opclient-namefollowed bydasadmin show −acclient-namecommands.

To setup the environment variables necessary for silo operations:setenv DAS_SERVER emasksetenv DAS_CLIENT aurorasetenv DAS_MEDIUM DLTsetenv ACI_MEDIA_TYPE DECDLT

To see a listing of all volsers available in the silo:dasadmin qvolsrange "" "" 10000

To see the current status of the drives in the silo:dasadmin listd

To change the allocation of a drive from client a4 to client aurora:dasadmin allocd DLT1 DOWN a4dasadmin allocd DLT1 UP aurora

SEE ALSOnsrjb (1m), jbconfig(1m), libstlstk (1m),mini_el(1m),ssi(1m), libstlibm (1m)

DIAGNOSTICSThe only available diagnostic information is error messages that may be printed out by dasadmin andlibstlemass in the course of normal operations.


DDMGR(1m) DDMGR(1m)

NAMEddmgr − Device detection manager. Manages auto-detection on local and remote storage nodes.

SYNOPSISddmgr [ −S ] [ −M ] [ −q ] [ −v ]

DESCRIPTIONddmgr is the main daemon for auto-detection that runs on the networker server machine. It spawns childprocesses (of dvdetect) for each storage node on which devices are to be detected.

It starts the child processes with the help of the nsrmon(1m) process, and depends on nsrmon to report onthe success or failure of the remote dvdetect process.

Once dvdetect on a storage node has finished its work of detecting devices, ddmgr takes up the task of cre-ating resources for these detected devices, and in case of jukeboxes, tries to find out the device mapping(element id to device path) by spawning another process, dtbind. dtbind determines the device mapping byloading each drive in the jukebox that was detected and then trying to access it via various device paths tillit finds the right one. This might take a long time depending on the type of the jukebox.

ddmgr is invoked by the nsrd process and is not be invoked on the command-line.

OPTIONS−q Tellsddmgr to detect and create device resources but not to enable them.

−M This option tellsddmgr that it has been invoked by the server and to direct messages to the dae-mon log.

−q This option tells ddmgr to run in the ’quiet’ mode without printing any messages.

−v This option is used to run ddmgr in the verbose mode for more debug messages.

EXIT STATUSExits with 0 on sucess and 1 on error. See error messages for more detail on errors.

SEE ALSOnsrmon(1m)

DIAGNOSTICSMost, if not all, of ddmgr error reports is preceded by the phrase "Detection process for host X reports", fol-lowed by the actual error message. This error message is based on the error reported by the nsrmon processmonitoring the dvdetect process, or in cases where nsrmon itself cannot be started, about the nsrmon pro-cess. The following are the error messages that ddmgr might produce along with their implications and pos-sible solutions.

remote dvdetect exec failure. Errno 76The remote storage node was unable to start the dvdetect process on the remote storage node. Thiscould happen for various reasons, like the dvdetect binary not having execute permissions, or morecommonly, the remote storage node not being configured to service requests from this server.

remote auto-detect feature not supportedAuto-detect was being performed on a host that does not support this feature. The client/storage-node should be 6.x or higher.

dvdetect process failed on signalThe remote dvdetect process was killed by a signal. This could happen even when the processencounters a memory fault. Check for core files in the nsr/cores directory.

dvdetect terminated due to timeoutThe dvdetect process was terminated because of its inactivity for a certain period of time. Thetimeout is set by default to 15 minutes. This is not user configurable.dvdetect process exited onsignal The local dvdetect process was killed by a signal. This could happen even when the processencounters a memory fault. Check for core files in the nsr/cores dir ectory.


DDMGR(1m) DDMGR(1m)

dvdetect exec failureThe ddmgr process was unable to start the dvdetect process on the server. Check for execute per-missions on the dvdetect binary.

nsrmon exec failureThe ddmgr process was unable to start the nsrmon process on the server. Check for execute per-missions on the nsrmon binary.

nsrmon process exited on signalThe nsrmon process exited on a signal. This could even happen whenthe process encounters amemory fault. Check the nsr/cores direcotry for a core file.

dvdetect failed with unknown errorddmgr was unable to determine the cause of the failure of the dvdetect process.

nsrmon failed. No info in the resdbThe nsrmon process exited without loggin any information about either the remote dvdetect pro-cess or itself. Ddmgr is unable to verify status of both.

nsrmon failed. Invalid request or hostnameThe nsrmon process was started with an invalid option or hostname. Check if the remote storagenode is reachable from the server.

nsrmon failed. Authorization failureThe nsrmon process could not get authorization from the networker server to talk to the remotestorage node.

nsrmon exited on resdb access failureThe nsrmon process encountered errors in reading the networker RAP database.

nsrmon exited on memory failureThe nsrmon process ran out of physical memory while processing. Add more memory.

nsrmon failed. Invalid request valueThe nsrmon process was asked to perform a request it is not familiar with.

process exited with errorThere was a problem with the detection process but ddmgr could not determine the exact cause ofthe failure.

RPC error Remote systemsThe nsrmon process was unable to connect to the remote host. This could be because of networkproblems, or if the networker processes were not installed on the remote system.







DESCRIPTIONdasadmin











EMCDISCOVER(1m) EMCDISCOVER(1m)

NAMEemcdiscover− Discover Symmetrix configuration information

SYNOPSISemcdiscover[ −v ]

DESCRIPTIONThe emcdiscoverprogram scans all devices on the host looking for Symmetrix devices and builds orrebuilds the Symmetrix host database.

You must runemcdiscoverbefore your first backup and anytime after you reconfigure your Symmetrix byadding or removing devices.

The emcdiscoverprogram must be run as the root user and on both the primary and secondary hostmachines involved in a backup.

OPTIONS−v Verbose: Instruct the shell to print commands and their arguments as they are executed.

SEE ALSOoraemcasm(1m),oraemcmap(1m)


ERASE(1m) ERASE(1m)

NAMEerase − erase a tape

SYNOPSISerase[ -slr ] -a b.t.l

DESCRIPTIONThe eraseprogram will send the SCSI ERASE command to the named device, using the LONG eraseoption unless the optional-s argument is specified.

The required-a argument must be used to select a specific ordinal SCSI address (seelibscsi(1m)) for thedevice that has the tape.

The optional-r argument sends a REWIND command to the named device prior to issuing an erase com-mand.

WARNINGSBe careful! This command destroys data! It doesnot prompt you to see whether you’re sure you want to dothi.s

SEE ALSOlibscsi(1m)


HPFLIP(1m) HPFLIP(1m)

NAMEhpflip − flip device type of HP optical disk drives

SYNOPSIShpflip [ -a b.t.l ] [ -r ]

DESCRIPTIONThe hpflip program reads a Vendor Unique mode page from an HP Optical disk drive, and toggles thedevice type setting to the appropriate device type. The device type can either be OPTICAL or DIRECTACCESS. Typically, most systems have drivers that can handle removable DIRECT ACCESS device types(often limited to 512 byte/sector formatted disks). Systems with these device types do not often have devicedrivers for additional OPTICAL device types. Thehpflip program enables you to control how an HP Opti-cal Disk Drive reports itself, and therefore makes the OPTICAL device available where it otherwise wouldhave required an additional device driver.

OPTIONS−a Selects a specific ordinal address (seelibscsi(1m)). This is a required option.

−r Resets the specified device to Optical, regardless of its current state. Without this argument, thedevice setting is toggled between OPTICAL and DIRECT ACCESS types.

SEE ALSOlibscsi(1m)


IBM_silo(8) IBM_silo(8)

NAMElibstlibm − shared library for communication to IBM 3494 silos

SYNOPSISlibstlibm.so (Solaris)

libstlibm.so.a (AIX)

DESCRIPTIONlibstlibm.xxx is a shared library that handles the communication betweennsrjb and the IBM silo driver(on AIX) or daemon (on Solaris). The IBM driver/daemon then handles the communication over thenetwork to the silo. There are no options, parameters or environment variables that affect the operation oflibstlibm. The correct path to this file should be entered when an IBM silo is configured usingjbconfig.The default values specified byjbconfig match the default locations chosen for the installation program,and in most cases can be accepted.

For NetWorker to work with the 3494, you must have first installed IBM’s Automated Tape Librarysupport.

On AIX, this will install a driver calledatldd (Automated Tape Library Device Driver). You may alsorequire IBM’s atape driver (Enhanced Tape and Medium Changer Device Driver) if you are using 3590drives in your 3494.

On Solaris, you will need to install thelmcpd package, (IBM Automated Tape Library Daemon) to use thesilo. Again, if you are using 3590 drives, you will also need to install theIBMtape driver. Note that whenyou are usingIBMtape, there will be two sets of device files that will access a given tape drive. There willbe the standard Solaris style/dev/rmt/Xmbn type, and there will be theIBMtape supported files of thetype /dev/rmt/Xstbn. You should use the IBM supported device files for proper operation of you tapedrives.

NOTE: Legato cannot supply these IBM drivers. They may be available on an IBM Device Driver ftp site(208.200.29.244), but this is just something we found, and is not necessarily a long-term IBM committedsite.

SEE ALSOnsrjb (1m), jbconfig(1m),dasadmin(1m), libstlemass(1m),ssi(1m),mini_el(1m), libstlstk (1m)

DIAGNOSTICSErrors in communication between the NetWorker server and the IBM 3494 silo are difficult to diagnose.The best method is to use the IBM supplied utilitymtlib to verify that you have properly configured the3494 to communicate with your host, and that the entire pathway from either the lmcp driver (on AIX) orthe lmcpd daemon (on Solaris) is functionong properly. Ifmtlib does not work, then there is no chance thatNetWorker will work.

If there are any questions about the connection between your host and the 3494, it is best to consult IBM, asthey support the connection between the the host and the silo. IBM supports both network and serial cableconnections to the silo. Since the nature of the connection is hidden from NetWorker by the driver/daemon,there is no difference to NetWorker between the two. Customers have successfully used both.


IELEM(1m) IELEM(1m)

NAMEielem− initialize element status

SYNOPSISielem [ -a b.t.l ] [ -r eladdr.nel]

DESCRIPTIONThe ielemprogram sends an INITIALIZE ELEMENT STATUS command to the named device.

Some changers support the ability to initialise element status for a range of elements. The command usedfor this is the Vendor Unique EXABYTE changer command:

INITIALIZE ELEMENT STATUS (with range)(command opcode 0xE7).

OPTIONS−a b.t.l Selects a specific ordinal SCSI address, whereb is the logical SCSI bus,t is the SCSI target, andl

is the SCSI logical unit number (LUN) on that target. Seelibscsi(1m). This is a required option.

−r eladdr.nelSpecifies the range of elements, whereeladdr is the starting decimal address (in the autochangersnumbering) of the element to start from, andnel is the number of elements of status to read. Thisoption can be used if your autochanger supports the Vendor Unique EXABYTE autochanger INI-TALIZE ELEMENT STATUS command.

SEE ALSOlibscsi(1m)


INQUIRE(1m) INQUIRE(1m)

NAMEinquire - list devices available

SYNOPSISinquire [ -a b.t.l ] [ −l ] [ −N ndmp option] [ −f filename]

DESCRIPTIONThe inquire program lists SCSI devices available. Theinquire program returns INQUIRY data either forthe named SCSI device (with the-a option), or for all SCSI devices attached to the system.

OPTIONS−a b.t.l Selects a specific ordinal SCSI address, whereb is the logical SCSI bus,t is the SCSI target, andl

is the SCSI logical unit number (LUN) on that target. Seelibscsi(1m).

−f Use with−N to provide a location to the configuration file of the NDMP Server. The config fileneeds to be created by running thendmpjbconf program before running theinquire program.The default location for the NetWorker Server on Solaris is/usr/lib/nsr. This location can bechanged while runningndmpjbconf.

The filename argument should be the path of the NDMP jukebox config file such as/usr/lib/nsr/ndmpjbconf_<ndmp hostname>

−l Performs a complete LUN search for all SCSI adapters in the system. This argument is acceptedon all systems, but does not have any effect on HP-UX systems. Due to the method used to scanfor available devices on HP-UX systems, all accessible devices are always shown, and the−loption has no additional effect. For all other systems, the normal behavior is to start checking atLUN 0 for SCSI deivces. The first empty LUN found will end the search for a given target ID.With the −l option, all LUNs present on all target IDs for all SCSI busses in the system will bechecked for devices. This can take avery long time and should therefore only be used when nec-essary. For example, a Fibre Channel adapter can support 126 target IDs, each of which may have80 or more LUNs. Checking all LUNs on this single adapter may take over 10 minutes.

−N For Solaris, performs an INQUIRY command to the NDMP Server with the provided NDMPServer information in the default location/usr/lib/nsr/ndmpjbconf_<NDMP_hostname>.Thislocation can be set by the LGTO_NDMP_LIBSCSI_CONFIG_FILENAME environment variableor by specifying−f <filename>. The ndmpjbconf program needs to be previously run to createthe above configuration file. If neither environment variable nor−f is provided,inquire will try toget NDMP Server information from/usr/lib/nsr/ndmpjbconffile.

The ndmp optionargument should bestandardfor devices on one of the Standard NDMP servers,netappfor devices on an NetApp NDMP server, orcelestra-hpfor devices on a Celestra HP NDMPserver. Howev er,celestrais still a valid option, for backward compatibility.

SEE ALSOlibscsi(1m)

LIMITATIONSThe inquire program always uses the built-in system drivers to test SCSI devices. The device type or pathname printed by theinquire program may be incorrect for devices that require special, third-party drivers.

The inquire program cannot retrieve device filenames for NDMP devices on an NDMP server. For aNetApp Filer, information for all drives cannot be retrieved.


JBCONFIG(1m) JBCONFIG(1m)

NAMEjbconfig − jukebox resource configuration tool

SYNOPSISjbconfig [ −sserver] [ -l ]

DESCRIPTIONThe jbconfig program provides an interactive script for configuring a jukebox (Media Autochanger Device)for use with a NetWorker server. The script pauses periodically for you to enter a response to a prompt. Ifyou want to accept the default choice displayed in braces, press [RETURN] or [ENTER].

After the jukebox is configured, use thensrcap(1m) command or the Registration window to enter theenabler code for your Autochanger Software Module. You must have a separate enabler code for each juke-box you want to use with NetWorker.

OPTIONS−sserver

Specifies the controlling server, whenjbconfig is being used from a storage node. To define ajukebox resident on a storage node, thejbconfig command must be run on the storage node. Seensr_storage_node(5) for additional information on storage nodes.

−l Performs a complete LUN search for all SCSI adapters in the system when performing Autodetec-tion. This argument is accepted on all systems, but does not have any effect on HP-UX systems.Due to the method used to scan for available devices on HP-UX systems, all accessible devices arealways shown, and the−l option has no additional effect. On all other systems, the normal behav-ior is to start checking at LUN 0 for SCSI deivces. The first empty LUN found will end the searchfor a given target ID. With the−l option, all LUNS present on all target IDs for all SCSI busses inthe system will be checked for jukeboxes. This can take avery long time and should thereforeonly be used when necessary. For example, a Fibre Channel adapter can support 126 target IDs,each of which may have 80 or more LUNs. Checking all LUNs on this single adapter may takeover 10 minutes.

CONFIGURATION DIALOGThe first questionjbconfig will ask you, is to select a type of jukebox to install.

1) Install a SmartMedia Jukebox.2) Install an Autodetected SCSI Jukebox.3) Install an Autodetected NDMP SCSI Jukebox.4) Install an SJI Jukebox.5) Install an STL Silo.

What kind of Jukebox are you installing?

Enter the number corresponding to the jukebox type you are installing.

An Autodetected SCSI Jukebox is any SCSI (Small Computer System Interface) based jukeboxconnected to a system that NetWorker can automatically detect.

Autodetected NDMP SCSI Jukebox is any SCSI (Small Computer System Interface) based juke-box connected directly to an NDMP Server, that NetWorker will automatically detect, with theprovided NDMP hostname, user-id, user-password, and jukebox handle. (See example).

An SJI Jukebox is a Standard Jukebox Interface compliant jukebox. This is a list of well knownSCSI based jukeboxes, plus any additional third party jukebox devices that adhere to this protocolthat the you may have added to the system.

If you select the second choice (Install an Autodetected SCSI Jukebox),jbconfig will print out alist of jukeboxes it detects on the system.



For example:These are the SCSI Jukeboxes currently attached to your system:

1) [email protected]: other, Vendor <AIWA>, Product <AL-17D>2) [email protected]: DLI Libra Series3) [email protected]: ARC-DiamondBack

Which one do you want to install?When this message appears, enter the number corresponding to the jukebox that you wish to con-figure.If you choose to install a Serial Jukebox,jbconfig will print a list of known Serial Jukeboxes andask which one you wish to select.

For example:Enter the number corresponding to the type of jukebox you are installing:1) Lago Datawheel (serial)2) STK-9709/Odetics (serial)3) ATL (serial)

Choice?

When this messages appears, you should enter the number corresponding to the jukebox thatyou want to configure.

If you choose to install an SJI compliant jukebox,jbconfig will print a list of known SJIJukeboxes and will prompt you for the appropriate type that you want to configure.

For example:Enter the number corresponding to the type of jukebox you are installing:

1) ADIC-1200c/ADIC-1200d2) ADIC-VLS3) ARC-DiamondBack4) Breece Hill5) DLI Libra Series6) Quantum DLT/Digital DLT7) EXB-10e/EXB-10h8) EXB-10i9) EXB-6010) EXB-12011) EXB-21012) EXB-21813) EXB-400 Series14) HP-C1553A/Surestore 12000e15) Metrum (SCSI)16) Qualstar17) Spectralogic18) STK-9704/Lago 34019) STK-9708/Lago 380 (SCSI) Datawheel20) IBM 7331/IBM 942721) ATL/Odetics SCSI22) HP-Optical 630MB/1.3GB23) other

Choice?

When this message appears, enter the number corresponding to the appropriate model, forexample, if you are installing an HP optical jukebox select the number "22".

For all jukebox types,jbconfig prompts you for thenameyou want to call this jukebox. Thisis a convenient way for you to identify the jukebox for yourself and NetWorker, for example,



Engineering Autochanger. NetWorker will store this name as a NetWorker resource (seensr_resource(5)). When defining a jukebox attached to a storage node,jbconfig prefixesthe hostname of the storage node to the beginning of the names using the remote device syn-tax ("rd=hostname:"). Seensr_storage_node(5) for additional information on storagenodes.

For all jukebox types,jbconfig prompts you for adescription of this jukebox. This isanother convenient way for you to identify the jukebox for yourself, for example, Engineer-ing 4 Drive DLT Autochanger on Rack #2.

For all jukebox types,jbconfig prompts you for the name of the control port associated withthe jukebox being configured. For Autodetected SCSI jukeboxes, the correct name is pre-sented as the default choice, because autodetection involves determining the correct name.The name is in the form oflibscsi devices (seelibscsi(1m)). For Serial jukeboxes, thedefault name varies based on the appropriate serial line name used for the platform whichNetWorker is located (for example,/dev/ttya for SunOS or Solaris, or/dev/tty0 for AIX).For SJI compliant jukeboxes, no default selection exists. The name you enter should eitherbe the device name for the device as described in any third party SJI compliant driverinstalled, or the format used for autodetected jukeboxes. A list of attached autochangers canbe obtained by running thechangers(1m) command.

Once a control port is entered,jbconfig will check to see if the model selected is a SCSI orSJI based jukebox. If the jukebox model is a SCSI or SJI based jukebox,jbconfig willattempt to query the jukebox about various internal parameters (for example, number of slotsand drives). If this query fails, it is possible that there is a device driver installation problemor hardware problem. If the jukebox is a Serial jukebox,jbconfig attempts to establish serialcommunications with it. If this fails, then there is either a setup problem or a hardware prob-lem.

Next, if the jukebox contains tape devices you are asked if automated cleaning of devices inthe jukebox should be turned on. If automated cleaning is enabled, the jukebox and alldevices in the jukebox are configured for automated cleaning. If the jukebox is successfullyinstalled, the information that pertains to device cleaning from the jukebox and all devices inthe jukebox are displayed.

For each physical drive in the jukebox,jbconfig asks whether the drive will be shared bymore than one device path. These device paths would be located on various storage nodeswithin a data zone, under the control of one NetWorker server. If the physical drive will beshared, the user is asked to enter the hardware id of the drive, which uniquely identifies thedrive. Seensr_device(5) for a description of the hardware id attribute. After this prompt isanswered, the user is prompted to enter information about each device path that will sharethe drive, which is described below.

Next, jbconfig asks for the hostname that owns the device path. This prompt allows thedevices within a jukebox to be controlled by more than one host. Note that all of the owninghosts must be storage nodes within the same data zone, under the control of one NetWorkerserver.

jbconfig then prompts you to enter the pathnames for all the media devices configured in thejukebox. You must add an entry for each device. If the device has not been configured forNetWorker, then you are prompted to enter the media device type.

If you select Autodetected SCSI jukeboxes, NetWorker determines the name of each mediadevice by sending inquiries for information to the jukebox. Not all jukeboxes support thiscapability, but many do (for example, the Exabyte 210). This inquiry does not take placewhen the owning host is different than wherejbconfig is running.

jbconfig then asks whether the devices to be configured are Network Data Management Pro-tocol (NDMP) devices. NDMP devices require a user name and password to be entered foreach device. The user name and password correspond to the entries set in the NDMP server.



Finally, NetWorker asks you if the jukebox supports a bar code label reader, and if so,whether NetWorker support for it should be turned on. If support for the bar code labelreader is turned on, then you are asked if the media volume labels should match the bar codelabels on the media. If you select this option, the label templates will not be used by thejukebox, and each media volume must have a readable bar code label.

If the jukebox has been configured successfully you will see the following message:

Jukebox has been added successfully

JBCONFIG FILEThe file /nsr/jbconfigis the jukebox models configuration file. This file can be used to configure a non-standard list of jukebox models.VECTOR-TYPE MODEL-NAME<NEWLINE> , where VECTOR-TYPE is either SJI (the StandardJukebox Interface) or ATL (RS232-based devices speaking the IGM-ATL serial communications protocol).The MODEL-NAME can be any string.

EXAMPLES(User entries are in italics).

Example 1)

# jbconfig1) Install a SmartMedia Jukebox.2) Install an Autodetected SCSI Jukebox.3) Install an Autodetected NDMP SCSI Jukebox.4) Install an SJI Jukebox.5) Install an STL Silo.

What kind of Jukebox are you installing? [2]<RETURN>These are the SCSI Jukeboxes currently attached to your system:1) [email protected]: EXB-210

Which one do you want to install?1<RETURN>Installing an ’EXB-210’ jukebox.

Name you would like to assign to the jukebox device?Engineering<RETURN>

Pathname of the control port for the jukebox device? [[email protected]]<RETURN>

Do you want automated device cleaning support enabled? (yes/no)yes<RETURN>

Will media drive 1 be shared by multiple device paths? (yes/no)no<RETURN>

Enter hostname that owns media drive 1: ? [host1]<RETURN>

Enter pathname of media drive 1: [/dev/rmt/0mbn]?<RETURN>

Should the drive be configured as a NDMP device? (yes/no)no<RETURN>


Enter hostname that owns media drive 2: ? [host1]<RETURN>

Enter pathanme of media drive 2: [/dev/rmt/1mbn]?<RETURN>

Should the drive be configured as a NDMP device? (yes/no)no<RETURN>

Following are attributes that define cleaning cartridge support for the jukebox ’Engineering’:

auto clean: Yesdefault cleanings: 12cleaning slots: 1

Cleaning cartridge volumes Slot number-------------------------- -----------



Cleaning Tape (12 uses left) 1

Make sure that the slots set aside for cleaning cartridges containing cleaning cartridges with the designatednumber of usable cleanings left.

Following are attributes that define the cleaning schedule for each device in the jukebox.

name: /dev/rmt/0mbndate last cleaned: Thu Apr 13 11:41:58 1995cleaning interval: 2 weekscleaning required: No

name: /dev/rmt/1mbndate last cleaned: Thu Apr 13 11:41:58 1995cleaning interval: 2 weekscleaning required: No

Verify that the values for these attributes are appropriate for your installation. If not check documentationon how to set up automated cleaning cartridge support.Jukebox has been added successfully

Example 2)

Here is an example of a SmartMedia autoloader configured with NDMP devices on a storage node.

# jbconfig -s serverOn a storage node, the hostname is a prefix to the jukebox name.Enter the hostname to use as a prefix? [brown.legato.com]<RETURN>using ’brown.legato.com’ as the hostname prefix


What kind of Jukebox are you installing? [1]<RETURN>Installing an SmartMedia jukebox.Name you would like to assign to the SmartMedia jukebox?myautoloader<RETURN>

Name of host machine for SmartMedia server? [brown.legato.com]<RETURN>Port number of SmartMedia server? [44444]<RETURN>

How many devices are to be configured (1 to 64)? [4]2<RETURN>Are devices NDMP devices? (yes/no)yes<RETURN>Enter hostname that owns logical device 1: ? [brown.legato.com]<RETURN>Enter name of logical device 1: ?stk1<RETURN>Enter NDMP user name: ?root<RETURN>Enter NDMP password (characters will not be echoed):password<RETURN>Enter hostname that owns logical device 2: ? [brown.legato.com]<RETURN>Enter name of logical device 2: ?stk2<RETURN>Enter NDMP user name: ?root<RETURN>Enter NDMP password (characters will not be echoed):password<RETURN>

Enter application name defined in SmartMedia for NetWorker? [NetWorker@server]<RETURN>Enter application key defined in SmartMedia for NetWorker? [<none>]<RETURN>



The barcode reader is enabled and volume labels are set to match barcode labels.

Jukebox has been added successfullyWould you like to configure another jukebox? (yes/no)no<RETURN>

Example 3)

Here is an example of configuring a Celestra/Solaris NDMP jukebox.

# jbconfig


What kind of Jukebox are you installing?3<RETURN>Enter NDMP Server name: ?amx<RETURN>Enter NDMP user name: ?root<RETURN>Enter NDMP password (characters will not be echoed):password<RETURN>Enter NDMP jukebox handle: ?/dev/rsjb0<RETURN>What is the NDMP type of ’amx’?

1) Celestra.2) NetApp.

Choice?1<RETURN>Communicating to devices on NDMP Server ’amx’, this may take a while...

These are the SCSI Jukeboxes currently attached to your system:1) [email protected]: Standard SCSI Jukebox, Vendor <HP>, Product <C7200-8000>2) [email protected]: Standard SCSI Jukebox, Vendor <ADIC>, Product <Scalar DLT 448>

Which one do you want to install?1<RETURN>

Installing an ’Standard SCSI Jukebox’ jukebox.Name you would like to assign to the jukebox device?hp<RETURN>

Pathname of the control port for the jukebox device? [[email protected]]<RETURN>

Do you want automated device cleaning support enabled? (yes/no)yes<RETURN>


Enter hostname that owns media drive 1: ? [boxster]amx<RETURN>

Enter pathname of media drive 1: ?/dev/rmt/0mbn<RETURN>

using ’rd=amx:/dev/rmt/0mbn’ as device name

Should the drive be configured as a NDMP device? (yes/no)yes<RETURN>

This media device has not been configured yet. Please select a media device type forrd=amx:/dev/rmt/0mbn.

1) 3480 21) sdlt



2) 3570 22) tz853) 3590 23) tz864) 4890 24) tz875) 4mm 25) tz886) 4mm 4GB 26) tz897) 4mm 8GB 27) tzs208) 4mm 12GB 28) tkz909) 4mm 20GB 29) dst (NT)10) 8mm 30) dst11) 8mm 5GB 31) dtf12) 8mm 20GB 32) himt13) 8mm AIT 33) LTO Ultrium14) 8mm AIT-2 34) qic15) 8mm Mammoth-2 35) SD316) 9490 36) vhs17) 9840 37) VXA-118) dlt 38) file19) dlt7000 39) logical20) dlt8000 40) optical

Choice?20<RETURN>Enter NDMP user name: ?root<RETURN>Enter NDMP password (characters will not be echoed):password<RETURN>Your jukebox has a bar code reader.Do you want bar code reader support enabled? (yes/no)yes<RETURN>Do you want volume labels to match bar code labels? (yes/no)no<RETURN>Following are attributes that define cleaning cartridge support for the jukebox ‘hp’:

auto clean: Yesdefault cleanings: 5cleaning slots: 19

Cleaning cartridge volumes Slot number-------------------------- -----------

19

Make sure that the slots set aside for cleaning cartridges contain cleaning cartridges. NetWorker must knowthe number of times it can use each cleaning cartridge. You can control how many times NetWorker willuse each cleaning cartridge by using the command: nsrjb -U (number of uses) -S (slot number). For moredetails please refer to nsrjb man pages.

Following are attributes that define the cleaning schedule for each device in the jukebox.

name: rd=amx:/dev/rmt/0mbndate last cleaned:cleaning interval: 2 weekscleaning required: No

Verify that the values for these attributes are appropriate for your installation. If not check documentationon how to set up automated cleaning cartridge support.

Jukebox has been added successfullyWould you like to configure another jukebox? (yes/no)no<RETURN>



SEE ALSOjbexercise(1m), nsr_device(5), nsr_jukebox(5), nsr_storage_node(5), nsr(5), nsrcap(1m).

DIAGNOSTICSunknown modelinvalid choice for ’model’ (35022)

Problem: The NetWorker system does not recognize the model chosen. If you added a/nsr/jbconfig* file after starting the daemons, you will see this error.Solution: Restart NetWorker.

root on computerhostis not on type:NSR’s administrator listProblem: The user ’root’ on the storage node ’host’ is not on the administrator list of the Net-Worker server.Solution: Add such an entry to the NetWorker server’s administrator list. Notethat the entry can be removed after this command completes.


JBEXERCISE(1m) JBEXERCISE(1m)

NAMEjbexercise − NetWorker jukebox exerciser

SYNOPSISjbexercise −m model −c control_port[ −V vendor_type] [ −CdsIv ] [ −D drive ] [ −Sslot ]

DESCRIPTIONThe jbexercisecommand tests the functionality of a jukebox. Before the command can be run, all contentsof the jukebox must be emptied except for media loaded in the first and last slots. These pieces of mediawill be moved around the jukebox as part of the various tests performed byjbexercise.

There are two major tests of functionality: drives and slots. Normally both the drive and slot tests are run.Individual component types can be tested by using the−d (for drives) and−s (for slots) options. In addi-tion, specific components can be singled out in the−D and−S options. When these are options are given,the only test run is on that particular component, i.e. if a specific slot is named the drives test is not run. Fordrives, the logical address of the component should be given. For slots, the physical address should begiven.

Upon startup, the program queries the user for the non-rewinding pathnames of the drives, if any, found inthe configuration of the jukebox. This query is not performed if the user is using a jukebox which does notrequire media to be ejected from a device, i.e. the device has automatic ejection capabilities.

The first test moves the media from the first slot to each of the drives. No operator intervention is required.

The second test loads the media from various slots to the first drive. The default is to test the media in thefirst and last slots in the jukebox. If a specific slot is being tested, the operator must first load that slot withmedia.

OPTIONS−C This makes jbexercise return the configuration of the jukebox. No tests are run.

−c Specify the control port which is used to interface with the jukebox (e.g. 1.5.0 from [email protected] could be found by issueing the ’inquire’ command).

−d Only drives are tested.

−D Only the specifieddrive is tested.

−I Return only an inventory of the jukebox. No tests are run.

−m Specify a jukebox model. (For a list of currently supported jukebox models run this commandwithout any arguments to print the usage string.)

−s Only slots are tested.

−S Only the specifiedslot is tested.

−v Verbose mode. Print out more information.

−V Specify a particular vendor id. This allows vendor to use the same driver for a number of jukeboxmodels.

SEE ALSOnsrjb (1m),nsr_jukebox(5).

DIAGNOSTICSMost diagnostic messages are specific to each type of jukebox. General messages include the following:

invalid <component> specifiedAn invalid id for the<component>was giv en. The id must be within the valid ranges of the juke-box configuration.<component>can be a drive, port or slot.

status incorrect, media presentThere is media loaded in a component but thecomponent statusoperation doesn’t indicate this.


JBEXERCISE(1m) JBEXERCISE(1m)

status incorrect, invalid slot locationThecomponent statusoperation is giving the incorrect source slot of the loaded media.

no drives found!!!No drives were listed in the configuration.

no slots found!!!No slots were listed in the configuration.


JBINFO(1m) JBINFO(1m)

NAMEjbinfo − test the SJI Jukebox Interface SJIINQ command on HPUX

SYNOPSISjbinfo devname

DESCRIPTIONThe jbinfo program tests the SJIINQ command on SJI compliant Jukeboxes This command prints informa-tion that identifies a Jukebox. If a Jukebox doesn’t support this feature, an appropriate error message will beprinted out.devnameargument should be any device name that can be used to reach a SJI compliant Juke-box driven by the system.jbinfo is usually used to determine whether or not a newly created device file isvalid. The information reported byjbinfo can also be used to construct a newjbcap entry for the jukebox.

EXAMPLE# jbinfo /dev/sjid1u1

Vendor: EXABYTEProduct: EXB-10EVersion: 1.5

Medium Transport Element Address: 11Number of Medium Transport Element: 1First Storage Element Address: 1Number of Storage Elements: 10First Import/Export Element Address: 0Number of Import/Export Elements: 0First Data Transfer Element Address: 0Number of Data Transfer Elements: 1

SEE ALSOlibsji(1m)


JBVERIFY(1m) JBVERIFY(1m)

NAMEjbverify − check jukebox/device configurations in NetWorker.

SYNOPSISjbverify [ −a ] [ −d { −i|−u } ] [ −D devicename]...[ −f filename] [ −F ] [ −h ] [ −H hostname]...[ −I Invoker] [ −j ] [ −J JB name]... [ −l ] [ −M ] [ −n ][ −N ] [ −P port ] [ −q ] [ −Q ] [ −r no. of retries][ −R ] [ −Sslot ] [ −sserver] [ −t ] [ −v ]... [ −Z ]

DESCRIPTIONjbverify verifies the devices defined in the NetWorker database, making sure that each one of them is con-figured properly by checking them for accessibility and usability. To do this, jbverify makes use of Net-Worker processes and requires that the NetWorker server (nsrd) be running on the server machine, and theNetWorker client (nsrexecd) be running on the client machines.

By default,jbverify checks all devices in the NetWorker database, but can be instructed to check only juke-boxes, only stand-alone drives or only local devices using the -j, -d and -l options respectively. Individualjukeboxes and drives can also be checked for by using the -J and -D options. Devices belonging to specifichosts can be checked using the -H option.

For jukeboxes,jbverify ensures proper configuration by loading a tape into each drive and unloading them,without performing any write operations on the tape. The only exception to this is if the -t option is used,explained below. A slot to be used for the test can be specified by using the -S option. If no slot is specified,jbverify goes through all the slots defined as available to NetWorker and loads the first one available.

Apart from checking for accessibility and usability,jbverify can run a series of tests on tapes loaded intothe drives being tested by calling on NetWorker’stapeexer program (see tapeexercise (1m)), when usedwith the -t option.

Running tapeexercise involves writing to the tape to determine the tape drive’s usability, so when -t is speci-fied, any volume which has a NetWorker label on it is immediately rejected as unusable and the next slot istried. If there are no non-NetWorker tapes in any of the slots, jbverify exits without doing any tests.

jbverify can be run on any storage node, and can be used to test any device on that storage node providedthe device has been configured in NetWorker. When run on the NetWorker server, it can be used to test anydevice on the network which has been configured in NetWorker. For a storage node which is not a Net-Worker server to be able to test devices other than its own, the nsrexecd on the target machine will have tobe started with the -s option with the invoking storage node as the argument, or have the invoking storagenode listed in the target machine’s ’servers’ file.

For example, if the NetWorker server is node NS, and there are two storage nodes Sto1 and Sto2: for Sto1to test devices on Sto2, the nsrexecd on Sto2 should be started as "nsrexecd -s NS -s Sto1." Or theservers/rservers file on Sto2 should have Sto1 listed as one of the valid servers.

SmartMedia devices are not tested by jbverify.

jbverify has extensive verbose messages built into it. In case of error in operation or inexplicablebehaviour, it is always helpful to use the -v options to diagnose the behaviour.

OPTIONS−a Tells jbverify to check all devices, even if they are disabled. By default, disabled devices are not

tested. This option is not supported at present.

−d This option tellsjbverify to check only stand-alone drives. No jukebox devices are tested.

−D This option is used to test a specific drive. The drive name should exactly match the name speci-fied in the NetWorker drive resource. Multiple drives can be specified by using the -D option mul-tiple times. If a jukebox drive is specified using this option, it is treated as a stand-alone drive.

−f Used to redirect jbverify’s output to a file. The argument is the file name to which the output is tobe redirected.



−F Reserved. This option is used internally by jbverify to indicate that this is a remotely forked jbver-ify.

−h Show the help options.

−H Tests the devices on the hostname mentioned. Use this option multiple times to test multiple hosts.Any other option specified on the command-line along with -H will be propagated to the remotehost being tested, except for the -D and -J options. When -H is used, only devices belonging to thathost are tested and hence only those -D and -J options which specify devices belonging to that hostare propagated forward.

−i Go into interactive mode. Used with -d for stand-alone devices. This option is useful when testingstand-alone devices on the local machine. If a particular stand-alone device does not have a tapeloaded, the -i option prompts the user to load a tape or cancel the operation so that it can skip tothe next drive. The -l option has to be specified with the -i option. Cannot be used with jukeboxes.

−I Reserved. Used internally by jbverify to specify the name of the invoking host machine to aremote jbverify.

−j Check jukebox devices only.jbverify checks only jukebox devices defined in the NetWorkerdatabase. All other devices are ignored.

−J This option is used to test a specific jukebox. The jukebox name should exactly match the namespecified in the NetWorker jukebox resource. Multiple jukeboxes can be specified by using the −Joption multiple times.

−l Check local devices only.

−M Reserved. Used internally by jbverify to indicate that it is being invoked by a NetWorker process.Messages are sent to the NetWorker server instead of being echoed to the stdout.

−n Perform tests in the no-op mode. Jbverify runs through the motions of testing the devices afterduly processing all given options but does not actually do the tests.

−N For a remote jbverify, put nsrexec into the same verbose mode as the jbverify. Usually redundant,but may be useful for debugging.

−P Reserved. Used internally by the jbverify process to indicate to a remote jbverify the port numberon which the server is listening.

−q Run both the local and the remote jbverify in the quiet mode.

−Q Run only the remote jbverify in the quiet mode. The results of the remote jbverify operation canstill be seen in the final status report printed out by the local jbverify. If -v is used on the com-mand-line with -Q, the local jbverify will run in the verbose mode while the remote jbverify runsquietly. -q and -v are mutually exclusive. Specifying both will result in jbverify running in level 1verbose mode.

−r Number of retries on error. Chiefly used on load and unload errors.jbverify will retry the numberof times specified if there is an error in operation.

−S Slot to be used for jukebox devices. The given slot will be used for loading tapes into jukeboxdevices during the test. If multiple jukeboxes are to be tested, make sure that the same slot in eachof those jukeboxes has a valid tape. If -t is specified, the tape in the slot has to be a non-NetWorkertape, else jbverify exits with an error.

−s Name of NetWorker server being tested.

−t Perform tapeexercise on tapes. see tapepexercise (1m) for details on tapeexercise. If -t is specified,there has to be a non-NetWorker tape in one of the slots for the exercise to proceed. If -S is speci-fied, the specified slot has to contain a non-NetWorker tape.

−u Run in unattended mode. Similar to the -i option and used for stand-alone devices only. If anydevice is not loaded with a tape, the -u option skips the device and goes on to the next one in thelist. Either -u or -i has to be specified with the -d option.



−v Run in verbose mode. Multiple -v options can be specified to increase the level of verbosity.Higher the level, more verbose the output. Currently has a maximum of 5.

−Z Reserved.

EXIT STATUSThe following are the error numbers with which jbverify could exit:ENWTAPE (501) : Found NetWorker tape when trying to run

tapeexercise.ELOADDETECT (502) : Unable to detect loaded state of a

device.EMEMORY (503) : Out of memory.ESRCEMPTY (504) : The source slot was empty.EDSTFULL (505) : The destination drive was full.EUNLOAD (506) : Error in unload.EUNKNOWN (507) : Unexpected error.ERDLABEL (508) : Error in read label operation.ESPAWN (509) : Error in spawn operation.EREAP (510) : Error in reaping tapeexercise

program.ELOADED (511) : Drive already loaded.ECONNECT (512) : Error in connect operation.EBASICTEST(111) : Error in basic test in

tapeexercise.EEOTTEST(222) : Error in EOT test in

tapeexercise.EFSFTEST(333) : Error in FSF test in

tapeexercise.

EXAMPLESTesting all devices without tapeexercise:

To test all stand-alone and jukebox devices, just run jbverify without any options:jbverify

To test all devices with verbose messages, use the-v option the required number of times:jbverify -v -v -v

Testing only stand-alone devices, in interactivemode:To test only stand-alone devices, use the-d option. -i sets the interactive mode:

jbverify -d -i -l

The -l option has to be specified when using the -i option since interactive mode is not supportedfor remote devices.

Testing only jukebox devices:To test only jukebox devices, use the-j option:

jbverify -j -v -v

Redirecting output to a file:To redirect the output of jbverify to a file, use the-f option:

jbverify -j -f output.jbv −v −v −v

Testing remote hostsTo test all the jukebox devices on hosts A and B, use the-H option:

jbverify −H A −H B −j −f outputfile

This tests only the jukebox devices on both hosts A and B and redirects the output to outputfile.



Running in quiet modeTo run jbverify in the quiet mode, use the−q option:

jbverify −q

This will result in only the final status report being printed. To run the local jbverify in the verbosemode, but all remote operations quietly, use the−Q option:

jbverify -v -v -v -Q

This will result in verbose output for all local operations but none for the remote ones. The statusof the remote operations can be seen in the final status report.

Specifying no. of retries on load/unload operations:To specify a certain no. of retries on errors, use the−r option:

jbverify −j −r 10 −S 12 −v

The above command makes jbverify use slot 12 of the jukebox for load and unload operations andmakes it retry 10 times on errors.

Running tapeexercise on tapes:To run tapeexercise on tapes loaded into devices, use the−t option:

jbverify −j −S 12 −t −v

FILES/nsr/res/nsr.res The NetWorker resource database.

SEE ALSOjbconfig(1m), jbexercise(1m),nsrjb (1m),nsr_device(1m),nsr_jukebox(5), nsr_storage_node(5),tapeexercise(1m)

DIAGNOSTICSThe following are the error messages that jbverify might produce along with their implications and possiblesolutions.

Bad resource database file!jbverify was unable to get the resource information about devices from the NetWorker RAPdatabase. Check if the NetWorker Server is up and running and if it is reachable from the currenthost.

Basic Test in tape exercise failed!The Basic Test in tapeexercise failed on the loaded tape. See tapeexercise (1m) for more details.

Can’t specify both -i and -u at the same time!The -i and the -u options are mutually exclusive. Choose one of them and retry operation.

Cannot use slot for stand-alone devices! Ignoring...The -S option is useful only for jukeboxes. This is just a warning that the option is being ignored.

Cannot run in interactive mode for remote devices-- use -l!The -i option is currently supported only for local devices. Specify -l to test only the local devices.

Could not connect to server! Quitting...The remote jbverify could not connect to the main jbverify for some reason. Examine other errormessages to establish cause.

Could not establish server socket! Quitting...jbverify could not open a socket to receive requests from remote jbverifys. Examine previous errormessages for exact cause of problem.

Could not extract control port info.jbverify was unable to parse the jukebox resource information it obtained about a jukebox from theRAP database. This might indicate a corruption of the RAP database in NetWorker. Check if youcan see the contents of the jukebox resource from the nwadmin GUI. Retry operation.



Couldn’t find control port in JB definition!jbverify was unable to parse the jukebox resource information it obtained about a jukebox from theRAP database. This might indicate a corruption of the RAP database in NetWorker. Check if youcan see the contents of the jukebox resource from the nwadmin GUI. Retry operation.

Could not find enabled drive <name> in database!A device was specified to be tested and jbverify could not find this device in the resource databaseof NetWorker. The most common reason would be incorrectly specifying a device name. Thedevice name has to exactly match the name given in the NetWorker device resource, including the"rd=..." prefix, if any.

Could not find jukebox <name> in database!A jukebox was specified to be tested and jbverify could not find this jukebox in the resourcedatabase of NetWorker. The most common reason would be incorrectly specifying a jukeboxname. The name has to exactly match the name given in the NetWorker jukebox resource, includ-ing the "rd=..." prefix, if any.

Could not get block size for this tape!jbverify could not find the defined blocksize for this tape. A default of 32k is usually assumed.

EOT Test in tape exercise failed!The EOT Test in tapeexercise failed on the loaded tape. See tapeexercise (1m) for more details.

Error in checkmedia operation on host <name>!The remote jbverify reported an error in checking the status of the device. See earlier error mes-sages for more information.

Error! Directory <name> doesn’t exist!This message is printed when processing a disk file drive and the said directory does not exist.

Error in eject tape from drive <name>! Skipping...There was a problem in ejecting the tape in the said drive. jbverify will skip testing this device andcontinue on with the next in line.

Error in read label operation! Cannot proceed with test!There was a problem while trying to read data off the loaded tape. Check previous error messagesto find the cause.

Error in resdb_query in getting device info.jbverify was unable to get the resource information about devices from the NetWorker RAPdatabase. Check if the NetWorker Server is up and running and if it is reachable from the currenthost.

Error in resdb_query in getting JB info.jbverify was unable to get the resource information about jukeboxes from the NetWorker RAPdatabase. Check if the NetWorker Server is up and running and if it is reachable from the currenthost.

Error in unload. Drive <num> (<name>), slot <num>There was an error in the unload operation of the said drive. Check previous error messages forpossible cause and error no. Try operation again in a higher verbose mode.

Error in unloading jukebox drives: <name>There was an error while trying to unload the said drive. Check other error messages for cause.

Error reported in eject tape from drive <name>! Device isoffline.There was an error reported by the NetWorker process during the eject operation, but the tapeseems to have been ejected; jbverify will continue to unload the tape to its slot.

FSF Test in tape exercise failed!The FSF Test in tapeexercise failed on the loaded tape. See tapeexercise (1m) for more details.



Failed to create xdr stream!This usually denotes lack of enough physical memory in the system. Check earlier error messagesfor more information.

Failed to detect loaded volume on drive <name> evenafter <num> tries. Giving up...jbverify failed to detect a loaded tape drive after putting a tape into the drive. Sometimes thismight happen if the drive is slow and the delay is too little. Try the operation again with a highnumber as the argument to the -r option or increase the load sleep attribute in the jukeboxresource.

Failed to get connection from remote jbverify!errno: <num>jbverify started a remote jbverify and is waiting for it to connect to it but has timed out withoutgetting a connection request. Examine other error messages to find the cause. One common causeis that the machine you are running jbverify on does not have the permission to request executionon the remote machine. To obtain permission, the nsrexecd on the remote machine has to bestarted with "-s <local machine name>." See example in the main section of this man page formore information.

Failed to get response for check media from remotehost <name>jbverify failed to get response from the remote jbverify on a request for checking the status of adevice. This could be because the remote jbverify was killed or terminated abnormally, the remotemachine went down, or just network problems. Check if you can ping the machine and retry oper-ation.

Failed to get stat packet!jbverify failed to receive an expected status packet from the remote jbverify. This could be becausethe remote jbverify was killed or terminated abnormally, the remote machine went down, or justnetwork problems. Check if you can ping the machine and retry operation.

Failed to read request packet from server!A remote jbverify failed to receive a request packet from the main jbverify. This could be becausethe main jbverify was killed or terminated abnormally, the machine went down, or just networkproblems. Check if you can ping the machine and retry operation.

Failed to redirect output to <name>. Errno <num>A system call failed. Run in verbose mode and contact support with error numbers and messages.

Failed to send FMEDIA on sock <num>! Errno <num>jbverify failed to send a request to check device status to the remote jbverify. This could bebecause the remote jbverify was killed or terminated abnormally, the remote machine went down,or just network problems. Check if you can ping the machine and retry operation.

Failed to spawn tapeexer!Failed to exec the NetWorker binary tapeexer. Check if the binary exists and if it has the adequatepermissions. Check other error messages for causes.

Failed to start nsrexec! errno: <num>jbverify failed to start the nsrexec process on the local machine. Examine previous error messagesfor exact cause. Some of the causes could be missing nsrexec binary, missing execute permissions,corrupt file etc.

Failed to start remote jbverify on <host>! errno <num>jbverify was unable to start jbverify on a remote machine. Check earlier messages for more infor-mation. Some of the causes could be that nsrexecd is not running on the remote machine, nsrexecdis of a version prior to 6.1, you are running jbverify on a machine which is not the server andwhich is not allowed to request execution on the remote machine. The last can be rectified by run-ning the remote nsrexecd with "-s <server> -s <machine>" option where <server> is the



NetWorker Server machine and <machine> is the machine on which you are running jbverify. Seeexplanation and example in the main section of this man-page for more details.

Have to specify -i or -u with -d option.The -d option has to be specified with either the interactive (-i) or the unattended mode (-u).Choose one of them and retry operation.

Invalid option specified: <option>.An invalid option was specified. Use the -h option to get a list of valid options.

Malloc errorSystem is out of physical memory. jbverify failed to allocate the required memory for an opera-tion. Exit some applications and retry the operation or increase the amount of memory on themachine.

NetWorker tape (<label>) in the drive. Cannot proceedwith test!The drive has a tape with the said NetWorker label on it. If jbverify is run with -t, it needs a tapewithout a NetWorker label on it to successfully run the tapeexercise program on it. If there is nonon-NetWorker tape in any of the slots, place one into one of the slots and retry the operation.

No block size found for this device: <name>!jbverify could not find a blocksize defined for this device in the NetWorker database. This usuallymeans that a default of 32k is assumed for this device.

No enabled stand alone devices found.The current configuration has no stand-alone device defined. This is just an informational mes-sage.

No enabled jukeboxes found in database.The current configuration has no jukeboxes defined. This is just an informational message.

No tape in slot <num>. Quitting...If -S was specified and there is no tape in the specified slot, jbverify posts this message and quits.Put a tape in the slot or specify another slot with a tape in it and retry operation.

Query resdb failed, err: <errmsg>.A RAP query to the NetWorker database failed. Check if the NetWorker Server is up and runningand if it is reachable from the current host.

Ran out of slots to choose from! Quitting...While trying to find a slot to use to load a tape into a jukebox device, jbverify has run out of slotsto try. If run with -t, jbverify needs to find a slot which has a tape without a NetWorker label on itas it will not overwrite NetWorker tapes even if they are no longer in the media database.

Received inv alid request from server:type: %djbverify received an unexpected request from the main jbverify. This could happen if the twomachines involved are running different versions of jbverify. Check and make sure that this is notso. This could also mean memory corruption. Retry operation at verbose level 5 and if error per-sists, send log to customer support at Legato.

Received unknown packet from remote host <name>!jbverify received an unexpected packet from the remote jbverify. This could mean memory corrup-tion. Retry operation at verbose level 5 and if error persists, send log to customer support atLegato.

SCO postion Test in tape exercise failed!The SCO position Test in tapeexercise failed on the loaded tape. See tapeexercise (1m) for moredetails.

Skipping disabled drive <name>jbverify, at present, does not test drives disabled in NetWorker. In the future, the -a option may beenabled to do so.



Skipping to next drive in list...After a load/unload error, jbverify is stopping the test of a drive and moving on to the next one inits list.

Slot <slot num> has Networker tape.The -t option was used and the slot from which the drive was loaded contained a NetWorker tape.If the -S option was used, this is a fatal error. If not, jbverify will try other slots to see if it can finda non-NetWorker tape.

Slot needs to be a valid number!The slot specified with the -S option has to be a real number.

Source slot empty! <slot num>The -S option was used but the specified slot did not contain a tape. Specify a slot which has a tapein it. If the -t option is also being used, specify a slot with a non-NetWorker tape in it.

Tapeexer executable not found!The tapeexer executable was not found. Check if it exists.

Tapeexer exited on signal <num>The tapeexer process was killed by the given signal.

Tapeexer exited abnormally with exit code <num>The tapeexer process exited abnormally with the given exit code.

Tapexercise on <name> exited without an exit status!jbverify was unable to get the exit status of the tapeexer process. This is a very rare case and mightnever happen unless the OS has a bug.

Unable to authenticate remote process!jbverify was unable to authenticate a connection request from the remote process.

Unable to get JB name! Skipping to next...jbverify was unable to find any name specified for the jukebox in the jukebox resource. Check thejukebox resource for any corruption and restore the NetWorker resource directory if needed.

Unable to find any devices in jukebox!jbverify was unable to find any devices configured for the jukebox. This is an error condition sinceit is not usually possible to have an enabled jukebox in NetWorker with no defined devices. CheckNetWorker configuration and run jbverify again.

Unable to get device info for <name>jbverify could not find any info for this device in the NetWorker database. Check that the name ofthe device matches exactly with the name defined in the NetWorker resource, including the "rd=..."prefix if it is a remote device/jukebox.

Unable to get JB name! Skipping to next...jbverify was unable to parse the jukebox resource information it obtained about a jukebox from theRAP database. This might indicate a corruption of the RAP database in NetWorker. Check if youcan see the contents of the jukebox resource from the nwadmin GUI. Retry operation.

Unable to load tape into drive <num> (<name>) as it seemsto be loaded!The said drive contains a tape even though jbverify must have unloaded it before trying the load.This might happen if the drives are not configured in the right order in the jukebox. Check if theorder of the drives is correctly configured in NetWorker.

Unable to to malloc for connlst! errno: <num>System is out of physical memory. jbverify failed to allocate the required memory for an opera-tion. Exit some applications and retry the operation or increase the anount of memory on themachine.



Unable to open <name>. Errno: <num>jbverify was unable to open the filename specified with the -f option. Check if you have permis-sions.

Unable to proceed to test drive <no>(<name>) inJB <name> as device is still loaded!

jbverify found the said drive to be loaded inspite of unloading it before accessing it. Check if anyother application is using this jukebox. Also check if the previous unload operation by jbverify failedby looking at the error messages or by running in higher verbose mode.

Unable to unload drive <num> (<name>)! May not beconfigured right!jbverify was unable to unload the said drive. This might happen if the drives are not configured inthe right order in the jukebox. Check if the order of the drives is correctly configured in Net-Worker.

Unknown state. Quitting...jbverify cannot determine the status of a load. This might happen with corrupted memory. Tryoperation again and contact support in case of failure.


LDUNLD(1m) LDUNLD(1m)

NAMEldunld − load or unload a tape device

SYNOPSISldunld { -u | -l } [ -a b.t.l ]

DESCRIPTIONThe ldunld program sends aload or unload command to a specified device.

OPTIONS-a b.t.l Selects a specific ordinal SCSI address, whereb is the logical SCSI bus,t is the SCSI target, andl

is the SCSI logical unit number (LUN) on that target (seelibscsi(1m)). This is a required option.

-u Unloads media from the specified device.

-l Loads media to the specified device.

SEE ALSOlibscsi(1m)


LGTOLIC(1m) LGTOLIC(1m)

NAMElgtolic − Legato license utility command

SYNOPSISlgtolic [ −s server] −c enabler_codelgtolic −i [ −m hostfile_dir]lgtolic [ −s server] −llgtolic [ −s server] −r [ −m hostfile_dir] [ −f output_file]lgtolic [ −s server] −u enabler_code−a authorization_codelgtolic [ −s server] −v enabler_code

DESCRIPTIONThe lgtolic command is used to manipulate Legato licenses that are stored within a license resourcedatabase. The license resource database is administered by a Legato license daemon. For a description ofthe license daemon, seelgtolmd(1m).

OPTIONS−a authorization_code

Authorizes a license with the specified authorization code, making the license permanent. Specifythe license to be authorized by using the−u option and the−a option. To obtain authorizationcodes for this product via the World Wide Web, simply point your web browser tolicense.legato.com. You will be asked to enter the enabler code for each authorization code thatyou request. For more details on product licensing, including other methods to obtain authoriza-tion codes, refer to the product Installation and Administration Guide and the latest Release Sup-plement.

−c enabler_codeCreates the license indicated by the specified enabler code. Enabler codes are listed on enablercertificates provided to you when you purchased this product. An authorization code is required tomake each license permanent.

−f output_fileCaptures customer registration data into the specifiedoutput_file. Once this text file is created,forward it to Legato customer support to register your product. If the−f output_fileoption is notspecified, the−r option encapsulates customer registration information into a file namedregis-ter.txt.Note: This option can only be used in conjunction with the−r option.

−i Prints out the hostid of the machine on which this command is running.

−l Lists all of the Legato product licenses currently stored within the license resource database.

−m hostfile_dirSpecifies the directory where the hostids file resides. If this option is specified, the program willuse the list of hostid(s) in the hostids file that resides in this directory to generate a compositehostid. This option is useful if the licensing manager is installed on a cluster machine or to forcethe hostid to be IP address-based instead of machine security ID-based on an NT machine. ForNetWorker, the typical directory for the hostids file is/nsr/res. For a stand-alone licensing man-ager running on a machine that does not have the NetWorker server installed, the typical path is/nsr/lic/res. The format for the list of hostids in a hostids file is:hostid1:hostid2:hostid3wherehostid is a hexadecimal string. This option must be used to specify a hostid file.

−r Creates or modifies customer registration data stored within the license resource database. The−foption captures this customer registration data into a text file.

−sserverSpecifies the hostname, RPC program number, and version for the license daemon whose databaseyou are targeting. License daemon information is displayed in the following format:

< hostname>:< rpc_number>:< version>

NetWorker Recovery Manager 6.1.1.Build.238 October 15, 2001 1

LGTOLIC(1m) LGTOLIC(1m)

Note: If you do not specify the−s serveroption, lgtolic uses the default values that map to thedaemon used by the product shipped. The current default for the hostname is localhost, the defaultfor the RPC program number is 390115, and the default for the RPC version number is 2. You canalso use environment variables to change these three defaults.LMD_HOSTNAME changes thedefault hostname of the machine where the license daemon is running.LMD_PROGNUMchanges the default RPC program number for connecting to the license daemon. You should neverhave to use this.LMD_VERSION changes the default RPC version number for connecting to thelicense daemon. The following example uses default hostname and RPC program number, butuses RPC version number 1 to list all licenses.Example:lgtolic -s "::1" -lTo specify a license daemon located on an alternative machine, use the−s< hostname> option.

−u enabler_codeUpdates an installed base license with the given base enabler specified at the command line.

−v enabler_codeDeciphers the specified enabler code. The generated output includes information about the licensename, type, serial number, and count.

NOTESThe daemon information provided by the−s option can also be obtained by using the following environ-ment variables:

LMD_HOSTNAME, The name of the host for the license daemon

LMD_PROGNUM, The program number for the license daemon

LMD_VERSION, The version number for the license daemon

DIAGNOSTICSProgram not registered

lgtolmd is not running.

Unknown hostEither the default hostname is invalid or the hostname specified with the−soption is invalid.

SEE ALSOlgtolmd(1m)


LGTOLMD(1m) LGTOLMD(1m)

NAMElgtolmd − Legato license daemon

SYNOPSISlgtolmd −p product−n version

DESCRIPTIONThe lgtolmd daemon is an RPC-based licensing service. This service allows applications to store andmanipulate license data. The RPC program number provided bylgtolmd is 390115. To support multipleinstances of the protocol, the version number is unique to each application. The required parameters aredetermined by each product’s installation script.

OPTIONS−p product

Specify the product that will be interfacing with the license daemon. The currently supportedproducts aregems(for GEMS default install directory /gems) andopt/SmartMedia (for Smart-Media default install directory /opt/SmartMedia) on UNIX platforms. For NetWorker,nsr/lic (forNetWorker default install directory /nsr) on UNIX platforms.

−n versionSpecify the version number. Some products use a unique version number. Currently, SmartMediauses version 2 and GEMS Storage Reporter uses version 3. Both GEMS and NetWorker use ver-sion 1. The future plan is to have all Legato products use the same license manager, that is, ver-sion number 1.

FILES/[product]/res/lgtolm.res

Attributes describing the license daemon’s license resources. This file should not be manuallyremoved or modified in any way.

/[product]/res/lictype.resFor internal use only. This file should not be manually removed or modified in any way.

/[product]/logs/lgtolmd.logLog file for diagnostic and informational messages on the license daemon. For example, if alicense has expired, this information will be printed to this log as well as to the console.

SEE ALSOlgtolic(1m)


LIBSCSI(1m) LIBSCSI(1m)

NAMElibscsi − SCSI device library

DESCRIPTIONThe SCSI device library is a private set of interfaces that NetWorker uses to communicate with SCSIdevices.Important: SCSI devices are named independently of the platform.

There are several functions in this library. The name of a SCSI device is identified as a combination of bus,target, and logical unit number (LUN) (b.t.l), whereb is the logical scsi bus,t is the SCSI target, andl is theSCSI LUN on that target. Donot assume that a logical SCSI bus number is related to any specific platformor hardware bus number. Rather, a logical SCSI bus number is a dense positive integer address space that isconsistent as long as the hardware configuration of the system remains the same. Target and LUN informa-tion is based upon the attached SCSI peripheral devices and their settings. Some platforms enable dynamicaddition and removal of SCSI devices, but may require flushing of cached device information (seelres-can(1m)).

PERMISSIONSTypically, if a device has no system driver, system privileges are not required for users to send commandsto this device. If a device has a system driver (for example, a tape drive), then system privileges are requiredin order to send a command to these devices.

SEE ALSOlrescan(1m)


LIBSJI(1m) LIBSJI(1m)

NAMElibsji − Standard Jukebox Interface (SJI) Library

DESCRIPTIONThe Standard Jukebox Interface (SJI) is a public set of interfaces that NetWorker uses to communicate withjukeboxes. Generally the function of this library is to convert SJI commands, as formed by NetWorker, tothe appropriate SCSI commands (since most autochangers are SCSI based). But the underlying attachmentto the jukebox is irrelevant to the functioning of this interface.

There are three entry points into this library:

void * sji_open ( char * devname )This opens a channel to an SJI compliant jukebox, nameddevname.A channel token, of typevoid* is returned if the channel is opened successfully, otherwise a channel token of typeNULL isreturned. The device namedevamecan be a specific ordinal SCSI address, for example,[email protected],whereb is the logical scsi bus,t is the SCSI target, andl is the SCSI logical unitnumber (LUN) on that target. For platforms that do not use Legato device drivers, the devicename can also be a platform-specific style device name, for example,/dev/sjid1u1.

int sji_cmd ( void *token, int cmd, void *arg )This sends an SJI command to the device opened by thesji_opencommand.

void sji_close ( void *token )This closes a channel to the device opened by the call made to thesji_opencommand.

FILESThe location of the SJI library varies from platform to platform.







DESCRIPTIONdasadmin











IBM_silo(8) IBM_silo(8)

NAMElibstlibm − shared library for communication to IBM 3494 silos

SYNOPSISlibstlibm.so (Solaris)

libstlibm.so.a (AIX)

DESCRIPTIONlibstlibm.xxx is a shared library that handles the communication betweennsrjb and the IBM silo driver(on AIX) or daemon (on Solaris). The IBM driver/daemon then handles the communication over thenetwork to the silo. There are no options, parameters or environment variables that affect the operation oflibstlibm. The correct path to this file should be entered when an IBM silo is configured usingjbconfig.The default values specified byjbconfig match the default locations chosen for the installation program,and in most cases can be accepted.

For NetWorker to work with the 3494, you must have first installed IBM’s Automated Tape Librarysupport.

On AIX, this will install a driver calledatldd (Automated Tape Library Device Driver). You may alsorequire IBM’s atape driver (Enhanced Tape and Medium Changer Device Driver) if you are using 3590drives in your 3494.

On Solaris, you will need to install thelmcpd package, (IBM Automated Tape Library Daemon) to use thesilo. Again, if you are using 3590 drives, you will also need to install theIBMtape driver. Note that whenyou are usingIBMtape, there will be two sets of device files that will access a given tape drive. There willbe the standard Solaris style/dev/rmt/Xmbn type, and there will be theIBMtape supported files of thetype /dev/rmt/Xstbn. You should use the IBM supported device files for proper operation of you tapedrives.

NOTE: Legato cannot supply these IBM drivers. They may be available on an IBM Device Driver ftp site(208.200.29.244), but this is just something we found, and is not necessarily a long-term IBM committedsite.

SEE ALSOnsrjb (1m), jbconfig(1m),dasadmin(1m), libstlemass(1m),ssi(1m),mini_el(1m), libstlstk (1m)

DIAGNOSTICSErrors in communication between the NetWorker server and the IBM 3494 silo are difficult to diagnose.The best method is to use the IBM supplied utilitymtlib to verify that you have properly configured the3494 to communicate with your host, and that the entire pathway from either the lmcp driver (on AIX) orthe lmcpd daemon (on Solaris) is functionong properly. Ifmtlib does not work, then there is no chance thatNetWorker will work.

If there are any questions about the connection between your host and the 3494, it is best to consult IBM, asthey support the connection between the the host and the silo. IBM supports both network and serial cableconnections to the silo. Since the nature of the connection is hidden from NetWorker by the driver/daemon,there is no difference to NetWorker between the two. Customers have successfully used both.


libstlstk(8) libstlstk(8)

NAMEssi − StorageTek silo interface module (Unix only)mini_el − event logger for use with ssi (Unix only)libstlstk − shared library for communication to ssi

SYNOPSISssi [ ACSLS server] [ socket] [ retry count ] &

mini_el [ −l logfile ] [ −d ] [ −h ] &libstlstk.so (Solaris)libstlstk.so.a(AIX)libstlstk.sl (HPUX)libstlstk.so.1(SGI)libstlstk.so.1(DYNIX/ptx)libstlstk.so (DECAXP)libstlstk.dll (NT i386)

DESCRIPTIONNOTE: in this document, the term "ACSLS server" will be used to indicate the name of the system that isrunning any one of StorageTek’s library manager programs: ACSLS on a Solaris or AIX host, LibraryStation on an MVS host, or Horizon Library Manager on a system running Windows NT or Windows 2000.

(Unix only)

Thessicommand is used indirectly bynsrjb to communicate with an ACSLS server.nsrjb loadslibstlstk, which handles the TCP calls to and fromssi. ssithen handles all of communication to and from theACSLS server. Starting with ACSLS version 5.3, it is possible to run NetWorker (either a server or astorage node) on the same host that ACSLS is running on.

ssi andmini_el must be running on the system on which jbconfig was run to create the jukebox resource.ssiandmini_el are almost always run as background processes, and are usually started automatically by thesystem.

In addition tossiandmini_el , a shared library file (usually calledlibstlstk.xxx where xxx is an operatingsystem dependent extension) is also required. An appropriate version of this library is installed as part ofNetWorker.

New in version 1.06 ofssi:

ssi now supports communication to more than one ACSLS server at any giv en time. Previous versions ofssiwere limited to a fixed TCP port for communication with NetWorker, and hence were limited to havingonly one instance running at a time. In version 1.06,ssiandlibstlstk have an added private communicationchannel that letslibstlstk ask any of the runningssi instances which ACSLS server it is connecting to, andto select the properssi for the current NetWorker silo. To make it easier to set up a configuration with thepossibilility of manyssiprocesses running, the command line forssihas changed.

While you can still startssi the same way as before - using the environment variable CSI_HOSTNAME toselect the ACSLS server to connect to - you can also simply specify the ACSLS server hostname on thecommand line. You may also specify the port number used for communication between NetWorker andthat particular instance ofssi. The allowed values for this port number are 50004 (for the first instance),50011 to 50019, and 50021 to 50099. Note that if you specify a port number that is already being used byan instance ofssi,the specified port cannot be used, and the next available port in the allowed range will beselected. If the port number is not specified, each successive instance ofssiwill take the next available portstarting from 50004 and going upwards. If there are no available ports in the range,ssiwill fail to load andshould display an error message. Note that specifying the port number is not necessary for normaloperation. You do not need to insure that a given ACSLS server is always accessed over a giv en port.NetWorker andssiuse the name of the ACSLS server to establish a connection on the fly.

If there is not a hostname specified on the ssi command line, beforessi is started, one environment variableMUST be set:CSI_HOSTNAME must be set to the name of the library server. If this variable is not



found,ssiwill exit with an error message.

mini_el is an event logger used byssi to maintain a log of certain events. It should be started beforessi.Multiple instances ofssiwill share a single instance ofmini_el. A header consisting of the ACSLS servername and the local TCP port thatssiwill be listening on is included at the start of any message placed intothe log by any instance ofssi

(NT only)

On NT, the software equivalent tossi and mini_el must be obtained from StorageTek as their product"Library Attach for NT". This package must be installed prior to configuring a Silo in NetWorker.

NOTE: Library Attach version 1.1 includes a portmapper function that will only install properly if theNetWorker services are not running. You should use Control Panel to stop the "NetWorker Backup andRecover Server" and the "NetWorker Remote Exec Service" before installing Library Attach. After LibraryAttach is installed, you should use Control Panel to start "NetWorker Remote Exec Service" and"NetWorker Backup and Recover Server".

NOTE: Since Legato does not supply "Library Attach for NT", we are unable to add the multiple ACSLShost functionality to our NT version of NetWorker.

(All platforms)

libstlstk.xxx is a shared library that handles the communication betweennsrjb andssior Library Attach.ssior Library Attach then handles the communication over the network to the library server (either ACSLS,Library Station or Horizon Library Manager). There are no options, parameters or environment variablesthat affect the operation oflibstlstk. The correct path to this file should be entered when an STK silo isconfigured usingjbconfig. The default values specified byjbconfig match the default locations chosen forthe installation program, and in most cases can be accepted.

OPTIONSmini_el

−l logfile Specifies the filename of the logfile to be created bymini_el. The default value is/nsr/logs/ssi_event.log.If present,logfile must be the complete path to the logfile. If the file doesnot exist, it will be created. If the file does exist, it will be appended to. If there is not a−lparameter, the default logfile/nsr/logs/ssi_event.logwill be used

−d Sets the debug flag.mini_el will output debug information

−h Displays usage information formini_el

PARAMETERSssi

ACSLS server is required if the CSI_HOSTNAME environment variable has not been set to the name ofthe system running ACSLS, LibraryStation or Horizon.

socket is only required if you need to specify the socket used for communication between NetWorker andssi

retry count is only required if you need to increase the retry count for communication betweenssiand theACSLS server due to network problems.

These parameters are not position sensitive. The command line code inssiuses the value of the parameterto decide which of the three acceptable parameters any giv en command line parameter actually is:

If the input parameter evaluates to zero using the atoi function, then it is assumed to be the ACSLSserver name. Only the first parameter that evaluates to zero will be used as the server hostname.

If the parameter evaluates to a number between 50004 and 50100, then it is assumed to be the socketvalue to be used. All parameters that fit this case will be saved as the socket value, so the lastmatching parameter on the command line will be the one actually used.



If a parameter evaluates to a number between 1 and 500, then it is assumed to be the retry count. Aswith the socket parameter, the last matching parameter on the command line wins.

Any parameter that falls outside of these cases will simply be ignored.

ENVIRONMENT VARIABLESssi

CSI_HOSTNAME (text, up to 256 chars, there is nodefault)If an ACSLS server name is not found on the command line,ssiwill use the hostname specified bythis variable. It is limited to 256 characters, and should simply be the hostname running thelibrary server program that you are trying to connect to. If neither the command line hostname northis environment variable specify a hostname forssi to use,ssiwill exit with an error message.

SSI_HOSTNAME (text, up to 256 chars, there is nodefault)This variable is intended for use on multi-homed systems. Normally,ssiuses thegethostbynamesystem function to determine the name to use for this side of the connection to the ACSLS server.On a system with several network interfaces, the name supplied by that function may not result inthe use of the network interface needed to communicate with the ACSLS server. On thesesystems, you can explicitly specify the exact name of the network interface thatssi will use toconnect to the ACSLS server. This variable needs to be set beforessi is started, and may bedifferent for various instances ofssi In all cases, a message will be logged in the event log statingif this environment variable was found, and if not, thatssiwill be using the hostname returned bygethostbyname.This is not an error message.

SSI_BASE_SOCKET(numeric, 0 < x < 64k, no default)If you need to restrict the socket values that ssi communicates on, this variable specifies thestarting number for ssi to use when it needs to open a socket to talk to the ACSLS server. Itappears that ssi will only open two sockets if this variable is set. The first, atSSI_BASE_SOCKETwill be used to connect to any host. The second, atSSI_BASE_SOCKET+ 1, will be used for direct communication to the ACSLS server. Note that there will still be thedefault sockets at 50001 and 50004 used to communicate between mini_el and ssi, but anycommunication between this host and the ACSLS server should occur using the two socketsstarting atSSI_BASE_SOCKET.

TIME_FORMAT (time format string,default = "%m-%d-%y %H:%M:%S")If you wish to see time values printed in a format other than the default of Month-Day-YearHour:Minute:Seconds, use this variable.

%m is replaced by the current month%d is replaced by today’s date%y is replaced by the current year%H is replaced by the current hour%M is replaced by the current minute%S is replaced by the current second

CSI_CONNECT_AGETIME (seconds, 0 < x < 31536000,default = 600)This will set the number of seconds for network connect aging purposes.

CSI_RETRY_TIMEOUT (seconds, 0 < x < ??, default = 4)This will set how longssiwill wait before retrying a network request.

CSI_RETRY_TRIES (numeric, 0 < x < 100, default = 5)This will set the number of timesssiwill retry a network message before reporting an error.



CSI_TCP_RPCSERVICE (boolean, default is TRUE)This sets whether ssi will use TCP sockets to connect with the library server.

CSI_UDP_RPCSERVICE (boolean, default is FALSE)This sets whetherssi will use UDP sockets to connect with the library server. SettingCSI_UDP_RPCSERVICE to TRUE will allow ssi to communicate with acsi that is running onthe same system

EXAMPLESNormal STK silo setup:

mini_el &ssi acsls1 &

− or −mini_el &setenv CSI_HOSTNAME acsls1ssi &

Connect to 3 different ACSLS servers:mini_el &ssi acsls1 &ssi acsls2 &ssi acsls3 &

− or −mini_el &setenv CSI_HOSTNAME acsls1ssi &setenv CSI_HOSTNAME acsls2ssi &setenv CSI_HOSTNAME acsls3ssi &

Connect to 3 different ACSLS servers over 3 differentnetwork interfaces:

mini_el &setenv SSI_HOSTNAME myhost_on_net1ssi acsls1 &setenv SSI_HOSTNAME myhost_on_net2ssi acsls2 &setenv SSI_HOSTNAME myhost_on_net3ssi acsls3 &

Connect to ACSLS server on socket 50025mini_el &ssi acsls1 50025 &

− or −setenv CSI_HOSTNAME acsls1mini_el &ssi 50025 &

To hav e mini_el use /nsr/logs/ssi.log.today as its log filemini_el -l /nsr/logs/ssi.log.today &ssi acsls1 &

FILES/nsr/logs/ssi_event.log

default logfile created/appended to bymini_el



SEE ALSOnsrjb (1m), jbconfig(1m),dasadmin(1m), libstlemass(1m), libstlibm (1m)

DIAGNOSTICSSeveral startup and shutdown messages along with any errors in communication between the NetWorkerserver and the ACSLS server will be logged in the logfile /nsr/logs/ssi_event.log (or other logfile as speci-fied on the mini_el command line). The messages from any onessi instance will be preceded by the nameof the ACSLS server that that instance will be communicating with plus the local TCP port number thatwill be used between NetWorker andssi.

For example:10-12-00 12:31:44 SSI[0]:[devlab-acsls/50004] ONC RPC: csi_init(): Initiation Startedsource csi_init.c; line 165

10-12-00 12:33:20 SSI[0]:[acsls2/50011] ONC RPC: csi_init(): Initiation CompletedONC RPC: csi_init(): ACSLS server acsls2 accessed through port 50011


LRESCAN(1m) LRESCAN(1m)

NAMElrescan− rescan for devices

SYNOPSISlrescan

DESCRIPTIONThe lrescanprogram tells the underlying SCSI library to discard any cached information and scan again fornew devices (seelibscsi(1m)).

SEE ALSOlibscsi(1m)


LRESET(1m) LRESET(1m)

NAMElreset − reset SCSI bus

SYNOPSISlrescanbusnumber

DESCRIPTIONThe lreset program tells the underlying SCSI library (seelibscsi(1m)) to reset the named logical scsi bus.You must have system privileges to execute this command.

WARNINGSThis command can cause the destruction of vital data since it will cause a SCSI bus reset. This may alsocrash your system. You should only use it as an extreme last resort to abort a hung command.

SEE ALSOlibscsi(1m)


LUSBINFO(1m) LUSBINFO(1m)

NAMElusbinfo − print SCSI information

SYNOPSISlusbinfo [ -v ]

DESCRIPTIONThe lusbinfo program prints a limited amount of information about the SCSI busses attached to the com-puter.

If you use the optional-v argument, additional information about the devices in the attached SCSI busseswill also printed.


LUSDEBUG(1m) LUSDEBUG(1m)

NAMElusdebug − set library debugging level

SYNOPSISlusdebugdebug-level

DESCRIPTIONThe lusdebugcommand sets a debugging level for the underlying NetWorker SCSI device drivers.

Debugging level 0 (zero) turns off debugging. Larger numbers enable greater levels of debugging.

The lusdebug level can now be specified as a bitmask - bit X set will show messages that are set to show atdebug level X+1. e.g. bit 0 set will cause messages at debug level 1 to be displayed. The exact level forany giv en message is listed at the end of the message in parentheses, such as (1m) for a message displayedfor debug level 8.

Using the bitmask allows you to display any or all levels of debugging information. The old method onlyallowed you to set the highest level you wished to see - all levels lower that the selected level were alwaysdisplayed, whether you wanted to see them or not.

You may still specify debugging levels the old way using by using the values old1 through old9. Theresults will be displayed using the new bitmask format.

Values can be entered in decimal (0 to 65535), hex (0x0 - 0xffff) or binary (0b0 - 0b1111111111111111).Zeros after the 0x or 0b prefixes are not required for binary or hex values.

Values that correspond to previous debug levels are:new new new

old decimal hex binary1 1 0x0001 0x00000000000000012 3 0x0003 0x00000000000000113 7 0x0007 0x00000000000001114 15 0x000f 0x00000000000011115 31 0x001f 0x00000000000111116 63 0x003f 0x00000000001111117 127 0x007f 0x00000000011111118 255 0x00ff 0x00000000111111119 511 0x01ff 0x000000011111111110 1023 0x03ff 0x000000111111111111 2047 0x07ff 0x000001111111111112 4095 0x0fff 0x000011111111111113 8191 0x1fff 0x000111111111111114 16383 0x3fff 0x001111111111111115 32767 0x7fff 0x011111111111111116 65535 0xffff 0x1111111111111111

Values corresponding to individual debug level are:new new new

old decimal hex binary1 1 0x0001 0x00000000000000012 2 0x0002 0x00000000000000103 4 0x0004 0x00000000000001004 8 0x0008 0x00000000000010005 16 0x0010 0x00000000000100006 32 0x0020 0x00000000001000007 64 0x0040 0x0000000001000000


LUSDEBUG(1m) LUSDEBUG(1m)

8 128 0x0080 0x00000000100000009 256 0x0100 0x000000010000000010 512 0x0200 0x000000100000000011 1024 0x0400 0x000001000000000012 2048 0x0800 0x000010000000000013 4096 0x1000 0x000100000000000014 8192 0x2000 0x001000000000000015 16384 0x4000 0x010000000000000016 32768 0x8000 0x1000000000000000

LIMITATIONSInvalid debug levels are quietly treated the same as debug level zero.

Debug level values greater than 65535 (0xffff, binary 0x1111111111111111) will be treated as 65535 (0ffff,binary..oh, you get the idea...).


LUSMODE(1m) LUSMODE(1m)

NAMElusmode − print out SCSI mode information

SYNOPSISlusmode

DESCRIPTIONThe lusmodeprogram prints out a rather large amount of MODE information about attached SCSI devicesattached to the system.


mini_el(8) mini_el(8)





(Unix only)












(NT only)




(All platforms)


OPTIONSmini_el




PARAMETERSssi








MM_DAT A(5) MM_DAT A(5)

NAMEmm_data − NetWorker media multiplexor data (tape and disk) format

DESCRIPTIONThis documents the data format that the NetWorker media multiplexor daemon,nsrmmd(1m), writes tolong term storage media such as tapes and optical disks. Seensr_device(5) andnsrmm(1m) for a discus-sion of supported device families and types. The format described here applies to any fixed record device,such as raw disks, or fixed record tape devices with file marks. NetWorker uses theeXternal Data Repre-sentation(XDR) standard to write media which can be interchanged among a wide variety of machines.Only the mechanism used to multiplex sav e set streams onto the storage media is described here; the for-mats of save set streams depend on the type of NetWorker client, and are described innsr_data(5).

A volumeis one physical piece of media such as a tape reel or disk cartridge. A tape volume is made up ofmultiple mediafiles, and each media file may contain several mediarecords. These media files and recordsshould not be confused with a client’s (for example UNIX or DOS) user files or records; the two do notnecessarily correspond. For example, a given media file or even a single media record may contain manysmall client user files. On the other hand, a single large client file may be split across several media files,and even across several volumes. Media files do not span volume boundaries. Save sets may span mediafiles and even volumes.

On most tapes, media files can be skipped very quickly by the device’s hardware or associated device driversoftware, and the hardware can detect when an end of a file has been reached. On some tapes, records canalso be quickly skipped forward. Otherwise, access to the media is sequential.

Media records are described by themrecord structure. Label records are fixed in size,MINMRECSIZEbytes. Other records can potentially be a larger size that must be some constant for the rest of the volume.NetWorker always writes, reads and skips data in units of full-sized media records. Eachmrecord containszero or moremchunks Thesemrecords are used for storing one or more client save sessions or used byNetWorker for synchronization and labelling. TheXDR format of a media file’smrecords andmchunksare as follows:

const MINMRECSIZE = 32768; /* minimum media record size */const MMAXCHK = 2048; /* maximum number of chunks in record */const MHNDLEN = 120; /* private area length for handlers */

enum mrec_version { /* mrecord version */MREC_VER5 = 0, /* older format mrecord */MREC_VER6 = 6 /* current format mrecord */

};

/** For media record format version 5, the data types lgui_t, lg_off64_t,* and lg_time64_t are defined as:*/typedef struct lgui_t unsigned long;typedef struct lg_off64_t unsigned long;typedef struct lg_time64_t unsigned long;

/** For media record format version 6, the data types lgui_t, lg_off64_t,* and lg_time64_t are defined as:*/typedef struct lgui_t { /* XDR encoded Unique Id. */

char _bytes[20];} lgui_t;typedef struct lg_off64_t unsigned long long;



typedef struct lg_time64_t unsigned long long;

typedef lgui_t ssid_t; /* save set id */typedef lgui_t volid_t; /* key for the volume database */

struct mchunk {ssid_t mc_ssid; /* owning save set id */lg_off64_t mc_low; /* 1st byte, relative to sav e stream */opaque mc_data<MINMRECSIZE>;/* chunk’s data */

};

struct mrecord {opaque mr_handler[MHNDLEN]; /* private to media handler */mrec_version mr_version; /* Media record version number */u_long mr_orec; /* record size */volid_t mr_volid; /* encompassing volume’s id */u_long mr_fn; /* encompassing file number */u_long mr_rn; /* record number within the file */u_long mr_len; /* record byte length */mchunk mr_chunk<MMAXCHK>;/* chunks of save streams */

};

The first field of anmrecord, mr_handler, is reserved for media-specific data (currently it is not used by any imple-mentation). Themr_version field is the version number of the media record format. The size of the rest of thefields in the media record depends on the version number. Themr_orec field is the size of the current record. Amedia record’s header fields,mr_volid , mr_fn , andmr_rn , are used to check the tape position and the data readfrom the record. The file numbers and record numbers start at zero and increment sequentially. The record numberis reset each time the file number is incremented. On disks, file numbers are always zero. Themr_len field is theactual number of valid bytes in this record, as opposed to the size of the device’s read or write request.

If file or record skipping is unreliable, NetWorker can still recover from isolated errors, at worst by rewinding andreading the tape from the start. If a volume can be physically unmounted or mounted without notice to the mediamanagement daemon, then the volume identifier in each record provides a quick way of verifying when this hap-pens, without the need for a full rewind and reading of the label in most cases.

The mchunks within anmrecord contain client data from one or more save sessions. Themc_ssidandmc_lowvalues are used to reconstruct the save streams from the chunks within the records. Themc_data field holds theactual data of each chunk. For a giv en sav e set,mc_low plus the length ofmc_data should equal the followingchunk’s value formc_low. Save sets may by intermingled arbitrarily within media records.

The first chunk of the first record of the first media file on the volume encapsulates thevolume label information;for some media, the second chunk contains additional volume information, for example, the media pool the volumebelongs to: Subsequent data in the first file is reserved for future expansion. The label may be duplicated in a secondfile for redundancy, in case the first copy of the label gets accidentally overwritten. The formats of the volume labeland additional label information are described by the followingXDR data structures:

const MVOLMAGIC = 0x070460; /* volume magic number */const NSR_LENGTH = 64; /* length of several strings */const RAP_MAXNAMELEN = 64; /* maximum length of attribute name */

struct mvollabel {u_long mvl_magic; /* medium volume verification number */lg_time64_t mvl_createtime; /* time at which volume labeled */lg_time64_t mvl_expiretime; /* time for volume to expire */



u_long mvl_recsize; /* expected size of mrecords */volid_t mvl_volid; /* medium volume id */string mvl_volname<NSR_LENGTH>;/* medium volume name */

};

struct vallist {vallist *next;string value<>; /* attribute value */

};

struct attrlist {attrlist *next;string name<RAP_MAXNAMELEN>;/* attribute name */vallist *values; /* attribute values */

};

/** Additional information may includes the following attributes* (listed by the name they are stored with):* "volume pool" : the media pool*/struct mvolinfo {

struct attrlist *mvi_attributes; /* any other information */};

Themvl_magic field must be equal toMVOLMAGIC in order for the chunk to represent a valid volume label. Ifthe volume label changes in the future, the new format will have another ‘‘magic’’ number, but the format describedhere must still be allowed. Themvl_volid is an internal identifier assigned and managed by the media manager.Themvl_volname is the volume name that is assigned when the media is first labeled. The time fields are in USTformat − the number of seconds elapsed since 00:00 GMT, January 1, 1970. Themvl_recsizeis the size of all sub-sequent media records found on the tape.

Themvp_pool is the pool name that is assigned when the media is first labeled. Different media pools allow admin-istrators to segregate their data onto sets of volumes. Media cannot be reassigned from one media pool to another.Pool names are a maximum ofNSR_LENGTH characters long.

Synchronization marks, calledschunks, are also written periodically to the media for each save set. Synchroniza-tion chunks are used byscanner(1m) when verifying or extracting directly from a volume. They are also used bynsrmmd when trying to recover from media errors during file recovery. The following XDR data structuredescribes a synchronization chunk:

typedef lgui_t clientid_t;

struct ssclone_t {lg_time64_t ssc_cloneid; /* unique UST stamp wrt ss_ssid */u_long ssc_flag; /* lots of status buried here*/u_long ssc_frag; /* not used, always 0 */

};

/** Synchronization chunk of the newer MREC_VER6 media format.*/struct schunk {

u_long ssi_gen; /* Not used. */ssid_t ssi_ssid; /* save set identifier */



ssid_t ssi_prev; /* non-zero iff continuation */u_long ssi_level; /* backup level*/lg_time64_t ssi_time; /* save time on client */lg_time64_t ssi_create; /* creation time on server */lg_time64_t ssi_insert; /* insertion time on server */lg_time64_t ssi_complete; /* completion time on server */clientid_t ssi_clientid; /* client name identifier */u_long ssi_flags; /* more details about this ss */string ssi_host<>; /* client name - save set owner */string ssi_name<>; /* symbolic name, for example "/usr" */lg_uint64_t ssi_size; /* actual number of bytes saved */lg_uint64_t ssi_nfiles; /* number of client files saved */u_long ssi_browse; /* browse time offset */u_long ssi_recycle; /* recycle time offset */struct attrlist *ssi_al; /* generic RAP attribute list */ssclone_t ssi_clones<>; /* information about this clone */

};

/** Synchronization chunk of the older MREC_VER5 media format.*/struct old_schunk {

opaque ssi_host[NSR_LENGTH]; /* save set host */opaque ssi_name[NSR_LENGTH]; /* symbolic name */u_long ssi_time; /* save time */u_long ssi_expiry; /* expiration date */u_long ssi_size; /* actual size saved */u_long ssi_nfiles; /* number of files */ssid_t ssi_ssid; /* ssid for this save set */u_long ssi_flag; /* various flags, see below */u_long ssi_info; /* volid or ssid, see below */

};

#define SSI_START 1 /* start of a save set */#define SSI_SYNC 2 /* synchronization point */#define SSI_CONT 3 /* continued from another volume */#define SSI_END 4 /* end of this save set */#define SSI_SSMASK 0x0000000f /* save set sync chunk type */#define SSI_LBIAS 0x10000000 /* the level is included in the flags */#define SSI_LMASK 0xff000000 /* mask to cover bits for level */#define SSI_LSHIFT 24 /* shift amount for the level */#define SSI_INCOMPLETE 0x00010000 /* not finished (aborted) */#define SSI_CONTINUED 0x00800000 /* continued save set series */

The ssi_ssidis the save set identifier of this save set. Thessi_timefield contains the create time of the save set inUST based on the client’s clock. Thessi_createfield contains the create time of the save set in UST based on theserver’s clock. Thessi_insertfield contains the time the save set was inserted into the media database in UST basedon the server’s clock. Thessi_completefield contains the completion time of the save set in UST based on theserver’s clock. Thessi_clientidandssi_hostare the client identifier and name of the index which contains this saveset. Traditionally this is the client identifier and name of the client where the save set originated. Thessi_nameisthe save set name to be presented to the user. These are both null-terminated strings, even though the fields are fixedlength in the older version media records. Thessi_sizeandssi_nfilesare the number of bytes and number of filessaved so far for this save set. Thessi_browseis the time offset in seconds from the save set insertion time to thetime this save set is no longer browsable. Thessi_recycleis the time offset in seconds from the save set insertiontime to the time this save set becomes recyclable. Thessi_alis the generic save set attribute.



Thessi_flagindicates the type of this synchronization chunk and other information about the save set. In the olderversion synchronization chunk, this field also contains the level of this save set. There are four basic types of syn-chronization marks that can be found from examiningssi_flag & SSI_SSMASK. SSI_START is used to mark thebeginning of a save set.SSI_SYNC marks a periodic synchronization point and is only written at an exact fileboundary in the save set.SSI_CONT indicates that this is the continuation of a save set that started on a differentvolume. Whenssi_flag & SSI_SSMASKis SSI_CONT, ssi_prevor ssi_infocontains the volume identifier for thesave set’s preceding volume. These synchronization chunks are used when a save set spans a volume boundary.SSI_ENDmarks the end of a save set.

On the new version of synchronization chunk, thessi_levelfield contains the save set backup level. On the olderversion of synchronization chunk. Should theSSI_LBIAS bit be set thenssi_flag& SSI_LMASK shifted to theright by the value of SSI_LSHIFTspecifies the level of the save set. TheSSI_INCOMPLETE bit indicates thatthis save set did not finish properly. This could be caused by a user interrupting an in progress save.

TheSSI_CONTINUED bit indicates that this save set is logically continued to or from another save set. These con-tinued save sets are used to handle very large save sets. If theSSI_CONTINUED bit is set andssi_flag &SSI_SSMASKis SSI_START, thenssi_prevor ssi_infogives the previous save set id that this save set was contin-ued from. If theSSI_CONTINUED bit is set andssi_flag & SSI_SSMASKis SSI_END, thenssi_prevor ssi_infogives the next save set id that this save set is continued to.

Thessi_expiryfield is the expiration date, in UST, for this save set. This field is zero if an explicit save set expira-tion time was not specified when the save set was created. This field no longer exists in the new synchronizationchunk.

SEE ALSOnsr_device(5), nsr_data(5), nsrmm(1m),nsrmmd(1m),nsrmmdbd(1m),nsr(1m),scanner(1m).

RFC 1014 XDR: External Data Representation Specification


MMINFO(1m) MMINFO(1m)

NAMEmminfo − NetWorker media database reporting command

SYNOPSISmminfo

[ −avV ] [ −o order ] [ −sserver] [ report ] [ query ] [ volname...]

< report >: [ −m | −p | −B | −S | −X | −r reportspec]< query >: [ −c client ] [ −N name] [ −t time] [ −q queryspec]

DESCRIPTIONThemminfo command reports information about NetWorker media and save sets. Themminfo commandcan produce several different reports depending on the flags specified. Several built-in reports can be speci-fied using short-hand flags. Custom reports can also be specified. The default report, along with the built-in reports printed by the use of the−v, −V, −m, −p, −S, −B, and−X flags, are described first below. Thecustom query and report generators, using the−q queryspecand−r reportspecoptions, are described in theCUSTOM QUERIES AND REPORTS section. Other options are described in theOPTIONS section.

Without any options,mminfo displays information about the save sets that completed properly during thelast twenty four hours, and are still contained in an on-line file index (browsable save sets). The followinginformation is printed for each save set: the containing volume name, the client’s name, the creation date,the size saved on that volume, the save set level, and the save set name. The size field is displayed in Bytes(B), KiloBytes (KB), MegaBytes (MB), GigaBytes (GB), TeraBytes (TB), PetaBytes (PB), or ExaBytes(EB). The save set level will display ‘full’, ‘incr’, ‘migration’ or 1 through 9, for full, incremental, migra-tion save sets, level 1 through 9, respectively. The level is only kept for scheduled saves and file migration;save sets generated by explicitly running thesave(1m) command (calledad hocsaves) do not have an asso-ciated level.

Specifying the−v flag causes three additional fields to be displayed: the creation time, the internal save setidentifier (ssid), and two flags. One character is used per flag.

The first flag indicates which part of the save set is on the volume. When the save iscompletely containedon the volume, ac is displayed. Anh is displayed when the save set spans volumes and thehead is con-tained on this volume. The remaining sections will be on other volumes. Anm is displayed when the saveset spans volumes and amiddle section is contained on this volume. The head and tail sections will be ondifferent volumes. There may be more than one middle section. At is displayed when thetail section of aspanning save set is contained on this volume. Again, the other sections will be on other volumes.

The second flag indicates the status of the save set. Ab indicates that the save set is in the on-line indexand isbrowsable via therecover(1m) command. Anr indicates that the save set is not in the on-line indexand isrecoverable via thescanner(1m) command. AnE indicates that the save set has been marked eligi-ble for recycling and may be over-written at any time. AnS denotes that the save set was scanned in (orrolled in). Rolled in save sets are not subject to the standard index management procedures and will remainin the file index until the user manually purges the save set. Ana indicates that the save was aborted beforecompletion. Aborted save sets are removed from the on-line file index bynsrck(1m). An i indicates thatthe save is still in progress. The−v flag prints aborted, purged, and incomplete save sets in addition to thecomplete, browsable save sets printed by default.

The −V flag displays even more detail than the−v flag, and is generally used for debugging. This formatalso displays information (such as, media file number and record number) that can be used to speed theoperation of thescanner(1m) command. Rather than displaying one line per save set per volume, threelines are displayed each time a section of a save set occurs within a file on a volume. A single save set willhave multiple index entries if it starts in one file on a volume and ends in another. This report contains allof the information reported via the−v flag, but, because of the additional detail, some of this information isreordered. The first line will contain the volume name, the client’s name, the size saved in that section, thesave set level, and the save set name. The size field lists the number of bytes that are contained in the sec-tion, rather than the total amount of the save set on this volume. The second line contains the followingfields: the internal save set identifier (ssid), the save time in seconds since 00:00:00 GMT, Jan 1, 1970, thecreation data and time of day, the internal save set identifier (ssid), the browse time, and the retention time.



The third line contains: the offset of the first and last bytes of the save set contained within section, themedia file number, the first record within the media file containing data for this save set, the internal volumeidentifier (volid), the total size of the save set, and the flags, described in the−v paragraph above, indicatingwhich part of the save set is contained in this media file (c, h, m, or t) and the save set’s status (b, r , a, or i).

The−p flag causesmminfo to display a report on the browse and retention times for save sets. Each line ofthe report displays the save set creation date, and the stored browse and retention dates (‘undef’ is dis-played when connecting to a downrev server), the save set identifier, the client’s name, and the save set’sname. The−v and−V options have no effect on the columns included in this report.

The−m flag causesmminfo to display the name of each volume in the media database, the number of byteswritten to it, the percent of space used (or the word ‘full’ indicating that the volume is filled to capacity),the retention (expiration) time, the number of bytes read, the number of times the volume has beenmounted, and the volume’s capacity. Volumes that are recyclable (seensrim(1m)) are flagged by anE inthe first column (meaningEligible for recycling). If a volume has been marked as manually-recyclable, anM is displayed instead of theE. If a volume is both manually-recyclable and eligible for recycling, anXwill be displayed. Archive and migration volumes are flagged by anA, also in the first column. If the vol-ume is not an archive or migration volume, and is not recyclable, no flag appears.

Specifying the−v flag with the−m flag causes three additional fields to be displayed: the internal volumeidentifier (volid), the number of the next file to be written, and the type of media.

Using a−V flag with the−m adds a column of flags to the output. There are currently two possible flags.Thed flag is set if the volume is currently being written (dirty). Ther flag is set if the volume is marked asread-only. If neither condition is present, the flags column will be empty.

The −S flag displays a long, multi-line save set report, which is used for debugging. The number of linesvaries per save set. Due to the length, there are no column headers. Instead, each attribute of the save set isdisplayed in a ‘name=value’ manner, except the client and save set name, which are displayed as‘client:name’, and the extended attributes, described below. The first line of each multi-line group starts onthe left margin and includes the save set identifier (ssid), save time as both a date/time string and secondssince 00:00:00 GMT, Jan 1, 1970, and the client and save set names. Subsequent lines for this save set areindented. If the save set is part of a save set series (a ‘continued save set’) and is not the first in the series,the save set identifier of the previous save set in the series in shown on the second line by itself. The nextline displays the level, the save set flags (in ‘ssflags’ format, as described in the table in theCUSTOMQUERIES AND REPORTS section), the save set size in bytes, the number of files in the save set, and thesave set insertion date. The next line displays the save set’s create, completion, browse and retention (akaexpiration) dates. The string ‘undef’ for any of the values on these two lines generally means an olderserver that does not store these values is being queried. If the client identifier is set, it is printed on the nextline. If the save set has extended attributes (such as the group to which the save set was a part or thearchive annotation), they are printed next, at most one attribute per line. The format of each extendedattribute is "name: values;". The clones or instances of the save set are shown last (every save set has atleast once instance). The first line of each clone shown the clone identifier, the date and time the instancewas created, and the per-clone flags (in ‘clflags’ format from theCUSTOM QUERIES AND REPORTStable). For each instance, each section of that instance is shows as a fragment line. The fragment lineshows the offset of that fragment from the beginning of the save set, the volume identifier (volid) containingthe fragment, the media file and record numbers of start of the fragment, an absolute positioning identifier(unused by existing servers), and the date of last access of the fragment. The−v and−V options have noeffect on this report.

The −X flag prepares a save set summary report instead of one or more lines per save set. Note that theentire media database must be examined to resolve this query, making it very slow and expensive. Thesummary lists the total number of save sets, and breaks the total down into several overlapping categoriessummarizing the save set types. The recent save set usage, if appropriate to the query, is also printed. Thecategories are the number of fulls, the number of incrementals, the number of other non-full, non-incremental saves, the number of ad hoc, archive, migration, empty and purged save sets, the number ofindex sav e sets, and finally, the number of incomplete save sets. For recent usage, the number of save setsper day is shown, up to a week ago, along with a summary of the week’s sav e sets and, if applicable, a



summary of the month’s sav e sets. For each line, the number of files (saved in the time interval specified),number of save sets, total size, average size per save set, and average size per file are listed. The percentageof the amount saved for incrementals versus fulls and the percentage of browsable files are also printed,when appropriate. The−v and−V options have no effect on the summary report.

The −B flag performs a canned query to output, in a convenient format, the list of bootstraps generated inthe previous five weeks. In this format, there is one line of output for each matched save set. Each linecontains the save date and time, save lev el, save set identifier (ssid), starting file number, starting recordnumber, and the volume. The equivalent query is described below in theEXAMPLES section. The−vand−V options have no effect on the bootstrap display.

OPTIONS−a Causes queries to apply to all complete, browsable save sets, not just those in the last 24 hours.

This option is implied by the−c, −N, −q, −m, and−o options, described below. When combinedwith a media-only report (−m or a custom report showing only media information),−a applies toall volumes, not just those with complete and browsable save sets.

−c clientRestricts the reported information to the media and/or save sets pertaining to the specifiedclient.

−m Displays a media report instead of the default save set report (in other words, a report about themedia containing save sets, not the save sets themselves).

−N nameRestricts the reported information to the media and/or save sets pertaining to the specified save setname.

−o orderSorts the output in the specified order. Before displaying the save sets, they are sorted by variousfields. Numeric fields are sorted least to greatest, other fields are sorted alphabetically.order maybe any combination of the letterscelmontR, representingclient, expiration date,length,medianame,name of save set,offset on media (file and record number),time, andRev erse, respectively.The default sorting order for save set reports ismocntl. The offset fields (file and record) are onlyconsidered when the−V option has been selected and for custom reports that show sav e set section(fragment) information. When applied to−m media-only reports, the length is the amount used onthe volume, the time is the last time the media was accessed, and the other order flags are ignored.

−q queryspecAdds the given query constraint to the list of constraints on the current query. Multiple−q optionsmay be given. See theCUSTOM QUERIES AND REPORTS section below for the syntax ofthequeryspec.

−r reportspecAppends the given report specification to the list of attributes to be displayed for the current query.Multiple −r options may be given. See theCUSTOM QUERIES AND REPORTS sectionbelow for the syntax of thereportspec.

−sserverDisplays volume and save set information from the NetWorker system onserver. Seensr(1m) fora description of server selection. The default is the current system.

−t time Restricts the reported information to the media and/or save sets pertaining to the save sets createdon or aftertime. Seensr_getdate(3) for a description of the recognized time formats. The defaultis ‘yesterday’.

−v Turns on the verbose display reports, described above.

−B Runs the canned query to report bootstraps which have been generated in the past five weeks, asdescribed above. This option is used bysavegrp(1m) when saving the server’s index and boot-strap.



−S Displays a long, multi-line save set report, as described above.

−V Displays additional verbose report output, as described above.

−X Prepares a summary report, as described above.

CUSTOM QUERIES AND REPORTSThe custom query and report options ofmminfo allow one to generate media and save set reports matchingcomplex constraints without resorting to pipelines and scripts. This section describes the syntax of customquery and report specifications, and gives some simple examples. Further examples are shown in theEXAMPLES section, below.

The custom query option,−q queryspec, is an extension to the short-hand query options, such as−c client,which allow you to make queries based on almost any media or save set attribute in the database, and allowvarious comparisons in addition to the simple equality comparison provided by the short-hand options. Theformat of aqueryspecis

[!] name[ comp value] [ , ... ]

wherenameis the name of a database attribute, listed in the table below,compis a valid comparator for theattribute, from the set ‘>’, ‘>=’, ‘=’, ’<=’, ’<’, andvalueis the value being compared. Leading and trailingspaces can be used to separate the individual components of the specification. The comparator and valuemust be specified for all but flag attributes. Generally numeric attributes allow all five comparators, andcharacter string attributes generally only allow equality. When comparing flags whose values are normally‘true’ and ‘false’, one may alternatively use the ‘[ ! ]name’ syntax. The ‘!name’ form is equivalent to‘name=false’, and ‘name’ by itself is equivalent to ‘name=true’. The comparisons in the specification areseparated by commas. If a time or a string contains commas, you must quote the value with single or dou-ble quotes. Quotes are escaped within a string by repeating them. The following is a valid string compari-son:

name="Joe’s daily, ""hot"" Save Set"

Note that command line shells also interpret quotes, so you will need to enclose the entire query withinquotes, and quote the single value inside the query, possibly with a different kind of quote, depending onthe shell. Except for multiple character string values, explained below, all of the specified constraints mustmatch a given sav e set and/or media volume before a line will be printed in the report. Multiple−q optionsmay be specified, and may be combined with the short-hand query constraints−c, −N and−t. The order ofthe above query constraints is unimportant.

Numeric constraints, except for identifiers (volume, save set and clone identifiers), allow ranges to be speci-fied, and all character string constraints allow multiple possible values to be specified. Note that times andlevels are considered to be numeric values, not character strings. The upper and lower bounds of a numericrange are specified as two separate constraints. For example,

%used>20,%used<80

matches volumes that are between 20% and 80% used. Each possible value of a given character stringattribute is specified as a separate equality constraint. For example,

client=pegasus,client=avalon

matches save sets from the client ‘pegasus’ or the client ‘avalon’.

The custom report option,−r reportspec, allows one to specify exactly which media and save set attributesshould be shown in the report, the order of the columns, the column widths, and where line breaks shouldbe placed. The format of areportspecis

name[ (width) ] [ , name[ (width) ] ... ]

wherenameis the name of a database attribute, listed below, and the optionalwidth, enclosed in parenthe-ses, specifies how wide the column should be. Leading and trailing spaces are ignored. The default col-umn width depends on the attribute; default widths are also shown in the table below. Multiple−r optionsmay be specified. The order of the columns in the report will be left to right, and correspond to the order ofthe attribute names specified. Each line of output will contain all of the data requested (you can cause line



breaks within a logical line by using thenewlineattribute name). If a value does not fit in the requestedcolumn width, subsequent values in the line will be shifted to the right (values are truncated at 256 charac-ters).

The table below lists all of the recognized attribute names, their valid range of query values (or ‘NA’ forattributes that are only valid for report specifications), their default column width in characters (or ‘NA’ forflag attributes that are only valid for query specifications), and a short description.

Numeric attributes (shown asnumberin the valid range column of the table) can be specified using any ofthe comparators listed above, and can be used in range comparisons.

The=id attributes are used for various identifiers (volume identifier, sav e set identifier, and so on) and onlyallow equality comparisons. In most cases, if the column is narrow (less that 50 characters), only the shortID is shown, which correponds to the ID used by downrev servers. If the column is wide enough, the fullID is shown. Client identifiers always display as full IDs, and clone identifiers always display as short IDs.

Flag attributes have the values ‘true’ or ‘false’, only apply as query constraints, and have correspondingflag summary strings for report specifications.

Timeattributes are specified innsr_getdate(3) format and are otherwise treated as numeric attributes (notethat you will need to quote times that contain commas). The special time ‘forever’, when used as an expira-tion date, means a save set or volume will never expire. The special time ‘undef’ is displayed when thetime is undefined. When output, times are displayed asMM/DD/YY HH:MM:SSfor numeric month, dayyear (last two digits), hours, minutes, and seconds, respectively. If the column is very narrow (less that 14characters), only the date is shown. Columns less than 17 characters wide drop the seconds, and columns17 or more characters wide print the full date.

Size and kbsizeattributes may have a scale factor appended to them: ‘KB’ for kilobytes, ‘MB’ forMegaBytes, ‘GB’ for GigaBytes, ‘TB’ for TeraBytes, ‘PB’ for PetaBytes, or ‘EB’ for ExaBytes. Thedefault scale (when no scale is explicitly specified) on query constraints for attributes is bytes; the defaultfor kbsizeattributes is kilobytes. The scale varies in reports, depending on the actual value.

String attributes may be any arbitrary character string, enclosed in quotes if necessary, as described abovein the query syntax paragraph.

attribute valuename range

width description

space NA 1 White space before the next column.newline NA 1 Line break(s) within a logical line.

Width is actually the number ofnewlines desired.

volume string 15 The volume name.volid =id 11 The unique volume identifier.barcode string 15 The volume barcode, when set.family string 4 The media family (for example, tape, disk).type string 7 The media type (for example, 8mm, optical).volflags NA 5 Volume summary flags,d andr ,

for dirty (in use) and read-only.state NA 3 Volume state summary,E, M , X andA,

meaning eligible for recycling,manually-recyclable, both, and archiveor migration volumes, respectively.

full flag NA Matches full volumes.inuse flag NA Matches in-use (dirty) volumes.volrecycle flag NA Matches recyclable volumes.readonly flag NA Matches read-only volumes.manual flag NA Matches manually-recyclable volumes.pool string 15 The pool containing the volume.



location string 15 The volume’s location.capacity size 8 The volume’s estimated capacity.written kbsize 7 Kbytes written to volume.%used number 5 Estimated percentage used, or ‘full’

or ‘full’ for volumes marked as full.read kbsize 8 Kbytes read (recovered) from the volume.next number 5 Next media file for writing.nrec number 5 Next media record for writing.volaccess time 9 Last time volume was accessed

(written to or labeled). Old serversdo not provide this value reliably.

volretent time 9 The date the last save set on thisvolume will expire.

olabel time 9 The first time the volume was labeled.labeled time 9 The most recent time the media

volume was (re)labeled.mounts number 6 Number of times the volume was mounted

explicitly (reboots do not count).recycled number 4 Number of times the volume

was relabeled.avail NA 3 Summary of volume availability, current

valid values,n meaning nearline(that is, in a jukebox), andov meaningthe volume is being managed by SmartMedia.

near flag NA Matches nearline volumes.smartmedia flag NA Matches volumes managed by SmartMedia.metric number 6 Volume speed and desirability metric

(unused by existing servers).savesets NA 6 Number of save sets on a volume.volattrs NA 31 The extended volume attributes.

name string 31 The save set name.savetime time 9 The save time (on the client).nsavetime NA 11 The save time, printed as seconds

since 00:00:00 GMT, Jan 1, 1970.sscreate time 9 The creation time (on the server).

If the client and server clocks are out ofsync, this time may be different from thesave time.

ssid =id 11 The unique save set identifier.level 0..9, 5 The backup level. Manual backups

full, incr, are printed as blank columnmigration values in reports.or manual

client string 11 The client name for the save set.attrs NA 31 The extended save set attributes.pssid =id 11 When part of a save set series, the

previous save set identifier in theseries, zero for the first or onlysave set in a series.

ssflags NA 7 The save set flags summary, one or morecharacters in the setCvrSENRiIF , forcontinued, valid, purged (recoverable),scanned-in (rolled-in), eligible for



recycling, NDMP generated, raw, incomplete,in-progress and finished (ended),respectively.

continued flag NA Matches continued save sets.recoverable flag NA Matches recoverable (purged) save sets.ssrecycle flag NA Matches recyclable save sets.incomplete flag NA Matches incomplete save sets.rolledin flag NA Matches rolled-in save sets.ndmp flag NA Matches NDMP save sets.raw flag NA Matches raw sav e sets.valid flag NA Matches valid save sets. All save sets

are marked ‘valid’ by current servers.sumflags NA 3 Per-volume save set summary flags,

as described for the−v report.fragflags NA 3 Per-section save set summary flags,

as described for the−V report.totalsize number 11 The total save set size.nfiles number 5 The number of the client’s files

in the save set.ssbrowse time 9 The save set’s browse time. This is

the time limit that the save set willremain browsable. ‘undef’ is displayedwhen connected to a downrev server.

ssretent time 9 The save set’s retention time(expiration time). This is the time limit thatthe save set will remain in the mediadatabase.

ssinsert time 9 The save set’s insertion time. This isthe time the save set was most recentlyintroduced into the database (for example, by abackup or by runningscanner(1m)).

sscomp time 9 The save set’s completion time. This isthe time the save set backup was completed.

clientid =id 9 The globally unique client identifier forthe host that was backed up in this save set.

copies number 6 The number of copies (instances orclones) of the save set, all with thesame save time and save set identifier.

cloneid =id 11 The clone identifier of one copy.clonetime time 9 The time a copy was made.clflags NA 5 The clone flags summary, from the

setais for aborted, incompleteand suspect (read error), respectively.

suspect flag NA Matches suspect save set copies, copiesthat had errors during file recovery.

annotation string 31 The (archive) sav e set’s annotation. In aqueryspec, the string is a regular expressionin the form used bygrep(1).

group string 12 The group of this save set. This is thegroup that backed up this save set.

first number 11 The offset of the first byte of thesave set contained within the section.

last NA 11 The calculated offset of the last byte



of the save set contained within thecurrent section.

fragsize NA 7 The calculated size of the currentsection of the save set.

sumsize NA 7 The calculated total size of all of thesections of the save set on this volume.

mediafile number 5 The media file number containingthe current section of the save set.

mediarec number 5 The media record number where thefirst bytes of the save set are foundwithin the current media file.

mediamark number 5 The absolute positioning data forthe current section (not used byexisting servers).

ssaccess time 9 The last time this section of the saveset was accessed (for backup or recover).

EXAMPLESIn the following examples, the equivalent short-hand and custom versions of the report are shown, when ashort-hand option exists for a given report or query.

Display all bootstraps generated in the previous five weeks, as reported bysavegrp(1m):mminfo −Bmminfo −N bootstrap −t ’5 weeks ago’ −ot

-r ’savetime(17),space,level(6),ssid’-r ’mediafile(6),mediarec(1m),space(3),volume’

Display information about all of the volumes:mminfo −mmminfo −a −r ’state,volume,written,%used,read,space’

-r ’mounts(5),space(2),capacity’

Display media information from volumesmars.001andmars.002:mminfo −m mars.001 mars.002mminfo −m -q ’volume=mars.001,volume=mars.002’

Display all save sets named/usr:mminfo −N /usrmminfo −q name=/usr

Display save sets named/usr, generated by clientvenus, in the past week:mminfo −N /usr −c venusmminfo −q ’name=/usr,client=venus’

Display save sets named/usr, generated by clientvenus, on volumemars.001:mminfo −N /usr −c venus mars.001mminfo −q ’name=/usr,client=venus,volume=mars.001’

Display a media report of all volumes written on in the past week:mminfo −m -t ’last week’mminfo −m -q ’savetime>=last week’

Display a media report of all non-full volumes, showing the percent-used, pool and location of each vol-ume:

mminfo −a −r ’volume,%used,pool,location’ -q ’!full’

Display a media report similar to the−m report but showing the barcode instead of the volume label:mminfo −a −r ’state,barcode,written,%used,read,space’

-r ’mounts(5),space(2),capacity’

Display a verbose list of the instances of all save sets with more than one copy, sorted by save time and



client name:mminfo −otc −v −q ’copies>1’

Display all archive sav e sets with an annotation of "project data" for the past four months.mminfo −q’annotation=project data’

-r"volume,client,savetime,sumsize,ssid,name,annotation"-t’four months ago’

FILES/nsr/mm/mmvolume

The save set and media volume databases (actually accessed bynsrmmdbd(1m)).

SEE ALSOgrep(1), nsr_getdate(3), nsr_layout(5), nsradmin(1m), nsrmmdbd(1m), recover(1m), savegrp(1m),scanner(1m).

DIAGNOSTICSno matches found for the query

No save sets or volumes were found in the database that matched all of the constraints of thequery.

invalid volume name ‘volname’The volume name given is not in a valid format. Note that volume names may not begin with adash. Queries that match no volumes will return the error ‘no matches found for the query’.

only one of −m, −B, −S, −X or −r may be specifiedOnly one report can be generated at a time. Use separate runs ofmminfo to obtain multiplereports.

invalid sorting order specifier, choose from ‘celmnotR’Only letters fromcelmnotRmay be used with the−o option.

only one −o allowedOnly one sorting order may be specified.

only one −s allowedOnly one server can be queried at one time. Use multiple runs ofmminfo to obtain reports frommultiple servers.

Out of MemoryThe query exhausted available memory. Try issuing it again, using the sorting order−om, or makethe query more restrictive (for example, list specific volumes, clients, and/or save set names).

invalid value specified for ‘attribute’The value specified is either out of range (for example, a negative number for a value that can onlytake positive numbers), the wrong type (an alphabetic string value specified for a numericattribute), or just poorly formatted (for example, non-blank characters between a close quote andthe next comma or a missing close quote).

value of ‘attribute’ is too longThe value specified forattribute is longer than the maximum accepted value. Query attributesmust have values less than 65 characters long.

non-overlapping range specified for ‘attribute’The range specified forattribute is a non-overlapping numeric range, and cannot possibly matchany sav e set or volume in the database.

unknown query constraint: attributeThe given queryattribute is not valid. See theCUSTOM QUERIES AND REPORTS table for alist of all valid attribute names.



need a value for query constraint ‘attribute’Theattribute is not a flag, and must be specified in the ‘name comparator value’ format.

constraint ‘attribute’ is only valid for reportsTheattributespecified for a query may only by used in report (−r) specifications. Calculated val-ues, flag summaries, save set extended attributes, and formatting tools (spaceandnewline) maynot be used in queries.

invalid comparator for query constraint ‘ attribute’The comparator used is not valid for the givenattribute. See theCUSTOM QUERIES ANDREPORTSsection for a list of the valid comparators forattribute.

query constraint ‘attribute’ specified more than onceThe given attribute was specified more than once with the same comparator, and is not a stringattribute (string attributes can match one of several specific values).

unknown report constraint: attributeThe given reportattribute is not valid; see theCUSTOM QUERIES AND REPORTS table for alist of all valid attribute names.

constraint ‘attribute’ is only valid for queriesTheattributespecified for a report is a flag matching attribute and may only be used in query (−q)specifications. See theCUSTOM QUERIES AND REPORTS table for the appropriate flagsummary attribute that one may use in reports of a given flag.

column width of ‘attribute’ is inv alidThe width specified forattribute is out of range. Column widths must be positive numbers lessthan 256.

missing close parenthesis after report constraint‘attribute’The width ofattribute is missing a close parenthesis.

missing comma after report constraint ‘attribute’There are non-blank characters after the width specification forattributewithout any comma pre-ceding them.

No data requested, no report generatedThe given report specification contains only formatting, no data attribute names.

LIMITATIONSYou cannot specify save set extended attributes as query constraints.

You cannot list several possible equality matches for numbers, only for strings.

Some queries, namely those that are not highly selective (few query constraints) and use a sorting orderwhere the volume name is not the primary sort key, still requiremminfo to retrieve the entire databasebefore printing any of it. Such queries use large amounts of memory inmminfo, but not, as was the casewith older versions, innsrmmdbd.

You cannot make a report that shows save set or media instances and a summary without runningmminfoat least twice.

You cannot specify query constraints that compare database attributes with each other.

You cannot make a report that uses-B flag with -c flag.


MMLOCATE(1m) MMLOCATE(1m)

NAMEmmlocate − NetWorker media location reporting command

SYNOPSISmmlocate[ −sserver] [ −l{ −n volname| −i volid | location}] [ −L ] [ −d location][ −c { −n volname| −i volid }] [ −u { −n volname|−i volid| location }]

DESCRIPTIONThe mmlocate command is used to access and manage the volume location information contained in themedia database. The information contained in a volume’s location field is meant to give the user an idea ofwhere the volume can physically be found. Other NetWorker commands will display the location alongwith the volume name (see theversionssub-command ofrecover(1m)). Any user can use this commandwith the -l (default) or-L options. The-c, -d and-u options are limited to NetWorker administrators (seensr(1m)). -l is assumed by mmlocate if a -L, -c, -d or -u option is not specified.

Runningmmlocate without any arguments lists all volumes and their locations for the specified server (ifyou do not specify a server, the current host is used).

Note that each timensrjb(1m) moves a piece of media inside a jukebox, the location of a volume is set tothe name of the jukebox. When using storage nodes, the name of the jukebox is used to indicate on whichnode the volume can be mounted. Hence, the first portion of this field containing the jukebox name shouldnot be changed. When using volumes on a storage node that are not contained within a jukebox, this fieldcan be used to indicate on which node a volume should be mounted, by giving it a value of any remotedevice on that node. Seensr_storage_node(5) for additional detail on storage nodes.

OPTIONS−c Clears the location field for the specified volume.

−d locationDeletes all volumes associated with the specified location. A confirmation prompt appears prior tothe deletion of each volume.

−i volid Restricts themmlocateoperation to the specified volume ID (volid).

−l Lists entries. Performs a database query using the suppliedvolume name, volume IDor location.

If a volume name or volume id is given, then only the volume’s location information is displayed.If a location is provided, then only volumes in that location are displayed. When the -l option isused without specifying any other options (volume name, volume id, or location), volumes withouta set location are displayed.

−L Lists all locationsfound in the database.

−n volnameRestricts the operation to the volume name (volname) listed.

−sserverAccesses theserver’s media database.

−u Updates the location for a volume. Locations are limited to a maximum length of 64 characters.The-n volnameor -i volid and locationoptions must be used in conjunction with the-u option.

EXAMPLESUpdate the media database to show that volume Offsite.011 is now at location ’Media Vault’

mmlocate −u −n Offsite.011 ’Media Vault’

Delete volumes at location ’Media Shelf 6’mmlocate −d ’Media Shelf 6’

Delete location information for volume NonFull.001mmlocate −c −n NonFull.001

List the location of volume NonFull.001mmlocate −n NonFull.001


MMLOCATE(1m) MMLOCATE(1m)

List all volumes stored in the location ’Media Vault’mmlocate ’Media Vault’

FILES/nsr/mm/mmvolume The media database.

SEE ALSOnsrmm(1m),mminfo(1m),nsr(1m),nsrjb (1m), recover(1m)nsr_storage_node(5)

DIAGNOSTICSServerserverdoes not support remote update

operations...If you are runningmmlocateagainst an old server, you are not allowed to use the -u or -c options.You must login to that server and run the mmlocate program there.


MMPOOL(1m) MMPOOL(1m)

NAMEmmpool − NetWorker media pool reporting command

SYNOPSISmmpool [ −sserver] [ volume...]

[ −d pool ] [ −l pool ] [ −L ]

DESCRIPTIONThemmpool command is used to access pool information stored in the NetWorker server’s media database.This command can also be used to delete all the volumes in a particular pool. If you specify one or morevolume names with themmpool command, the report shows the pool that each named volume belongs to.By default, all volumes and their pools are displayed.

You cannot change the pool to which a volume belongs without relabeling the volume, which destroys alldata stored on the volume. Pools are configured through a NetWorker administration tool, such asnwad-min(1m) or nsradmin(1m). These tools are used to create and modify uniquepool (seensr_pool(5))resources.

OPTIONS−d pool

Deletes all volumes for the given pool. The user will be prompted for deletion of each volume.

−l pool Lists all volumes and the pools to which they belong. If a pool is specified,mmpool only lists thevolumes belonging to that pool.

−L Lists the names of all pool resources configured on the server.

−sserverSpecifies the NetWorker server to act upon. Seensr(1m) for a description of server selection.

FILES/nsr/mm/mmvolume (UNIX)

The media database on the server.

SEE ALSOnsr(1m),nsr_device(5), nsr_pool(5), nsradmin(1m),nsrjb (1m),nsrmm(1m)nwadmin(1m).


MMRECOV(1m) MMRECOV(1m)

NAMEmmrecov − recover a NetWorker media index

SYNOPSISmmrecov [ −q | −v ]

DESCRIPTIONThemmrecov command is used in recovering from the loss of a NetWorker server’s critical files.mmre-cov restores the media index and the server’s resource files. Typical events causing such disasters are acci-dental removal of these files by a user or a disk crash on the NetWorker server itself. Seensr_crash(1m)for a discussion of general issues and procedures for NetWorker client and server crash recovery if you arerunning NetWorker for UNIX.

mmrecov is used to recover the NetWorker server’s media database and resource files from the media(backup tapes or disks) when the media database or resource files have been lost or damaged. Note that thiscommandoverwritesthe server’s existing media index. Themmrecov command is not used to recoverNetWorker clients’ online indexes; you must use thensrck(1m) command for this purpose.

The NetWorker system must be fully installed and correctly configured prior to using this command. If anyof the NetWorker software is lost, re-install NetWorker from the distribution files before you runmmrecov.Use the same release of NetWorker, and install it in the same location as it was before the software was lost.

The mmrecov program extracts the contents of abootstrapsave set, which contains the media index andresource files. Oncemmrecov is done running, you shut the NetWorker server down, move the recoveredresource files into place, and restart the server. At this point, the file indexes for the server and client maybe restored by usingnsrck.

Whenmmrecov is started, it will ask for the device from which the bootstrap save set will be extracted.Next, it will ask for the bootstrap save set identifier. This number is found in the fourth column (labeledssid) of the last line of the bootstrap information sheet printed bysavegrpandmminfo -B, an example ofwhich is shown below:

Jun 17 22:21 1992 mars’s NetWorker bootstrap information Page 1

date time level ssid file record volume6/14/92 23:46:13 full 17826163 48 0 mars.16/15/92 22:45:15 9 17836325 87 0 mars.26/16/92 22:50:34 9 17846505 134 0 mars.2 mars.36/17/92 22:20:25 9 17851237 52 0 mars.3

In the example above, thessidof the most recent bootstrap save set is ‘17851237’. If you are cloning savesets, your bootstrap save set is also cloned, and you need to use thesecondto last save set. See theRECOVERING FROM CLONE MEDIA section for an example of boostrap information with clonedsave sets.

Next, mmrecov prompts for the file and record location of the bootstrap save set. Both values may defaultto zero if they are not known. Note, however, that specifying the correct file and record numbers will allowNetWorker to more quickly locate the bootstrap save set. The file and record locations are the fifth andsixth columns of the bootstrap information sheet. In the example above, the values for the file and recordlocations are 52 and 0, respectively. Finally,mmrecov will prompt that the volume (‘mars.3’ in the exam-ple above) containing the selected bootstrap save set be inserted into the specified device. The ssid, filelocation, record location, and the physical volume must be determined by the user from the printed sheet,sincemmrecov has no way of determining this information. On the other hand, if the volume containingthe bootstrap is not known, the-B option ofscanner(1m) can be used to determine the file and record loca-tions.

If the bootstrap save set spans more than one volume, multiple volume names are printed. The orderprinted is the order required bymmrecov. In the example above, the third save set produced on 6/16/92begins on volume ‘mars.2’ and spans to volume ‘mars.3’. If a bootstrap save set spans volumes,mmrecov



will prompt for the name of the device where the next volume has been loaded when an end-of-volumeoccurs. The volume is then scanned, and the bootstrap save set extracted.

After the volume scan completes,mmrecov will complete. At this point, if your original server resourcefiles were lost, you must shut down the NetWorker server, move the new resource files into place, andrestart the NetWorker server. Now the indexes can be recovered.

In order to recover the indexes for the server and client, you must runnsrck -L7. This command willreconstruct complete indexes from the save sets generated by the server’s sav e schedule. Since the save setsmay be spread across multiple volumes,nwadmin(1m) ornsrwatch(1m) should be run, and the volumesmounted as they are requested.

When nsrck completes, the message "completed recovery of index for client ’<client-name>’" is dis-played. Once a NetWorker client’s index is recovered, that client can start recovering its files usingrecover. Note that it is not necessary for the server’s index to be restored before the client indexes may berestored.

As stated earlier, the NetWorker resource files are saved as part of the bootstrap save set. If your resourcefiles were also deleted, you may quickly replace them by copying or moving them from /nsr/res.R to/nsr/res. Before restoring them to /nsr/res, the daemons must be shut down (seensr_shutdown(1m)).

Sometimes it is neccessary to recover the NetWorker server onto a new machine, for example, after a majorhardware failure. When this occurs, the NetWorker Licensing software will detect the move. Once theNetWorker server has been moved to a new machine, it must be re-registered with Customer Support within15 days of the move, or the server will disable itself. After disabling itself, you will only be able to recoverfiles; new backups cannot be performed until the server is re-registered. Notifications will be sent by theNSR Registration notification, warning of the need to re-register the product.

RECOVERING FROM CLONE MEDIAIf you are runningmmrecov with clone media only, for example, at a remote site, you will need to performthe recovery using a slightly different method. When selecting the bootstrap identifier, make sure that youare using the information associated with the cloned save set: the last save set listed in the bootstrap output.Consider the following list of save sets:

Jun 17 22:21 1996 mars’s NetWorker bootstrap information Page 1

date time level ssid file record volume6/14/96 23:46:13 full 17826163 48 0 mars.16/14/96 23:46:13 full 17826163 12 0 mars_c.16/15/96 22:45:15 9 17836325 87 0 mars.26/15/96 22:45:15 9 17836325 24 0 mars_c.26/17/96 22:20:25 9 17851237 52 0 mars.36/17/96 22:20:25 9 17851237 6 0 mars_c.3

In the example above, the ssid of the most recent bootstrap save set is ‘17851237’. The cloned save setresides onmars_c.3and the values for the file and record locations are 6 and 0, respectively.

If you lost your resource files and need to use the ones restored frommmrecov, the NetWorker serverneeds to be shut down so that you can replace the installation resource files with your recovered ones.

Once the original resource files are in place, the NetWorker server should be restarted. After it is restarted,you may recover the indexes for the server and clients by issuing thensrck -L7 command. This commandqueries the media database for the index backups and restores the indexes for the server and each client. Ifall clone volumes needed are online when the index recovery proceeds,nsrck will complete on its own.

If some of the volumes are not online, thennsrck will attempt to recover the index from the original vol-ume it was backed up to, and therefore request the original media. In the example bootstrap output above,mars_c.1and mars_c.3would both need to be online. If volumemars_c.3was the only volume online,thennsrck would also requestmars.1. To finish recovering the server’s index in this case, you need to per-form the following steps:



1. Note what volumes are needed for recovery and delete them from the media database.

nwadmin(1m) or nsrwatch(1m) lists the volumes needed for recovery in thePendingmessagespanel. Usenwadmin(1m) ornsrmm(1m) to delete the volumes from the media database.

Given the scenario in the example above where onlymars_c.3was mounted, we would have todeletemars.1from the media database, for example,nsrmm -d mars.1.

2. Restart the server to terminate the index recovery in progress.

Usensr_shutdown(1m) to bring the server down. Runnsrd(1m) to start the server again.

3. Recover the server’s index by usingnsrck -L7 servername.

Whennsrck completes, the message "The index is now fully recovered" appears.

OPTIONS−q Quiet. Displays only error messages.

−v Verbose. Generates debugging information.

FILES/nsr If this was a symbolic link when the bootstrap save set was created, it needs to be re-created manu-

ally prior to runningmmrecov.

/nsr/resThis directory and its contents are saved as part of the bootstrap save set.mmrecov restores thisdirectory, and then renames it to/nsr/res.R. The original directory is temporarily renamed to/nsr/res.orgwhile the bootstrap save set is being recovered.

/nsr/mm/mmvolumeThe NetWorker server’s media index sav ed as part of the bootstrap save set, and unconditionallyrecovered bymmrecov.

BUGSThe namemmrecov is misleading; as a result,mmrecov is often used when it is not needed. A name like"recover_server_media_database_or_resource_files_when_missing" is more descriptive. Note that any partof the bootstrap save set contents are recoverable using normal recover procedures provided that theserver’s on-line index, resource files, and media index are intact.

To recover files that are not in the on-line file index (for example, files saved after the last run ofsavegrp),scannermust be used to rebuild the media and on-line file indexes from the contents of the volumes gener-ated between the time of the last run ofsavegrpand the loss of the original index.

SEE ALSOmminfo(1m),nsr_crash(1m),nsr(1m),nsrck(1m),nsrd(1m),nsr_client(5), nsr_schedule(5),nsr_shutdown(1m), recover(1m),save(1m),savefs(1m),savegrp(1m),scanner(1m),nsrindexasm(1m),nsrmm(1m),nsrmmdbdasm(1m),nwadmin(1m),nsrwatch(1m),nsr_getdate(3)


MSENSE(1m) MSENSE(1m)

NAMEmsense − get mode sense data

SYNOPSISmsense -ab.t.l [ -p pagecode]

DESCRIPTIONThemsenseprogram will send a MODE SENSE command to the named device.

OPTIONSThe required-a argument must be used to select a specific ordinal SCSI address (seelibscsi (1m)).

The optional-p pagecodeargument may be used to select a specific mode page, else all pages are fetched(code 0x3f). This argument must be specified in hexadecimal notation.

BUGSThe output is not readable. It is intended as input topmode (1m).

SEE ALSOlibscsi(1m),pmode(1m)


NDMPJBCONF(1m) NDMPJBCONF(1m)

NAMEndmpjbconf − NDMP Server information configuration tool

SYNOPSISndmpjbconf

DESCRIPTIONThe ndmpjbconf program is an user-interactive program to get information of the NDMP server, which hasa jukebox (Media Autochanger Device) directly attached, including NDMP server name, user name andpassword on the NDMP server, and jukebox handle on the NDMP server. With the provided information,the user will be able to issue a set of scsi commands from NetWorker Server to communicate with theNDMP server remotely.

NOTE: The ndmpjbconf program is only required to run before issuing NetWorker "inquire -N -f ..." and allsji commands. However, it is not necessary to run this program before running jbconfig.

A NDMP configuration file will be generated in a location as the user prefers, or the default location. Thename of the NDMP config file will be "ndmpjbconf_<NDMP server name>". A default config file named"ndmpjbconf" with the same content as the above config file will also be generated in the same directorythat the user chose. This default config file will be used by inquire command if -f is not specified.

The format of the NDMP config file is a one-line text file:

server user password jukebox_handle

EXAMPLE# ndmpjbconf

Enter NDMP Server name:hawaii<RETURN>Enter NDMP user name:root<RETURN>Enter NDMP user password (characters will not be echoed):password<RETURN>Enter NDMP Jukebox handle:mc0<RETURN>Enter location for NDMP Configuration file? [/usrlib/nsr]<RETURN>

PERMISSIONSRequire super user privilege to execute the ndmpjbconf program in order to create directory for storing theconfig file. The NDMP config file will be generated with super user read-only permission.

SEE ALSOinquire (1m)


NWADMIN(1m) NWADMIN(1m)

NAMEnwadmin, networker − graphical administration interface to NetWorker

SYNOPSISnwadmin [ −sserver]networker [ −sserver]

DESCRIPTIONnwadmin is an X Window System application. It is used to administer and monitor NetWorker servers.

The server’s name may be specified with the−s serverargument. When no server is specified,nwadminuses the server selection rules found innsr(1m). When multiple NetWorker servers are accessible, theymay be selected from within thenwadmin command.

The nwadmin command is used to administer NetWorker servers. Clients may be added and deleted.Schedules controlling save lev els may be created and modified. Groups of clients may be formed and con-trolled together. Directives controlling how data is saved may be defined and changed. Cloning of savesets and recover by sav e sets may be specified. Cloning of entire backup volumes may also be specified.The notification messages may be displayed. In general, there is a panel for each component.

The current state of NetWorker servers may be monitored. The amount of data saved, the number of clientssaving, and requests for mounting and unmounting volumes, are among the items displayed. This informa-tion is displayed on the console panel.

The graphical interfaces for backup and recover are available throughnwbackup(1m) andnwrecover(1m),respectively.

A complete explanation of thenwadmin command may be found in theNetWorker Administrator’s Guide.

OPTIONS−sserver

Set the current NetWorker server toserver.

FILES/usr/lib/X11/app-defaults/Networker

The X11 resources fornwadmin.

NOTEnetworker is linked tonwadmin.

SEE ALSOnsr(1m),nsradmin(1m),nwbackup(1m),nwrecover(1m),The NetWorker Administrator’s Guide


SAVEGEMS(1m) SAVEGEMS(1m)

NAMEsavegems, newgems − backup a GEMStation with NetWorker, migrate to a new version of GEMS

SYNOPSISsavegems<save options>

newgems[ −n ] −V version−E directory

DESCRIPTIONsavegemssaves the information in a GEMStation to a NetWorker server (seesavegrp(1m)). This commandis intended to be specified as thebackup command value for the client resource that is configured tobackup a GEMS server (seensr_client(5)). savegemswill backup a GEMStation by bringing down theGEMS server software, exporting the configuration data for each GEMS subsystem, then backing-up theexported data using the NetWorkersavecommand. Oncesavehas completed, the GEMStation is broughtup. A directory namedbackup will be created during the backup process (seeFILES section below). Thisdirectory will contain the files exported from GEMS. This directory should be explicitly listed as a valuefor thesave setattribute in theNSR client resource (seensr_client(5)). After thesavecommand completesthe backup, thebackup directory is deleted.

Before thesaveprocess is started,savegemsverifies the integrity of the exported files. If the exported filesappear to be intact, thesaveprocess executes, and the success of the save set is determined by the executionof the savecommand. However, if the integrity of any exported file is in question, thesavegemsprocesswill not executesave, but will print a save set failed message. A failed backup will also terminate with anabnormal exit code. The cause of a failed GEMStation backup can be determined by perusing the log fileslisted below.

newgemscreates the files necessary to migrate from an existing GEMS installation to a new GEMSrelease. Thedirectory argument is a directory where the files used in the migration process will be stored.Whennewgemsis executed, log files (seeFILES below) will be stored in this directory as well, instead ofthe location used bysavegems.

To recover a successfully backed-up GEMS server, or complete the GEMS upgrade process, seerecgems(1m).

OPTIONSsavegems

Please see thesave(1m)man page for a description of the options supported bysavegems.

newgems−n Do not restart the GEMStation after the files have been exported.

−V version versionis the currently installed version of GEMS. This flag and value are irrelevant, butare required for backward-compatibility.

−E directory Specify the directory where export and log files will be written. The path specified mustbe an existing directory.

FILES/gems/logs/savegems.out.log

The standard output of the export processes. This file contains information useful for tracking theprogress of a backup. When executingnewgems, this file will bedirectory/newgems.out.log.

/gems/logs/savegems.err.logThe error output of the export processes. This file contains information useful for determining thecause of a failed backup. When executingnewgems, this file will bedirectory/newgems.err.log.

/gems/tmp/backupThe directory where exported data files are stored. This directory should be specified as the solesave set value for the client resource used to backup a GEMStation.

1.3.Build.149a May 04, 2001 1


SEE ALSOnsr_client(5), recgems(1m), recover(1m),save(1m),savefs(1m),savegrp(1m)

DIAGNOSTICSExit Codes

0 Normal exit. This means that the backup executed normally, and a save set was correctly createdon the server.

<>0 Abnormal exit. This means that one or more files created during the backup process failed to bebacked-up, and/or a save set was not correctly created on the server.

Messageshost: /gems/tmp/backup level=level, size time countfiles.

This message (with the appropriate client host name, level, total save set size, elapsed time, andfile count) is printed wheneversavegemsis run bysavegrp(1m) and exits normally.

host: /gems/tmp/backup: failedMessages of this form indicate that the GEMStation backup failed. The log files listed aboveshould be perused to determine the cause of the failure.

Log MessagesSeveral messages may be present in the log files listed above. Diagnostics pertaining to integrity checks onexported files begin with ‘Status:’, ‘Warning:’, or ‘Error:’. Status messages indicate successful verificationof a given file, while warning messages indicate a condition that can be safely ignored. Error messages indi-cate a condition that must be corrected to perform a successful backup.

NOTESThe savegrp(1m)command may retry a failed GEMStation backup depending on the value of theclientretries attribute in theNSR group resource (seensr_group(5)). If the value of this attribute is greater thanzero, the process described in this document is invoked one more than the value of theclient retriesattribute. Determining the cause of an initial failure may be difficult since the log files created are overwrit-ten upon eachsavegemsinvocation.

1.3.Build.149a May 04, 2001 2

NSR(1m) NSR(1m)

NAMENSR − introduction and overview of NetWorker

DESCRIPTIONNetWorker facilitates the backup and recovery of files on a network of computer systems. Files and filesys-tems may be backed up on a scheduled basis. Recovery of entire filesystems and single files is simplifiedby use of an on-line index of sav ed files.

NetWorker uses a client-server model to provide the file backup and recover service. At least one machineon the network is designated as the NetWorkerserver, and the machines with disks to be backed up areNetWorkerclients. Fiv e daemons provide the NetWorker service, control access to the system, and provideindex and media support. On the clients, there are special programs to access the file systems and commu-nicate with the NetWorker server.

The NetWorker system has several parts. Commands and files are only briefly mentioned here; see theappropriate reference manual page for more detailed information. Each command has a manual page entryin section 8. The files and their formats are explained in section 5 manual pages.

The Legato NetWorker Administrator’s Guideprovides information on configuring and administering aNetWorker system. It includes many examples and rationales for setting up and running a successfulbackup operation.

INSTALLATIONHow NetWorker is installed depends on the architecture of the machine upon which you are installing. Fordetailed installation instructions, see theLegato NetWorker Installation Guidefor your specific platform.

nsr_ize(1m) The NetWorker installation script. The script will install both clients and servers. Thensr_ize script can also be used to de-install NetWorker. Note that some systems use othermethods for installing and de-installing NetWorker, in which case thensr_izescript will notexist.

nsr_layout(5)Describes where NetWorker programs, files, and manual pages are installed.

SERVER DAEMONSNetWorker uses a client-server model to provide a backup and recover service. The following daemonsencompass the server side of NetWorker.

nsrd(1m) The main NetWorker daemon.nsrd handles initial communication with clients, and startsand stops the other NetWorker server daemons.

ansrd(1m) Theagentnsrd process, spawned bynsrd in response to a recovery, clone, or other session.Theansrd daemon is invoked on an as-needed basis and is only present when there are ses-sions active to the NetWorker server. Modern versions ofsave(1m) do not require use ofanansrd daemon.

nsrindexd(1m)This server daemon provides access to the NetWorker on-line index. The index holdsrecords of saved files. The index allows clients to selectively browse and choose files torecover without having to access the backup media.

nsrmmdbd(1m)The media management database daemon provides an index of sav e sets and media. Thensrmmdbd daemon provides a much coarser view of the saved files than doesnsrindexd,and therefore the resultant index is usually much smaller.

nsrmmd(1m) The media multiplexor daemon provides device support for NetWorker. When more thanone client is saving files, the data from each client is multiplexed. During recovery opera-tions, the data is demultiplexed and sent back to the requesting clients. When the multipledevices are enabled, several of these daemons may be active simultaneously.


NSR(1m) NSR(1m)

ADMINISTRATIONNetWorker is administered viaresourcesandattributes. Every resource has one or more attributes associ-ated with it. For example, a device is a NetWorker resource type; an attribute of devices is the devicetype,for example, 4mm or 8mm. The NetWorker resource format is documented innsr_resource(5). There isalso a manual page for each NetWorker resource in section 5 of the manual.

Resource files are not normally edited by hand. Rather, a NetWorker tool (usuallynwadmin(1m) ornsrad-min(1m)) is used to modify resource files dynamically so that values can be checked and changes can bepropagated automatically to the interested programs. The following are tools that are used to administervarious aspects of NetWorker.

nwadmin(1m)Monitors the activity of and administers NetWorker servers. Thenwadmin command is anX Window System application, using a Motif look and feel. Thenwadmin command ismost users’ primary interface to NetWorker.

nsradmin(1m)A curses(3) based tool for the administration of NetWorker servers.

nsrwatch(1m)A curses(3) based tool to monitor the activity of NetWorker servers.

nsrmm(1m) Media manager command. Thensrmm command is used to label, mount, unmount, deleteand purge volumes. Mount requests are generated bynsrmmd, and displayed bynwadminor nsrwatch. The size of the on-line user file indexes may be controlled by deleting andpurging volumes.

nsrjb (1m) The NetWorker jukebox-controlling command. When dealing with a jukebox,nsrjb , ratherthannsrmm, should be used to label, load, and unload the volumes contained within a juke-box.

nsrim(1m) Automatically manages the on-line index. It is usually run periodically bysavegrp.

mminfo(1m) Provides information about volumes and save sets.

nsrck(1m) Checks and repairs the NetWorker on-line index. It is run automatically whennsrd starts upif the databases were not closed cleanly due to a system crash.

nsr_shutdown(1m)A shell script used to safely shut down the local NetWorker server. Thensr_shutdownscript can only be run by the super user.

SAVING FILESNetWorker supports both scheduled and manual saving of files and filesystems. Each client may be sched-uled to save all or part of its filesystems. Different clients may be scheduled to begin saving at differenttimes.

save(1m) A command-line-based tool used to back up a specified file or group of files. Thesavecom-mand may be run manually by users and administrators, or automatically bysavegrp.

nwbackup(1m)A Motif-based tool for backing up files. Thenwbackup command is the graphical equiv-alent ofsave.

savegrp(1m) Used to initiate the backup of a group of client machines. Usually started automatically bythe NetWorker server. Thesavegrpcommand also backs up the clients’ on-line file indexes,which are stored on the server. When backing up the server itself, abootstrapsave set isalso created.

nsrexec(1m) The agent savegrp process, spawned bysavegrp. The nsrexec command monitors theprogress of NetWorker commands.


NSR(1m) NSR(1m)

nsrclone(1m)The NetWorker save set/volume cloning command. Usingnsrclone, clones, or exact repli-cas, of save sets or entire volumes can be made. Clone data is indistinguishable from theoriginal data, except for the NetWorker media volumes upon which the data reside.

nsrexecd(1m)NetWorker-specific remote execution service which runs on NetWorker clients. Used bysavegrpto startsaveandsavefson client machines.

savefs(1m) Used bysavegrp to determine characteristics of a client, and to map the save setAll to thecurrent list of all save sets on a client.

RECOVERING FILESNetWorker maintains an on-line index of user files that have been saved. Users may browse the index andselect files for recovery. This information is used to build a representation of the file heirarchy as of anytime in the past. NetWorker then locates the correct volume and recovers the requested files.

recover(1m) Browses the on-line user file index and selects files and filesystems to recover.

nwrecover(1m) A Motif-based tool for recovering files. Thenwrecover command is the graphical equiv-alent ofrecover.

mmrecov(1m) Used only for disaster recovery. Recovers the specialbootstrapindex and the server’s on-line file index. Therecover or nwrecover commands are used to recover other on-linefile indexes.

scanner(1m) Verifies correctness and integrity of NetWorker volumes. Can also recover complete savesets and rebuild the on-line file and media indexes.

nsr_crash(1m) A man page describing crash recovery techniques.

nsrinfo(1m) Used to generate reports about the contents of a client’s file index.

APPLICATION SPECIFIC MODULESIn order to process user files in an optimal manner, NetWorker provides theASM mechanism. Patternmatching is used to select files for processing by the differentASMs. The patterns and associatedASMs aredescribed innsr(5). Thesavecommand keeps track of whichASMs were used to process a file so thatrecovermay use the sameASMs to recover the file.

uasm(1m) UNIX filesystem specific save/recover module. Theuasmman page documents the gen-eral rules for allASMs. Theuasmcommand and its man page actually comprise severaladditionalASMs, includingcompressasm, mailasm, andxlateasm, to name a few.

nsrindexasm(1m)Processes the on-line user file indexes.

nsrmmdbasm(1m)Processes the on on-line media database.

SERVER LOCATIONOn large networks there may be several NetWorker servers installed. Each NetWorker client commandmust select a server to use.

For server selection, the client commands are classified into two groups:administrationandoperation. Theadministration commands includenwadmin, nsrwatch, andmminfo. The operation commands includesave, savefs, andrecover. Both groups of commands accept a−s serveroption to explicitly specify a Net-Worker server.

When a server is not explicitly specified, the operation commands use the following steps to locate one.The first server found is used.

1) The local machine is examined to see if it is a NetWorker server. If it is, then it is used.

2) The machine where the current directory is actually located is examined to see if it is a NetWorkerserver. If it is, then it is used.


NSR(1m) NSR(1m)

3) The machine specified with the−c option is examined to see if it is a NetWorker server. If it is, thenit is used.

4) The list of trusted NetWorker servers is obtained from the local machine’snsrexecd(1m). Eachmachine on the list is examined to see if it is a NetWorker server. The first machine determined to bea NetWorker server is used.

5) A broadcast request is issued. The first NetWorker server to respond to the request is used.

6) If a NetWorker server still has not been found, then the local machine is used.

The administrative commands only use step 1.

SECURITYBefore a save is allowed, there must be anNSR client resource created for the given client. Before a recov-ery is allowed, the server validates client access by checking theremote accessattribute in theNSR clientresource (seensr_client(5)).

The savegrp(1m) command initiates thesave(1m) command on each client machine in an NSR group byusing thensrexecd(1m) remote save execution service. See thensrexecd(1m) man page for details. Forbackward compatibility with older versions of NetWorker,savegrp(1m) will fall back on using thersh(1)protocol for remote execution ifnsrexecdis not running on a particular client.

Access to the NSR resources through thensradmin(1m) ornwadmin(1m) commands is controlled by theadministrator attribute on the NSR server resource (seensr_service(5)). This attribute has a list of namesof the users who have permission to administer that resources. Names that begin with an ampersand (&)denote netgroups (seenetgroup(5)). Also names can be of the formuser@hostor user=user,host=hosttoauthorize a specific user on a specific host.

ROOT PRIVILEGESThe system administrator can grant root privileges to specific groups of users by changing the mode of aNetWorker program to setuid-root and setgid-group. (Seechgrp(1) andchmod(1) for more details.)

When a user invokes a program that is both setuid-root and setgid-group, he may retain root privileges ifone of the following is true:

1. The user’s name and the program’s group name are identical.

2. One of the process’s supplementary group id names is identical to the program’s group name.(Seegetgroups(2) for more details.)

3. The user’s name is an element of the netgroup whose name is identical to the program’s groupname. (Seegetgrnam(3) for more details.)

For example, the mode and group owner of therecover command can be changed such that thels outputlooks like:

-rws--s--x 1 root staff 548808 Apr 18 16:04 recoverA user invoking this command will retain root privileges if (1) his name is ‘‘staff’’, or (2) he is a member ofthe group ‘‘staff’’, or (3) his name appears as an element of the netgroup ‘‘staff’’.

Granting root privileges may be applied to the following NetWorker programs:nsrexec(1m), nsr-ports(1m), recover(1m), nwretrieve(1m), nwrecover(1m), nwadmin(1m), nsrclone(1m), nsrssc(1m),nsrmm(1m), mmpool(1m), mmlocate(1m), nsrjb(1m), nsrinfo(1m), nsrstage(1m), nsrcap(1m),save(1m), nsrpmig(1m), nwbackup(1m), nsrck(1m), nsrim(1m), jbconfig(1m), nsrcnct(1m),andscan-ner(1m).

NAMING AND AUTHENTICATIONAs described above, the NSR server only accepts connections initiated from the machines listed as clientsor listed in theremote accesslist (for recovering). Since machines may be connected to more than onephysical network and since each physical network connection may have numerous aliases, the policiesbelow are used as a compromise between security and ease of use. For further information about naming inthe UNIX environment, refer togethostent(3) or other documentation on name services.

A client determines its own name as follows. First the client’s UNIX system name is acquired via the


NSR(1m) NSR(1m)

gethostname(2) system call. The UNIX system name is used as a parameter to thegethostbyname(3)library routine. The client declares its name to be the official (or ‘‘primary’’) name returned bygethostby-name. This name is passed to the NetWorker server during connection establishment.

A server authenticates a client connection by reconciling the connection’s remote address with client’sstated name. The address is mapped to a list of host names via thegethostbyaddr(3) library function.Next, the client’s stated name is used as a parameter togethostbynameto acquire another list of hostnames. The client is successfully authenticated only if a common name between the two lists exists.

The NetWorker server maps a client’s name to an on-line index database name by resolving the client’sname to the official name returned bygethostbyname. This mapping takes place both at client creationtime and at connection establishment time.

To ensure safe and effective naming, the following rules should be employed:

1) The NetWorker clients and servers should access consistent host name databases. NIS (YP) and theDomain Name System (DNS) are naming subsystems that aid in host name consistency.

2) All hosts entries for a single machine should have at least one common alias among them.

3) When creating a new client, use a name or alias that will map back to the same official name that theclient machine produces by backward mapping its UNIX system name.

SEE ALSOrsh(1), gethostname(2), gethostent(3), netgroup(5), nsr(5), nsr_layout(5), nsr_resource(5), ypfiles(5),ypmake(5), mminfo(1m),nsr_crash(1m),nsr_ize(1m),nsr_service(5), nsr_shutdown(1m),nsradmin(1m),nsrck(1m),nsrclone(1m),nsrd(1m),nsrexecd(1m),nsrim(1m),nsrindexasm(1m),nsrindexd(1m),nsrinfo(1m),nsrjb (1m),nsrls(1m),nsrmm(1m),nsrmmd(1m),nsrmmdbasm(1m),nsrmmdbd(1m),nsrwatch(1m),nwadmin(1m),nwbackup(1m),nwrecover(1m), recover(1m),mmrecov(1m),save(1m),savefs(1m),savegrp(1m),scanner(1m),uasm(1m).TheLegato NetWorker Administrator’s Guide


NSR(5) NSR(5)

NAMEnsr − NetWorker directive file format

DESCRIPTIONThis man page describes the format of.nsr directive files. These files are interpreted bysave(1m) andApplication Specific Module (ASM) programs, during NetWorker backup processes. This format is alsoused in thedirectiveattribute of thensr_directive(5) resource.

Directives control how particular files are to be backed-up, how descendent directories are searched, andhow subsequent directives are processed. For each file backed-up, anyASM information required torecover that file is also backed-up. This enablesrecover(1m), or anyASM directly invoked, to recover a filecorrectly, even if the current directives hav e changed since the file was backed-up. Seeuasm(1m) for ageneral description of the variousASMs.

The .nsr directive file in each directory is parsed before anything in that directory is backed up, unless Net-Worker is being run inignore mode. Each line of a.nsr directive file, and each line of thedirectiveattribute, contains one directive. Any text after a "#" character until the end of the line is treated as a com-ment and discarded. Directives appear in one of three distinct forms:

[+] ASM[args...] : pattern...save environment<< dir >>

The three forms are referred to asASM specifications,save environmentdirectives, and<< dir >> direc-tives, respectively.

UseASM specifications (name and any arguments) to specify how files or directories with a matchingpat-tern are backed-up. When apatternmatches a directory, the specifiedASM is responsible for handling thedirectory and its contents. Anypatternor ASM arguments requiring special control or white space charac-ters should be quoted using double quotes (" ).

A colon (:) is used as the separator between theASM specification (and any arguments) and thepatternspecification list. Thepatternlist for eachASM specification consists of simple file names or patterns. Thepatterncannot be ".." and must not contain any "/" characters (all names must be within the current direc-tory). The string "." can be used to match the current directory. Standardsh(1) file pattern matching (*,[...], [!...], [x-y], ?) can be used to match file names. If a "+" precedes theASM name, then the directive ispropagated to subdirectories. When a directory is first visited, it is searched for a.nsr file. If one is found,it is then read. Each.nsr file is only read once. When starting a save at a directory below/, any .nsr fileson the normalized path of the current working directory are read before any files are saved to catalog anypropagated directives.

The following algorithm is used to match files to the appropriateASM specification. First the.nsr file inthe current directory (if any) is scanned from top to bottom for anASM specification without a leading "+"whosepattern matches the file name. If no match is found, then the.nsr in the current directory is re-scanned for anASM specification with a leading "+" whosepatternmatches the file name (for clarity, werecommend placing all propagating ("+") directives after all the non-propagating directives in a.nsr file).If no match is found, then the.nsr file found in the ".." directory (if any) is scanned from top to bottomlooking for a match with anASM specification that has a leading+. This process continues until the.nsrfile in the "/" directory (if any) is scanned. If no match is found (or a match is found with anASM specifi-cation whose name is the same as the currently runningASM), then the currently runningASM will handlethe save of the file.

Usesave environmentdirectives to change howASM specifications and future.nsr files are used. Thesaveenvironmentdirectives do not take any file patterns. They affect the currently runningASM and subsequentASMs inv oked below this directory. There are three different possiblesave environmentdirectives that canbe used:


NSR(5) NSR(5)

forgetForget all inherited directives (those starting with a "+" in parent directories).

ignoreIgnore subsequent.nsr files found in descendent directories.

allow Allow .nsr file interpretation in descendent directories.

The << dir >> directive can be used to specify a directory where subsequentASM specifications from thecurrent.nsr file should be applied. This directive is intended to be used to consolidate the contents of sev-eral .nsr files to a single location or directory. Thedir portion of this directive must resolve to a validdirectory at or below the directory containing this directive or subsequentASM specifications will beignored. Relative path names should be used for file names to ensure the interpretation of subsequentASMdirectives is consistent, even if a directory is mounted in a different absolute part of the filesystem.

Theremust be a<< dir >> as the first directive in a directive file used in conjunction with the−f option tosave(1m), savefs(1m) or with anASM program. Also, when<< dir >> directives are used in this manner,whether first or later in the file, absolute path names should be used to ensure appropriate interpretation.Absolute path names should also be used for each directory specified within thedirective attribute of theNSR directive resource (seensr_directive(5)).

When a<< dir >> directive is used, subsequent directives are parsed and logged for later use. When adirectory specified bydir is opened, anysave environmentdirectives specified for that directory (for exam-ple,allow, ignore, andforget) are processed first. If theASM is not currently ignoring.nsr files and a local.nsr file exists, the file is read and processed. Finally, any of the nonsave environmentdirectives specifiedfor that directory are handled as if they where appended to the end of a.nsr file in that directory. If multi-ple << dir >> specifications resolve to the same directory, then the corresponding save directives are han-dled logically in "last seen first" order.

EXAMPLESHaving a/usr/src/.nsrfile containing:

+skip: errs *.o+compressasm: .

will cause all files (or directories) located in the/usr/srcdirectory namederrs or *.o (and anything con-tained within them) to be skipped. In addition, all other files contained in the/usr/src directory will becompressed during save and will be set up for automatic decompression on recover.

Having a/var/.nsrfile containing:compressasm: adm .nsrnull: * .?*

causes all files (or directories) and their contents located within the/var directory and anything containedwithin them (except for those files located in the/var/admdirectory and the.nsr file itself) to be skipped,although all the names in the directory would be backed-up. In addition, sincecompressasmis asearchingdirective (seeuasm(1m)), the files contained within the/var/adm directory will be compressed duringbackup and will be set up for automatic decompression on recover.

The following is an example of using the/.nsr file as a master save directive file for the entire filesystem byusing<< dir >> directives to consolidate the variousASM save directives to a single location:

# Master NetWorker directive file for this machine<< ./ >># /mnt and /a are used for temporary fs mounting# and need not be saved

skip: mnt a+skip: core errs dead.letter *% *˜

# Don’t bother saving anything within /tmp<< ./tmp >>

skip: .?* *


NSR(5) NSR(5)

<< ./export/swap >>swapasm: *

# Translate all mailboxes. Also, use mailasm to save each# mail file to maintain mail file locking conventions and# to preserve the last file access time.<< ./usr/spool/mail >>

xlateasm: .mailasm: *

# Allow .nsr files to be interpreted in /nsr, even if we# are currently ignoring .nsr files. NetWorker# applications (such as nsrindexd) set up their own private# .nsr files which save index files more intelligently.<< ./nsr >>

allow# We can rebuild any .o files in /usr/src# from sources except those in /usr/src/sys.<< ./usr/src >>

+skip: *.o<< ./usr/src/sys >>

forget

FILES.nsr save directive file in each directory

SEE ALSOsh(1), nsr_directive(5), nsrindexasm(1m),nsrmmdbasm(1m), recover(1m),save(1m),savefs(1m),uasm(1m).


NSR_ARCHIVE_REQUEST(5) NSR_ARCHIVE_REQUEST(5)

NAMEnsr_archive_request − NetWorker resource type ‘‘NSR archive request’’

SYNOPSIStype: NSR archive request

DESCRIPTIONEach NSR archive request is described by a single resource of typeNSR archive request (seensr_resource(5)). To edit the NSR archive request resources for a NetWorker server type:

nsradmin -c "type:NSR archive request"

See thensradmin(1m) manual page for more information on using the NetWorker administration program.The archive request resource may also be edited using thenwadmin(1m) command.

This resource allows administrators to set up an archive to occur later or to set up frequent archives of a setof data. The administrator can run an archive on a specified client within the next 24 hours. The archive isexecuted via thensralist(1m)) command.

ATTRIBUTESThe following attributes are defined for resource typeNSR archive request. The information in parenthe-ses describes how the attribute values are accessed.Read-only indicates that the value cannot be changedby an administrator.Read/write means the value can be set as well as read.Hidden means it is anattribute of interest only to programs or experts. Hidden attributes can only be seen when the hidden optionis turned on innsradmin(1m) or by selecting the details Menu Item in the View Menu for a particular win-dow in nwadmin(1m). Choicemeans that the value of the attribute can only be one from a list specific tothat attribute (for example, status can be start now or start later).Dynamic attributes have values whichchange rapidly.Encrypted attributes contain data that is not displayed in its original form. The assump-tion is that the data is sensitive in nature and needs to be protected from accidental disclosure. Severaladditional attributes (for example, administrator) are common to all resources, and are described innsr_resource(5).

annotation (read/write)This attribute contains the annotation text associated with the archive sav e set generated from thisarchive request.Example:annotation: Product Release 4.1;

archive clone pool (read/write)This attribute indicates the archive clone media pool the archive request should use when cloningthe archive sav e set generated by this archive request.Example:archive clone pool: Archive clone;

archive completion (read/write)A notification action to be executed to send status of the archive request to.Example:archive completion: /usr/ucb/mail -s "Product Archive" systemadmin;

archive pool (read/write)This attribute can be used to override the normal media pool selection applied to the archive sav eset generated from the archive request. Selecting a pool will direct the archive to that media pool.Example:archive pool: Archive;

client (read/write)This attribute indicates what NetWorker archive client the archive request is to be executed on.Example:client: neptune;

clone (read/write)This attribute controls whether the archive sav e set generated by the archive request is to becloned. A value ofYesimplies the archive sav e set should be cloned. A value ofNo does notimply cloning.Example:clone: No;



cloned (read/write, hidden)This attribute is unused.Example:cloned: No;

completion time (read/write, hidden)This attribute indicates when the archive request completed. The format is "day-of-week monthday hours:minutes:seconds year".Example:"Thu Oct 22 17:00:37 1994";;

directive (read/write)This attribute specifies the directive to use when running the archive. The default value is nothingselected. The valid choices for the directive resource are names of the currently defined ‘NSRdirective’ resources, seensr_directive(5).Example:directive: Default with compression;

grooming (read/write)This attribute indicates any grooming actions to be taken once the archive sav e set generated bythe archive request has been created, verified, and cloned. A value ofnoneimplies no action. Avalue ofremoveimplies the files and directories specified in thesave setattribute will be removedvia thermdir (2) andunlink (2) system calls.Example:grooming: none;

log (read/write, hidden)This attribute contains any information pertaining to the execution of thensralist command.Example:log:; name (read/write) This attribute specifies the name of this NetWorkerarchive request.Example:name: Product Source Tree;

save set(read/write)This attribute lists the path names to be archived on the archive client. The names should be sepa-rated by a comma and a space (", ").Example:save set: /product/src, /accounting/db;

start time (read/write)This attribute determines when the archive request will be run. Thestatus attribute (see above)must be set tostart later for the archive request to be scheduled. The 24 hour clock format is"hours:minutes".Example:start time: 3:33;

status (read/write, choice)This attribute determines if an archive request should be run. No value implies the archive requestis not scheduled. Selectingstart nowcauses the archive request to be run immediately. Selectingstart latercauses the archive request to be run at the time specified by thestart time attribute.Example:status:;

verified (read/write, hidden)This attribute is unused.Example:verified: No;

verify (read/write, choice)This attribute indicates the archive request should verify the archive. Seensr_archive(5) for moreinformation on archiving. Selecting theYeschoice causes the verification to occur. Selecting theNo choice will not cause any verification. If the user also requests that the archive sav e set becloned, the verification is done on the clone since the cloning operation will have verified the orig-inal archive sav e set.Example:verify: Yes;

EXAMPLENote: the hidden options are not shown in this example.

A resource to define an archive request, called Product:



type: NSR archive request;name: Product Source;

annotation: Product Release 3.0;status: Start later;

start time: "2:00";client: space;

save set: /product/source;directive: Default with compression;

archive pool: Archive;verify: Yes;clone: Yes;

archive clone pool: Archive Clone;grooming: none;

archive completion: mail -s Product Source Archive productteam;

SEE ALSOnsr(5), nsr_directive(5), nsr_resource(5), nsradmin(1m),nwadmin(1m), rmdir (2), unlink (2).


NSR_CLIENT(5) NSR_CLIENT(5)

NAMEnsr_client − NetWorker resource type ‘‘NSR client’’

SYNOPSIStype: NSR client

DESCRIPTIONEach NSR client is described by a single resource of typeNSR client (seensr_resource(5)). To edit theNSR client resources for a NetWorker server type:

nsradmin -c "type:NSR client"

See thensradmin(1m) manual page for more information on using the NetWorker administration program.The client resource may also be edited using thenwadmin(1m) command.

For each NetWorker client, this resource describes which files should be saved, the schedule used to savethese files, which directive should be used to omit files from the save, how long the files’ index entriesshould be kept in the on-line file index and the media index, and who is allowed to back up, browse, andrecover this client’s files. A client may have more than one resource describing it.

ATTRIBUTESThe following attributes are defined for resource typeNSR client. The information in parenthesesdescribes how the attribute values are accessed.Read-only indicates that the value cannot be changed byan administrator.Read/write means the value can be set as well as read.Hidden means it is an attribute ofinterest only to programs or experts. Hidden attributes can only be seen when the hidden option is turnedon in nsradmin(1m) or by selecting the details Menu Item in the View Menu for a particular window innwadmin(1m). Dynamic attributes have values which change rapidly.Encrypted attributes contain datathat is not displayed in its original form. The assumption is that the data is sensitive in nature and needs tobe protected from accidental disclosure. Several additional attributes (for example, administrator) are com-mon to all resources, and are described innsr_resource(5).

name (read-only, single string)This attribute specifies the hostname of this NetWorker client.Example:name: venus;

server (constant, single string)This attribute specifies the hostname of this client’s NetWorker server. The server‘s hostname willbe used as the default value.Example:server: jupiter;

archive services(read/write, choice)This attribute determines if this system can use archive services. This attribute can only be set ifarchive support has been enabled on the server. The choices are enabled or disabled.Example:archive services: enabled;

schedule (read/write, choice)This attribute specifies the name of the schedule controlling the backup levels for the save setslisted in the ‘save set’ attribute. The default value is ‘Default’. Any currently defined schedulenames may be used, seensr_schedule(5).Example:schedule: Default;

browse policy (read/write, choice)This attribute specifies the name of the policy controlling how long entries will remain in thisclient’s on-line file index. The default value is ‘Month’. Any currently defined policy name maybe used as long as the period defined by the policy is not longer than the retention policy’s period,seensr_policy(5).Example:browse policy: Month;



retention policy (read/write, choice)This attribute specifies the name of the policy controlling how long entries will remain in themedia index before they are marked as recyclable. The default value is ‘Year’. Any currentlydefined policy name may be used as long as the period defined by the policy is not shorter than thebrowse policy’s period, seensr_policy(5).Example:retention policy: Year;

directive (read/write, choice)This attribute specifies the directive to use when backing up the client. The default value isNULL. The valid choices for the directive resource are names of the currently defined ‘NSRdirective’ resources, seensr_directive(5).Example:directive: Unix with compression directives;

group (read/write, choice list)This attribute specifies the group this client is a member of. The group controls the start time forautomatic backups. The value may be one of the currently defined ‘NSR group’ resources, seensr_group(5). The default value is ‘Default’.Example:group: Default;

save set (read/write, list)This attribute lists the path names to be saved for this client. The names should be separated bycomma space (, ). The default value is ‘All’. On Unix clients, ‘All’ refers to the mounted file sys-tems. On DOS clients, ‘All’ refers to file systems that have been specified on the client via the‘Change Automatic Backup’ selection of theNetWorker for DOS.By default, all of a DOSclient’s hard disks are backed up.

When a client needs to have different file systems saved on different schedules, a client resource isneeded for each set of file systems on a particular schedule. For all the client resources with thesame name in a group, a given path name may only appear once. When a client resource lists thesave set ‘All’, it must be the only client resource with its name belonging to its group.Example:save set: /, /usr, /usr/src;

priority (hidden, read/write, choice)This attribute controls the backup priority of this client. Priority 1 is the highest, 1000 is the low-est. Automated savegrp’s will attempt to back up clients with higher priorities before clients withlower priorities. Note that this is only one factor used to determing the next client. Thesavegrpcommand has many parameters to consider, and may choose a lower priority client while trying tobalance the load.Example:priority: 500;

remote access (read/write, string list)This attribute controls who may back up, browse, and recover a client’s files. By default thisattribute is an empty list, signifying that only users on the client are allowed to back up, browse,and recover its files. Additional users, hosts, and netgroups may be granted permission to accessthis client’s files by adding their names to this attribute. Netgroup names must be preceded by anampersand (’&’). Input of the form <user>@<host> or <host>/<user>, grants access to theclient’s files to the specified users. The <user> and/or <host> may be a wild card, "*". If a username is a wild card, it means all users at the host are granted access to the client’s data. When ahost name is a wild card, that user on all hosts is granted access to the client’s data. All users on ahost may also be granted access to the client’s data by just listing the host’s name, that is, <host> isequivalent to *@<host> or <host>/*. Note that this attribute does not override file system permis-sions, the user still needs the necessary file system permissions to back up, browse, or recover afile. The following example grants access to the client’s data for all users that satisfy at least oneof the following criteria, <user name, user’s hostname, server’s domain> is a member of the net-group "netadmins", the user is from the host mars, the user is from the host jupiter, the user’s nameis sam from host pluto, or the user’s id is root from any host.Example:remote access: &netadmins, mars, *@jupiter, sam@pluto, */root;



remote user (read/write, string)This attribute has several uses. For those clients that are accessed via thersh(1) protocol (newclients usensrexecd(1m) instead), this attribute specifies the user login name the NetWorkerserver will use to authenticate itself with the client. The default value is NULL, implying that‘root’ should be used. Whensavegrp-p (seesavegrp(1m)) is run on the NetWorker server, theserver runs commands on the client to determine which files to save. Note that when thensrex-ecd(1m) protocol is used to access the client, the remote user attribute is not used for authentica-tion.Certain clients, such as NetWare fileservers, use this attribute along with thepasswordattribute,below, to gain access to the files being backed up. Other clients that back up application data, suchas Sybase databases, use this attribute along with the password to gain access to the applicationdata. There may be a different value of this attribute for each resource that describes the sameclient.

NDMP clients use this attribute along with thepasswordattribute to configure access to a NDMPserver. The same username (remote user attribute) andpassword should be configured in thedevice resource as they are configured for the NDMP server.Example:remote user: operator;

password (read/write, encrypted)The savegrpcommand uses this attribute when initiating thesavefsandsavecommands on theclient’s machine. Thesavefsand savecommands use the password to gain access to the filesbeing backed up. If a password is given, then the "remote user" attribute for the client resourcemust also be defined. There may be a different value of this attribute for each resource thatdescribes the same client.

This attribute does not need to be set for existing Unix clients that are not backing up any applica-tion specific data.This attribute is also used in conjunction with theremote userattribute to configure access to aNDMP server.

backup command (read/write, string)The remote command to run to back up data for this client and save sets. This command can beused to perform pre and post backup processing and defaults to thesavecommand. The valuemust not include a path and must start with the prefix "save" or "nsr".Example:backup command: savemsg;

executable path (read/write, string, hidden)This attribute specifies the path to use when the NetWorker server is executing commands on theclient. When no path is specified, the "remote user’s" $PATH is used.Example:executable path: /etc/nsr;

server network interface (read/write, string, hidden)The name of the network interface on the server to be used for saves and recovers.Example:server network interface: mars-2;

aliases (read/write, string list, hidden)This attribute is a list of aliases (nicknames) for the client machine that queries can match. If thislist is empty, match on client name alone.Example:aliases: mars;

owner notification (read/write, hidden)A notification action to be executed to send the contents of status messages to the owner/primaryuser of a machine (for example, savegrp completion messages).Example:owner notification: /usr/ucb/mail -s "mars’ owner notification" carl@mars;

statistics (constant, hidden, dynamic)This attribute contains three values: the size of the client’s on-line file index in kilobytes, the num-ber of kilobytes actually used, and the number of entries in the index.Example:



statistics: elapsed = 1761860, index size (KB) = 776,amount used (KB) = 680, entries = 2216;

index save set (update-only, hidden, dynamic)This attribute specifies the client file index sav e set to purge when the index operation is set topurging oldest cycle.Example:index sav e set: /;

index path (read/write, hidden)This attribute is used to allow the NetWorker administrator to balance NetWorker online file indexdisk utilization across multiple disk partitions. If set, this attribute contains the full path to thedirectory containing the client’s online file index. Note that the last component of the path mustmatch thenameattribute of the client resource (see above). If left blank, the index path defaults tothe path /nsr/index/name, wherenameis the name attribute from the client resource.Example:index path: /disk2/index/venus;

index message (update-only, hidden, dynamic)This attribute contains the ending status message for the previous index operation. This attribute istypically blank, indicating that the previous operation completed successfully.Example:index message:;

index operation start (update-only, hidden, dynamic)This attribute contains the starting time of the current index operation. This attribute is a nullstring ("") when the operation is ‘Idle’. The format is weekday followed by hour and minutes.Example:index operation start: Wednesday 02:45;

index progress (update-only, hidden, dynamic)This attribute contains the progress the index has made towards finishing the current task. Thisattribute is blank when the operation is ‘Idle’. The progress is expressed as a percentage.Example:index progress: 45;

index operation (update-only, hidden, dynamic)This attribute contains the current index operation. It is normally ‘Idle’.Example:index operation: Reclaiming space;

parallelism (read/write, hidden)This attribute specifies the maximum number of saves that should be run at the same time for theclient.Example:parallelism: 2;

archive users (read/write, string list)This attribute specifies a list of users that are allowed to use the archive services on the client.This attribute can only be set if archive support has been enabled on the server. To schedule anarchive request for a client, root (or equivalent) must be on that client’s Archive users list, or elseroot@client must be in the server’s Administrator list. If no users are listed and the client residesin same machine as the server, only administrators and the local root user (that is, root@server) areallowed to use the archive services on the client. A value of ’*’ implies any user is allowed toarchive or retrieve data. The ’/’ and ’@’ characters are not allowed as part of the user name.Example:archive users: paul;

application information (read/write, hidden,string list)This attribute contains client application information. The use of this attribute is client specificand should be utilized as indicated by the documentation received with the product. NDMPclients fill in various parameters and values in this attribute separated by an equals sign (’=’).Example:application information: HIST=yes;

NDMP (read/write, choice)This attribute indicates whether or not the client resource is configured for NDMP backups. If theclient is used for NDMP backups, theremote userandpasswordattributes must be filled in. The



application information attribute may also be used.Example:ndmp: yes;

storage nodes (read/write, string list)This attribute is an ordered list of storage nodes for the client to use when saving its data. Its savesare directed to the first storage node that has an enabled device and a functional media daemon,nsrmmd(1m). The default value of ’nsrserverhost’ represents the server. Seensr_storage_node(5) for additional detail on storage nodes.

clone storage nodes (read/write, string list)This attribute is an ordered list of storage nodes for the storage node to use when its data is beingcloned. Cloned data originating from the storage node will be directed to the first storage nodethat has an enabled device and a functional media daemon,nsrmmd(1m). There is no defaultvalue. If this attribute has no value, the server’s’clone storage nodes’will be consulted. If thisattribute also has no value, then the server’s’storage nodes’attribute will be used to select a targetnode for the clone. Seensr_storage_node(5) for additional detail on storage nodes.

EXAMPLESNote: The hidden attributes are not shown in these examples.

A resource to define a client, called venus, backing up all of its files to the NetWorker server mars:

type: NSR client;name: venus;

server: mars;archive services: Disabled;

schedule: Full Every Friday;browse policy: Month;

retention policy: Quarter;directive: Unix with compression directives;

group: Default;save set: All;

remote access: ;remote user: ;

password: ;backup command: ;

aliases: venus, venus.legato.com;archive users: ;storage nodes: nsrserverhost;

clone storage nodes: ;

The resources for a client backing up different file systems on different schedules:

type: NSR client;name: saturn;


schedule: Default;browse policy: Month;

retention policy: Quarter;directive: ;

group: engineering;save set: /, /usr, /usr/src;

remote access: venus, sam@*, jupiter/john;remote user: operator;




aliases: saturn.legato.com;archive users: ;storage nodes: nsrserverhost;


type: NSR client;name: saturn;


schedule: Full on 1st Friday of Month;browse policy: Month;

retention policy: Quarter;directive: Unix standard directives;

group: Default;save set: /usr/src/archive;

remote access: sam@venus, &netadmins, root@*;remote user: operator;


aliases: saturn.legato.com;archive users: ;storage nodes: nsrserverhost;


SEE ALSOrsh(1), ruserok(3), nsr(5), nsr_schedule(5), nsr_directive(5), nsr_group(5), nsr_policy(5),nsr_storage_node(5), save(1m),savegrp(1m),savefs(1m),nsradmin(1m),nsrexecd(1m),nwadmin(1m)


NSR_CRASH(1m) NSR_CRASH(1m)

NAMEnsr_crash − How to recover from a disaster with NetWorker

DESCRIPTIONNetWorker can be used to recover from all types of system and hardware failures that result in loss of files.

When a NetWorker client has lost files, therecover command can be used to browse, select, and recoverindividual files, selected directories, or whole filesystems. If the networkerrecover command is lost ordamaged, it will have to be copied either from a NetWorker client or from the NetWorker distributionmedia.

When recovering a large number of files onto a filesystem that was only partially damaged, you may notwant to overwrite existing versions of files. To do this, wait untilrecover asks for user input to decide howto handle recovering an existing file. You can then answerN meaning ‘‘always no’’ to cause recover toavoid overwriting any existing files, orn if you want to protect this file but you wantrecover to ask againon other files.

If you do want to replace the existing version of a file or set of files with the saved versions, answerY or ywhenrecover asks if it should overwrite existing files (Y means ‘‘always yes’’ for future overwrite cases;ymeans just overwrite this one file).

For more information on using therecovercommand, see therecover(1m) manual page.

If the NetWorker server daemons or commands are lost, it may be necessary to re-install the server from theNetWorker distribution media. Once the NetWorker server is installed and the daemons are running, otherNetWorker server files can be recovered using therecover command. When re-installing NetWorker youmust be sure to install the/nsr directory in exactly the same place as it was originally installed. Themachine used to recover files may be different that the one used to save the files, but it must have the samehostname as the original machine. Recovery of the NetWorker server and client indexes requires that thedestination machine be of the same kind as the one used to save the indexes.

If the NetWorker server’s media database is lost, it will be necessary to recover thebootstrap from media.mmrecov recovers thebootstrap which contains the media database and the NetWorker server resourcefiles. Since the resource files cannot be restored on top of the ones the NetWorker server is using, it is nec-essary to shut down NetWorker, rename the recovered resource files, and restart NetWorker. The save setidentifier and other information about the bootstrap save set is printed bysavegrpat the end of each sched-uled save. It can also be displayed usingmminfo -B or scanner -B.

See thesavegrp(1m),mminfo(1m), andscanner(1m) man pages for more details.

If the index of any NetWorker server or client is lost, the index must be recovered from backup mediabefore therecover command can be used to browse and recover files that were saved from that client. Torecover the NetWorker server or any other client’s index once the media database and server resource fileshave been recovered, use thensrck command. Thensrck command recovers the lost index for a Net-Worker server or client by locating theindex:clientnamesave set produced by thesavegrp(1m) commandat the end of a scheduled save.nsrck queries the media database to determine which save sets to extractfrom which volumes to recover the index to the latest time. See thensrck (1m) man page for more details.

To summarize, these are the steps you must do to recover your server aftermmrecovcompletes.

1. Shut down your NetWorker server (nsr_shutdown -a).

2. Change to the/nsrdirectory (cd /nsr).

3. Save the temporary resource directory created when you reinstalled the NetWorker server (mv resres.save).

4. Move the recovered resource directory into place (mv res.R res).

5. Restart the NetWorker server (cd / ; nsrd).

6. After verifying that the recovered resources are valid, remove the temporary resource directory(rm -r /nsr/res.save).


NSR_CRASH(1m) NSR_CRASH(1m)

7. Recover your server and client indexes (nsrck -L7).

NOTE: Themmrecov command is only used to recover the NetWorker server’s media database andresource files. Usensrck to recover the server and client indexes.

Once the media database and server resource files have been recovered, you may recover any of your serveror client indexes in any order. It is not necessary to recover the server’s index before recovering the clients’indexes. Moreover, if your clients have the NetWorker client installed, you may run on-demand and sched-uled saves once the media database and server resource files have been recovered. However, you will notbe able to browse the saves for a client until you recover the client’s file index. You may use save setrecover to recover files before a client’s file index has been recovered.

See therecover(1m) man page for details on running recover by sav e set.

If the server is damaged so badly that it will not run at all, you will need to follow the manufacturer’sinstructions for re-installing and rebooting a multiuser system. Once you have the system up and runningin multiuser mode, you can re-install NetWorker (that is extract NetWorker from the distribution media andinstall it, usingnsr_ize(1m), pkgadd(1M), or any other installation utility depending on your system), usemmrecov to recover the media database and resource files, and usensrck to rebuild the on-line indexes forthe server and each client. Finally, you will want to recover files which previously existed on the machine,but which do not exist on the manufacturer’s distribution media. This may include system files which hadbeen customized, a specially tailored kernel, new special device entries, locally developed software, andusers’ personal files.

SEE ALSOnsr_ize(1m),nsr_layout(5), nsr(1m),nsrck(1m), recover(1m),savegrp(1m),mmrecov(1m),scanner(1m)


NSR_DAT A(5) NSR_DAT A(5)

NAMEnsr_data − Data formats for NetWorker Save and Recover

DESCRIPTIONAll data in the NetWorker system is encoded using theeXternal Data Representation(XDR) standard.When files are passed between client (seesave(1m) andrecover(1m)) and server (seensrd(1m)) and media(seensrmmd(1m)), they are represented as asavestream, which is encoded as a linked list ofsavefiles.There are currently 2 differentsavefileformats. A magic number at the start of each file indicates the par-ticular type of the followingsavefilethus allowing for self identifyingsavestreamscontaining more thanonesavefiletype. Logically eachsavefileconsists of some header information followed by file data. Theoriginal savefile1format uses a doubly wrapped set of client attributes describing the file attributes and thefile data is encoded as abuck etlist. The newersavefile2format uses an alternate singularly wrapped clientattributes with the file data encoded as a bucket-less succession of self describing sections each containing atype, a length, and bytes of data. The file data section of a file is terminated by an ending section with atype of 0 (NSR_ASDF_END).

The XDR language description of the OS independent portion of thesavestreamdata structures is shownbelow.

const NSR_IDLEN = 1024; /* length of file id */const NSR_MAXNAMELEN = 1024; /* max length of file system name */const NSR_MAXCATTRSIZE = 8192; /* max size of client specific attributes */const NSR_MAXBUCKETDAT A = 8192; /* max size of file bucket’s data (w/o slop) */const NSR_MAXBUCKETSIZE = 9000; /* max total size of file bucket (w/ slop) */const NSR_MAXCLNTSIZE = 16384; /* max size of a clntrec */

typedef opaque fileid<NSR_IDLEN>; /* file identifier */typedef string nsrname<NSR_MAXNAMELEN>; /* file name type */typedef opaque clientattr<NSR_MAXCATTRSIZE>; /* client attributes */typedef opaque wraposaverec<NSR_MAXCLNTSIZE>;/* wrapped osaverec */typedef nulong_t checksum; /* 4 bytes for checksum */typedef u_long sfid_t; /* savefile id (offset) */

struct id {string id_str<>; /* id string */id *id_next; /* next such structure */

};

struct asmrec {id *ar_info; /* name and args to ASM */nsrname *ar_path; /* not currently used */asmrec *ar_next; /* next such structure */

};

const NSR_MAGIC1 = 0x09265900; /* older format using buckets & ssaverec’s */

struct osaverec {nsrname sr_filename; /* name of this file */fileid sr_fid; /* client specific file id */asmrec *sr_ar; /* ASM list for this file */u_long sr_catype; /* client specific attribute type */clientattr sr_cattr; /* client specific file attributes */

};

struct ssaverec {



sfid_t sr_id; /* savefile id in the savestream */u_long sr_size; /* size of encoded savefile */nulong_t sr_savetime; /* savetime of this saveset */wraposaverec sr_wcr; /* a wrapped osaverec */

};

/** File data for older style savestream is logically* expressed as a linked list of file buckets.*/struct bucketlist {

bucket bl_bucket;bucketlist *bl_next;

};

/** XDR description of the original savefile1 format.*/struct savefile1 {

u_long sf_magic; /* magic number (must be NSR_MAGIC1) */u_long sf_chksumtype; /* file checksum type */ssaverec sf_saverec; /* wrapped file attributes */bucketlist *sf_data; /* file data in buckets */checksum sf_checksum; /* checksum value */

};

/** New sav estream defines and structures.*/const NSR_MAGIC2 = 0x03175800; /* newer bucketless format */

const NSRAPP_BACKUP = 1; /* backup application name space */const NSRAPP_HSM = 2; /* HSM application name space */const NSRAPP_ARCHIVE = 3; /* Archive application name space */

struct saverec {sfid_t sr_id; /* savefile id in the savestream */u_long sr_size; /* size of encoded savefile */nulong_t sr_savetime; /* savetime of this saveset */nulong_t sr_appid; /* application id */nsrname sr_filename; /* name of encoded file */fileid sr_fid; /* client specific file id */asmrec *sr_ar; /* ASM list for this file */u_long sr_catype; /* client specific attribute type */clientattr sr_cattr; /* client specific file attributes */

};

/** Defines for self describing data sections.* The NSR_ASDF_END type defines the end of the file data.* The NSR_ASDF_FILE_DAT A_TYPE type has the file data preceded by an* nulong_t that is the relative offset from the last block into the file.*/const NSR_ASDF_END = 0x0; /* end of ASDF data */



const NSR_ASDF_FILE_DAT A_TYPE = 0x100; /* normal file data */

/** Describes a section of NetWorker "file data" when* using ASM Structured Data Format (ASDF) sections.*/struct asdf_hdr {

nulong_t typevers; /* type of file data */nulong_t length; /* section length */

};

/** Pseudo XDR description of the newer savefile2 format.* The new sav efile2 format uses the unwrapped saverec structure* and a "bucketless" file data format that is based on ASDF.* The data portion ends with a 0 sized section of type NSR_ASDF_END.*/struct savefile2 {

u_long sf_magic; /* magic number (must be SF_MAGIC2) */u_long sf_chksumtype; /* file checksum type */saverec sf_saverec; /* new sav erec structure */<asdf_hdr & data> /* ASDF section sans buckets */

...<asdf_hdr & data> /* ASDF section sans buckets */<asdf_hdr.typevers = 0> /* final ASDF section type = NSR_ASDF_END */<asdf_hdr.length = 0> /* final ASDF section len = 0 */checksum sf_checksum; /* checksum value */

};

SEE ALSOmm_data(5), nsr(1m),nsrmmd(1m),nsrd(1m), recover(1m),save(1m),xdr (3n)RFC 1014 XDR Protocol Spec


NSR_DEVICE(5) NSR_DEVICE(5)

NAMEnsr_device− NetWorker resource type "NSR device"

SYNOPSIStype: NSR device

DESCRIPTIONEach storage device used by a NetWorker server is described by a single resource of typeNSR device. Seensr_resource(5) for information on NetWorker resources. To edit theNSR deviceresources run:

nsradmin -c "type:NSR device"Be sure to include quotation marks and to insert a space between "NSR" and "device". Seensradmin(1m)for information on using the NetWorker administration program. The mounting and unmounting of indi-vidual volumes (tapes or disks) is performed using thensrmm(1m), nsrjb (1m), andnwadmin(1m) com-mands.

ATTRIBUTESThe following attributes are defined for resource typeNSR device. The information in parenthesesdescribes how the attribute values are accessed.Read-only indicates that the value cannot be changed byan administrator.Read/write indicates a value that can be set as well as read.Hidden indicates a hiddenattribute of interest only to programs or experts. These attributes can only be seen when the hidden optionis turned on innsradmin(1m), or if theDetails View option is selected in theMedia Deviceswindow innwadmin(1m). Static attributes change values rarely, if ever.Dynamic attributes have values that changerapidly. For example, an attribute marked(read-only, static) has a value that is set when the attribute iscreated and never changes.

name (read-only, static)The name attribute specifies the path name of the device. Only non-rewinding tape devices aresupported. For systems that support "Berkeley style" tape positioning,use the BSD tape devicename. The name given to Optical disks is typically the name given to the "c" partition of the rawdevice.A logical device type has been defined to facilitate interaction with external media managementservices. When interacting with external media management services, the device name may bedetermined by the media management service associated with the device where a volume isloaded. The logical device is used to define a NetWorker device resource. The number of deviceresources that can exist is limited by the number of volumes managed by the service that Net-Worker may access simultaneously. The name given to a logical device is not related to any spe-cific device, but is required to be a unique name for the device. For logical devices both the mediatype and the family are set tological. The name, type, and family are determined after the mediamanagement service has loaded a volume into a device in response to a request made by Net-Worker. The name, type, and family of the actual device are then stored in the attributeslogicalname , logical type ,andlogical family , respectively. The association between the logical deviceand the actual device only exists when the volume is loaded into the device and allocated for useby NetWorker.When defining a remote device on a storage node, include the prefix "rd=hostname:", in the pathname; wherehostnameis the system to which the device is directly attached (the storage node).For more information, seensr_storage_node(5).Example: name: /dev/rmt/0hbn;

description (read/write)This attribute is used to store a brief description about the device. The description is used to helpadministrators identify the device, and it can be in any format.Example: description: DLT8000 tape drive in Engineerning Lab rack #2;

media type (read-only, static)This attribute indicates the type of media a device uses. Themedia typevaries depending on theOperating System/platform. Potential values, their meaning, and default capacities are:4mm − 4mm digital audio tape (1 GB);8mm − 8mm video tape (2 GB);8mm 5GB− 8mm videotape (5 GB);dlt − digital linear tape cartridge (10 GB);vhs− VHS data grade video tape (14 GB);



3480− high-speed cartridge tape (200 MB);qic − quarter inch data cartridge (150 MB);himt −half inch magnetic tape (100 MB);tk50 − DEC TK50 cartridge tape (94 MB);tk70 − DEC TK70cartridge tape (296 MB);optical − optical disks, Write Once Read Many (WORM), ErasableOptical Disks (EOD), or standard UNIX files are supported;file − file device type, standard UNIXfile system is supported;logical − used when interacting with an external media management ser-vice.Example: media type: 8mm 5GB;

enabled (read-write)This attribute indicates whether a device is available for use. The value for this attribute is eitheryesor no. If the value is set tono, no volumes may be mounted into the device. This value cannotbe changed if a volume is mounted.Example: enabled: yes;

shared devices (read-write, hidden)This attribute enables or disables all devices that have the same value for theirhardware idattribute, and so are sharing the same physical drive. Possible values areenable all, disable allordone. After the value is set to eitherenable allor disable all, and the action is performed, thevalue will be reset todone. The action will enable or disable as many devices as it can, regardlessof any error conditions. For example, it is not possible to disable a device that has a mounted vol-ume. So when this attribute is set todisable all, as many devices as possible will be disabled,excluding those with mounted volumes. For such cases, an error message will be logged.Example: shared devices: done;

read only (read-write)This attribute indicates whether a device is reserved for read only operations, such as recover orretrieve. The value for this attribute can be eitheryesor no. If the value is set toyes, only readoperations are permitted on the device. This value cannot be changed if when the volume ismounted.Example: read only: yes;

target sessions (read/write)This attribute indicates the target number of sessions that will write to a device. When all deviceson a host have the same value for this attribute, sessions are assigned to a device, until the device’starget sessionsis reached; then sessions are assigned to the next device on the host. Once alldevices have reached theirtarget sessions,new sessions are assigned equally across all devices.When this attribute has different values for devices on a host, and thensrmmd(1m) has not yetbeen assigned to a device, then sessions are assigned to annsrmmd(1m) based on the lowestattribute value among the host’s devices. Once thensrmmd(1m) is assigned to a device, thetar-get sessionsvalue for the assigned device is used.Use higher values to multiplex more clients onto each tape. This attribute is not a maximum num-ber for a device, but is used for load-balancing.Example: target sessions: 3;

NDMP (read-only)The NDMP attribute is used to note which devices are associated with NDMP servers. Thisattribute cannot be changed after the resource has been created. The resource must be deleted andrecreated if the user needs to change this attribute for this device. The same username (remoteuserattribute) andpasswordshould be configured in the device resource as they are configuredfor the NDMP server.Example: NDMP: yes;

remote user (read/write, string)The remote user attribute is used when the NDMP attribute is set to a value of yes. The valueentered for this attribute should be the username configured for the NDMP server.Example: remote user: root;



password (read/write, encrypted)This attribute is used in conjunction with theremote userattribute to configure access to a NDMPserver.Example: password: ;

media family (read-only, static, hidden)The media family attribute describes the class of storage media, as determined from the mediatype. The only legal values are:tape− tape storage device;disk − disk storage device;logical −used when interacting with an external media management service.Example: media family: tape;

message (read-only, dynamic, hidden)This attribute specifies the last message received from the NetWorker server regarding this device.The values for this attribute may include information on the progress or rate of the operation.Example: message: "Tape full, mount volume mars.017 on /dev/nrst8";

volume name (read-only, dynamic, hidden)This attribute monitors the mounting and unmounting of volumes for a device. When a volume ismounted, the value is the volume name, otherwise there is no value.Example: volume name: mars.017;

write enabled (read/write, dynamic, hidden)This attribute indicates whether writing to the current volume is allowed. The value for thisattribute may be set toyesor no. This value can only be set when a volume is not mounted.Example: write enabled: no;

volume operation (read/write, dynamic, hidden)Thevolume operationattribute manipulates the media (volume) currently located inside thedevice. This attribute can be set to one of the following values:Unmount, Mount , Verify label,Verify write time , Label, Label without mount, Eject, or Monitor device. Each of these opera-tions may require parameters to be set.When the value isUnmount, NetWorker releases the device. TheUnmount operation is asyn-chronous.When the value isMount , NetWorker mounts the loaded volume into the device. TheMountoperation is asynchronous.When the value isVerify label, the volume’s label is read by NetWorker, and the attributesvol-ume labelandvolume expiration are set. TheVerify label operation is synchronous, and there-fore the operation may take a long time to complete.When the value isVerify write time, the volume’s label is read by NetWorker, and the attributesvolume label, volume expiration, andvolume write time are set. TheVerify write time opera-tion is synchronous, and therefore the operation may take a long time to complete.When the value isLabel or Label without mount, the volume receives a new label as determinedby the attributes below. When the value isLabel, the volume is then mounted. These operationsare asynchronous.When the value isEject, NetWorker ejects the volume from the device. TheEject operation isasynchronous.When the value isMonitor device and the device is idle (no volume loaded into the device), Net-Worker will periodically check the device to determine whether a volume has been loaded into thedevice. When a volume containing a readable NetWorker label is loaded, the volume is placedinto the NetWorker media database. The volume can then be written to by NetWorker if the vol-ume is mounted with write permissions turned on; otherwise, the volume is mounted as read only,and cannot be written to by NetWorker. When a volume without a readable NetWorker label isloaded into the device, the device’sunlabeled volume loadedattribute is set toyes, and the vol-ume may be labeled at a later date. TheMonitor device operation is never performed on jukeboxdevices, because NetWorker only monitors non-jukebox devices.



volume label (read/write, dynamic, hidden)This attribute is set by theVerify label operation and can be performed before theLabel opera-tion. If this attribute is blank during the labeling process, then the volume’s current label is reused.

volume default capacity (read/write, static, hidden)This attribute is used by theLabel operation when thevolume current capacityattribute is blank.A non-blank value is used to override the default capacity associated with the media type. Thevalue of this attribute must end with K, M, or G, where K represents Kilobytes, M representsMegabytes, and G represents Gigabytes.This hidden attribute can be modified by a user, and can be used to override default sizes whenusing devices (and/or tapes) with different capacities than the defaults.Example: To override the default capacity of a tape drive to 10 Gb for all future volume labeloperations, set the value as follows:volume default capacity: 10G;

volume current capacity (read/write, dynamic, hidden)If the attribute’s value is non-blank, it determines the capacity of a volume during theLabel opera-tion. Its format is the same asvolume default capacity.Example: volume current capacity: 5G;

volume expiration (read/write, dynamic, hidden)This attribute is set by theVerify label operation and can also be used by theLabel operation.The value for this attribute is specified innsr_getdate(3) format. A blank value causes the defaultexpiration to be used during labeling.Example: volume expiration: next year;

volume pool (read/write, hidden)This attribute indicates the pool that a mounted volume belongs to. If this attribute is set during aLabel or Label without mount operation, this value will indicate the pool a volume is beingassigned to. Seensr_pool(5) for more information on volume pools.Example: volume pool: Default;

NSR operation (read-only, dynamic, hidden)This attribute indicates the current operation being performed by a device. The valid values forthis attribute are:Idle, Write , Read, Eject, Verify label, or Label.Example: NSR operation: Write;

minor mode (read-only, dynamic, hidden)This attribute indicates the current state of a device. TheNSR operationattribute is the majormode. The valid values for this attribute are:idle, reading, writing , rewinding, moving forward,moving backward, error , done, writing eof, or finding eom.Example: minor mode: moving forward;

statistics (read-only, dynamic, hidden)This attribute reports the statistics for the operation of this device. The statistics include:the time of operation ("elapsed"), the number of errors ("errors"), the last writing rate ("lastrate"), the max number of concurrent clients ("max clients"), the number of file marks written("file marks"), the number of rewinds ("rewinds"), the number of files skipped ("files skipped"),the number of records skipped ("records skipped"), the current file number ("current file"), thecurrent record number ("current record"), the relative nhumber of files being spaced over ("seekfiles"), the relative number of records being spaced over ("seek records"), the total estimatedamount read/written on the volume, in KB ("estimated KB", to be implemented in a futurerelease), the total amount read/written on the volume, in KB ("amount KB"), the current amountread/written on this file, in KB ("file amount KB"), and the current number of sessions assigned tothis device ("sessions").

cleaning required (read/write)This attribute indicates whether a device needs to cleaned. The value for this attribute may beeitheryesor no. If the value of this attribute changes from yes to no and the value ofdate last



cleanedattribute is not updated, then thedate last cleanedattribute is set to the current time. Net-Worker might set this attribute to yes if at the time the device is next scheduled to be cleaned it isnot available to be cleaned. In this case, the following message is displayed:device cleaningrequired . This message indicates that the device needs to be cleaned. This attribute can only beused for a device whosemedia family is tapeandjukebox deviceis yes. For all other devices thevalue of this attribute is alwaysno.

cleaning interval (read/write)This attribute indicates the amount of time from thedate last cleaneduntil the next scheduledcleaning for the device. This value can be specified indays, weeks, or months. One day, week, ormonth is implied if a number is not specified. If this attribute is set anddate last cleanedis blank,date last cleanedis set to the current time. This attribute may only be used for a device whosemedia family is tapeandjukebox deviceis yes.Example: cleaning interval: 2 weeks;

date last cleaned (read/write)This attribute indicates the time and day a device was last cleaned. Input may be in any formatacceptable tonsr_getdate(3). Some values acceptable tonsr_getdate(3) are relative, for exam-ple,now. For that reason all input is converted intoctime(3) format, weekday, month, day, time,year. As noted in the description ofcleaning requiredandcleaning interval , the value of thisattribute might be set automatically by NetWorker. This attribute can only be used for a devicewhosemedia family is tapeandjukebox deviceis yes.

volume block size (read-only, dynamic, hidden)This attribute indicates the block size of the currently mounted volume.

volume id (read-only, dynamic, hidden)This attribute indicates the volume id for the currently mounted volume.

long volume id (read-only, dynamic, hidden)This attribute indicates the volume id for the currently mounted volume in the long globallyunique format.

accesses (read-only, hidden)This attribute indicates the total number of operations performed on the device since it was config-ured as a NetWorker device. Changes to this attribute are propagated to all devices that have thesamehardware id value.

access weight (read/write, hidden)This attribute indicates the weight of a single operation performed on the device. The "accesses"attribute will be incremented by "access weight" each time an operation is performed on thedevice. The higher the weight, the less often the device will be selected for new operations.Changes to this attribute are propagated to all devices that have the samehardware id value.

consecutive errors (read-only, dynamic, hidden)This attribute indicates the current number of consecutive errors on a device. Changes to thisattribute are propagated to all devices that have the samehardware id value.

max consecutive errors (read/write, hidden)This attribute indicates the maximum number of consecutive errors allowed before disabling thedevice. Changes to this attribute are propagated to all devices that have the samehardware idvalue.

operation arg (read-only, dynamic, hidden)This attribute indicates extra parameters to be used during device operations. Parameters arepacked into a string and parsed by the associated operation’s function.

ev ent tag (read/write, single number, hidden)This attribute contains the tag (unique identifier) of the last notification event sent to thensrd (1m)daemon. The tag is used to clear the previous event. This attribute is used to pass informationbetween NetWorker programs, and should not be changed manually by the administrator.



volume message (read-only, dynamic, hidden)This attribute indicates the result of the last volume operation.

volume write time (read-only, dynamic, hidden)This attribute indicates the time that a save set was first written to the volume.

volume flags (read/write, hidden)This attribute displays the new flags for the volume being operated on. This attribute is used dur-ing "Label" or "Label without mount" operations.

jukebox device (read/write, dynamic, hidden)This attribute indicates the media device that is part of a jukebox device. This value can be eitheryesor no.

volume error number (read only, dynamic, hidden)This attribute indicates the last error number reported for this device. This is a numeric valueencoded with the source, severity and the actual error number. Processes check for this value onlyon error in a media operation when the media operation is known to update this field, e.g., a labelverify. The error number is not reset on a successful media operation, so it is not an indication ofthe status of the last media operation, but just the last error number reported for this device.

unlabeled volume loaded (read only, dynamic, hidden)This attribute indicates whether a volume loaded into the device has a readable NetWorker volumelabel. This value can be eitheryesor no. This attribute is set toyeswhen NetWorker is monitoringthe device, a volume is loaded into the device, and the volume does not have a valid NetWorkerlabel that can be read by this device. This attribute is set tono when the volume in the device islabeled or ejected from the device.

auto media management (read-write)This attribute indicates whether "automated media management" is enabled for a device. For juke-box devices this value is alwaysno. Seensr_jukebox(5) for a description ofauto media man-agementfor a jukebox. For non-jukebox devices, this value can be eitheryesor no. If this valueis set toyes, then any recyclable volumes loaded into the device might be automatically re-labeledby NetWorker for re-use, and unlabeled volumes loaded into the device can be automaticallylabeled. When NetWorker is labeling a volume that is not expected to have a valid NetWorkerlabel, it verifies that the volume is unlabeled before labeling the volume. A volume is consideredto be unlabeled if the volume does not contain a label that may be read by this device.Note: If a volume contains a label, but the label is written at a density that cannot be read by theassociated device, the volume is considered to be unlabeled. If the volume contains data writtenby an application other than NetWorker, it most likely does not have a label recognizable by Net-Worker, and the volume is considered to be unlabeled. With this attribute enabled, care should betaken when loading any volume considered to be unlabeled or recyclable into the device. The vol-ume might be re-labeled and the data previously on the volume over-written by NetWorker.When this attribute is set toyesfor a device, and the device is idle (no tape loaded into the device),NetWorker will monitor the device and wait for a volume to be loaded. See the description ofMonitor device in the discussion of thevolume operationattribute.Example: auto media management: yes;

logical name (read-only, hidden, no create)This attribute indicates the name of the actual device associated with the logical device. Thisattribute is only used for logical devices.Example: logical name: /dev/rmt/0hbn;

logical type (read-only, hidden, no create)This attribute indicates the actual device type associated with the logical device. The values thatcan be associated with this attribute are the values that are valid for the attributemedia type . Theonly exception is that the value of this attribute cannot be set tological. This attribute is only usedfor logical devices.Example: logical type: 8mm 5GB;



logical family (read-only, hidden, no create)This attribute indicates the family of the actual device currently associated with the logical device.The values that can be associated with this attribute are the values that are valid for the attributemedia family . The only exception is that the value of this attribute cannot be set tological. Thisattribute is only used for logical devices.Example: logical family: tape;

connection process id (read only, hidden, no create)This attribute indicates the process identifier maintaining the connection with an external mediamanagement service.External media management services often require a connection to be maintained while an appli-cation is using allocated resources. If the connection is not maintained the service may attempt toreclaim any resources allocated to an application. This may include unloading a volume currentlymounted into a device. Therefore, while NetWorker has a volume mounted into a device beingmanaged by such a service, a process must maintain an open connection with the media manage-ment service.

connection message (read only, hidden, no create)This attribute records any error message(s) reported upon exit by a process maintaining a connec-tion with an external media management service.

connection status (read only, hidden, no create)This attribute records the exit status reported by a process maintaining a connection with an exter-nal media management service. A status of zero indicates that the process exited successfully. Anon-zero status indicates an error occured while the process was exiting.

hardware id (read/write)This attribute represents the unique identification of a shared physical drive, which can beaccessed by multiple device resources. Each device resource that shares the same physical drivemust have the same value for this attribute. It can only be updated when the device is disabled andnot within a jukebox resource. When a value is defined for this attribute, corresponding devicemessages will contain a number that uniquely represents thehardware id attribute, and will bevisible in administrator commands, such asnwadmin(1m) andnsrwatch(1m). This number iden-tifies the devices that share the same physical drive.

save mount timeout (read/write, hidden, no create)This attribute indicates the timeout value for an initial save mount request for the storage node onwhich a device is located. If the request is not satisfied within the indicated time, the storage nodewill be locked from receiving save processes for the "save lockout" time. Seensr_storage_node(5) for a description of storage nodes. This attribute can be used for localdevices as well, but "save lockout" cannot be changed from its default value of zero. Hence, localdevices cannot be locked out from save requests.

save lockout (read/write, hidden, no create)This attribute indicates the number of minutes a storage node will be locked from receiving saveassignments after it reaches thesave mount timeouttime during a save mount request. A value ofzero indicates that the node will not be locked. This attribute cannot be changed for local devices.

autodetect id (read/write, hidden)This attribute is for identifying auto-detected devices. It is used by NetWorker programs only, andshould not be changed manually by the administrator.

EXAMPLEA complete example follows:

type:NSR device;name:/dev/nrst8;message:writing, donevolume name:mars.017;



media family:tape;media type:8mm 5GB;enabled:Yes;shared devices:done;write enabled:Yes;read only:No;target sessions:4;volume label:mars.017;volume default capacity:;volume current capacity:5000 MB;volume expiration:"Thu Sep 21 17:23:37 1996";volume pool:Default;volume flags:;volume operation:;volume write time:;volume block size:32 KB;volume id:32449;accesses:199;access weight:1;consecutive errors:0;max consecutive errors:20;operation arg:;volume message:;NSR operation:;minor mode:idle;jukebox device:Yes;statistics:elapsed = 257572, errors = 0, last rate = 397,max clients = 3, file marks = 22, rewinds = 4,files skipped = 1976, records skipped = 0,current file = 2389, current record = 162,seek files = 0, seek records = 0,estimated kb = 0, amount kb = 6273,file amount kb = 6273, sessions = 1;cleaning required:No;cleaning interval:2 weeks;date last cleaned:"Tue Apr 11 15:10:32 1995";auto media management:No;unlabeled volume loaded:No;logical name:;logical type:;logical family:;connection process id:;connection message:;connection status:;hardware id:;save mount timeout:30;save lockout:0;

FILES/nsr/res/nsr.res− this file should never be edited directly. Usensrmm(1m),nsradmin(1m), ornwad-min(1m) instead.

SEE ALSOnsr_getdate(3), ctime(3), nsr_resource(5), nsr_pool(5), nsr_schedule(5), nsr_service(5),nsr_storage_node(5), nsr(1m),nsrmmd(1m),nsrmm(1m),nsradmin(1m),nwadmin(1m).


NSR_DIRECTIVE(5) NSR_DIRECTIVE(5)

NAMEnsr_directive − NetWorker resource type ‘‘NSR directive’’

SYNOPSIStype: NSR directive

DESCRIPTIONEach NSR directive is described by a single resource of typeNSR directive (seensr_resource(5)). To editthe NSR directive resources for a NetWorker server usensradmin(1m) ornwadmin(1m). See the corre-sponding manual page for more information on the use of these NetWorker administration programs.

These resources are used by the NetWorkerasmfamily of commands when processing files, seeuasm(1m)and nsr(5). Directives can be used to improve the efficiency of backups by controlling which files getsaved and specifying special handling on certain types of files.

ATTRIBUTESThe following attributes are defined for resource typeNSR directive. The information in parenthesesdescribes how the attribute values are accessed.Create-only indicates that the value cannot be changedafter the resource has been created.Read/write means the value can be updated by authorized administra-tors. Hidden means it is an attribute of interest only to programs or experts, and these attributes can onlybe seen when the hidden option is turned on innsradmin(1m) or the details view is enabled innwad-min(1m). Dynamic attributes have values which change rapidly. Sev eral additional attributes (e.g. admin-istrator) are common to all resources, and are described innsr_resource(5).

name (create-only)The names of directive resources are displayed as choices when creating or updating NetWorkerclient resources, seensr_client(5). The name can generally be chosen at the administrator’s con-venience, but it must be unique for this NetWorker server. The directive resource named ‘Unixstandard directives’ may be modified, but it may not be deleted. Other directives can only bedeleted if no clients or archive lists are using them.Example:name: Unix standard directives;

directive (read/write)This attribute contains the rules defining the directive. The value of this attribute is similar to thecontents of a.nsr file except that absolute path names must be specified for each<< path>> direc-tive. Seensr(5) for more information on the format of NetWorker directives.Example:directive: "<< / >> skip : core";

NOTENetWorker comes with four directive resources already defined, "Unix standard directives", "Unix withcompression directives", "DOS standard directives", and "NetWare standard directives". The first two aremeant for use with clients running on Unix platforms. "DOS standard directives" is intended for use withclients on machines running DOS. The last directive, "NetWare standard directives", is meant for use withclients running on NetWare platforms. There may also be two other directives "Default" and "Default withcompression". These are old names for "Unix standard directives" and "Unix with compression directives",respectively. NetWorker will remove the directive resources using the old names when they are no longerof use.

EXAMPLEAn example NSR directive resource, named ‘Unix directive’, follows:

type: NSR directive;name: Unix directive;

directive: "<< / >>

+skip : coreskip : tmp

<< /usr/spool/mail >>mailasm : *


NSR_DIRECTIVE(5) NSR_DIRECTIVE(5)

<< /nsr >>allow

";

SEE ALSOnsr(5), nsr_resource(5), savegroup(1m),savefs(1m),uasm(1m),nsradmin(1m),nwadmin(1m).


NSR_GETDATE(3) NSR_GETDATE(3)

NAMEnsr_getdate − convert time and date fromASCII

SYNOPSIS#include <sys/types.h>

time_t nsr_getdate(buf)char *buf;

DESCRIPTIONThensr_getdate()routine converts most common time specifications to standard UNIX format. It takes acharacter string containing time and date as an argumant and converts it to a time format.

The character string consists of zero or more specifications of the following form:

tod A tod is a time of day, which is of the formhh[:mm[:ss]] (or hhmm) [meridian] [ zone]. If nomeridian −amor pm− is specified, a 24-hour clock is used. Atod may be specified as justhh fol-lowed by ameridian. If no zone (e.g. GMT) is specified, the current timezone, as determined bythe second parameter,now, is assumed.

date Adate is a specific month and day, and possibly a year. The acceptable formats aremm/dd[ /yy]andmonthname dd[, yy] If omitted, the year defaults to the current year. If a year is specified as anumber in the range 70 and 99, 1900 is added. If a year is in the range 00 and 30, 2000 is added.The treatment of other years less than 100 is undefined. If a number not followed by a day or rela-tive time unit occurs, it will be interpreted as a year if atod, monthname, anddd have already beenspecified; otherwise, it will be treated as atod. This rule allows the output fromdate(1) orctime(3) to be passed as input tonsr_getdate.

day A day of the week may be specified; the current day will be used if appropriate. Aday may bepreceded by anumber,indicating which instance of that day is desired; the default is1. Neg ativenumbersindicate times past. Some symbolicnumbersare accepted:last, next, and the ordinalsfirst throughtwelfth (secondis ambiguous, and is not accepted as an ordinal number). The sym-bolic numbernext is equivalent to2; thus,next monday refers not to the immediately comingMonday, but to the one a week later.

relative timeSpecifications relative to the current time are also accepted. The format is [number] unit; accept-able units areyear, month, fortnight , week, day, hour, minute, andsecond.

The actual date is formed as follows: first, any absolute date and/or time is processed and converted. Usingthat time as the base, day-of-week specifications are added; last, relative specifications are used. If a date orday is specified, and no absolute or relative time is given, midnight is used. Finally, a correction is appliedso that the correct hour of the day is produced after allowing for daylight savings time differences.

Nsr_getdateaccepts most common abbreviations for days, months, etc.; in particular, it will recognize themwith upper or lower case first letter, and will recognize three-letter abbreviations for any of them, with orwithout a trailing period. Units, such asweeks, may be specified in the singular or plural. Timezone andmeridian values may be in upper or lower case, and with or without periods.

SEE ALSOctime(3), date(1), ftime(3c), localtime(2), time(2).

BUGSThe grammar and scanner are rather primitive; certain desirable and unambiguous constructions are notaccepted. Worse yet, the meaning of some legal phrases is not what is expected;next weekis identical to2weeks.

The daylight savings time correction is not perfect, and can get confused if handed times between midnightand 2:00 am on the days that the reckoning changes.

Becauselocaltime(2) accepts an old-style time format without zone information, passingnsr_getdatea cur-rent time containing a different zone will probably fail.


NSR_GROUP(5) NSR_GROUP(5)

NAMEnsr_group − NetWorker resource type ‘‘NSR group’’

SYNOPSIStype: NSR group

DESCRIPTIONEach NetWorker group is described by a single resource of typeNSR group (seensr_resource(5)). Toedit theNSR group resources for a NetWorker server type:

nsradmin −c "type:NSR group"or use thenwadmin(1m) GUI. See thensradmin(1m) manual page for more information on using theNetWorker administration program.

These resources control when a group of NetWorker clients begin saving data and whether backups arestarted automatically each day. EachNSR client resource (seensr_client(5)) lists the groups of which thatclient (or save sets for that client) is a member. Groups can only be deleted if no clients are members ofthem.

ATTRIBUTESThe following attributes are defined for resource typeNSR group. The information in parenthesesdescribes how the attribute values are accessed.Create-only indicates that the value cannot be changed byan administrator once the resource is created.Read/write means the value can be set as well as read at anytime. Choice indicates that the value can only be selected from a given list.Yes/nomeans only a yes or nochoice is possible.Static attributes change values rarely, if ever.Dynamic attributes have values whichchange rapidly.Hidden means it is an attribute of interest only to programs or experts, and these attributescan only be seen when the hidden option is turned on innsradmin(1m). For example, an attribute marked(create-only, static)has a value which is set when the attribute is created and never changes. Several addi-tional attributes (e.g. administrator) are common to all resources, and are described innsr_resource(5).

name (create-only)This attribute contains the name of the group defined by this resource. The name must be uniquefor this NetWorker server, but otherwise can be anything that makes sense to the administrator.This name will appear as a choice attribute of eachNSR client andNSR pool(5) resource. TheNSR group resource named ‘Default’ may be modified, but it may not be removed. The name canonly be specified when the group is created.Example:name: marketing;

autostart (read/write, choice)The autostart attribute determines if this group will be saved automatically every day. It may beone of three values:Enabled, Disabled or Start now. When the value isEnabled, the membersof this group will start saving data at the time specified in thestart time attribute. When the valueis Disabled, the member of this group will not automatically start saving their data. When theStart now value is specified, the member clients will start saving their data immediately. Theattribute will then return to its prior value.Example:autostart: Enabled;

autorestart (read/write, choice, hidden)This controls whether this group should be automatically restarted when an incomplete run (due toa power failure or administrator intervention) is noticed during NetWorker server startup. Like theautostart attribute, setting this attribute’s value toRestart now causes NetWorker to restart thegroup immediately. Enablingautorestart only has an effect ifautostart is also enabled.

stop now (read/write, choice, hidden)Setting this value to ‘True’ when this group is running causes this group to abort all of its savesimmediately. Once the group is stopped, the value is set back to ‘False’. These are the only validvalues.

start time (read/write)The start time attribute specifies the time of day when this group will start saving. The NetWorkerserver’s local time is used. The time is specified as "hours:minutes". Note that the quotes may be



necessary when using character-based administration tools such as thensradmin program becauseof the colon in the value. The hours may range from 0 to 23 (using a 24 hour clock) and the min-utes range from 0 to 59.Example:start time: "4:53";

last start (read/write, hidden)The last time this group was started. This attribute is for informational purposes only, and chang-ing it has no effect on the system.

interval (read/write, static, hidden)The interval time specifies how often this group is to be run automatically by NetWorker. Manu-ally starting a group overrides the interval. The default value is 24:00, which means run once aday.

force incremental (read/write, static, hidden, choice)Setting this attribute to ‘Yes’ will force an incremental level of a sav e group, when theintervalattribute is less than 24 hours. The default value is ‘Yes,’ which means force an incremental if thegroup is run more than once a day. A value of ‘No’ would allow more than one full per day.

client retries (read/write)The number of times failed clients should be retried beforesavegroupgives up and declare themfailed. Zero means don’t retry. Abandoned saves are not retried, because they may eventuallycomplete. A client’s sav e sets are retried bysavegroupwhenever sav e group would otherwise notbe able to start a new sav e set. That is, savegroup prefers to start new sav e sets first, and onlyretries when there is nothing else to do.Example:client retries: 1;

clones (read/write, static, yes/no, choice)Setting this value to ‘Yes’ causes saves of this group to automatically make a clone of every saveset backed up. The save set clones will be sent to the pool named in theclone poolattribute.

clone pool (read/write, static, choice)The pool to which save set clones should be sent when ‘clones’ is ‘Yes’. Only pools of type‘Backup Clone’ are allowed (seensr_pool(5)).

options (read/write, static, hidden)The values specify flags with which this group will be run. The values No Monitor, No indexsave, No sav e, Verbose, Estimate, and Preview map to the thesavegroupcommand line flags -m,-I, -n, -v, -E, and -p respectively. Some of these values (Preview and No save) are automaticallyreset when a run ofsavegroupcompletes normally.Example:options: Verbose;

level (read/write, hidden, choice)This is an explicit level the savegroup will use when started automatically by NetWorker.Thishidden attribute can be modified by a user.This value is not cleared automatically, i.e. if one setsthis attribute to full this savegroup will run with a full level until this value is manually cleared.When not specified (the normal case), the NSR Schedule for each client filesystem will be used todetermine the level. Manually runningsavegroup from the command line overrides this value.The choices are the standard level identifiers ‘full’, ‘consolidate’, ‘incr’, ‘skip’ and the numberlevels ‘1’ through ‘9’.

printer (read/write, static, hidden)The printer to which the bootstrap save set information will be printed, if one is generated by therun of this group.This hidden attribute can be modified by a user.If an invalid printer name isspecified, bootstrap information will be included in the savegroup completion information pipedthrough thesavegroup completionnotification (seensr_notification(5)).Example:printer: ps;



schedule (read/write, choice, hidden)The schedule to use for determining what level of sav e to perform.This hidden attribute can bemodified by a user.This value is not cleared automatically, i.e. if one sets this attribute to a partic-ular schedule, all clients which are part of this group will have their schedules overridden until thisvalue is manually cleared. This overrides the schedule specified for individual clients. Seensr_schedule(5)).

schedule time (read/write, hidden)An explicit time can be specified when looking at a schedule to determine which level of sav e toperform. A null value (normal setting) means use the current date to determine the level.Example:schedule time: "3:00 am 01/11/93";

inactivity timeout (read/write, static, hidden)The number of minutes that thesavegroupcommand waits for any kind of activity on the serverbefore concluding that asavegroupdescendant is hung.This hidden attribute can be modified bya user. Once a hang is detected,savegroup prints a message indicating that asave is beingaborted, kills or aborts the backup, and moves on to its next task. Inactivity is defined as the lasttime a client has sent data to the server. If a client has a very large filesystem and an incrementalis run, it is possible forsavegrp to abort a save set that only appears to be hung. In these cases,the inactivity timeout should be increased to accomodate the particular client.Example:inactivity timeout: 30;

work list (read/write, dynamic, hidden)The list of saves still not completed. These come in sets of 3 values: the client name, the level ofsave, and the path to save.Example:work list: mars, incr, /usr, mars, incr, /g, mars, venus, /usr

completion (read/write, dynamic, hidden)The status of each save set that has been completed. These come in sets of 4 values: the clientname, the path saved, a status message (succeeded, failed, or unexpectedly exited), and the outputfrom the save.Example:completion: "mars", "/usr", "succeeded", "mars: / level=full, 6577 KB 00:06:41625 files"

status (read-only, dynamic, hidden)The current status of this NSR group. Currently, this can have the values ‘idle’, ‘running’ and‘cloning’. The value ‘idle’ is set when the group is not active, it is ‘running’ while backups are inprogress, and it is ‘cloning’ when the backups are complete and clones are automatically beingmade.

EXAMPLEThe default NSR group resource automatically starts its members at 33 minutes past 3 o‘clock in the morn-ing:

type: NSR group;name: Default;

autostart: Enabled;start time: "3:33";

administrator: root;

A complete example follows, with the hidden attributes shown with values reasonable for an active group:

type: NSR group;name: Default;


options: Restartable;printer: lp2;



inactivity timeout: 30;work list: mars, incr, /g, mars, incr, index,

venus, incr, /usr, venus, incr, index,jupiter, full, /, jupiter, full, /usr,jupiter, full, index

completion: mars, /, succeeded,"mars: / level=incr, 31 KB 00:01:01 72 files",

mars, /usr, succeeded,"mars: /usr level=incr, 2 KB 00:00:48 5 files",

venus, /, succeeded,"venus: / level=incr, 7711 KB 00:04:37 29 files";

administrator: root, &operator;

SEE ALSOnsr(1m),nsr_notification(5), nsr_pool(5), nsr_resource(5), nsr_schedule(5), nsradmin(1m),nwadmin(1m),savegroup(1m).


NSR_IZE(1m) NSR_IZE(1m)

NAMEnsr_ize − NetWorker installation and removal

SYNOPSISnsr_ize[ −i | −r | −u ] [ −c | −t | −l | −s ] [ −kmnqxv ]

DESCRIPTIONnsr_ize is an interactive program that installs or removes NetWorker software and files to or from amachine. In either case, the user is prompted with a series of questions. Most questions also providedefault answers that produce the desired results in a standard environment.

Since this command installs all NetWorker software in one directory per machine architecture (the defaultis /usr/bin), one should plan for the NetWorker client software to consume 13 Mbytes of disk space permachine architecture installed. The NetWorker server software requires an additional 8 Mbytes permachine architecture.

The NetWorker server also requires disk space for its on-line indexes of all files saved by the NetWorkerclients. On the average, each instance of a file saved to the NetWorker server requires 200 bytes of diskspace. Generally, 2% - 5% of the total amount of data backed up has to be allocated to the online index.

The script modifies many administrative files, including/etc/rpc. If YP is being used, then the administra-tor should consider modifying the YP master’s/etc/rpc file with the same modifications thatnsr_izemakesto the local copy.

When invoked with no options, this command makes an intelligent guess as to whether the user is installingor removing NetWorker software, and confirms its guess with the user. If the−s , −t , −l or −c flags are notspecified when installing,nsr_izewill prompt and ask whether the machine being installed is going to be aNetWorker server.

OPTIONS−c Performs NetWorker client side−only operations.

−i Installs the NetWorker software and associated files.

−k Kills daemons if running, without confirming.

−l Performs NetWorker license management−only operations.

−m Does not remove orinstall man pages.

−n Does not perform any actions that actually change the filesystem, just print what would be done ifan installation or removal took place.

−q Quiet. Does not print as many reassuring messages as in the normal case.

−r Removes the NetWorker software and associated files.

−s Performs NetWorker server−side operations.

−t Performs NetWorker Storage Node side−only operations.

−u Prepares for a software upgrade. NetWorker software is removed, but associated files are pre-served.

−v Sets verbose mode.

−x Sets debugging flag.

FILES/nsr This symbolic link is created bynsr_ize when installing the NetWorker server; it points to the

directory where the NetWorker server keeps the on-line indexes of clients’ saved files. Alterna-tively, /nsr may be a directory that exists before invoking this command.

SEE ALSOnsr(1m)


NSR_IZE(1m) NSR_IZE(1m)

Legato NetWorker Installation Guide UNIX Version


NSR_JUKEBOX(5) NSR_JUKEBOX(5)

NAMEnsr_jukebox − NetWorker resource type ‘‘NSR jukebox’’

SYNOPSIStype: NSR jukebox

DESCRIPTIONEach jukebox known to NetWorker is described by a single resource of typeNSR jukebox. A jukeboxmay also be used to keep track of the resources, volumes and devices, being managed by an external mediamanagement service that are available to this NetWorker server. An example of an external media manage-ment service is OpenVault. This resource describes the physical characteristics of a jukebox. Seensr_resource(5). To edit the NSR jukebox resources for a NetWorker server, type:

nsradmin -c "type:NSR jukebox"or use thenwadmin(1m) GUI. See thensradmin(1m) manual page for more information on using theNetWorker administration program.

ATTRIBUTESThe following attributes are defined for resource typeNSR jukebox. The information in parenthesesdescribes how the attribute values are accessed.Create-only indicates that the value cannot be changed byan administrator, except when the resource is created.Read-only indicates that the value cannot bechanged by an administrator.Read/write means the value can be set as well as read at any time.Choicelist means that any number of values can be chosen from the given list.Yes/nomeans only a yes or nochoice is possible.Single string means that only a single value is allowed.Number means that onlynumeric values are allowed.Static attributes change values rarely, if ever.Dynamic attributes have valueswhich change rapidly.Hidden means it is an attribute of interest only to programs or experts, and theseattributes can only be seen when the hidden option is turned on innsradmin(1m) (or the View Detailsmenu pulldown innwadmin(1m)). For example, an attribute marked(read-only, dynamic) has a valuewhich cannot be changed by the administrator but which may change each time it is retrieved from the Net-Worker server due to underlying state changes. Several additional attributes (for example, administrator)are common to all resources, and are described innsr_resource(5).

name (create-only, single string)This attribute specifies the name of this jukebox. The value of this attribute may follow the"rd=hostname:" syntax of a remote device, when the jukebox is defined on a storage node. Seensr_storage_node(5) for additional detail on storage nodes.Example:name: Huntington;

description (read/write)This attribute is used to store a brief description about the jukebox. The description is used to helpadministrators identify the jukebox, and it can be in any format.Example:description: DLT Changer drive in Engineerning Lab;

model (create-only, single string)This attribute specifies the jukebox model.Example:model: ADIC-VLS;

physical slots (read-only, list of numbers, hidden)This attribute specifies the first and last physical slot numbers in the jukebox. The first slot num-ber must be less than or equal to the last slot number. The numbers must also be specified as twoseparate attribute values.

For Silo Tape Libraries (STL), this attribute is equal to the number of volumes allocated to thisNetWorker server,nsrjb (5) −a or −x. The number of physical slots changes as volumes are addedor removed from the STL.Example:physical slots: 1, 54;

control port (read/write, single string)This attribute specifies the path of the control port for the jukebox robotics. Control commands(load slot 47 into drive b, for example) are sent to the jukebox via the control port.



For an STL, this attribute specifies the information required to set up a connection to the STLserver. Form and contents of the attribute depend on the type of the STL, but most often it merelycontains the hostname of STL server.

The value of this attribute may follow the "rd=hostname:" syntax of a remote device, when thejukebox is defined on a storage node. Seensr_storage_node(5) for additional detail on storagenodes.Example:control port: [email protected];

devices (read/write, list of strings)This attribute lists the device pathnames of the devices in the jukebox. Each entry that appears inthis attribute must have a correspondingNSR deviceresource. There must be the same number ofentries in thedevicesattribute as there are physical drives in the jukebox, if none of the drives arebeing shared by multiple device resources. In addition, they must be listed in the same order asthey are physically installed in the jukebox. The entries are specified as separate attribute values.Example:devices: /dev/rmt/0mbn, /dev/rmt/1mbn;

number devices (read/write, single number, hidden)The number of configured devices in the jukebox. This value corresponds to the number of entriesin thedevicesattribute.Example:number devices: 2;

number drives (read/write, single number, hidden)The number of unique physical drives configured in the jukebox. When multiple device resourcesshare a physical drive, each drive is represented by a uniquehardware id attribute, that is speci-fied in all of the device resources sharing the same drive.Example:number drives: 2;

device hardware ids (read-only, hidden)The hardware ids of the jukebox’s devices. This attribute will have ahardware id entry for eachof the entries in thedevicesattribute of the jukebox resource. The entries of those devices sharinga physical drive will have the same value.

idle device timeout (read/write, hidden)This attribute specifies the number of minutes to wait before unmounting a volume in an idledevice. Setting this attribute’s value to zero disables unmounting idle volumes. The function ofthis attribute only applies to SmartMedia jukeboxes, or silo and native jukeboxes with device shar-ing enabled.Example:idle device timeout: 10;

SmartMedia update interval (read/write, hidden)This attribute specifies the number of hours between calls to update the SmartMedia server’sdatabase. The SmartMedia database contains information copied from the NetWorker mediadatabase. The information includes the pool to which a volume belongs, whether the volume isfull, and so forth. This information is used by the SmartMedia server when selecting a volume forwriting. Since this information may change over time, it is necessary to periodically make surethat the data replicated in the SmartMedia server’s database is current. This attribute determinesthe time period between attempts to update the SmartMedia server’s database. This attribute onlyapplies to SmartMedia jukeboxes.Example:SmartMedia update interval: 12;

write enabled (read/write, yes/no, hidden)This attribute indicates whether writing can be done to the mounted volume. This attribute is onlyused during a jukebox ‘‘Load’’ operation. This attribute is used to pass information between Net-Worker programs, and should not be changed manually by the administrator.Example:write enabled: Yes;

bar code reader (read/write, yes/no)This attribute indicates whether NetWorker should use the barcode label from the media if thejukebox has a barcode label reader. This should only be enabled if the jukebox has a barcode label



reader.Example:bar code reader: No;

match bar code labels (read/write, yes/no)This attribute indicates whether NetWorker should use the barcode label instead of a label tem-plate when labeling media volumes. This should only be enabled if the jukebox has a barcodelabel reader and the attribute "bar code reader" is enabled.Example:match bar code labels: No;

volume expiration (read/write, single string, hidden)This attribute specifies the expiration time of a volume currently being labeled. For jukeboxesinteracting with external media management services, this attribute specifies the minimum expira-tion time for the volume to be loaded. This attribute is used to pass information between Net-Worker programs, and should not be changed manually by the administrator.

av ailable slots (read/write, list of numbers)This attribute specifies the slots containing volumes available to automatically satisfy NetWorkerrequests for writable volumes. When automatically selecting a writable volume,nsrjb (1m) willonly consider volumes from the list of available slots. The slots are specified as a list of ranges,one range per attribute value. A range may be a single slot number or a pair of slot numbers sepa-rated by a dash. The first number of a pair must be less than or equal to the second.

For Silo Tape Libraries, this attribute is automatically updated when adding or removing volumes,nsrjb (1m)−a or −x.

When satisfying requests to mount a particular volume (that is, by its volume name) or slot, all ofthe volumes in the slots listed inphysical slotscan be used. This allows the jukebox to be parti-tioned, with saves restricted to a group of volumes while all of the volumes contained within thejukebox are accessible for recovers.Example:available slots: 1-10;

enabler code (read-only, single string, hidden)This attribute lists the enabler code for theNSR licenseresource (seensr_license(5)) correspond-ing to this jukebox resource. A jukebox cannot be used until a license enabler has been loaded tocontrol that jukebox.Example:enabler code: 123456-123456-123456;

enabled slots (read-only, single string, hidden)The value of this attribute is the number of slots enabled for this jukebox. This attribute’s value isset by the server when an enabler code is loaded to the jukebox.Example:enabled slots: 8;

operation (read/write, choice list, hidden)This attribute shows the operation currently being performed on the jukebox. This attribute is usedto pass information between NetWorker programs, and should not be changed manually by theadministrator.Example:operation: Load;

operation message (read-only, single string, hidden)This attribute displays an error message after a jukebox operation fails.Example:operation message: ;

operation device (read/write, single string, hidden)This attribute is used to pass the name of the device to which the current operation refers. Thisattribute is used to pass information between NetWorker programs, and should not be changedmanually by the administrator.Example:operation device: /dev/rmt/0mbn;

operation slots (read/write, single string, hidden)This attribute is used to pass the slots on which the current operation will be performed. Thisattribute is used to pass information between NetWorker programs, and should not be changed



manually by the administrator.Example:operation slots: 1-10;

operation options (read/write, single string, hidden)This attribute is used to pass the mode of the volume used when the current operation will be per-formed,nsrjb (5) −o option. This attribute is used to pass information between NetWorker pro-grams, and should not be changed manually by the administrator.Example:operation options: manual;

operation barcodes (read/write, list of strings,hidden)This attribute is used to pass the volume tags or barcodes on which the current operation will beperformed. This attribute is used to pass information between NetWorker programs, and shouldnot be changed manually by the administrator. This attribute is only used for Silo Tape Librariesand is only defined on platforms which provide support for Silo Tape Libraries.Example:operation barcodes: A01B, A0/3−5/B;

operation response(read/write, choice list, hidden)This attribute designates a default response to questions that may be asked while performing theoperation. This attribute is used to pass information between NetWorker programs, and should notbe changed manually by the administrator.Example:operation response: Yes;

operation report mode(read/write, choice list,hidden)This attribute designates the amount of output generated during the execution of the operation.This attribute is used to pass information between NetWorker programs, and should not bechanged manually by the administrator.Example:operation report mode: verbose;

operation label state (read/write, choice list,hidden)This attribute designates whether a volume being labeled is to be recycled or is expected to beunlabeled. If a volume is to be recycled, it must already have a NetWorker label. You can recyclea volume while it is being mounted. This attribute is used to pass information between NetWorkerprograms, and should not be changed manually by the administrator.Example:operation label state: recycle;

operation volume capacity (read/write, single string,hidden)This attribute specifies the capacity of a volume being labeled. This attribute is used to pass infor-mation between NetWorker programs, and should not be changed manually by the administrator.Example:operation volume capacity: 10G;

operation volume type (read/write, choice list,hidden)This attribute specifies types of volumes that may be considered when allocating a volume. It isonly used when interacting with an external media management service. This attribute is used topass information between NetWorker programs, and should not be changed manually by theadministrator.Example:operation volume type: 8mm, dlt;

operation ineligible (read/write, hidden)This attribute specifies volumes which are ineligible for the current operation. Only used wheninteracting with an external media management service. This attribute is used to pass informationbetween NetWorker programs, and should not be changed manually by the administrator.Example:operation ineligible: ;



operation task (read/write, choice list, hidden)This attribute designates a secondary task or operation to be performed with the current operation.For example, choosing themount after label task will cause the volume to be mounted after it hasbeen labeled. Currently, this attribute is only used when interacting with an external media man-agement service. This attribute is used to pass information between NetWorker programs, andshould not be changed manually by the administrator.Example:operation task: mount after label;

operation result (read only, hidden)Some jukeboxes support multiple simultaneous operations. This attribute is similar to theopera-tion messageattribute except that this attribute reports error messages for multiple operations. Itmaintains error messages for the last 32 simultaneous operations that failed for this jukebox. Theinstance of the operation that failed is stored with any error message generated by the operation.This attribute is used to pass information between NetWorker programs, and should not bechanged manually by the administrator.

operation instance(read/write, single string, hidden)This attribute designates an instance number to be associated with the operation. The instancemust be unique for all current operations. The value used must be obtained from theoperationnext instanceattribute. This attribute is used to pass information between NetWorker programs,and should not be changed manually by the administrator.Example:operation instance: 3;

operation next instance(read only, single string,hidden)This attribute designates an instance number to be associated with the next simultaneous opera-tion. The value of this attribute should be applied to the next simultaneous operation request forthis jukebox. This attribute is used to pass information between NetWorker programs, and shouldnot be changed manually by the administrator.Example:operation next instance: 3;

operation instances (read only, hidden)This attribute reports the instance number of every simultaneous operation currently executing.This attribute is used to pass information between NetWorker programs, and should not bechanged manually by the administrator.Example:operation instances: 1, 2;

operation hostname (read only, single string, hidden)This attribute designates the name of the machine on which the operation is to be executed. Thisattribute is only used for those jukeboxes which support devices attached to multiple hosts. Thehost machine may be inferred from other attributes for the operation, such asoperation device. Ifa device is specified, the operation will be executed on the host for the device. Otherwise the hostwill be inferred from the name of the jukebox, unless a value is specified for this attribute. Thisattribute is used to pass information between NetWorker programs, and should not be changedmanually by the administrator.Example:operation hostname: host1;

operation dev hostname (read only, single string,hidden)This attribute designates the name of the machine from which a device is to be selected for theoperation. It applies to shared jukeboxes, which can have drives attached to multiple hosts. Thisattribute is used to pass information between NetWorker programs, and should not be changedmanually by the administrator.Example:operation dev hostname: host1;

operation template(read/write, single string, hidden)This attribute shows the template that the label operation will use. The verify operation sets this tothe volume name found on a piece of media. This attribute is used to pass information betweenNetWorker programs, and should not be changed manually by the administrator.



Example:operation template: Default;

operation volume pool (read/write, choice list,hidden)This attribute specifies the default volume pool to use when labeling. This attribute is used to passinformation between NetWorker programs, and should not be changed manually by the adminis-trator.Example:operation volume pool: NonFull;

operation source pool (read/write, choice list,hidden)This attribute specifies the pool from which a volume may be selected when recycling a volume.This attribute is only supported on jukeboxes for volumes being managed by an external mediamanagement package. This attribute is used to pass information between NetWorker programs,and should not be changed manually by the administrator.Example:operation source pool: Default;

operation uses left (read/write, single string,hidden)This attribute sets the number of times a cleaning cartridge may be used. This attribute is used topass information between NetWorker programs, and should not be changed manually by theadministrator.Example:operation uses left: 12;

volumes(read/write, list of strings, hidden)This attribute contains a list of resident volume names. The order corresponds to the slot number.This attribute is used to pass information between NetWorker programs, and should not bechanged manually by the administrator.Example:volumes: mars.001, mars.002, mars.003, mars.004;

volume ids (read/write, list of strings, hidden)Every volume labeled by NetWorker is assigned a volume identifier, often referred to as a volid.This attribute contains a list of volume identifiers for the resident volumes. The volume identifiersstored could be the new long volume ids or the older and shorter volume ids. The type of volumeidentifiers stored depends on whether the storage node on which the device belonging to the juke-box resides on, supports the new long volume id or not. The order corresponds to the slot number.This attribute is used to pass information between NetWorker programs, and should not bechanged manually by the administrator.Example:volumes: 24198, 24199, 24200, 24197;

volume cartridge ids (read/write, list of strings,hidden)Some jukeboxes track volumes that are managed by external media management services. Theremay be multiple volumes on the same media, for example, a volume on each side of an opticaldisk. This attribute is used to track the identifier for each cartridge on which a volume resides.The order corresponds to the slot number. This attribute is used to pass information between Net-Worker programs, and should not be changed manually by the administrator.

loaded volumes(read/write, list of strings, hidden)This attribute contains the names of the volumes currently loaded on the jukebox devices. Theorder is with respect to thedevicesattribute. This attribute is used to pass information betweenNetWorker programs, and should not be changed manually by the administrator.Example:loaded volumes: mars.089, mars.003;

Using the names specified in the previousdevicesattribute,mars.089is loaded in ‘/dev/rmt/0mbn’andmars.003is loaded in ‘/dev/rmt/1mbn’.

loaded bar codes(read/write, list of strings, hidden)This attribute contains the barcodes of the loaded volumes, if the use of barcodes is enabled for thejukebox. The order is with respect to thedevicesattribute. This attribute is used to pass



information between NetWorker programs, and should not be changed manually by the adminis-trator.Example:loaded barcodes: 12345, 67890;

Using the names specified in the previousdevicesattribute, the volume with barcode 12345 isloaded in ‘/dev/rmt/0mbn’ and the volume with barcode 67890 is loaded in ‘/dev/rmt/1mbn’.

loaded slots (read/write, list of numbers, hidden)This attribute contains the slot numbers of the loaded volumes. The order is with respect to thedevicesattribute. This attribute is used to pass information between NetWorker programs, andshould not be changed manually by the administrator.Example:loaded slots: 48, 3;

Using the names specified in the previousdevicesattribute, the volume in slot 48 is loaded in‘/dev/rmt/0mbn’ and the volume in slot 3 is loaded in ‘/dev/rmt/1mbn’.

ev ent tag (read/write, single number, hidden)This attribute contains the tag (unique identifier) of the last notification event sent to thensrd(1m)daemon. The tag is used bynsrjb (1m) to clear the previous event. This attribute is used to passinformation between NetWorker programs, and should not be changed manually by the adminis-trator.Example:ev ent tag: 6319962287;

ev ent message(read/write, single string, hidden)This attribute contains the text of the last notification event sent to thensrd(1m) daemon. Thensrjb (1m) command will send a notification event tonsrd when operator intervention is neededbefore nsrjb can proceed. This attribute is used to pass information between NetWorker pro-grams, and should not be changed manually by the administrator.Example:ev ent message: could not unload device /dev/rmt/1mbn into slot 4;

messages (read/write, list of strings, hidden)This attribute contains a log of messages reflecting previous operationsnsrjb (1m) has done. Gen-erally, an entry is made each timensrjb is invoked and for each mechanical operation. Each entryis time stamped. This attribute is used to pass information between NetWorker programs, andshould not be changed manually by the administrator.Example:messages: 04/01/91 01:15:08 loaded slot 4 into drive a;

minimum space (read/write, single string, hidden)This attribute contains the low water mark for remaining space. When the remaining space on thevolumes contained in the available slots is less than the minimum space, a notification is sent tonsrd. This hidden attribute can be modified by a user.

The minimum space may be specified as a number of gigabytes or megabytes. Either ‘G’ or ‘g’may be used for gigabytes, ‘M’ or ‘m’ for megabytes.Example:minimum space: 7g;

jukebox options (read-only, list of strings, hidden)This attribute contains a list of the options for this jukebox. This option is automatically set afterjukebox creation.Example:jukebox options: two_sided;

auto clean (read/write, yes/no)This attribute specifies whether automatic cleaning of devices in the the jukebox is enabled.Example:auto clean: Yes;

cleaning slots (read/write, list of numbers)This attribute designates a range of slots in the jukebox that has been set aside for cleaning car-tridges. A range may be a single slot number or a pair of slot numbers separated by a dash. If apair of slot numbers is given, the first number of the pair must be less than or equal to the second.Only one range of slots may be set aside for cleaning cartridges. Ifauto clean is set tono, thevalue of cleaning slots is ignored and these slots may contain regular volumes. Whenauto clean



is set toyes, the range of slots specified for this attribute are assumed to contain cleaning car-tridges, and the range of slots specified byav ailable slotsand this attribute must not overlap.

For Silo Tape Libraries this attribute should not be changed directly. This attribute is automaticallyupdated, when adding (nsrjb −U) or removing (nsrjb −x) cleaning cartridges.

Example:cleaning slots: 9-10;

default cleanings(read/write, single number)This attribute designates the number of uses assigned to a new cleaning cartridge during an inven-tory of a jukebox bynsrjb (1m). A cleaning cartridge is considered to be new when a slot set asidefor cleaning cartridges that was empty is discovered to be full during an inventory of a jukebox.Example:default cleanings: 12;

auto media management(read-write)This attribute indicates whether automated media management for the jukebox is enabled. Thevalue can beyesor no. If the value is set toyes, then unlabeled volumes in the jukebox may beautomatically labeled by NetWorker. NetWorker verifies that the volume is unlabeled beforelabeling the volume. A volume is considered to be unlabeled if the volume does not contain alabel that may be read by the device in the jukebox into which the volume is loaded. Note that ifthe volume contains a label, but the label is written at a density that cannot be read by the devicethe volume is considered to be unlabeled. If the volume contains data written by an applicationother than NetWorker, it most likely does not have a label recognizable by NetWorker and the vol-ume is considered to be unlabeled. With this attribute enabled, care should be taken when loadingany volume considered to be unlabeled into the jukebox. The volume may be re-labeled and thedata previously on the volume over-written by NetWorker. For devices in a jukebox the value oftheirauto media managementattribute is alwaysno.Example:auto media management: yes;

STL device names (read/write, list of strings)This attribute lists the corresponding Silo device names of the devices listed in thedevicesattribute. If several device resources are sharing the same physical Silo drive, as indicated by acommonhardware id value, this attribute will only have an entry for each of the physical drives.This attribute is only used for Silo Tape Libraries and is only defined on platforms which providesupport for Silo Tape Libraries.

STL interface lib (read/write, single string)The path name of the dynamically linked interface library. This attribute is only used for Silo TapeLibraries and is only defined on platforms which provide support for Silo Tape Libraries.Example:STL interface lib: /usr/lib/libstl.so.1;

STL device sharing (read/write, single string)This attribute specifies how to handle device sharing. Device sharing means automatic, loaddependent, device switching of devices in a Silo Tape Library between different hosts connected tothe library. This feature can only be used if it is supported by theSTL interface lib. Possible val-ues for this attribute are an empty string (device sharing disabled) or "perm-max", where perm andmax are numbers with perm < max. perm is the number of devices, which can be reserved perma-nently, that is, do not have to be released, when not in use. max is the maximum of devices, whichcan be reserved. This attribute is only used for Silo Tape Libraries and is only defined on plat-forms which provide support for Silo Tape Libraries.Example:STL device sharing: 2-4;

STL barcodes (read-only, list of strings, hidden)The barcodes of the volumes in the library, which are available for NetWorker. This attributemaintains the volume names used by the Silo Tape Libraries for the corresponding volumes in thevolumesattribute. This attribute is only used for Silo Tape Libraries and OpenVault virtual juke-boxes. The attribute is only defined on platforms which provide support for Silo Tape Libraries orOpenVault.



STL device reservation (read-only, list of strings,hidden)This list contains the reservation state of shared devices in a tape library. The possible states are"Yes" (device is reserved), "No" (device is not reserved) and "Error" (an error occurred duringrelease of this device). The order of the reservation state matches the ‘devices’ attribute. Thisattribute is only used for Silo Tape Libraries with device sharing enabled and is only defined onplatforms which provide support for Silo Tape Libraries.

application name(read/write, encrypted, hidden)This attribute is only used for OpenVault jukeboxes. OpenVault requires any application to iden-tify itself when submitting a request. This is the name used by this server to identify itself toOpenVault when submitting a request to access resources listed in this jukebox.

application key (read/write, encrypted, hidden)This attribute is only used for OpenVault jukeboxes. OpenVault requires any application to iden-tify itself when submitting a request. This is the key used by this server to identify itself to Open-Vault when submitting a request to access resources listed in this jukebox.

read hostname (read/write, single string)The hostname that is used in selecting a storage node for recover and read-side clone requests. Forrecover requests, if the required volume is not mounted, and the client’s "storage nodes" attributedoes not match one of the owning hosts in the jukebox, then this attribute is used. For clonerequests, if the required volume is not mounted, then this attribute is used.

jukebox lock (read only, single string, hidden)This attribute synchronizes access to resources in a jukebox that supports multiple simultaneousoperations. This attribute is used to lock the entire jukebox. When locked, this is the only opera-tion that has access to the jukebox. The value is equal to the instance number assigned to thesimultaneous operation holding the lock. This attribute is used to pass information between Net-Worker programs, and should not be changed manually by the administrator.Example:jukebox lock: 10;

device locks (read only, hidden)This attribute synchronizes access to device resources in a jukebox that supports multiple simulta-neous operations. Each value for this attribute is a list of three numbers separated by "-". The firsttwo numbers identify a range of devices locked. Each number identifies a device by the corre-sponding order the devices are listed in thedevicesattribute. The third number is the instancenumber assigned to the simultaneous operation which holds the lock. This attribute is used to passinformation between NetWorker programs, and should not be changed manually by the adminis-trator.Example:device locks: 1-1-10;

volume/slot locks (read only, hidden)This attribute synchronizes access to volume and slot resources in a jukebox which supports multi-ple simultaneous operations. Each value for this attribute is a list of three numbers separated by"-". The first two numbers identify the range of volumes/slots locked. The third number is theinstance number assigned to the simultaneous operation which holds the lock. Holding a lockgrants access to the slot and any volume associated with the slot. This attribute is used to passinformation between NetWorker programs, and should not be changed manually by the adminis-trator.Example:volume/slot locks: 1-5-10;

autodetect id (read/write, hidden)This attribute is for identifying auto-detected devices. It is used by NetWorker programs only, andshould not be changed manually by the administrator.

EXAMPLEA resource defining a jukebox namedHuntingtonis shown. Themodelattribute specifies a ‘Exabyte 210’jukebox. The control port attribute specifies the bus, target, and LUN id for the robotics device



‘[email protected]’. Thedevice attribute lists the pathnames of the two tape devices in the jukebox,‘/dev/rmt/0mbn’ and ‘/dev/rmt/1mbn’. Since the jukebox has a bar code reader, the two bar code yes/noattributes are both set to ‘Yes’. Theavailable slotsattribute lists the slots to consider when automaticallyselecting a volume to load for writing. The available slots are 2 through 11. The hidden attributes are dis-played. auto cleanis yes so automatic cleaning of devices is enabled for this jukebox.cleaning slotsis setto slot 1. This slot is reserved for a cleaning cartridge.

type: NSR jukebox;name: Huntington;

model: EXB-210;physical slots: 1-11;

control port: [email protected];devices: /dev/rmt/0mbn, /dev/rmt/1mbn;

number devices: 2;number drives: 2;

device hardware ids: "", "";idle device timeout: 10;

SmartMedia update interval: 12;write enabled: Yes;

bar code reader: Yes;match bar code labels: Yes;

volume expiration: ;available slots: 2-11;

enabler code: 012345-6789ab-cdef00;operation: ;

operation message: ;operation device: ;

operation slots: ;operation ports: ;

operation options: ;operation barcodes: ;operation response: ;

operation report mode: ;operation label state: ;

operation volume capacity: ;operation volume type: ;

operation ineligible: ;operation task: ;

operation result: ;operation instance: ;

operation next instance: 2;operation hostname: ;

operation dev hostname: ;operation template: ;

operation number uses: ;operation volume pool: ;operation source pool: ;

volumes: -, -, -, -, -, -, -, -, -, -, -;volume ids: "", "", "", "", "", "", "", "", "", "", "";

STL barcodes: ;STL device sharing: ;

STL device reservation: ;STL interface lib: ;

ev ent tag: ;



ev ent message: ;messages: "09/12/97 11:50:56 CREATED";

minimum space: 7g;jukebox options: ;

auto clean: Yes;cleaning slots: 1;

default cleanings: 12;auto media management: Yes;

reset class: initialize unload;application name: ;

application key: ;read hostname: hostname;

jukebox lock: ;device locks: ;

volume/slot locks: ;

SEE ALSOnsr(5), nsr_device(5), nsr_storage_node(5), nsradmin(1m),nsrd(1m),nsrjb (1m),nwadmin(1m).


NSR_LABEL(5) NSR_LABEL(5)

NAMEnsr_label − NetWorker resource type ‘‘NSR label’’

SYNOPSIStype: NSR label

DESCRIPTIONEach NSR label template is described by a single resource of typeNSR label (seensr_resource(5)). Toedit the NSR label resources for a NetWorker server, type:

nsradmin -c "type:NSR label"or use thenwadmin(1m) GUI. See thensradmin(1m) manual page for more information on using theNetWorker administration program.

This resource describes the templates used to generate volume labels.

ATTRIBUTESThe following attributes are defined for resource typensr_label. The information in parentheses describeshow the attribute values are accessed.Read-only indicates that the value cannot be changed by an adminis-trator. Read/write means the value can be set as well as read.Choicemeans that the value of the attributecan only be one from a list specific to that attribute (for example, separator can be ’-’, or ’.’). Several addi-tional attributes (for example, administrator) are common to all resources, and are described innsr_resource(5).

fields (read/write, list of strings)This attribute specifies the constituent fields of alabel template. When generating a volume name,the current value of each field is concatenated. The first field is considered the most significant,the last field the least. If there is aseparator(see below) defined, then it will be placed betweenfields as they are concatenated to form a volume name. The fields are separated by commas.

There are four different types of fields: ‘numeric range’, ‘lower-case range’, ‘upper-case range’,and a ‘list of strings’. A ‘list of strings’ consists of space (‘ ’) separated strings. The other typesare specified as starting and ending values separated by a dash (‘-’) . The starting and ending val-ues of a range must have the same number of characters.

Thenextattribute (see below) contains the current position or value of each field. After a volumename has been assigned to a volume, thenextattribute is incremented. When the ending value isreached, the current value will wrap around to the starting value. A ‘list of strings’ field is incre-mented by selecting the next string in the list. A numeric range field is incremented by adding 1 toits current value. Lower-case and upper-case ranges are incremented by moving on to the next let-ter in the least significant position. In the example below, after aa.99, the next label would beab.00.Example:fields: aa-zz, 00-99;

name (create only, single string, static)This attribute specifies the name of this label template. The label template is referred to by itsname in the jukebox resource, seensr_jukebox(5).Example:name: Default;

next (read/write, single string)This attribute specifies the next volume name to use. After it is assigned to a volume, the next vol-ume name will be generated and remembered here. The attribute consists of a component for eachof the specified fields and the separator.Example:next: aa.00;

Using the separator and field attributes shown above, thenextattribute would show: next: aa.01;This would be followed by: next: aa.02;separator (read/write, single choice, null ok) Thisattribute specifies the character to use to separate the label fields. It may be one of ‘.’, ‘_’, ‘:’, ‘-’or NULL.


NSR_LABEL(5) NSR_LABEL(5)

Example:separator: .;

EXAMPLESA label resource namedengineeringis shown below. (Hidden options are not shown.) There are tworange-type fields defined, the first ranging from ‘aa’ to ‘zz’, the second from ‘00’ to ‘99’. Theseparatorattribute has the value ‘.’ and it will be inserted in between the two fields. Thenextattribute holds the nextname that will be generated by this template. Afteraa.00 is used, the00 will be incremented. The newname will beaa.01. After 98 more names have been generated, thenextattribute will hold the nameaa.99.When this name is incremented, thenextattribute will holdab.00. After generating 67,500 more names,thenextattribute will holdzz.99. This will be followed byaa.00.

type: NSR label;name: engineering;fields: aa-zz, 00-99;

separator: .;next: aa.00;

A label resource namedaccountingis shown below. The field attribute defines five component fields. Theseparatorattribute has the value ‘.’. It will be inserted in between adjacent fields. Thenextattribute holdsthe next name that will be used with this template. After0.23.aa.AA.firstis used, the fifth field will beincremented. The new name will be0.23.aa.AA.second. This will be followed by0.23.aa.AB.first. After1349 more volume names, the name will be0.23.aa.ZZ.second. This will be followed by0.23.ab.AA.first.After using9.45.zz.ZZ.second, the name will wrap around to0.23.aa.AA.first.

type: NSR label;name: accounting;fields: 0-9, 23-45, aa-zz, AA-ZZ, first second;

separator: .;next: 0.23.aa.AA.first;

SEE ALSOnwadmin(1m),nsradmin(1m),nsrjb (1m),nsrmm(1m),nsr(1m),nsr_jukebox(5)


NSR_LAYOUT(5) NSR_LAYOUT(5)

NAMEnsr_layout - NetWorker file layout

SYNOPSIStype: NSR layout

DESCRIPTIONThe NetWorker server filesystem has a directory called/nsr that contains log files, on-line indexes, andconfiguration information. This directory can be created in any filesystem with /nsr set up as a symboliclink to the actual directory (this is determined at installation time). The format of this directory is as fol-lows:

/nsr/logsContains server logging messages. The files in this directory are in ASCII format.

/nsr/resContains the configuration files for various components of the NetWorker server. For example, theserver stores configuration files in/nsr/res/nsr.resand/nsr/res/nsrjb.res.

/nsr/mmContains the media index. Information about the contents of this index file can be printed with thensrls(1m) command. See thensrmm(1m) andmminfo(1m) manual pages on how to view andmanipulate the media index information.

/nsr/indexThis directory contains subdirectories with names that correspond to the NetWorker clients thathave sav ed files. Each index directory contains files that allow the NetWorker server to provide anon-line database of the client’s sav ed files. The most important element is thedb6 directory whichcontains the NetWorker save records and access indexes to those records. The disk space utilizedby the index grows with the number of files saved by the NetWorker service. Administratorsshould plan to use about 200 bytes per saved file instance placed in this index. There are no prac-tical limits on the maximum size of an online index, except that it must reside entirely within asingle file system.

The format of thedb6 directory is subject to change, and is accessible only through an RPC inter-face tonsrindexd(1m). However, thensrls(1m) command can be used to obtain some usefulstatistics from this directory. Thensrck(1m) command is used for checking and rebuilding indexfiles as well as recovering index files from backup media.

The data in the files in thedb6 directory are stored in platform-independent order, so these filesmay be migrated from one NetWorker server to another. Moving the media database from oneNetWorker server to another of unlike architecture is not currently supported.

The files in thedb6 directory include the files listed below. Note that these files are for the inter-nal use of the server and are not to be modified or changed for any purposes.

<savetime>.recThese files contain the index records for each file saved at the savetime, where<savetime>is a hexadecimal representation of the time.

<savetime>.k0These files contain the keys on the<savetime>.recfile based on file name.

<savetime>.k1These files contain the keys on the<savetime>.recfile based on inode. These may bezero length files if the client file index is for a windows client.

<savetime>.sipThis is a save-in-process file and only exists when a save has been started and is not yetcomplete. Once the save is complete, this file is renamed to<savetime>.rec.


NSR_LAYOUT(5) NSR_LAYOUT(5)

v6hdr This file contains a summary of all the<savetime>.recfiles that exist in a client’sdb6directory.

v6journalThis file contains updates to thev6hdr file that are waiting to be merged into thev6hdrfile. Any index operation includes the entries here as well as the ones in thev6hdr.

v6ck.lckThis file is a lock thatnsrck uses to ensure that only onensrck operates on a client’sindex at any giv en time.

v6hdr.lckThis file locks thev6hdr for reading and thev6journal for reading and writing.

v6tmp.ptrThis file refers to the working directory in which conversion and recovery will take place.

tmprecovThis is a working directory in which conversion and recovery take place.

recoveredThis directory contains intermediate results of an index that has been converted or recov-ered. The results here are complete and will be integrated into the file index whennsrckis run against this client file index.

/nsr/coresContains directories that correspond to the NetWorker server daemons and certain executables.Each directory may contain core files from NetWorker server daemons or executables that haveabnormally terminated.

/nsr/driversThis directory may contain any device drivers for use with NetWorker.

/nsr/tmpThis directory contains temporary files used by the Networker system.

The executables for the NetWorker system are usually installed in the/usr/etc or /usr/bin, directoriesthough alternate locations may be chosen when thensr_ize(1m) installation script is run. Alternate loca-tions are not available on SCO systems that install withcustom(ADM). Seepkgadd(1M) for details onalternate executable locations for Solaris 2.x.

When executables for more than one architecture are installed, the non-native architectures are by defaultput in the directory/export/exec/arch/etc, wherearch refers to a given architecture name. A different loca-tion to install non-native executables can be chosen at installation time.

FILES/nsr NetWorker indexes, log files, and configuration information.

/usr/etc, /usr/bin Where NetWorker executables for the native architectures are normally installed.

/export/exec/arch/etc Where NetWorker executables for non−native architectures are normally installed.

/usr/bin, /usr/sbin, /usr/lib/nsrWhere NetWorker executables for Solaris 2.x are normally installed.

SEE ALSOnsrck(1m),nsrindexd(1m),nsrls(1m),nsrmm(1m),mminfo(1m),nsr_ize(1m).


NSR_LICENSE(5) NSR_LICENSE(5)

NAMEnsr_license − NetWorker resource type ‘‘NSR license’’

SYNOPSIStype: NSR license

DESCRIPTIONA resource of typeNSR licenseis used to describe each feature enabled in your NetWorker installation.Seensr_resource(5) for more information on NetWorker resources. To inspect the NSR license resourcestype:

nsradmin −c "type:NSR license"or use thenwadmin(1m) GUI. NSR license resources may be created, enabled and authorized from theGUI, but thensrcap(1m) command must be used to update an existing license resource. Seensrad-min(1m) for more information on using the NetWorker administration program.

ATTRIBUTESThe following attributes are defined for resource typeNSR license. The information in parenthesesdescribes how the attribute values are accessed.Create-only indicates that the value cannot be changed byan administrator, except when the resource is created.Read/write means the value can be changed at anytime by authorized administrators.Read-only means the value cannot be changed by authorized adminis-trators. Static attributes change values rarely, if ever.Dynamic attributes may change often.Hiddenmeans it is an attribute of interest only to programs or experts, and these attributes can only be seen whenthe hidden option is turned on innsradmin(1m) or by selecting the details Menu Item in the View Menufor the Server Registration window innwadmin(1m). For example, an attribute marked(create-only,static) has a value which is set when the attribute is created and never changes. Several additionalattributes (for example, administrator) are common to all resources, and are described innsr_resource(5).

name (create-only, static)This attribute holds the name of the license resource.

enabler code (create-only, static)This code is identical to the code entered into thensrcap(1m) command to enable the featurenamed in this resource. The enabler code consists of 18 hexidecimal digits, printed in groups of 6digits.Example:enabler code: 123456-123456-123456;

host id (read-only, dynamic)The unique host id associated with the computer or licensed operating system on which theenabler has been loaded. This value will often be an 8 digit hexidecimal number; however, otherformats are possible, depending on specific platform requirements.Example:host id: 7260d859;

expiration date (read-only)The date on which this enabler will expire, if the enabler is an evaluation enabler or an otherwiseun-registered license enabler. The enabler expires at 12:00:01 am on the date listed. The specialprefix G means that a grace period has been allowed for this enabler. Enablers with the graceperiod allowed should be registered immediately. If the enabler has been registered, and theauthcodeis filled in with a valid value, the value will be as shown in the example, below.Example:expiration date: Authorized - no expiration date;

auth code (read/write)An 8 digit hexidecimal value used to permanently authorize an enabler. The unique, valid autho-rization code for an enabler is obtained from Legato by registering each purchased license enabler.Evaluation enablers cannot be permanently authorized. If the server’s host id changes, all autho-rization codes will immediately be invalidated, and the enablers must be re-registered with Legatoto obtain new authorization codes.Example:auth code: abcdef00;


NSR_LICENSE(5) NSR_LICENSE(5)

license type (create-only, hidden)A special code, used internally to describe the specific feature or features enabled by this licenseenabler.Example:license type: J16;

checksum (read/write, hidden)A coded checksum used to maintain consistency of a NSR license resource, and between licenseresources.

EXAMPLEBelow is a complete NSR license resource for an authorized base enabler:

type: NSR license;name: NetWorker Advanced/10;

enabler code: 123456-123456-123456;host id: 7260d859;

expiration date: Authorized - no expiration date;auth code: abcdef00;

license type: B10;checksum: xxxxxxxxxxxxxxxxxxxxxx;

FILES/nsr/res/nsr.res− this file should never be edited directly. Usensradmin(1m) instead.

SEE ALSOnsr_resource(5), nsr(1m),nwadmin(1m),nsradmin(1m),nsrcap(1m)


NSR_MIGRATION(5) NSR_MIGRATION(5)

NAMEnsr_migration − NetWorker resource type ‘‘NSR migration’’

SYNOPSIStype: NSR migration

DESCRIPTIONEach NSR migration client is described by a single resource of typeNSR migration (seensr_resource(5)).To edit the NSR migration resources for a NetWorker server type:

nsradmin -c "type:NSR migration"

See thensradmin(1m) manual page for more information on using the NetWorker administration program.The client resource may also be edited using thenwadmin(1m) command.

For each NetWorker migration client, this resource describes which files should be saved, the schedule usedto save these files, which directive should be used to omit files from the save, the group that files will bepre-migrated with, the high-water mark for migration, the low-water mark for migration, the minimum lastaccess time for file migration, the minimum file size for migration, a list of file owners and groups toinclude or exclude for migration, and a list of file name patterns to skip migration of. A client may havemore than one resource describing it.

ATTRIBUTESThe following attributes are defined for resource typeNSR migration. The information in parenthesesdescribes how the attribute values are accessed, and how they are entered in the migration setup.Read-only indicates that the value cannot be changed by an administrator.Read/write means the value can beset as well as read.Hidden means it is an attribute of interest only to programs or experts, and theseattributes can only be seen when the hidden option is turned on innsradmin(1m) or by selecting the detailsMenu Item in the View Menu for a particular window innwadmin(1m). Dynamic attributes have valueswhich change rapidly.Encrypted attributes contain data that is not displayed in its original form. Theassumption is that the data is sensitive in nature and needs to be protected from accidental disclosure.Sin-gle String indicates that only one value can be entered.List indicates that multiple values can be entered.Choice indicates that the value can be selected from a checkbox selection. Several additional attributes (forexample, administrator) are common to all resources, and are described innsr_resource(5).

client (read-only, single string)This attribute identifies the HSM client whose save sets are to be placed under migration control.This name must be a valid name for an existing NSR client resource.Example:client: elantra;

directive (read/write, choice)Directives tell the client how to migrate certain files. The choices are defined by the set of existingdirectives. The default value is NULL. The valid choices for the directive resource are names ofthe currently defined ‘NSR directive’ resources, seensr_directive(5).Example:directive: Unix with compression directives;

enabled (read/write, choice)This attribute determines whether the save set named in this resource should be automaticallymigrated. A resource can be disabled to temporarily keep migration from occuring. On update,any ongoing migration operations will complete. This attribute has no affect on recall operations.Example:enabled: No;

file group (read/write, list)A list of the groups whose files should be migrated. A leading dash (‘-’) in a group name indicatesnegation, in which case all groups except the named group’s files will be migrated.Example:file group: staff, developers;

file owner (read/write, list)A list of the users whose files should be migrated. A leading dash (‘-’) in a user name indicatesnegation, in which case all users except the named user’s files will be migrated.



Example:file owner: pgupta, dbinder;

group (read/write, list)The groups this client/saveset is part of for pre-staging migrated files. The choices are defined bythe set of existing groups.Example:group: Default;

highwater mark % (read/write, single string)The point at which files should start being replaced by stubs, measured as the percentage of avail-able space used on the filesystem. Migration (stub replacement) will continue until the lowerwater mark is reached.Example:high water mark (%): 90;

last access time(read/write, single string)Migrate only those files that have not been accessed in the past specified relative time. If thisvalue is empty, the last access time is not considered.Example values: 5 days ago, 1 month ago, 1 second ago.Example:last access time: 7 days ago

low water mark % (read/write, single string)The point at which files should be stop being replaced by stubs, measured as the percentage ofavailable space used on the filesystem.Example:low water mark (%): 80;

minimum file size (KB) (read/write, single string)Migrate only those files that are larger than the specified size (in Kilobytes). Setting this value tozero causes file size to not be considered.Example:minimum file size (KB): 5;

name (read-only, single string)This attribute identifies the NetWorker client and save set whose migration attributes are stored inthis resource.Example:name: venus All;

fingerprint size (KB) (read/write, single string)The fingerprint size specifies the size (in kilobytes) of the HSM stub file that remains on the clientfilesystem after migration. Read and write operations within the fingerprint area do not cause thefile to be recalled. The default fingerprint size is 32 KB. The minimum value for the fingerprintsize is 0 KB. This attribute is only valid for XDSM HSM on Solaris. XDSM filesystems mayround the fingerprint size to a different value when migrating a file.Example:fingerprint size (KB): 8;

file owner (read/write, list)A list of the users whose files should be migrated. A leading dash (‘-’) in a user name indicatesnegation, in which case all users except the named user’s files will be migrated.Example:file owner: karl, cohrs;

file group (read/write, list)A list of the groups whose files should be migrated. A leading dash (‘-’) in a group name indicatesnegation, in which case all groups except the named group’s files will be migrated.Example:file group: staff, developers;

preserve (read/write, list)A list of regular expressions, in the client shell syntax (for example, /bin/sh syntax on Unix). Afile name which matches any pattern in the list will be preserved and will never be migrated.Example:preserve: *.exe *.dll;

save set (read/write, list)This attribute lists the path names of filesystems or sub-trees within filesystems to put undermigration control for this client. The names should be separated by comma space (, ). The defaultvalue is ‘All’. On Unix clients, ‘All’ refers to all mounted filesystems, except for the filesystems /,



/usr, /var (and /opt on solaris), where migration of files is not allowed.Example:save set: /usr/src, /spare;

statistics (read/write, hidden, list)A list of statistics about recent migration activity for the save set(s) managed using this resource.The first value is the last update time. Subsequent groups of values contain a save set name andstatistics. The statistics are currently a date, K-bytes pre-migrated, files pre-migrated, K-bytesstubbed, files stubbed, K-bytes de-migrated and files de-migrated.

update statistics(read/write, hidden, choice)This controls whether the statistics in this resource should be updated to match the current valueson the client. Selecting "Yes" causes the statistics to be updated, but the attribute value will notactually change.Example:update statistics: No;

EXAMPLESNote: The hidden options are not shown in these examples.

A resource to define an HSM client, called elantra, migrating and recalling files in the /test filesystem froma NetWorker server:

client: elantra;directive: Unix with compression directives ;enabled: Yes;

file group: staff, developers;file owner: pgupta, dbinder;

group: Default;high water mark (%): 90;

last access time:;low water mark (%): 80;

minimum file size (KB): 5;name: "elantra:/test";

preserve: *.exe *.dll;save set: /test;

statistics:;type: NSR migration;

A resource to define an HSM client, called elantra, migrating and recalling files owned by the user "pgupta"in all filesystems except for /, /usr, and /var (and /opt on solaris), to the NetWorker server jupiter:

client: elantra;directive: Unix with compression directives ;enabled: Yes;

file group:;file owner:;

group: Default ;high water mark (%): 90;

last access time:;low water mark (%): 60;

minimum file size (KB): 10;name: "elantra:All";

preserve:;save set: All;

statistics:;type: NSR migration;

SEE ALSOnsr(5), nsr_directive(5), nsr_group(5), savegrp(1m),savefs(1m),nsrpmig(1m),nsrmig(1m),nsrhsmck(1m),nsradmin(1m),nsrexecd(1m),nwadmin(1m)


NSR_NOTIFICATION(5) NSR_NOTIFICATION(5)

NAMEnsr_notification − NetWorker resource type ‘‘NSR notification’’

SYNOPSIStype: NSR notification

DESCRIPTIONA resource of typeNSR notification is used for each combination of an event, priority, and action handledby the NetWorker notification system. A NetWorker notification consists of a single event type, a singlepriority, and a message. The notification system posts each message to the action of eachNSR notificationresource (by executing the command listed in the action, with the message on standard input) that includesthat event type and priority. Seensr_resource(5) for more information on NetWorker resources. To editthe NSR notification resources type:

nsradmin −c "type:NSR notification"or use thenwadmin(1m) GUI. Seensradmin(1m) for more information on using the NetWorker adminis-tration program.

ATTRIBUTESThe following attributes are defined for resource typeNSR notification. The information in parenthesesdescribes how the attribute values are accessed.Create-only indicates that the value cannot be changed byan administrator, except when the resource is created.Read/write means the value can be changed at anytime by authorized administrators.Choice list means that any number of values can be chosen from thegiven list. Single stringmeans that only a single value is allowed.Static attributes change values rarely, ifev er. Hidden means it is an attribute of interest only to programs or experts, and these attributes can onlybe seen when the hidden option is turned on innsradmin(1m) or expert mode (−x) in nwadmin(1m). Forexample, an attribute marked(create-only, static)has a value which is set when the attribute is created andnever changes. Several additional attributes (for example, administrator) are common to all resources, andare described innsr_resource(5).

action (read/write, single string)The value is a command line to be executed when the given event occurs. The command line isrun (seepopen(3s)) with the event information connected to standard input. Typical actions are tolog the message with thesyslog(3) package, or send electronic mail to a system operator.Example:action: /usr/ucb/mail −s "savegroup completion" root;

ev ent (create-only, choice list, hidden)Each value is a class of events that will trigger the given notification. More than one class may beselected. Valid values are:Media for events related to the media multiplexor subsystem,Savegroup for events generated by thesavegrp(1m) command (usually the nightly automaticbackups),Index for events related to the on-line file index subsystem,Registration for eventscaused by changes in the product’s registration status (for example, a license that will soon timeout), andServer for other NetWorker server events, such as restarting.Example:ev ent: Media;

name (create-only, static)This attribute holds the name of the notification resource.

priority (create-only, choice list, hidden)Each value is a priority at which the notification will be triggered. More than one priority may beselected. The valid values in increasing priority order areInfo − supplies information about thecurrent state of the server;Notice − an important piece of information;Warning − informationabout a non-fatal error;Waiting − the server is waiting for an operator to perform a routine task,such as mounting a tape;Critical − the server detected an error condition that should be fixed by aqualified operator;Alert − a sev ere error condition that demands immediate attention;Emergency− a condition that may cause NetWorker to fail unless corrected immediately.Example:priority: Notice;


NSR_NOTIFICATION(5) NSR_NOTIFICATION(5)

EXAMPLEA complete example follows with two resources, one for mail and one using the syslog mechanism:

type: NSR notification;name: savegroup completion;

administrator: root;action: \

/usr/ucb/mail -s \"savegroup completion\" root;ev ent: Savegroup;

priority: Notice;

type: NSR notification;name: log default;

administrator: root;action: /usr/ucb/logger -p daemon.notice -f -;ev ent: Media, Savegroup, Index, Server, Registration;

priority: Info, Notice, Warning, Waiting,Critical, Alert, Emergency;

FILES/nsr/res/nsr.res− this file should never be edited directly. Usensradmin(1m) instead.

SEE ALSOnsr_resource(5), nsr_service(5), nsr_device(5), nsr(1m),nsrmm(1m),syslog.conf(5), syslog(3),nsradmin(1m),nsrmmd(1m),nwadmin(1m)


NSR_POLICY(5) NSR_POLICY(5)

NAMEnsr_policy − NetWorker resource type ‘‘NSR policy’’

SYNOPSIStype: NSR policy

DESCRIPTIONEach NetWorker policy is described by a single resource of typeNSR policy (seensr_resource(5)). Toview theNSR policy resources for a NetWorker server, enter nsradmin at the command prompt to start thensradmin program. At the nsradmin prompt, enter:

nsradmin>print type:NSR policySeensradmin(1m) for more information on using the NetWorker administration program.

These resources control how long entries remain in a client’s on-line file index and when to mark a save setas recyclable. EachNSR client resource (seensr_client(5)) uses two policies, a browse policy and a reten-tion policy. Policies can only be deleted if no clients are using them.

Each policy defines an amount of time. The amount of time is determined by theperiodand thenumber ofperiods.

ATTRIBUTESThe following attributes are defined for resource typeNSR policy. The information in parenthesesdescribes how the attribute values are accessed.Create-only indicates that the value cannot be changed byan administrator once the resource is created.Read/write means the value can be set as well as read at anytime. Several additional hidden attributes (for example, administrator) are common to all resources, and aredescribed innsr_resource(5).

name (create-only)This attribute contains the name of the policy defined by this resource. The name must be uniquefor this NetWorker server, but otherwise can be anything that makes sense to the administrator.This name will appear as a choice attribute of each NSR client resource. TheNSR policyresources named "Quarter" and "Year" may be modified, but may not be removed. The name canonly be specified when the group is created.Example:name: life cycle;

comment (read/write)This attribute is provided for the administrator to keep any explanatory remarks or supplementaryinformation about the policy.

number of periods(read/write)The number of periods attribute specifies the number of base units to use.Example:number of periods: 3;

period (read/write)The period attribute determines the base unit for this policy. It may be one of four values:Days,Weeks, Months or Years. A week is defined as 7 days, a month as 31 days, and a year as 366days.Example:period: Months;

EXAMPLEThe followingNSR policy resource named "Quarter" defines a period of 3 months, or one quarter of a year:

type: NSR policy;name: Quarter;

period: Months;number of periods: 3;

SEE ALSOnsr(1m),nsrim(1m),nsr_resource(5), nsr_client(5), nwadmin(1m),nsradmin(1m).


NSR_POOL(5) NSR_POOL(5)

NAMEnsr_pool − NetWorker resource type ‘‘NSR pool’’

SYNOPSIStype: NSR pool

DESCRIPTIONEach NSR pool is described by a single resource of typeNSR pool(seensr_resource(5)). To edit the NSRpool resources for a NetWorker server type:

nsradmin −c "type:NSR pool"Be careful to include the quotes and the space between ‘‘NSR’’ and ‘‘pool’’. See thensradmin(1m) man-ual page for more information on using the NetWorker administration program.

These resources are used by NetWorker to determine what volumes save sets should reside on dependingupon the characteristics, for example, Group or Level, of the save. Consult yourNetWorker Administrator’sGuidefor more guidelines on using pools.

There are four types of pools.Backuppools accept data fromsavegrpand manual backups.Archivepoolsaccept archive data. Data cloned from a backup pool can be directed to abackup clonepool. Likewise,archive data can be cloned to anarchive clonepool.

There are four pools shipped pre-enabled with NetWorker. TheDefaultpool is meant to collect any backupdata not directed to a pool a user creates with selection criteria. Any archive data not directed to a poolwith selection criteria is collected in theArchivepool. When cloning data, the user must select a destina-tion pool for the operation. TheDefault clonepool is available for users to clone backup data to. TheArchive clonepool is available for users to clone archive data to.

There are also a few pools shipped with NetWorker that are not enabled by default. TheFull andNonFullpools can be used to segregate full level backups from other backups, for example, fulls versus incremen-tals. TheOffsitepool can be used to generate offsite backups, because no index entries are stored for themedia pool and will not be referenced during normal recovers. Note that one can also clone media to pro-duce copies of data to be taken offsite. Save sets that are generated without index entries can still be recov-ered using the ‘‘Save Set Recover’’ feature ofnwadmin(1m) orrecover(1m).

ATTRIBUTESThe following attributes are defined for resource typeNSR pool. The information in parentheses describeshow the attribute values are accessed.Create-only indicates that the value cannot be changed after theresource has been created.Read/write means the value can be updated by authorized administrators.Yes/nomeans only a yes or no choice is possible.Choice indicates that the value can only be selected froma giv en list. Hidden means it is an attribute of interest only to programs or experts, and these attributes canonly be seen when the hidden option is turned on innsradmin(1m) or if theDetails Viewoption is selectedin theMedia Poolswindow innwadmin(1m).

archive only (read/write, yes/no, hidden, create)If yes is selected, only archive sav es are allowed to this pool.This hidden attribute can be modi-fied by a user.Example:archive only: no;

auto media verify (read/write, yes/no, choice)If set toyes, NetWorker verifies data written to volumes from this pool. Data is verified by re-positioning the volume to read a portion of the data previously written to the media and comparingthe data read to the original data written. If the data read matches the data written, verificationsucceeds; otherwise it fails. Media is verified whenever a volume becomes full while saving and itis necessary to continue onto another volume, or when a volume goes idle because all save setsbeing written to the volume are complete. When a volume fails verification, it is marked full soNetWorker will not select the volume for future saves. The volume remains full until it is recycledor a user marks it not full. If a volume fails verification while attempting to switch volumes, allsave sets writing to the volume are terminated.Example:auto media verify: yes;



clients (read/write, choice)Whatclients(nsr_client(5)) are allowed in this pool. If agroup is specified, only clients that aremembers of that group are allowed to be listed.Example:clients: mars;

devices (read/write, choice)This attribute lists theONLY devices that volumes from this pool are allowed to be mounted onto.If no devices are listed, volumes from this pool may be mounted on any device.Example:devices: /dev/nrst8;

groups (read/write, choice)Whatgroups(nsr_group(5)) are allowed in this pool.Example:groups: Accounting;

label template (read/write, choice)Determine whatlabel template(nsr_label(5)) is referenced when generating volume names forthis pool.Example:label template: Accounting;

levels (read/write, choice)What levels(nsr_schedule(5)) are allowed in this pool.Example:levels: full;

name (create-only)The names of pool resources are used when labeling volumes and when determining what volumesa sav e set should reside on. The name can be chosen at the administrator’s convenience, but itmust be unique for this NetWorker server. The pool resources namedDefault, Default Clone,Archive, andArchive Clone cannot be modified or deleted. The pool resource namedFull andNonFull cannot be deleted. Other pools can only be deleted if no volumes still reference them.Example:name: Accounting;

recycle from other pools (read/write, yes/no, choice)This attribute determines whether or not a given pool can recycle volumes from other pools whenit exhausts all its write-able and recyclable volumes.Example:recycle from other pools: yes;

recycle to other pools(read/write, yes/no, choice)This attribute determines whether or not a given pool allows other pools to recycle its recyclablevolume for their use.Example:recycle to other pools: yes;

save sets (read/write, choice)What save sets (nsr_client(5)) are allowed in this pool. Save sets can be matched using the regu-lar expression matching algorithm described innsr_regexp(5)).Example:save sets: /, /usr, C:\windows\system, *.JPG ;

status (read/write, hidden, choice)If set toenabled, this pool is considered for determining what pools a save set should be saved towhen performing backup volume selection. If set toclone, this pool is considered only as the des-tination of cloning operations. If set todisabled, this pool is completely ignored.This hiddenattribute can be modified by a user.Example:status: enabled;

store index entries (read/write, yes/no, choice)If set to yes, entries are made into the file indexes for the backups. Otherwise, only mediadatabase entries for the save sets are created.Example:store index entries: yes;

volume type preference(read/write, choice)This attribute is used as a selection factor when a request is made for a write-able volume. Thepreferred type will be considered first within a priority level such asjukeboxor stand alone device



.Example:volume type preference: 4mm;

EXAMPLEA complete NSR pool resource, named ‘Default’, follows:

type: NSR pool;archive only: No;

clients: ;devices: ;groups: ;

label template: Default;levels: ;name: Default;

save sets: ;status: Enabled;

store index entries: Yes;auto media verify: Yes;

recycle from other pools: Yes;recycle from other pools: Yes;volume type preference: 4mm;

SEE ALSOnsr(5), nsr_label(5), nsr_resource(5), nsradmin(1m), nwrecover(1m), recover(1m), savegroup(1m),savefs(1m),uasm(1m)


NSR_REGEXP(5) NSR_REGEXP(5)

NAMEnsr_regexp − regular expression syntax

DESCRIPTIONThis manual page describes the regular expression handling used in NetWorker. The regular expressionsrecognized are described below. This description is essentially the same as that fored(1).

A regular expression specifies a set of strings of characters. A member of this set of strings is said to bematched by the regular expression.

Form Description

1. Any character except a special character matches itself. Special characters are the regular expres-sion delimiter plus a backslash(\), brace([), or period(.) and sometimes a carat(ˆ), asterik(* ), ordollar symbol($), depending upon the rules below.

2. A . matches any character.

3. A \ followed by any character except a digit or a parenthesis matches that character.

4. A nonempty strings, bracketed string[s] (or [ˆs]) matches any character in (or not in)s. In s, \ hasno special meaning and] may only appear as the first letter. A substringa-b, with aandb inascending ASCII order, stands for the inclusive range of ASCII characters.

5. A regular expression of form 1 through 4 followed by* matches a sequence of 0 or more matchesof the regular expression.

6. A bracketed regular expressionx of form 1 through 8,$x$, matches whatx matches.

7. A \ followed by a digitn matches a copy of the string that the bracketed regular expression begin-ning with thenth $x$ matched.

8. A regular expressionx of form 1 through 8 followed by a regular expressiony of form 1 through 7matches a match forx followed by a match fory, with thex match being as long as possible whilestill permitting ay match.

9. A regular expression of form 1 through 8 preceded byˆ (or followed by$), is constrained tomatches that begin at the left (or end at the right) end of a line.

10. A regular expression of form 1 through 9 picks out the longest among the leftmost matches in aline.

11. An empty regular expression stands for a copy of the last regular expression encountered.

SEE ALSOed(1), nsr_client(5).


NSR_RESOURCE(5) NSR_RESOURCE(5)

NAMEnsr_resource − NetWorker resource format

SYNOPSISresource::= attribute list <blank line>attribute list::= attribute[ ; attribute]*attribute::= name[ : value[ , value]* ]name, value::= <printable string>

DESCRIPTIONThe NetWorker system uses files containingresourcesto describe itself and its clients. Each resource rep-resents a component of the NetWorker system that might need administration. Devices, schedules, andclients are examples of NetWorker resources. The system administrator manipulates resources to controlthe NetWorker system. The file and the resources in them are accessible through thenwadmin(1m) andnsradmin(1m) programs. They can also be viewed with a normal text editor.

The files all share a common format. The same format is used by thensradmin(1m) program. Eachresource is described by a list of attributes, and ends in a blank line. Each attribute in the attribute list has aname and an optional list of values. The attribute name is separated from the attribute values by a colon (:),attribute values are separated by commas (,), and each attribute ends in a semicolon (;). A comma, semi-colon or back-slash (\) at the end of a line continues the line. A line beginning with a pound-sign (#) is acomment and the rest of the line is ignored. The back-slash character can also be used to escape the specialmeaning of other characters (comma, semicolon, pound-sign, and back-slash).

The attribute name and values can contain any printable character. Upper and lower case is not distin-guished on comparisons, and extra white space is removed from both ends but not from inside of names andvalues. For example,

Name: this is a test;matches

name : This Is A Test ;but is different than

Name: this is a test;

In the following example resource, there are eight attributes. They aretype, name, server, schedule,directive, group, save set, andremote access. Theremote accessattribute has no value.


server: earth;schedule: Default;directive: Unix standard directives;

group: Default;save set: All;

remote access: ;

In the following resource, there are six attributes. Theadministrator attribute has three values:&engi-neering, root, andoperator. Note that the three values are separated by commas. Theaction attribute hasone value:incr incr incr incr incr full incr . Note that this is a single value − it just happens to havespaces separating its words.

type: NSR schedule;action: incr incr incr incr incr full incr;

administrator: &engineering, root, operator;name: engineering servers;

override: ;period: Week;



SPECIAL ATTRIBUTESEach NetWorker resource includes seven special attributes:type, name, administrator , hostname, ONCprogram number, ONC version number, andONC transport. The type andname attributes are nor-mally visible, but the others attributes are hidden. That an attribute is hidden indicates that it is infrequentlyused and perhaps esoteric. Hidden attributes should usually not be changed by the user.

The type attribute defines which other attributes a resource can contain. For example, a resource with typeNSR clientwill always include the attributeserver, while a resource of typeNSR scheduledoes not.

The name attribute is a descriptive name of the object that a resource represents. In the first exampleabove, thenameattribute is the name of the NetWorker client machine. In the second example, thenameattribute describes a schedule used to back up the the servers in the engineering department.

Theadministrator attribute is the list of users that have permission to modify or delete this resource. Thisattribute is inherited from thetype: NSR resource when a new resource is created. The administrator of theNSR resource also controls who has permission to create and delete NetWorker resources.

The hostname attribute specifies the hostname of the machine on which the service that controls thisresource is running. It is used internally and cannot be changed by the administrator.

The remaining attributes (ONC program number, ONC version number, andONC transport) specifythe Open Network Computing information for this service. They should never be changed manually.

In some cases, the resource identifier will be visible. Although it may look like an attribute, it is an internalvalue that is set and used by the NetWorker system to provide unique identification of each resource. Whennew resources are created in theedit command ofnsradmin(1m), the resource identifier attribute should beleft off. This signals that this is a new resource and a new identifier will be assigned.

NetWorker resources are implemented by the Legato Resource Administration Platform, which is describedin theresource(5) manual page. This flexible architecture means that in future releases of NetWorker, moreresource types or attributes may be added, and the administration tools in this release will automatically beable to use them. To make this possible, each server providestype descriptorsthat are used internally todescribe the attributes of each type, between the administration tools and the services. These type descrip-tors may cause limitation on the values, such as only allowing a single value, allowing no value, or onlynumeric values.

RESOURCE TYPESThis release of NetWorker defines the following types of resources:

NSR This resource describes a NetWorker server. It contains attributes that control administrator autho-rization, information about operations in progress, and statistics and error information about pastoperations. For more information see thensr_service(5) manual page.

NSR clientThis resource describes a NetWorker client. It includes attributes that specify the files to save,which schedule to use, and which group this client belongs to. There may be more than one clientresource for a NetWorker client. This allows a client to save files on different schedules. For moreinformation see thensr_client(5) manual page.

NSR deviceThis resource type describes a storage device. It includes attributes that specify a particular devicename (for example, /dev/nrst1), media type (for example, 8mm), and the name of the currentlymounted volume. It also provides status and statistics on current and past operations. For moreinformation see thensr_device(5) manual page.

NSR directiveThis resource describes a directive. Directives control how a client’s files are processed as they arebeing saved. For more information see thensr_directive(5), nsr(5) anduasm(1m) manual pages.

NSR groupThis resource specifies a logical grouping of NetWorker clients and a starting time. Each day, atthe specified time, all members of the group will start their saves. For more information see the



nsr_group(5) manual page.

NSR jukeboxThis resource type describes a jukebox. It includes attributes such as the jukebox model, the firstand last slot numbers in the jukebox, and the names of the devices within the jukebox. For moreinformation see thensr_jukebox(5) manual page.

NSR labelThis resource type specifies a template describing a sequence of names to be used when labelingvolumes. For more information see thensr_label(5) manual page.

NSR licenseThis resource contains licensing information for each feature currently enabled in this NetWorkerinstallation. It contains various enabler and authorization codes that are used by NetWorker to val-idate licensed capabilities. For more information see thensr_license(5) andnsrcap(1m) manualpages.

NSR notificationA notification specifies an action to be performed when a particular type of NetWorker event takesplace. For more information see thensr_notification(5) manual page.

NSR policyPolicy resources are used as part of the index management process in NetWorker. These policiescontrol how long entries remain in a client’s on-line file index and when to mark a save set as recy-clable. For more information see thensr_policy(5) manual page.

NSR poolThis resource type is used by NetWorker to determine what volumes save sets should reside onbased on the characteristics of the save (for example, group or level). For more information seethensr_pool(5) manual page.

NSR scheduleSchedule resources define a sequence of save lev els and an override list. The override list is madeup of pairs of levels and dates. The level controls the amount of data saved when a client isbacked up. For more information see thensr_schedule(5) manual page.

NSR stageEach stage resource describes a staging policy. The resource includes attributes that define controlparameters for the policy, and devices managed by the policy. For more information see thensr_stage(5) manual page.

FILES/nsr/res/nsr.res Holds the NetWorker server’s resources.

SEE ALSOresource(5), nsr(5), nsr_client(5), nsr_device(5), nsr_directive(5), nsr_group(5), nsr_jukebox(5),nsr_label(5), nsr_license(5), nsrcap(1m),nsr_notification(5), nsr_policy(5), nsr_pool(5),nsr_schedule(5), nsr_service(5), nsr_stage(5), nsr(1m),nwadmin(1m),savegroup(1m),savefs(1m),nsradmin(1m),uasm(1m).


NSR_SCHEDULE(5) NSR_SCHEDULE(5)

NAMEnsr_schedule − NetWorker resource type "NSR schedule"

SYNOPSIStype: NSR schedule

DESCRIPTIONEach NetWorker schedule is described by a single resource of typeNSR schedule(seensr_resource(5)).To edit theNSR scheduleresources for a NetWorker server, type:

nsradmin −c "type:NSR schedule"or use thenwadmin(1m) GUI. Seensradmin(1m) for more information on using the NetWorker adminis-tration program.

This resource describes a sequence of levels controlling the amount of data saved by NetWorker clients (seensr_client(5)). There is oneNSR scheduleresource for each NetWorker schedule.

ATTRIBUTESThe following attributes are defined for resource typeNSR schedule. The information in parenthesesdescribes how the attribute values are accessed.Read-only indicates that the value cannot be changed byan administrator.Read/write means the value can be set as well as read. Several additional hiddenattributes (e.g., administrator) are common to all resources, and are described innsr_resource(5).

name (read/write)This attribute specifies the schedule’s name. The schedule is referred to by its name in clientresources.Example:name: monthly_fulls;

period (read-only)This attribute specifies the length of the schedule’s period. It may be either "Week" or "Month"."Week" schedules repeat every 7 days and start on Sunday. "Month" schedules start over at thefirst of each month. The default is "Week."Example:period: Month;

action (read/write)This attribute specifies the sequence of save lev els making up the schedule. One entry is used foreach day of the schedule. The entries must be separated by whitespace, i.e., blanks or tabs. Thevalid levels are "consolidate," "full," "incr," "skip," and the numbers 1 through 9. The actions con-solidate, full, incr, and skip may be abbreviated "c," "f," "i," and "s," respectively.

When the action attribute does not contain enough entries to account for every day in the period,NetWorker will repeat the list of actions when the end of the action list is reached.Example:action: f i i i i i i;

override (read/write)This attribute specifies a list of actions and dates overriding the actions specified in theactionattribute. The format of the override specification isaction date. action must be one of "full,""incr," "skip," or one of the numbers 1 through 9.datemust be either afixed dateor recurringdate. Fixed dateis of the form "month/day/year." Month and day are 2 digit numbers, year maybe either 2 or 4 digits. If the year is 2 digits, numbers in the range 70-99 are assumed to be offsetsfrom 1900, those in the range 00-69 are assumed to be offset from 2000.Recurring dateis of theform ‘[ number] weekdayev ery [number] period’. numbercan be a number (1, 2, 3, etc.) or anordinal (first, second, third, etc.), and it is optional.weekdaymust be one of "monday," "tuesday,""wednesday," "thursday," "friday," "saturday," "sunday."periodmust be one of "week," "month,""quarter," or "year." Action/date pairs are separated by commas (‘,’).Example:override: full 1/1/1994, full first friday every 2 week;

EXAMPLEThe following defines aNSR scheduleresource named "Default." The Default schedule may be modified,but it may not be deleted. Each NetWorker server must have a Default schedule. This schedule has aperiod of one week, does a full save on Sunday, followed by 6 incremental saves. There are no override


NSR_SCHEDULE(5) NSR_SCHEDULE(5)

actions specified.

type: NSR schedule;name: Default;

period: Week;action: f i i i i i i;

override:;

The following defines a schedule named "quarterly." It has a period of one month. The action attributespecifies level 5, 9, and incremental saves. In the override attribute, full saves are specified for the first dayof each quarter. Note that there are only 7 entries in the action attribute. Upon reaching the end of the list,NetWorker will start over at the beginning of the list, performing a level 5 sav e.

type: NSR schedule;name: quarterly;

period: Month;action: 5 incr incr incr 9 incr incr;

override: f 1/1/1994, f 3/1/1994, f 6/1/1994, f 9/1/1994, f 1/1/1995;

SEE ALSOnsr(1m),savefs(1m),mminfo(1m),nsradmin(1m),nsr_client(5), nsr_policy(5), nsr_resource(5), nwad-min(1m).


NSR_SERVICE(5) NSR_SERVICE(5)

NAMEnsr_service − NetWorker server resource type ‘‘NSR’’

SYNOPSIStype: NSR

DESCRIPTIONEach NetWorker server is described by a resource of typeNSR. Seensr_resource(5) for general informa-tion on NetWorker resources. To edit theNSR resource use the command:

nsradmin −c "type:NSR"or use thenwadmin(1m) GUI. Seensradmin(1m) for information on using the NetWorker administrationprogram.

ATTRIBUTESThe following attributes are defined for theNSR resource. The information in parentheses describes howthe attribute values are accessed.Read-only indicates that the value cannot be changed by an administra-tor. Read/write means the value can be set and read.Choice list means that any number of values can bechosen from the given list.Static attributes change values rarely.Dynamic attributes have values whichcan change rapidly.Hidden means it is an attribute of interest only to programs or experts, and theseattributes can only be seen when the hidden option is turned on innsradmin(1m) or details option is set innwadmin(1m). For example, an attribute marked(read-only, static) has a value which is set when theresource is created and never changes, or is changed only by the server.

name (read-only, static)This attribute specifies the hostname of this NetWorker server.Examplename: mars;

version (read-only, dynamic)This is the software version of the NetWorker server daemon,nsrd(1m). This includes a slash andthe number of clients currently licensed.Example:version: NetWorker 4.1 Turbo/110;

save totals (read-only, dynamic, hidden)Save statistics. A string containing the total number of save sessions, the number of saves witherrors (if any) and the total number of bytes saved (if any). This attribute is updated after eachsave session completes.Example:save totals: "37 sessions, 457 MB total";

recover totals (read-only, dynamic, hidden)Recovery statistics. A string containing the total number of recover sessions, the number of recov-ers with errors (if any) and the total number of bytes recovered (if any). This attribute is updatedafter each recover session completes.Example:recover totals: "347 sessions, 48 MB total";

totals since (read-only, dynamic)The time statistics collection started. This is usually the last time the NetWorker server wasrebooted.Example:totals since: "Fri Jun 1 09:35:02 1992";

NSR operation (read-only, choice list, hidden)This attribute is currently unused and is provided for backward compatibility.

parallelism (read/write, static)This attribute sets the number of concurrent save sessions that this server will allow. The valuecan be set by an administrator. Use higher values for better performance on a fast system with lotsof main memory and swap space. Use lower values to avoid overloading a slow system, or systemswith little main memory and/or swap space.Warning : due to defects in some versions of UNIX,high values of parallelism may cause the system to lock up.Example:parallelism: 4;



session statistics (read-only, dynamic, hidden)This attribute reports the statistics of each active session. There are 14 values for each set ofstatistics, namely,id (session’s unique identifier),name (session’s name),mode (read, write,browse),pool (current pool),volume (current volume),rate kb (current data transfer rate for savesession),amount kb (current amount read/written by session),total kb (total amount to be readby session),amount files (current number of files recovered; to be implemented in a futurerelease),total files (current number of files to recover; to be implemented in a future release),con-nect time (time session has been connected),num volumes (number of volumes to be used byrecover session),used volumes(number of volumes processed by recover session), andcomple-tion (running, complete, or continued)Example:sessions statistics: ;

manual saves (read/write, hidden)This attribute allows the administrator to disable manual backups to the server. Scheduled back-ups continue to work normally.

volume priority (read/write)If a NetWorker server has volumes in locally managed jukeboxes and volumes being managed bySmartMedia, this attribute allows the administrator to assign a priority for volume selection whensaving data. Determines whether the server has a preference for volumes being managed by Smart-Media, SmartMedia Priority, or whether the server has a preference for volumes in a locallymanaged jukebox,NearLine Priority. The default value isNearLine Priority.Example:volume priority: NearLine Priority;

SmartMedia save mount (read/write)This attribute controls the form of the request made to SmartMedia to mount a volume for savingdata. Setting this attributes value tovolume by characteristicscauses NetWorker to request avolume meeting specified criteria, and lets SmartMedia select an appropriate volume from allmedia which satisfy the criteria specified. When this attribute’s value is set to,volume by name ,NetWorker will request the volume by name and SmartMedia mounts the volume requested. Thedefault value isvolume by characteristics .Example:SmartMedia save mount: volume by characteristics;

license server (read/write, hidden)The name of the server on which a Legato license manager is installed and running. This attributeused to be called "GEMS server". You can set the value to a GEMStation where a GEMS licensemanager is running or to a machine where a Legato license manager is running.Example:licenseserver: jupiter;

message (read-only, dynamic, hidden)The last message of any kind logged. A time stamp is included at the start of the string.Example:message: "Mon 12:25:51 Tape full, mount volume mars.001 on /dev/nrst1";

message list (read-only, dynamic, hidden)A list of recent messages, with a time stamp and a string message for each value.Example:message: "Mon 12:25:51 Tape full, mount volume mars.001 on /dev/nrst1";

server message (read-only, dynamic, hidden)Lists recent concise general messages about the status of the server.Example:message: "Tape full, mount volume mars.001 on /dev/nrst1";

sequence number (read-only, dynamic, hidden)The sequence number of the corrosponding server message.

server message time (read-only, dynamic, hidden)The time at which the server message was generated.

server message priority(read-only, dynamic, hidden)The priority of the server message. Currently these are for Networker internal use.



server message source(read-only, dynamic, hidden)This attribute gives the networker component that generated the message on the server. This can bea client, a device etc.

server message category(read-only, dynamic, hidden)This attribute is the category the server message belongs to. Currently these are for Networkerinternal use.

session (read-only, dynamic, hidden)The value of this attribute is a list of session information strings. Each string includes the Net-Worker client name, type of operation (saving, browsing, or recovering) and information about thesave set, including name, number of bytes, and number of files. All sizes and rates are in bytes persecond, Kilobytes (1024), Megabytes (a thousand Kilobytes), etc.Example:session: "venus:/usr saving to mars.001 20MB",

"mars:/usr/src done saving 24MB";"

session message (read-only, dynamic, hidden)Concise session message string of the session attribute described above.Example:session: "venus:/usr saving to mars.001",

"mars:/usr/src done saving";"

session client name(read-only, dynamic, hidden)The name of each client for which each session is active.

session rate (read-only, dynamic, hidden)The rate of data transfer for the active session.

session rate label (read-only, dynamic, hidden)Unit of data transfer for the active sessons.

session device name (read-only, dynamic, hidden)The name of the device on which the session is active.

pending (read-only, dynamic, hidden)A list of events pending with the NetWorker event notification system (seensr_notification(5)).The first three fields are the time, priority, and event name.Example:pending: "Fri 14:40:15 alert: media mount of mars.001 suggested on /dev/nrst1";

status (read-only, dynamic, hidden)A list of status flags for the NetWorker server. These flags are only for use by NetWorker server-side programs (for example,savegrp) and list various features enabled in the running server. Theformat is currentlyname=boolean(true or false). The listed features and their states can change atany time.

statistics (read-only, dynamic, hidden)A list of strings of the formname=numberthat give a number of server statistics.

types created (read-only, static)A list of all the other resource types this NetWorker server can create and about which clients canquery.Example:types created: NSR device, NSR group;

administrator (read/write, static)This is a list of names (or netgroups) of users who are allowed to administer NetWorker. Onlyusers in this list can see or modify this attribute. Normally this list is inherited by all otherresources on this server. Administrators can change the values of attributes for the resource whichlists them. The administrator list also determines who can add and delete resources of the otherNSR types. The user "root" on the local host of the server is always an administrator. Entriesspecifying other administrators are of the form:



user, user@host, user@domain, group@host, group@domain, host/user & netgroup, oruser_attribute=value[, ...]

whereuser is a user name;host is a host name;group is a user group name;domainis a domainname;user_attributecan beuser, group, host, domain, or domaintype (type of the domain,NISor WINDOMAIN )

valuecan be any string delimited by white space. If the value has space in it, then it can be quotedwith double quotes. The value may contain wild cards, "*". Entering just a user name allows thatuser to administer NetWorker from any host (equivalent touser@* or */useror user=user). Net-group names are always preceded by an "&".

The following example grants NetWorker administrative privileges to, "root" from any host, theuser "operator" from the hosts "jupiter" and "mars", the user "admin" from any host, and all <username, user’s hostname, server’s domain> in the netgroup "netadmins".Example:administrator: root, operator@jupiter, mar/operator, admin@*, &netadmins;

The following example grants NetWorker administrative privileges to the user "root" on host"pluto", users in group "Backup Operators" from windows domain "Accounting". and user "joe" inNIS domain "YP.fubar.COM".Example: administrator: "user=root,host=pluto", "group=\"Backup Operators\",domain=Accounting, domaintype=windomain", "user=joe, domain=YP.fubar.COM, domain-type=NIS";

contact name (read/write, static)This attribute is used for product licensing/registration purposes. It must be specified before print-ing the registration information from the registration window.Example:contact name: contact_name;

company (read/write, static)This attribute is used for product licensing/registration purposes. Your company name must bespecified before printing the registration information from the registration window.Example:company: Legato;

street address (read/write, static)This attribute is used for product licensing/registration mailing purposes. Specify your mailingstreet address.Example:street address: 3145, Porter Drive;

city/town (read/write, static)This attribute is used for product licensing/registration mailing purposes.Example:city/town: Palo Alto;

state/province (read/write, static)This attribute is used for product licensing/registration mailing purposes.Example:state/province: CA;

zip/postal code (read/write, static)This attribute is used for product licensing/registration mailing purposes.Example:zip/postal code: 94304;

country (read/write, static)This attribute is used for product licensing/registration mailing purposes.Example:country: USA;

phone (read/write, static)This attribute is used for product licensing/registration purposes. This attribute must to be speci-fied before printing the registration information from the registration window.Example:phone: 650-812-6100;



fax (read/write, static)This attribute is used for product licensing/registration purposes.Example:fax: 650-812-6031;

email address (read/write, static)This attribute is used for product licensing/registration purposes.Example:email address: [email protected];

server OS type (read/write, static)This attribute is used for product licensing/registration purposes.Example:server OS type: Solaris;

purchase date (read/write, static)This attribute is used for product licensing/registration purposes. It specifies the purchase date ofthe product enabler code. This attribute must be specified before printing the registration informa-tion from the registration window.

product serial number (read/write, static)This attribute is used for product licensing/registration purposes. It must be specified before print-ing the registration information from the registration window.

mm op message(read/write, dynamic, hidden)This attribute lists the descriptive message for the most recently completed media database opera-tion. The NetWorker program (such asnsrmm(1m)) that requested the operation clears thisattribute as soon as it has read the result. An administrator should never change this attribute man-ually.

mm operation value (read/write, dynamic, hidden)This attribute is used by programs such asnsrmm(1m) to pass the desired media database opera-tion location or flags to the NetWorker server. The value is automatically cleared when the opera-tion completes. An administrator should never change this attribute manually.

mm operation (read/write, choice list, dynamic,hidden)This attribute is used by programs such asnsrmm(1m) to pass the media database operation typecurrently desired to the NetWorker server. The possible choices are: purge volume, purge save set,delete volume, delete save set, mark volume, mark save set, unmark volume, unmark save set,specify volume location, specify volume flags, and specify save set flags. The server serializessuch operations and performs the appropriate queries onnsrmmdbd(1m). The value is automati-cally cleared when the operation completes. An administrator should never change this attributemanually.

mm operation id (read/write, dynamic, hidden)This attribute is used by programs such asnsrmm(1m) to pass the desired media database opera-tion identifier to the NetWorker server. The value is automatically cleared when the operationcompletes. An administrator should never change this attribute manually.

nsrmon info (read/write, dynamic, hidden)This attribute is used by programs such asnsrmon(1m) to pass information about remote daemonrequests to the NetWorker server. The value is automatically cleared when the request completes.An administrator should never change this attribute manually. Seensr_storage_node(5) for adescription of storage nodes and remote daemons.

nsrmmd count (read-only, dynamic, hidden)This attribute is used by programs such asnsrd(1m) to track the number and location of the mediadaemons,nsrmmd(1m).

nsrmmd polling interval (read/write, hidden)This attribute specifies the number of minutes between polling events of a remotensrmmd(1m).nsrd(1m) polls a remotensrmmd(1m) at this interval, to determine whether it is running. If itdetermines from this poll that the daemon is no longer running, it will restart thensrmmd(1m),



with a delay set by the ‘nsrmmd restart interval’; see thensrmmd restart interval description. Seensr_storage_node(5) for additional details on this attribute and storage nodes.

nsrmmd restart interval (read/write, hidden)This attribute specifies the number of minutes between restart attempts of a remotensrmmd(1m).Whennsrd(1m) determines that a remotensrmmd(1m) has terminated, it periodically attempts torestart the remote daemon. A value of zero for this attribute means the daemon should be restartedimmediately. Seensr_storage_node(5) for additional details on this attribute and storage nodes.

nsrmmd control timeout (read/write, hidden)This attribute specifies the number of minutesnsrd(1m) waits for storage node requests.

enabler code (read/write, dynamic, hidden)This attribute specifies the enabler code for the base enabler of the server software.

SS cutoff size (read/write, hidden)This attribute sets the default "save set cut off size" to be used when saving. A blank value usesthe built in default value. A non blank value for this attribute consists of a number followed byKB, MB, or GB signifying Kilobytes, Megabytes, or Gigabytes. Note that this field only affectsclients older than version 6.0. Continuation save sets have been eliminated for version 6.0 andabove.

EXAMPLEA complete example follows:

type: NSR;name: mars;

version: NetWorker 4.1 Turbo/110;save totals: "84 sessions, 3597 MB total";

recover totals: 1 session;totals since: "Fri Oct 14 12:41:31 1994";

NSR operation: Idle Write Read Verify label Label ;parallelism: 4;

manual saves: Enabled Disabled ;message: \

"Mon 14:37:25 media alert event: recover waiting for 8mm tape mars.001";message list: \

"Mon 07:10:12 media info: loading volume man.001 into /dev/nrst11""Mon 07:10:33 /dev/nrst11 mount operation in progress""Mon 07:11:15 /dev/nrst11 mounted 8mm 5GB tape man.001

session: "mars:george browsing","mars:/home/mars starting recovery of 9K bytes";

session statistics: ;pending: \

"Mon 14:40:15 media alert: recover waiting for 8mm tape mars.001";status: disabled=false, jukebox=true, dm=true,

archive=true, cds=true, turbo=true,single=false;

statistics: elapsed = 257415, saves = 1176, recovers = 12,save KB = 12050007, recover KB = 28272839,bad saves = 0, bad recovers = 0,current saves = 1, current recovers = 0,max saves = 12, max recovers = 1, mounts = 0,recover delays = 0, saving daemons = 0,recovering daemons = 0, idle daemons = 0;

types created: NSR device, NSR group, NSR directive,NSR notification, NSR client, NSR policy,



NSR schedule, NSR pool, NSR label, NSR jukebox,NSR license, NSR archive client,NSR archive list;

administrator: root;contact name: Technical Support;

company: Legato;street address: 3145, Porter Drive;

city/town: Palo Alto;state/province: CA;

zip/postal code: 94304;country: USA;

phone: 650-812-6100;fax: 650-812-6220;

email address: [email protected];purchase date: ;

product serial number: ;mm op message: ;

mm operation value: ;mm operation: ;

mm operation id: ;nsrmon info: ;

nsrmmd count: "mars:2";nsrmmd polling interval: 3;nsrmmd restart interval: 2;

nsrmmd control timeout: 5;enabler code: ;

SS cutoff size: ;

FILES/nsr/res/nsr.res− this file should never be edited directly. Usenwadmin(1m) ornsradmin(1m) instead.

SEE ALSOnetgroup(5), nsr(5), nsr(1m),nsr_device(5), nsr_group(5), nsr_notification(5), nsr_resource(5),nsr_storage_node(5), nsradmin(1m),nsrd(1m),nsrmm(1m),nsrmmdbd(1m),nsrmon(1m),nwad-min(1m), recover(1m),save(1m).


NSR_SHUTDOWN(1m) NSR_SHUTDOWN(1m)

NAMEnsr_shutdown − stop a NetWorker server’s processes

SYNOPSISnsr_shutdown[ −a ] [ −A ] [ −d ] [ −n ] [ −q ] [ −s ] [ −v ]

DESCRIPTIONnsr_shutdown terminates NetWorker processes on a NetWorker server. This command is simpler than theprocedure of usingps(1), grep(1), andkill (1).

OPTIONS−a Terminates all daemons; this is the same as using the−A, −d, and−soptions.

−A Terminates anynsralist(1m) processes.

−d This is the default option; it terminates the server daemons. These may includensrd(1m),ansrd(1m), nsrindexd(1m), nsrexecd(1m), nsrib(1m), nsrmmd(1m), and nsrmmdbd(1m).Since savegrp(1m), nsrexec(1m), andnsralist(1m) processes depend on the service daemons,they are also terminated.

−n Echoes thekill (1) command without actually invoking it.

−q Perform the shutdown quietly; don’t prompt for confirmation.

−s Terminates anysavegrp(1m) (andnsrexec(1m)) processes.

−v Verbose: Instruct the shell to print commands and their arguments as they are executed.

SEE ALSOansrd(1m), nsralist(1m), nsrd(1m), nsrexec(1m), nsrexecd(1m), nsrindexd(1m), nsrmmd(1m), nsrm-mdbd(1m),savegrp(1m).


NSR_STAGE(5) NSR_STAGE(5)

NAMEnsr_stage − NetWorker resource type ‘‘NSR stage’’

SYNOPSIStype: NSR stage

DESCRIPTIONEach staging policy used by a NetWorker server is described by a single resource of typeNSR stage. Seensr_resource(5) for information on NetWorker resources. To edit theNSR stageresources run:

nsradmin -c "type:NSR stage"Be careful to include the space between ‘‘NSR’’ and ‘‘stage’’ and the surrounding quotes. Seensrad-min(1m) for information on using the NetWorker administration program.

ATTRIBUTESThe following attributes are defined for resource typeNSR stage. The information in parentheses describeshow the attribute values are accessed.Read-only indicates that the value cannot be changed by an adminis-trator. Read/write means the value can be set as well as read.Hidden means it is an attribute of interestonly to programs or experts, and these attributes can only be seen when the hidden option is turned on innsradmin(1m) or if theDetails Viewoption is selected in theStagewindow in nwadmin(1m). Staticattributes change values rarely, if ever.Dynamic attributes have values which change rapidly. For exam-ple, an attribute marked(read-only, static) has a value which is set when the attribute is created and maynever change. Additional attributes (for example, administrator) are common to all resources, and aredescribed innsr_resource(5).

name (read-only, single string)The name attribute specifies the staging policy name.

enabled (read/write, choice)The enabled attribute determines whether or not save sets are automatically staged from devicesassociated with this policy. It also enables and disables the periodic recover space operations. Itmay be one of two values:Yes, No

max storage period (read/write)Specifies the maximum number ofdaysfor a save set in a given volume before it isstagedto a dif-ferent volume.

high water mark % (read/write)The point at which save sets should bestaged,measured as the percentage of available space usedon the file system. Staging will continue until the lower mark is reached.Example:high water mark (%): 90;

low water mark % (read/write)The point at which the staging process shouldstop,measured as the percentage of available spaceused on the file system.Example:low water mark (%): 80;

Save set selection (read/write)Save set selection criteria for staging. It may be one of four values:largest save setsmallest save setoldest save setoryoungest save set.

Destination pool (read/write)The pool to which save sets should be sent (seensr_pool(5)).

Devices (read/write, multiple choice)This attribute lists thefile typedevices associated with this policy.


NSR_STAGE(5) NSR_STAGE(5)

Recover space interval (read/write, hidden)The number ofhoursbetween recover space operations for save sets with no entries in the mediadatabase form file devices.

Fs check interval (read/write, hidden)The number ofhoursbetween file system check operations.

Start now (read/write)Updating this aribute will cause the selected operation to be triggered immediately on all devicesassociated with this policy. The attribute value will not actually change. Operation can be one ofthe following:Check fs- check file system and stage data if neccessary.Recover space- recover space for save sets with no entries in the media database.Stage all save sets- stage all save sets to the destination pool.

EXAMPLES

Note: the hidden options are not shown in the first example.

The following example shows a resource that defines a stage policy called ‘test stage1’. Save sets will bestaged from device ‘/disk/fd0’ to pool ‘Default Clone’ when the file system is 90% full or 7 days after thedate of the backup, whichever comes first. The largest save set will be the first to stage to the destinationpool:

type: NSR stage;name: test stage1;

autostart: Enabled;max storage period: 7;high water mark (%): 90;low water mark (%): 85;save set selection: largest save set;destination pool: Default Clone;

devices: /disk/fd0;start now: ;

The following example shows a resource that defines a stage policy called ‘test stage2’. Save sets will bestaged from device ‘/disk/fd2’ to pool ‘Default’ when the file system is 95% full or 14 days after the date ofthe backup, whichever comes first. The smallest save set will be the first to stage to the destination pool.The file system will be checked every 1 hour and a staging operation will be triggered if necessary. Arecover-space operation (for save sets with no entries in the media database) will be triggered every 3 hourson all devices associated with the policy:

type: NSR stage;name: test stage2;

autostart: Enabled;max storage period: 14;high water mark (%): 95;low water mark (%): 80;save set selection: smallest save set;destination pool: Default;

devices: /disk/fd2;recover space interval: 3;

fs check interval: 1;start now: ;

administrator: root@omni;hostname: omni;

SEE ALSOnsrstage(1m),nsrclone(1m),nsradmin(1m),nwadmin(1m).


NSR_STORAGE_NODE(5) NSR_STORAGE_NODE(5)

NAMEnsr_storage_node − description of the storage node feature

SYNOPSISThe storage nodefeature provides central server control of distributed devices for saving and recoveringclient data.

DESCRIPTIONA storage nodeis a host that has directly attached devices that are used and controlled by a NetWorkerserver. These devices are calledremote devices,because they are remote from the server. Clients maysave and recover to theseremote devicesby altering their "storage nodes" attribute (seensr_client(5)). Astorage nodemay also be a client of the server, and may save to its own devices.

The main advantages provided by this feature are central control of remote devices, reduction of networktraffic, use of faster local saves and recovers on a storage node, and support of heterogeneous server andstorage node architectures.

There are several attributes which affect this function. Within the NSR resource (seensr_service(5)) thereare the "nsrmmd polling interval", "nsrmmd restart interval" and "nsrmmd control timeout" attributes.These attributes control how often the remote media daemons (seensrmmd(1m)) are polled, how longbetween restart attempts, and how long to wait for remote requests to complete.

Within the "NSR device" resource (seensr_device(5)) the resource’s name will accept the"rd=hostname:dev_path" format when defining aremote device.The "hostname" is the hostname of thestorage node and "dev_path" is the device path of the device attached to that host. There are also hiddenattributes called "save mount timeout" and "save lockout," which allow a pending save mount request totimeout, and a storage node to be locked out for upcoming save requests.

Within the "NSR client" resource (seensr_client(5)), there are "storage nodes" and "clone storage nodes"attributes. The "storage nodes" attribute is used by the server in selecting a storage node when the client issaving data. The "clone storage nodes" attribute is used during cloning, to direct cloned data from a volumeon the storage node (the node represented by this client resource).

The "NSR jukebox" resource (seensr_jukebox(5)), contains the "read hostname" attribute. When all of ajukebox’s devices are not attached to the same host, this attribute specifies the hostname that is used inselecting a storage node for recover and read-side clone requests. For recover requests, if the required vol-ume is not mounted, and the client’s "storage nodes" attribute does not match one of the owning hosts in thejukebox, then this attribute is used. For clone requests, if the required volume is not mounted, then thisattribute is used.

INSTALL AND CONFIGUREIn order to install a storage node, choose the client and storage node packages, where given the choice. Forthose platforms that do not have a choice, the storage node binaries are included in the client package. Inaddition, install any appropriate device driver packages. If not running in evaluation mode, a storage nodeenabler must be configured on the server for each node.

As with a client, ensure that thensrexecd(1m) daemon is started on the storage node. To define a device ona storage node, from the controlling server define a device with the above mentioned "rd=" syntax. For aremote jukebox (on a storage node), runjbconfig(1m) from the node, after adding root@storage_node tothe server’s administrator list, (where root is the user runningjbconfig(1m) and storage_node is the host-name of the storage node). This administrator list entry may be removed afterjbconfig(1m) completes.

In addition tojbconfig(1m), when runningscanner(1m) on a storage node, root@storage_node must be onthe adminstrator list.

When a device is defined (or enabled) on a storage node, the server will attempt to start a media daemon(seensrmmd(1m)) on the node. In order for the server to know whether the node is alive, it polls the nodeev ery "nsrmmd polling interval" minutes. When the server detects a problem with the node’s daemon orthe node itself, it attempts to restart the daemon every "nsrmmd restart interval" minutes, until either thedaemon is restarted or the device is disabled (by setting "enabled" to "no" in the device’s "enabled"attribute).

UNRELEASED October 15, 2001 1

NSR_STORAGE_NODE(5) NSR_STORAGE_NODE(5)

In addition to needing a storage node enabler for each storage node, each jukebox will need its own jukeboxenabler.

OPERATIONA storage node is assignable for work when it is considered functional by the server -nsrexecd(1m) run-ning, device enabled,nsrmmd(1m) running, and the node is responding to the server’s polls. When aclient save starts, the client’s "storage nodes" attribute is used to select a storage node. This attribute is alist of storage node hostnames, which are considered in order, for assignment to the request.

The exception to this node assignment approach is when the server’s index or bootstrap is being saved -these save sets are always directed to the server’s local devices, regardless of the server’s "storage nodes"attribute. Hence, the server will always need a local device to backup such data, at a minimum. These savesets can later be cloned to a storage node, as can any sav e set.

If a storage node is created first (by defining a device on the host), and a client resource for that host is thenadded, that hostname is added to its "storage nodes" attribute. This addition means the client will back upto its own devices. However, if a client resource already exists, and a device is later defined on that host,then the client’s hostname must be added manually to the client’s "storage nodes" attribute. This attribute isan ordered list of hostnames; add the client’s own name as the first entry.

The volume’s location field is used to determine the host location of an unmounted volume. The serverlooks for a device or jukebox name in this field, as would be added when a volume resides in a jukebox.Volumes in a jukebox are considered to be located on the host to which the jukebox is connected. The loca-tion field can be used to bind a stand-alone volume to a particular node by manually setting this field to anydevice on that node (using the "rd=" syntax). For jukeboxes which do not have all of their devices attachedto the same host, see the previous description of the "read hostname" attribute.

There are several commands that interact directly with a device, and so must run on a storage node. Theseinclude jbconfig(1m), nsrjb (1m) andscanner(1m), in addition to those in the device driver package.Invoke these commands directly on the storage node rather than on the server, and use the server option ("-sserver_host", where server_host is the controlling server’s hostname).

CLONING FUNCTIONA single clone request may be divided into multiple sub-requests, one for each different source machine(the host from which save sets will be read). For example, suppose a clone request must read data fromvolumeA and volumeB, which are located on storage nodes A and B, respectively. Such a request would bedivided into two sub-requests, one to read volumeA from storage node A and another to read volumeB fromstorage node B.

A clone request involves two sides, the source that reads data and the target that writes data. These twosides may be on the same host or on different hosts, depending on the configuration. The source host isdetermined first and then the target host. If the volume is mounted, the source host is determined by its cur-rent mount location. If the volume is not mounted at the time of the clone request and it resides in a juke-box, then the source host is determined by the value of the jukebox’s "read hostname" attribute.

Once the source host is known, the target host is determined by examining the "clone storage nodes"attribute of the client resource of the source host. If this attribute has no value, the "clone storage nodes"attribute of the server’s client resource is consulted. If this attribute has no value, the "storage nodes"attribute of the server’s client resource is used.

LIMITATIONSA server cannot be a storage node of another server.

SEE ALSOjbconfig(1m), mmlocate(1m), nsr_client(5), nsr_device(5), nsr_jukebox(5), nsr_service(5), nsr-clone(1m),nsrexecd(1m),nsrjb (1m),nsrmmd(1m),nsrmon(1m),scanner(1m).


NSRADMIN(1m) NSRADMIN(1m)

NAMEnsradmin − NetWorker administrative program

SYNOPSISnsradmin [ −c ] [ −i file ] [ −sserver] [ −p prognum] [ −v version] [ query]

nsradmin [ −c ] [ −i file ] [ −d resdir . . . ] [ −t typefile] [ query]

nsradmin [ −c ] [ −i file ] [ −f resfile. . . ] [ −t typefile] [ query]

DESCRIPTIONThensradmin command is a command-line based administrative program for the NetWorker system. Nor-mally nsradmin monitors and modifies NetWorker resources over the network. Commands are entered onstandard input, and output is produced on standard output.

If nsradmin is started without a query it uses a default query that selects all resources involved in Net-Worker products.

OPTIONS−c Uses thetermcap(5) andcurses(3) packages to implement a full-screen display mode, just like

thevisual command described below. (UNIX Only)

−d resdirUses the NetWorker resource databaseresdir instead of opening a network connection. Thedatabaseresdir must be in directory format. This should be used sparingly, and only when theNetWorker server is not running. Multiple−d andresdir arguments can be used to startnsradminwith access to more than one database at a time.

−f resfileSimilar to the−d resdir option except that it opens a resource file, rather than a resource directory.Some configuration databases are stored in file format, others in directory format.

−i file Takes input commands fromfile instead of from standard input. In this mode, the interactiveprompt will not be printed.

−sserverOpens a connection to the named NetWorker server instead of allowing administration of allservers. Useful to limit the number of resources if there are many servers, or to administer whentheRAP location service is not working.

−p programUses the given RPC program number instead of the standard program number. The standard num-ber is 390109. This option is generally used only for debugging.

−t typefileUses the alternate filetypefileto define RAP types.

−v versionBinds to the NetWorker RAP service with the given version number. The default is 2. This optionis generally used only for debugging.

query If a query is specified (in the form of an attribute list), the edit operation is performed on theresults of the query. SeeCOMMANDS for more information on how the edit command works.

RESOURCESEach NetWorker resource is made up of a list of named attributes. Each attribute can have zero or morevalues. The attribute names and values are all represented by printable strings. Upper and lower case is notdistinguished on comparisons, and spaces are ignored except inside the names and values.

The format for specifying attributes and attribute lists is:

attribute::= name[ : value[ , value]* ]An attribute is a name optionally followed by a colon, followed by zero or more values, with val-ues separated by commas. A comma at the end of a line continues the line.



attribute list::= attribute[ ; attribute]*An attribute list is one or more attributes separated by semicolons. A semicolon at the end of aline continues the line. The list is ended by a newline that is not preceded by a comma or semi-colon.

Here is an example of an attribute list:

name: mars;type: NSR client;remote access: mars, venus, jupiter;

For more information on attributes, attribute lists and the NetWorker resource types, see theresource(5),nsr_resource(5), andrap(1m) manual pages.

COMMANDSAt each input prompt,nsradmin expects a command name and some optional arguments. Command namescan be shortened to the smallest unique string (for example, p for print). Command arguments are alwaysspecified in the form of an attribute list. Most commands operate on a set of resources returned by aquery.The query is specified as an attribute list which is used to match resources with the following rules:

1) The resource must match all the given attributes.

2) If more than one value is specified the resource can match any one of the values.

3) Values in a query may be in the form ofed(1) style regular expressions. A pattern match isattempted against all resources that contain the specified attribute.

4) If an attribute is specified with no value the resource must contain an attribute of that name.

Thus, a query:type:NSR device;name:mars, venus;test

will match all resources that have atype attribute with the valueNSR deviceand aname attribute with avalue of eithermars or venus, and an attributetestwith any value.

If the query has only one name and no values (for example, if there is no semi-colon or colon in it), then theprogram tries to guess a more reasonable query. If the name is a host name, then the query will select all theresources on the given host. Otherwise, the name will be interpreted as a type name, and all resources ofthat given type will be selected.

bind [query]Bind to the service that owns the resource described byquery. If no query is specified, queries aresent to theRAP Resource Directory, and update, create, and delete commands to the service thatowns the resource being changed. On failure, the previous service will continue to be used.

createattribute listCreate a resource with the given attributes. One of the attributes must betype to specify a Net-Worker type that can be created. Thetypes command can be used to find out which NetWorkertypes a server supports.

delete[query]Delete the resources that match the current query. If aquery is specified, it becomes the currentquery.

edit [query]Edit the resources that match the current query. If aquery is specified, it becomes the currentquery. If the environment variableEDITOR is set, then that editor will be invoked, otherwisevi(1)will be started. When the editor exits,nsradmin applies update, delete and create operations basedon the changes to the resources. Be careful to not edit the resource identifier attribute, and to writethe file out before exiting the editor. (UNIX Only)



help [command]Print a message describing a command. If no command name is given a synopsis of all of the com-mands is printed.

option [list]This command enables some options to change the display of resources. With no arguments it dis-plays the current options; with a list of options it turns the specified ones on. The options are:Dynamic displays all dynamic attributes, even the normally hidden ones.Hidden displays allattributes, even the normally hidden ones.Resource IDdisplays the resource identifier on eachresource, a number that is used internally to provide sequencing and uniqueness.

print [query]Print the resources that match the current query. If aquery is specified, it becomes the currentquery. If a name has been specified for the the current show list, only the attributes for the speci-fied name in the show list will be displayed.

quitExitsnsradmin.

server [servername]Bind to the given NetWorker server name. If no server is specified, the RAP location service willbe used. On failure, the previous server will continue to be used.

show[name; ...]If a name list (really an attribute list with no values) is specified, add those names to the show list.Only these attributes will be displayed in subsequentprint commands. If no name list is given theshow list is cleared, resulting in all attributes being shown.

types Print a list of all known types.

unset[list]This command turns off the specified option.

updateattributesUpdate the resources given by the current query to matchattributes.

visual [query]Enter a full-screen mode using thecurses(3) package to step through commands in a perhaps moreuser-friendly manner than the command line interface. You can get this mode directly using the−c command line argument. (UNIX Only)

. [query]If a query is specified, this command will set the current query without printing the results of thequery. Otherwise, it will display the current query, show list, server binding, and options.

? [command]Same as thehelp command above.

EXAMPLESprint type:NSR device

Print all resources of typeNSR deviceand make this the current query.

show type; nameSet the show list to display only the attributestype andname.

deleteDelete all resources that match the current query.

delete type:NSR device; hostname: marsDelete the resource with attributes:type: NSR deviceandhostname: mars.

edit type:NSR notificationEdit all resources of typeNSR notification.



SEE ALSOed(1), vi(1), curses(3), nsr_resource(5), termcap(5), nsr(1m) rap(1m).

DIAGNOSTICSThe following exit status values are meaningful:

0 Interactive mode exited normally.

1 There was a usage or other non-query related error.

2 When reading input from a file (−i file), one or more RAP operations failed. This status is neverreturned interactively.


NSRALIST(1m) NSRALIST(1m)

NAMEnsralist − NetWorker archive request executor

SYNOPSISnsralist −R archive request name

DESCRIPTIONThensralist command is used to execute an archive request (seensr_archive_request(5)). Thensralistcommand is run automatically bynsrd(1m), as specified by each archive request resource.

Thensralist command will set up an RPC connection tonsrexecd(1m) to runnsrarchive(1m) on the spec-ified client. Ifnsrexecdis unavailable,nsralist will fall back on using thercmd(3) protocol and the client-sidershd(1m).

Thensralist monitors the execution of the archive command and stores any output in the log of the archiverequest. Thensrarchive command running on the client updates the server with its progress, includingwhether or not optional verification and cloning operations have completed successfully. Seensrclone(1m)for more information on cloning.

OPTIONS−R archive request name

This option specifies which archive request is supposed to be run.

FILES/nsr/tmp/al.request_name

A lock file to keep multiple runs of the same archive list from running simultaneously.SEE ALSO

nsrarchive(1m),nsrclone(1m),nsrexecd(1m),nsr_archive_request(5).


NSRARCHIVE(1m) NSRARCHIVE(1m)

NAMEnsrarchive − archive files to long term storage with NetWorker

SYNOPSISnsrarchive [ −BinpqvxVy ] [ −b pool ] [ −C clone pool] [ −f directive filename] [ −G remove] [ −N name] [ −R name] [ −sserver] [ −T annotation] [ −W width ] [ path . . . ]

DESCRIPTIONnsrarchive archives files, including directories or entire filesystems, to the NetWorker server (seensr(1m)).The progress of an archive can be monitored using the X Window System basednwadmin(1m) program orthe curses(3X) basednsrwatch(1m) program for other terminal types. Use ofnsrarchive is restricted tousers on the administrator and archive users lists.

If no path arguments are specified, the current directory is archived.nsrarchive archives a directory byarchiving all the files and subdirectories it contains, but it does not cross mount points or follow symboliclinks.

The directive files (seensr(5)) encountered in each directory is read by default, The files contain specialinstructions directing how particular files are to be archived (that is, compressed, skipped, etc.). These filesare named ’.nsr’.

Each file in the subdirectory structures specified by thepath arguments is encapsulated in a NetWorkerarchive stream. This stream of data is sent to a receiving process (seensrd(1m)) on the NetWorker server.Entries are added to the media database for the archive sav e set. The data eventually resides on a long termstorage medium (seensrmmd(1m)).

Details about handling media are discussed innsrmm(1m) andnsr_device(5).

If the grooming optionremoveis requested, all files and directories archived are removed. If verification isrequested, the files will not be removed if the verification failed. Likewise, the files will not be removed if arequested cloning operation fails. The user is prompted for confirmation before the files are removedunless the−y option is supplied.

If the user does not supply a−T option on the command line, the user is prompted to enter an annotationfor the archive.

OPTIONS−b pool

Specify a destination pool for the archive sav e set. This option overrides the automatic pool selec-tion which normally occurs on the server.

−B Force archive of all connecting directory information from root (‘‘/’’) down to the point of invoca-tion. The connecting directory information is always archived even without this option if clientfile index is generated.

−C clone poolGenerate a clone of this archive sav e set to the specifiedclone pool.

−E Estimate the amount of data which will be generated by the archive, then perform the actualarchive. The estimate is generated from the inode information, and thus the data is only read once.

−f filenameThe file from which to read default directives (seensr(5)). A filenameof - causes the defaultdirectives to be read from standard input.

−i Ignore directive files as they are encountered in the subdirectory structures being archived.

−G removeGroom the files after they hav e been archived. If cloning or verification is requested, no groomingis performed until those operations have completed successfully. The user is prompted forremoval of top-level directories unless they option is supplied.nsrarchive creates a temporaryfile which contains a list of all files and directories to be removed. The temporary file is placed in/tmpunless the environment variableTMPDIR is set.


NSRARCHIVE(1m) NSRARCHIVE(1m)

−n No archive. Estimate the amount of data which will be generated by the archive, but do not per-form the actual archive.

−N nameThe symbolic name of this archive sav e set. By default, the firstpath argument is used as thename.

−v Verbose. Cause thensrarchive program to tell you in great detail what it is doing as it proceeds.

−p Exit with status 0. Used by server to determine if client installed properly.

−q Quiet. Display only summary information and error messages.

−R nameThis option should only be used by thensralist program, which handles executing archiverequests. Updates to the named archive request resource occur when this option is specified.

−sserverSpecify which machine to use as the NetWorker server.

−T annotationArchive sav e sets can be annotated with arbitrary text. This option specifies an annotation for thearchive sav e set being generated.

−V Verify the archive sav e set after it completes.

−W widthThe width used when formatting summary information output.

−x Cross mount points.

−y Answer yes to any questions.

SEE ALSOcurses(3X), nsr_getdate(3), nsr(5), nsr(1m),nsr_service(5), nsr_device(5), nsrmm(1m),nsrmmd(1m),nsrd(1m),nsrwatch(1m),nsrretrieve(1m),nwadmin(1m).


0 Normal exit.Non-zero

Abnormal exit.


NSRCAP(1m) NSRCAP(1m)

NAMEnsrcap− update the capabilities of a NetWorker installation

SYNOPSISnsrcap [ −vn ] { −c | −d | −u } enabler-code

DESCRIPTIONThe nsrcap program is primarily used to enable new features on NetWorker. You can also usensrcap toupgrade or downgrade NetWorker software features that are currently being used. (Upgrades and down-grades should be performed carefully. Read the options descriptions below). Enablers are separate fromthe NetWorker software, and are specified by an 18-digit code, usually displayed as 3 groups of 6 digitseach. To enable a new feature, thensrd(1m) program must be running on the system where the NetWorkerserver software is installed. To enable a new feature you must be logged in to the NetWorker server asadministrator or root. Thensrcap program is run once for each feature you want to enable by specifyingthe 18-digit code each time. If no errors occur, no output is printed. You can inspect the enablers currentlyloaded by viewing the NSR license resources usingnwadmin(1m) ornsradmin(1m).

OPTIONS−c Causesnsrcap to enable a feature that is not currently installed, using the specified enabler code.

You can only load a feature once; an error is returned if you attempt to load the enabler more thanonce. You can only specify one of the−c, −d, or −u options.

−d Causesnsrcap to downgrade an existing Base or Jukebox enabler. After you downgrade theenabler, you cannot return to the previous level enabled on the system.Do notuse the-u optionunless instructed to do so by Legato Technical Support. You must specify one of the−c, −d, or −uoptions.

−u Causesnsrcap to enter an enabler that upgrades an existing Base or Jukebox enabler. After youupgrade the enabler, you cannot return to the previous level enabled on the system.Do notuse the-u option unless instructed to do so by Legato Technical Support.

−v Causesnsrcap to display more verbose information, describing the enabler being loaded. Youmust specify one of the−c, −d, or −u options.

−n No load. Causesnsrcap to inspect the enabler code for validity. When you specify the-n option,the enabler code you enter on the command line is inspected and verified, but is not entered intothe NetWorker server’snsr_licenseresource.

SEE ALSOnwadmin(1m), jbconfig(1m),nsradmin(1m),nsrd(1m).

DIAGNOSTICSenabler-code is too long

Enabler codes must be 18 digits in length. The code entered is longer than 18 digits, and is invalid.

invalid code: xxxxxx-xxxxxx-xxxxxxThe 18-digit code entered on the command line is invalid. Re-check the enabler-code on yourenabler sheet.

cannot find a jukebox resource to enableThe code word entered is a jukebox license enabler, but there are no jukebox resources to enable.You need to runjbconfig(1m) to complete the jukebox installation before runningnsrcap.

found a jukebox, but it had more than N slots.Jukebox enablers can only enable jukeboxes with at mostN physical slots, whereN is the type ofjukebox enabler. Either the jukebox was installed incorrectly, or you need to obtain a larger juke-box enabler.

this enabler-codeis already assignedThe enabler-code entered is already loaded onto the system and cannot be used again for anupgrade.


NSRCAP(1m) NSRCAP(1m)

no appropriate jukeboxes to upgradeAn upgrade was attempted, but no jukebox resources were found. Only use the−u option forjukeboxes when upgrading from one jukebox level to another, not on the initial installation. Youalso need to runjbconfig(1m) before runningnsrcap.

this enabler-codepreviously loadedThe enabler-code entered has been loaded onto the system previously and cannot be used again.You need to purchase a new enabler-code for the upgrade.

don’t know how to upgrade this enablerdon’t know how to downgrade this enabler

The enabler-code entered is not for a base or jukebox enabler. These are the only types of enablersthat you can currently upgrade or downgrade.

base enabler must be loaded before upgradingbase enabler must be loaded before downgrading

You cannot perform an upgrade or downgrade until a base product has been installed. Install abase enabler, and then perform the upgrade or downgrade.

cannot find the enabler to upgradeA jukebox upgrade was attempted, but the license enabler for the jukebox is not currently loaded.You must use the−c option for the initial installation of a jukebox enabler, not the−u option.

RPC error: Program not registeredThensrcap program requires that the NetWorker daemons be running. Start your NetWorker dae-mons (cd /; nsrd) and re-run thensrcap program. Ifnsrd is already running, you have probablyreached a resource limit on the server (for example, not enough memory, or no more processes).

RAP error: user login nameneeds to be of thetype:NSR administrator list.Your login name is not listed in the administrator’s list for the server. You need to be a validadministrator to runnsrcap.

RAP error: ...Various other errors can be returned fromnsrd if the enabler is invalid. For example, if you try toload a base enabler onto a system that already has a base enabler loaded, or if you attempt to loada jukebox enabler before the jukebox has been completely installed. The specific problem will fol-low theRAP error prefix.


NSRCAT(1m) NSRCAT(1m)

NAMEnsrcat − NetWorker notification redirector for tty devices

SYNOPSISnsrcat [-n]

DESCRIPTIONThe nsrcat command appends a carriage return to all newlines. This allows NetWorker notification mes-sages can be redirected to the/dev/consoleor /dev/ttydirectory on systems with tty drivers that do notappend a carriage return to output lines. This command reads text messages from standard input, appends acarriage return to the newline character, and writes the message to standard out.

OPTIONS−n Indicates that the codeset is to be converted from UTF-8 to the user’s native character encoding.

EXAMPLEStype: NSR notification;name: Log default;

action: nsrcat > /dev/console;

SEE ALSOconsole(4), tty (4), nsr_notification(5), nsr(5).


NSRCK(1m) NSRCK(1m)

NAMEnsrck − NetWorker index consistency check, repair, and recovery program

SYNOPSISnsrck [ −qMv ] | [ −R [ −Y ] ] [ −L check-level[ −t date] | −X [ −x percent] | −C | −F | −m | −c ] [ −Ttempdir] [ clientname. . . ]

DESCRIPTIONnsrck is used to check the consistency of the NetWorker online index of clients’ save records. Normally,nsrck is started automatically and synchronously by thensrindexd(1m) program whennsrindexd starts.You can modify thensrck modes to allow normal users to runnsrck and retain root privileges (seensr(1m)for more details).

Whennsrindexd starts up, it determines whether any further checking of a client’s index is necessary. Thisphase checks certain internal state of the index database, and if that state is consistent, avoids furtherpasses. This phase also reports any suspicious-looking index names (that is indexes whose names cannotbe mapped into network addresses). These online file indexes are then checked more rigorously.

nsrck detects whether any client indexes need to be converted and does the proper conversion. Convertingthe indices takes free space on the volume that contains the indices; if there is not sufficient free space, youmay use the−T tempdir flag to specify a different directory which the conversion will use as its workspace. You may also manually convert client indices by issuing thensrck commmand manually.

There are seven different checking levels supported bynsrck. If client names are supplied, the check isperformed on the given client names. If no names are given, the checks are performed for all clientindexes. The check levels work as follows for each client checked:

Level 1 validates the online file index header, merging a journal of changes with the existing header.

Level 2 does a level 1 check and checks the online file index for new and cancelled saves. New sav es areadded to the online file index, and cancelled saves are removed.

Level 3 does a level 2 check and reconciles the online file index with the online media index. Records thathave no corresponding media save sets are discarded.

Level 4 does a level 3 check and checks the validity of the online file index’s internal key files. If any ofthese key files are invalid, they are rebuilt.

Level 5 does a level 4 check and verifies the digest of individual save times against their key files.

Level 6 does a level 5 check and extracts each record from each save time, verifying that each record can beextracted from the database. The digest of each save time is re-computed and compared against the storeddigest, and the internal key files are rebuilt.

Level 7 recovers an online file index from backup media, rebuilds the internal key files, and rebuilds theindex header. The−t date option may be used to recover the index as of a specific time. Note that recover-ing the index to a specific time adds the entire contents of the index as of that time to the current index con-tents. This option allows browsing of save sets that have passed their browse policy and are still recover-able. The save sets referred to by the recovered index will be marked as browsable. They will remainbrowsable for the length of time they were originally browsable.

Checks of a higher level generally take longer than checks at a lower level. Checks at a higher level pro-vide a more thorough checking of the online file index. Level 7 is used when the online file index on diskneeds to be recovered from backup media. Thensrck program is restartable at any time during its execu-tion. Therefore, it can survive system crashes or exhaustion of resources without losing data.

OPTIONS−C This option validates the client’s online file index header. It is identical to specifying the−L 1

option.

−c This option is the same as using−L 2.

−F This option is the same as using−L 2.


NSRCK(1m) NSRCK(1m)

−t date Recover the index as of the specifieddate (in nsr_getdate(3)format). This option is only validwith the−L 7 option.

−T tempdirSpecifies a different directory to use for conversion. This is useful if your client indexes are on filesystems that are nearly full. It will enable the conversion to use the tempdir specified as a workspace for converting indexes. It is not recommended to use /tmp, since its contents are lost if yourmachine is rebooted.

−L levelSpecifies the level of checking to use. The valid levels are 1-7.

−M Master mode (not advised for manual operation). This advisesnsrck that it is being run bynsrd(1m) or another NetWorker daemon and should log messages with timestamps, and performany other behavior expected bynsrd.

−m Forcesnsrck to check and rebuild the media database b-tree indexes, instead of checking a client’sonline file index.

−q Quiet mode. All advisory messages are suppressed.

−v Verbose mode. Advisory messages are emitted.

−R Removes the index for the client. This is valid only when the−Y option is also specified. If thensrck is not in master mode, the user will be prompted with a warning indicating which online fileindexes will be completely removed and given an opportunity to kill the commmand if this wasnot what the user intended.

−X This is the same as using−L 3

−x percentThis is the same as using−L 1. The "percent" value is ignored, but permitted. This allows cus-tomer scripts using this option to continue working.

−Y Used in conjunction with−R to remove online file indexes. Using this flag means that you reallydo wish to remove the online file index(es). If you fail to use this flag with the−R option, you willbe warned that you need to add the−Y flag to thensrck command.

FILES/nsr/index/ clientname /db6/nsrck.lck

nsrck locks this file thereby insuring that only one copy ofnsrck is checking a client’s index.

/nsr/index/clientname

/nsr/index/clientname/db6

SEE ALSOnsr_layout(5), nsr_policy(5), hosts(5), nsr(1m),nsrd(1m),nsrindexd(1m),nsrmmdbd(1m),nsrim(1m),savegrp(1m)

DIAGNOSTICSchecking index forclientname

Informative message that the files associated with the named client are being inspected.

WARNING no valid savetimes - cross-check not performedfor clientnameDuring a cross-check, no save sets were found for this client. Since this situation can occur duringdisaster recovery,nsrck avoids deleting the entire contents client index.

cross-checking index forclientnameDisplayed when the−L 3 option is in effect.

completed checkingcountclientsDisplayed as the program finishes, provided some form of checking was accomplished.


NSRCK(1m) NSRCK(1m)


NSRCLONE(1m) NSRCLONE(1m)

NAMEnsrclone − NetWorker save set cloning command

SYNOPSISnsrclone

[ −v ] [ −sserver] [ −b pool ] { −f file | volname...}

nsrclone[ −v ] [ −sserver] [ −b pool ] −S { −f file | ssid... }

nsrclone[ −v ] [ −sserver] [ −b pool ] −V { −f file | volid... }

and on HSM enabled server:

nsrclone[ −v ] [ −sserver] [ −b pool ] −c client−N saveset

DESCRIPTIONThensrcloneprogram makes new copies of existing save sets. These copies are indistinguishable from theoriginal, except for the volume(s) storing the copies. The copies are placed on different media volumes,allowing for higher reliability than a single copy provides. The copies may be made onto any kind ofmedia (for example, save sets on an 8mm tape may be copied to a set of optical disks). However, all mediaused as the destination of annsrcloneoperation must be in aclone pool. Seensr_pool(1m) for a descrip-tion of the various pool types.

Although the command line parameters allow you to specify volume names or volume identifiers,nsrclonealways copies complete save sets. Save sets that are only partially contained on a specified volume will becompletely copied, so volumes may be requested during the cloning operation in addition to those specifiedon the command line. Note thatnsrclone does not perform simplevolume duplication, but rather, copiesfull save sets to a set of destination volumes in a given pool. If the first destination volume chosen cannothold all of the save sets to be copied, another volume will be chosen. This allows you to use different kindsof media for each copy, allowing for variable sized volumes, such as tapes.

If you use the−c and−N options together,nsrclonewill create asuper-fullclone for the given client saveset. A super-full clone is a feature that is supported only with HSM (most useful on HSM enabled servers).It automatically creates a clone of the most recent complete full backup of the named client and save set,along with any HSM migration save sets referenced to by the backup. Super-full clones should be clonedto a volume from a pool of typeMigration Clone. If no migration save sets are referenced by the mostrecent backup, only the full save set is cloned.

The nsrclone program, in conjunction withnsrmmd(1m), guarantees that each save set will have at mostone clone on a given volume. When you specify a volume name or identifier, the copy of the save sets onthat volume are used as the source. When save sets are specified explicitly, those with existing multiplecopies are automatically chosen (copies of save sets that exist on volumes in a jukebox are chosen overthose that require operator intervention). You can also specify which copy (clone) of a save set to use as thesource, (see the−Soption,in the options section).

Cloning between storage nodes is accomplished by annsrmmd(1m) on the source node reading from a vol-ume, and anothernsrmmd(1m) on the target node writing to a volume. The source node is determined bythe location of a source volume, which is given by where the volume is currently mounted or by its "loca-tion" field if unmounted (seemmlocate(1m)). The target node of a clone is determined by the "clone stor-age nodes" attribute of the storage node’s client resource, the "clone storage nodes", or the "storage nodes"attribute of the server’s client resource in decending priority. Seensr_storage_node(5) andnsr_client(5)for additional detail on how these attributes are used and on other storage node information.

OPTIONS−b pool

Specifies the media pool to which the destination clones should be sent. The pool may be anypool currently registered withnsrd(1m) that has its status set toclone. The possible values can beviewed by selecting the Pools menu item from the Administration menu ofnwadmin(1m). If you



omit this option, the cloned save sets are automatically sent to theDefault Clonepool.

−c clientSpecifies a client whose save sets should be considered in the creation of a super-full clone. Thisoption must always be specified in conjunction with the−N option.

−f file Instructsnsrclone to read the volume names, volume identifiers or save set identifiers from the filespecified, instead of listing them on the command line. The values must be listed one per line inthe input file. Thefile may be "-", in which case the values are read from standard input.

−N savesetSpecifies which save set name is used to create a super-full clone. This option must always bespecified in conjunction with the−c option.

−sserverSpecifes a NetWorker server to migrate save sets from. Seensr(1m) for a description of serverselection. The default is the current system.

−v Enable verbose operation. In this mode, additional messages are displayed about the operation ofnsrclone, such as save sets that cross volumes, or save set series expansions.

−S Causesnsrclone to treat subsequent command line parameters as save set identifiers, not volumenames. Save set identifiers are unsigned numbers. You can find out the save set identifier of asave set using themminfo -v command (seemminfo(1m)). The−S option is useful when youwant to copy individual save sets from a volume or all save sets matching anmminfo query (seethe examples below). The save set identifiers may also specify exactly which copy of a sav e setwith multiple copies to use as the source. To specify exact copies, use thessid/cloneidformat foreach save set identifier. In this case, thessidand thecloneidare unsigned numbers, separated by asingle slash (/). You can find out thecloneidfor a particular copy by using themminfo -S report,or a custom report.

−V Causesnsrclone to treat subsequent command line parameters as volume identifiers, not volumenames. Volume identifiers can be found using themminfo -mv report, for example. This optioncan not be used in conjunction with the−Soption.

EXAMPLESCopy all save sets on the volumemars.001to a volume in theOffsite Clonepool:

nsrclone −b ’Offsite Clone’ mars.001

Copy all complete save sets created during the previous weekend (recall thatnsr_getdate(3) dates withouttime-of-day match midnight at the beginning of that day). Only complete save sets can be copied bynsr-clone(1m):

nsrclone -S ‘mminfo −r ssid \-q ’!incomplete,savetime>last saturday,savetime<last monday’‘

Copy a specific clone of a specific save set:nsrclone -S 1538800517/770700786

SEE ALSOnsr_getdate(3), nsr_pool(5), nsr_storage_node(5), mminfo(1m),nsr(1m),nsrd(1m),nsrmmd(1m),nwadmin(1m).

DIAGNOSTICSThe exit status is zero if all of the requested save sets were cloned successfully, non-zero otherwise.

Several messages are printed signalling thatnsrd(1m) is unavailable for cloning data; these are self-explanatory. You may also see a message from the following list.

adding save set series which includesssidIf running in verbose mode, this message is printed whennsrclonenotices that a requested saveset is continued, requiring the entire series to be cloned (even if only part of the series was speci-fied in the command line parameters).



Cannot contact media databaseThe media database (and most likely other NetWorker services as well) on the named server is notanswering queries. The server may need to be started, or if it was just started, it needs to finish itsstartup checks before answering queries.

cannot clone save setnumber, series is corruptThe given sav e set is part of a save set series (used for saving very large files or filesystems), butnot all of the save sets in the series were found in the media database. This can happen if, forexample, you relabel a tape that contains part of a save set series.

cannot clone backup and archive data togetherArchive and backup data is fundamentally different and cannot be cloned to the same pool. Youneed to runnsrclonetwice, once to clone the backup save sets and once more for the archive sav esets.

cannot open nsrclone session withserverThis message is printed when the server does not accept clone sessions.

cloning not supported; upgrade requiredAnother enabler is required to use this feature.

cloning requires at least 2 devicesCloning requires at least one read/write device and one read-only or read/write device, since datais copied from one volume directly to another.

serverdoes not support cloningThe named server is not capable of cloning.

each clone host needs at least two enabled devicesWhen cloning between two storage nodes that share the same physical drive, each node must haveat least two enabled devices.

error, no valid clones of ssidnumberThe listed save set exists, but cannot be cloned because there are no complete copies of the saveset. The save set was either aborted or is in progress. Only complete save sets can be copied.

error, user usernameneeds to be on administrator listerror, user usernameneeds to be on archive users list

Only NetWorker administrators are allowed to make clones of backup save sets. NetWorkeradministrators are listed in the NSR server resource, seensr_service(5) for more information. Forservers with archive capability, users listed in the NSR archive client’s user list are allowed toclone archive sav e sets.

no complete, full backup ofclient:savesetDuring an attempt to create a super-full clone,nsrclonecould not find a complete, full backup ofthe requested save set.

no complete save sets to cloneOnly complete save sets can be copied, and no complete save sets were found matching therequested command line parameters.

numberis not a valid save setThe given sav e set identifier is not valid. Two forms are understood: simple save set identifiersand those with a cloneid specified. Simple save sets are unsigned numbers. The save set with thecloneid form is specified as two unsigned numbers separated by a single slash (/).

pool is not a cloning poolThe pool specified with the−b pooloption is not a clone pool. You must always use a pool with atype of "Backup Clone" or "Archive Clone" for the−b option.

save setnumberdoes not existThe given sav e set (from a−Ssave set list) does not exist. Verify your save set identifiers usingmminfo(1m).



save setnumbercrosses volumes; requestingadditional volumesThis message is printed in verbose mode when volume names or IDs were specified, but the givensave set is only partially resident on the listed volumes. Since only complete save sets can becloned,nsrcloneautomatically requests additional volumes.

save set clonenumber/cloneiddoes not existA specific clone of a save set was specified, but that save set has no clones with that clone identi-fier. Verify your save set identifiers usingmminfo(1m).

volumename-or-numberdoes not existThe given volume (either a volume name or a volume id specified in the−V option) does not existin the media database.

waiting 30 seconds then retryingA temporary error occurred andnsrclonewill automatically retry the request until the condition iscleared. For example, an error will occur if all devices are busy saving or recovering andnsrclonemust wait for these devices become available.

Warning: No candidate migration save sets ofclient:savesetIf you are usingnsrcloneto create a super-full clone, and the most recent full backup of the namedclient and save set does not refer to any migration save sets, this warning is printed asnsrclonebegins cloning the full backup.


NSRD(1m) NSRD(1m)

NAMEnsrd − daemon providing the NetWorker service

SYNOPSISnsrd [ −k virtual-service-name]

ansrd [ commentary ]

DESCRIPTIONThe nsrd daemon provides an RPC-based save and recover service. This service allows users to save,query for, and recover their files across a network. The RPC program number provided bynsrd is 390103.

Normally nsrd is invoked from a startup shell script (for examplerc.local, rc.boot) at boot-time, andshould never need to be started directly by a user. After it is started,nsrd starts up the other daemons itneeds to provide the NetWorker service.

Thensrd command must be run on a machine with appropriate resources. These resources include devices(for example, tape drives) which are under the control of the media multiplexor software (seensr-mmd(1m)), and sufficient disk space for the index daemons, (seensrindexd(1m) andnsrmmdbd(1m)) tomaintain the index of sav ed user files and volumes with corresponding files.

Each time a backup, recover, or another session begins,nsrd starts the program,ansrd, to process therequested session. Theansrd program is called anagent. The agent is in charge of monitoring thatbackup, recover, or another session, and automatically exits when a session completes. Usingps(1) oranother process monitoring tool, you can inspect the subsequent parameters ofansrd to see what kind ofsession it is monitoring. If necessary, agents can be forcibly terminated to abort a backup or recover ses-sion. Agents cannot be run directly; they can only be started bynsrd.

Whennsrd is started with the−k option, it checks to see whether it has been installed as a cluster serviceand that the virtual host which owns/nsr/res matchesvirtual-service-name.If either of these validationsteps fails,nsrd exits immediately. (To check whether NetWorker has been installed as a cluster service,nsrd checks for a file calledNetWorker.clustersvr in the directory containing thensrd binary. To checkthat/nsr/res is owned byvirtual-service-name, nsrd queries the cluster management software.)

If the −k option is not used when starting NetWorker in a cluster, the server assumes the identity of the vir-tual host which owns/nsr/res. If no virtual host owns/nsr/res, thennsrd will not start.

OPTIONS−k virtual-service-name

Instructs nsrd to start up in cluster failover mode usingvirtual-service-nameas its host-name/identity. This option is used by the NetWorker cluster control script which starts NetWorker.

FILES/nsr/logs/daemon.log

The file to whichnsrd and other NetWorker daemons send information about various errorconditions that cannot otherwise be logged using the NetWorker event mechanism.

/nsr/res/nsr.resAttributes describing the NetWorker service and its resources (Seensr_service(5)).

/nsr/res/nsrjb.resAttributes describing the jukebox resources of the NetWorker service.

NetWorker.clustersvrIf this file exists in the directory containing NetWorker’s daemons, it indicates that the Net-Worker server has been installed as a cluster service.

SEE ALSOnsr(1m),nsr_service(5), nsrmmd(1m),nsrmmdbd(1m),nsrindexd(1m),ps(1), rc(1m).


NSREXEC(1m) NSREXEC(1m)

NAMEnsrexec − remotely execute NetWorker commands on NetWorker clients

SYNOPSISnsrexec[ −a auth] [ −vR ] [ −T hangseconds] [ −c client ] [ −f file | - ] [ −N ] [ −− ... ]

DESCRIPTIONThe nsrexeccommand is run only by other NetWorker commands. It is used to remotely execute com-mands on NetWorker clients runningnsrexecdand monitor the progress of those commands.

OPTIONS−a the authentication type to use.

−T the inactivity timeout.

−c the client on which to run the command.

−R a filesystem probe is being performed.

−v verbose output has been requested.

−f data is read from a file or the standard input.

−N client is an NDMP client.

−− (double dash) information after the double dash is used for descriptive purposes only.

SEE ALSOnsr(5), nsr(1m),nsrexecd(1m),savegrp(1m)


NSREXECD(1m) NSREXECD(1m)

NAMEnsrexecd − NetWorker client execution service

SYNOPSISnsrexecd[ −sserver[ −sserver... ]] [ −f serverfile] [ −p savepath] [ −i ]

DESCRIPTIONnsrexecd is used by NetWorker servers to perform automatic operations on NetWorker clients. It is cur-rently used bysavegrp(1m) to start saves, pre-migrations and storage node functions on NetWorker clientmachines. When migration is enabled on a NetWorker client,nsrexecdmonitors disk usage, starts migra-tion runs usingnsrmig(1m), and also manages file recall usingnsrib(1m). When storage node functionsare in use,nsrexecdstartsnsrmmd(1m) daemons andnsrjb (1m) commands on the host, and responds topolling requests from the server. Seensr_storage_node(5) for additional detail on storage nodes. Thensrexecdservice is normally started at boot time on each NetWorker client machine. Since NetWorkerservers are usually expected to be clients of themselves,nsrexecdruns on all NetWorker servers as well.

The nsrexecd service exports an RPC-based service to remotely execute NetWorker operations. Allrequests must be authenticated, and can optionally be restricted to specific NetWorker servers. Onlysaverequests (for example,save(1m) or savefs(1m)), migration requests (for example,nsrpmig(1m) or nsr-mig(1m)), and storage node requests are allowed.

When command execution is requested,nsrexecdfirst verifies that the request is authenticated, and that itcomes from a valid NetWorker server. Next,nsrexecdverifies that the command is a save or migrationcommand (for example,save(1m) ornsrpmig(1m)). It then executes the specified command from the Net-Worker binary directory. This directory is normally determined by the location of thensrexecdexecutable,but can be specified on the command line.

OPTIONS−i As part of the NetWorker server authentication, the server’s network address is mapped to a name.

The name is then reverse-mapped to a network address. The server is authenticated if and only ifthe original network address matches the reverse-mapped address. The−i flag skips the addresscomparison thereby allowing workarounds to misconfigured or misfeatured naming systems. Thisoption should be used with care since it may allow the NetWorker client to send its data to anunauthorized machine.

−f serverfileSpecifies a file containing a list of NetWorker servers which can initiate saves. This file should listone server name per line. If no−f or −s options are specified,nsrexecdlooks for a default file inthis same format (or Mac preferences on the Mac client). The location of this default file is listedin theFILES section of this man page.

−p savepathTells nsrexecdto look for save commands in thesavepathdirectory, rather than the default (thedirectory in whichnsrexecdexists).

−sserverAllows save requests to be initiated only by the given NetWorker server. Multiple−s options maybe given to allow access by several NetWorker servers. If a NetWorker server has multiple net-work interfaces, it is often best to list the hostname corresponding to each network interface, toavoid failed saves.

FILES/nsr/res/nsrla.res The file with attributes describing the NetWorkernsrexecdservice and

its resources (seensr_service(5))./nsr/res/servers The file containing the default list of servers that can back up or migrate

the NetWorker client.

SEE ALSOnsr_service(5), nsr_storage_node(5), nsrib(1m),nsrmig(1m),nsrpmig(1m),nsrports(1m),save(1m),save(1m),savefs(1m),savegrp(1m)


NAME

nsrfile - NetWorker module to aid in writing shell script ASMs

SYNOPSIS

nsrfile -s [-N asmname] [-C command] <ASM args> file... nsrfile -r [-F] [-N asmname] [-C command] <ASM args> file...

DESCRIPTION

The nsrfile command is an external ASM (Application Specific Module) that makes it easy to implement shell script based ASMs for saving and recovering databases, raw devices, and other site-specific special data.

It is intended that nsrfile be invoked by a shell script which is in turn invoked by uasm(8) during save(8) or recover(8) operations.

In save mode, nsrfile creates a NetWorker protocol save record on standard-out for each file given on the command line. A command (see -C option below) is executed for each file name and the data produced by that command is converted into NetWorker protocol save record data and emitted on standard-out. In this way, nsrfile can be used to create a save record from the data produced by any UNIX command. Save mode looks like:

save-command | nsrfile -s ==> save record

In recover mode, nsrfile reads a NetWorker protocol save stream from standard-in and executes a command for each save record in the stream. The save data for each save record is written to the standard-input of the executing command. Recover mode looks like:

save record ==> nsrfile -r | recover-command

OPTIONS

See uasm(8) for a general description of ASM's and the <ASM args>.

1 of 4

-s Save mode.

-r Recover mode.

-F Don't create the target file. This flag is only valid in recover mode (with the -r flag). This flag is used when the data being recovered contains the information necessary to create the target file (e.g. tar or cpio data).

-C command Execute the given command for each file in either save or recover mode. In save mode (-s flag), the command is executed once for each file given on the command line. The stdout of the command is piped to the stdin of nsrfile which produces a NetWorker save record on stdout. In recover mode (-r flag), the command is executed once for each file in the recover data stream. The stdout of nsrfile is connect through a pipe to the stdin of the command. The command can include an argument % which will be replaced with the current pathname on each invocation of command.

-N asmname Pretend to be the ASM called asmname. This option allows nsrfile to take the name of the shell script ASM that invoked it so that warnings and error messages are prepended with asmname instead of nsrfile.

file ... List of files and directories for command, see option - C.

IMPLEMENTING ASM SHELL SCRIPTS

Several support procedures are necessary to implement an ASM shell script. The hideasm(8) command is implemented as a shell script ASM and it can be used as the source of these shell procedures.

asm_echo Works like the shell echo command, but it writes to stderr instead of stdout. This procedure should be used instead of echo throughout the ASM script because stdout is used for the save record output in save mode and to pipe to a command in recover mode.

asm_error Like asm_echo above, but it prepends the name of the

2 of 4

ASM script to the message.

asm_setup Parses the ASM script arguments and breaks them into several shell variables. This procedure expects as arguments all of the arguments to the ASM shell script.

The shell variables defined by asm_setup are:

Nsrfile_args The arguments that nsrfile should be invoked with. These arguments are derived from the shell script arguments.

Files The list of files and directories to operate on.

Mode The ASM mode. The value of this variable can be SAVE, RECOVER, or COUNT.

Myname The name that the ASM shell script was invoked with. The directory path is stripped off. Exec_path The directory path name that should be used to explicitly invoke nsrfile or other ASM executables.

EXAMPLE

To implement a shell script ASM that uses tar(1) to back up and recover files and directories the following shell script (plus the shell procedures described above) would be used:

asm_setup $0 $* if [ $Mode = SAVE ]; then $Exec_path/nsrfile -C "tar cf - %" $Nsrfile_args $Files elif [ $Mode = RECOVER ]; then $Exec_path/nsrfile -F -C "tar xBf -" $Nsrfile_args $Files else # $Mode = COUNT $Exec_path/nsrfile $Nsrfile_args $Files fi

If this tarasm script were executed with the command line:

tarasm -s foo

the Mode variable would be set to SAVE and nsrfile would run as though the following command line had been typed:

tar cf - foo | nsrfile -N tarasm -s foo

3 of 4

SEE ALSO

nsr(5), hideasm(8), save(8), recover(8), uasm(8).

4 of 4

NSRHSMCK(1m) NSRHSMCK(1m)

NAMEnsrhsmck − check and correct consistency for an HSM-managed filesystem

SYNOPSISnsrhsmck [ −cdfMv ] [ −sserver] path

DESCRIPTIONThe nsrhsmck command checks and corrects consistency between HSM migrated files stored in Net-Worker and on-disk files. There are four situations handled bynsrhsmck.

The first situation occurs when the stub for a migrated file is renamed. This is known as the rename case.In this situation, the stub with the original filename no longer exists. The corrective action taken bynsrhsmck in this instance is to update the HSM file index to reflect the new name.

The second situation occurs when a symbolic link is created that points to the same name in the NetWorkerIB namespace as another symbolic link. This is known as the duplicated link case. The corrective actionfor a duplicated link is to replace the duplicate with another symbolic link. The replacement must point tothe original symbolic link, and not directly into the NetWorker IB namespace.

The third situation occurs when the stub pointing to a migrated file has been deleted. This is known as thepossible delete case. The term "possible" is used here, because the stub may reappear at some later point intime, possibly if a recover of the stub is done using NetWorker. The corrective action for a possible dele-tion is to mark the index entry for the migrated file as possibly deleted. Note that if a file marked as possi-bly deleted is detected on disk before the index entry can be deleted, the index entry will be unmarked as apossible deletion.

The fourth situation handled bynsrhsmck is an index entry that is marked as a possible deletion, but haspassed the 60-day expiration period. This is known as the expired delete case. The corrective action forthis case is to remove the expired entries from the HSM file index. Before an entry is deleted from theHSM file index, a final check is made to make sure the file does not exist on disk.

You must specify a path on the command line whennsrhsmck is run. Only files and index entries that fallunder the path specified will be examined for consistency.

OPTIONS−c Scans the HSM file index, and deletes index entries that are marked as possibly deleted but have

passed the 60-day expiration period.

−d Scans the HSM file index, and marks any possible deletions that are detected.

−f Scans the filesystem on disk, and searches for duplicated links and renames them.

−M Master mode (not advised for manual operation). This advisesnsrhsmck that it is being run bynsrexecd(1m) or another NetWorker daemon and should log messages with timestamps, and per-form any other behavior expected bynsrexecd.

−n Does not correct inconsistencies found, but just reports them.

−sserverUseserveras the NetWorker server.

−v Increases the verbosity level incrementally. This flag may be specified up to three times to achievethe maximum verbosity. Note that verbose mode can produce an extremely large quantity of out-put, and is not recommended for use in most situations.


0 Normal exit.1 Abnormal exit.

SEE ALSOnsr(5), nsr(1m),nsr_migration(5), nsrexecd(1m),nsrmig(1m),nsrpmig(1m)


NSRHSMCLEAR(1m) NSRHSMCLEAR(1m)

NAMEnsrhsmclear − destroy outstanding HSM events and sessions

SYNOPSISnsrhsmclear[ −vn ]

DESCRIPTIONIf killed, HSM processes may leave sessions, tokens, and file locks. Other processes trying to access alocked file will block until the lock is released.nsrhsmclearwill release those locks.

nsrhsmclear destroys all XDSM HSM tokens and sessions that may have outstanding file locks, and maybe associated with HSM filesystem read, write, or truncate requests. Those requests will be aborted with anI/O error.

Note:nsrhsmclearshould be used with caution because file system clients and HSM processes will receiveI/O errors.

OPTIONS−n Print tokens and sessions, but do not destroy any.

−v Print verbose descriptions of tokens.

SEE ALSOnsrhsmd(1m),nsrhsmls(1m),nsrpmig(1m),nsrmig(1m),nsr_migration(5),


0 Normal exit.1 An error occurred.


NSRHSMD(1m) NSRHSMD(1m)

NAMEnsrhsmd − XDSM HSM event daemon

nsrhsmrc − recall migrated file

SYNOPSISnsrhsmd [ -v ] [ -C parallelism ] [ −s <NWserver> ]

nsrhsmrc NWserver client session token fsid ino igen ofsid ofid savetime path

DESCRIPTIONnsrhsmd listens for XDSM file system events that may require recalling migrated files.nsrhsmd is startedby nsrexecd if Migration resources are configured for the client.

nsrhsmrc is started by nsrhsmd to recall each migrated file.

OPTIONS−v Verbose output

−C parallelismSpecifies the maximum number of recalls that will be executed in parallel. Default is 8.

−s <NWserver>Specifies which machine to use as the NetWorker server (by hostname or IP address).

NWserverThe Networker server for the migrated file.

client The Networker client which originally migrated the file.

session tokenThe XDSM session and token that describe the file system event that caused the recall of themigrated file. The session and token need an exclusive lock on the file.

fsid ino igenThe file system identifier, inode number, and inode generation number of the requested file.

ofsid oinoThe file system identifier and inode number of the file when it was premigrated.

savetimeThe savetime identifying the save set where the file data is stored.

path The name of the file when it was premigrated.

FILES/nsr/logs/daemon.log The file to whichnsrhsmd , nsrhsmrc , and other NetWorker daemons

send information about various error conditions.

/nsr/res/nsrla.res Attributes describing the NetWorkernsrexecdservice and its resources.Local Migration resources are described here, as well as in the Net-worker server’s/nsr/res/nsr.resfile.

SEE ALSOnsrhsmls(1m),nsrpmig(1m),nsrmig(1m),nsr_migration(5), nsrexecd(1m)


NSRHSMLS(1m) NSRHSMLS(1m)

NAMEnsrhsmls − list contents of a XDSM HSM configured directory

SYNOPSISnsrhsmls[ −abim ] [ −lv ] [ −R | −r <level> ] [ −s <NWserver> ] [<path> ]

DESCRIPTIONnsrhsmls is a NetWorker XDSM HSM-specific "ls" command to display the premigration and migrationstatus of a file or directory, and other information.

It works much like an "ls" command, but it will list HSM-related information. For example, it will indicatewhether or not a file has been premigrated and migrated, number of disk blocks currently being used by thefile, the file size and full filename.

If none of the display options (-a, -b, -i, or -m ) are specified,nsrhsmls will display migration summaryinformation for each file. The migration summary output is as follows:

The first character shows the migration status of the file:

r File is resident (it has not been migrated)m File has been migrated- Unable to determine file’s migration statusd Is a directoryx Is not a file or a directory

The second character shows the premigration status of the file:

n File has not been premigratedp File has been premigrated

The next entry gives the number of disk blocks, including indirect blocks, currently being used by the file.

The next entry gives the file size in bytes.

OPTIONS−l For the migration summary display, separates the displayed columns with "|".

−v Displays more detailed migration information.

−R Runs down all subdirectories in recursive manner.

−r <level>runs down the specified level of subdirectories in a recursive manner. Lev el 1, the default, works inthe current directory. Lev el 2 causes nsrhsmls to run down one level below the current directory,and so on.

−s <NWserver>Specifies which machine to use as the NetWorker server (hostname or IP address).

−a Prints the allocation information. The allocation information shows which regions of a file havespace allocated on the file system. A hole may be caused by migration of a file or by a creation ofa unix hole.

−b Prints the XDSM attribute in hexadecimal format. The XDSM attribute is used by Networker tofind the data that was premigrated. The attributes used by Networker include: the Networkerserver, the Networker client, save set, file system identifier, inode number, and a managed regionmap.

−i Prints the XDSM handle in hexadecimal format. The handle information contains the file systemidentifier, the inode number, and the inode generation of the file.

−m Prints the managed region map. The managed region map indicates which regions of a file haveread, write, and truncate events enabled for HSM.


NSRHSMLS(1m) NSRHSMLS(1m)

EXAMPLESTo display migration status of files upto three levels down in the /export/home/hsmfs directory, with theNetWorker server being "jupiter", you would enter the following command:

nsrhsmls -r 4 -s jupiter /export/home/hsmfs

SEE ALSOls(1), nsrpmig(1m), nsrmig(1m).


NSRHSMNFS(1m) NSRHSMNFS(1m)

NAMEnsrhsmnfs − perform HSM operations on NFS mounted file systems

SYNOPSISnsrhsmnfs migrate[ (nsrpmig and nsrmig options) ]

nsrhsmnfs ls[ (nsrhsmls options) ]

nsrhsmnfs recall[ (nsrhsmrecall options) ]

DESCRIPTIONnsrhsmnfs is a NetWorker XDSM HSM-specific command used to perform HSM commands on NFSmounted file systems.nsrhsmnfssubmits the command to execute on the NFS server. If the file system isnot NFS mounted,nsrhsmnfsexecutes the command locally.

nsrhsmnfs translates pathnames if necessary. Output of the invoked commands is the output ofnsrhsm-nfs. No translation of the output is done bynsrhsmnfs.

nsrhsmnfs migrate invokes nsrpmig, followed by nsrmig. The -l 0 option is automatically added to thensrmig command to set the low water mark to zero percent. Without this option, file migration would onlytake place if file system usage is above the low water mark.

nsrhsmnfs lsinvokes nsrhsmls with the specified options.

nsrhsmnfs recallinvokes nsrhsmrecall with the specified options.

SEE ALSOnsrhsmls(1m), nsrhsmrecall(1m), nsrpmig(1m), nsrmig(1m).


NSRHSMD(1m) NSRHSMD(1m)

NAMEnsrhsmd − XDSM HSM event daemon

nsrhsmrc − recall migrated file

SYNOPSISnsrhsmd [ -v ] [ -C parallelism ] [ −s <NWserver> ]

nsrhsmrc NWserver client session token fsid ino igen ofsid ofid savetime path

DESCRIPTIONnsrhsmd listens for XDSM file system events that may require recalling migrated files.nsrhsmd is startedby nsrexecd if Migration resources are configured for the client.

nsrhsmrc is started by nsrhsmd to recall each migrated file.

OPTIONS−v Verbose output

−C parallelismSpecifies the maximum number of recalls that will be executed in parallel. Default is 8.

−s <NWserver>Specifies which machine to use as the NetWorker server (by hostname or IP address).

NWserverThe Networker server for the migrated file.

client The Networker client which originally migrated the file.

session tokenThe XDSM session and token that describe the file system event that caused the recall of themigrated file. The session and token need an exclusive lock on the file.

fsid ino igenThe file system identifier, inode number, and inode generation number of the requested file.

ofsid oinoThe file system identifier and inode number of the file when it was premigrated.

savetimeThe savetime identifying the save set where the file data is stored.

path The name of the file when it was premigrated.

FILES/nsr/logs/daemon.log The file to whichnsrhsmd , nsrhsmrc , and other NetWorker daemons

send information about various error conditions.

/nsr/res/nsrla.res Attributes describing the NetWorkernsrexecdservice and its resources.Local Migration resources are described here, as well as in the Net-worker server’s/nsr/res/nsr.resfile.

SEE ALSOnsrhsmls(1m),nsrpmig(1m),nsrmig(1m),nsr_migration(5), nsrexecd(1m)


NSRHSMRECALL(1m) NSRHSMRECALL(1m)

NAMEnsrhsmrecall − recall HSM files to file system

SYNOPSISnsrhsmrecall [ −qRuv ] [ −I includefile ] [ path1 path2 ... ]

DESCRIPTIONnsrhsmrecall is a NetWorker XDSM HSM-specific command used to recall one or more files from nearlinestorage to the file system.nsrhsmrecall can be used before reading a large number of files. For two ormore files,nsrhsmrecall is usually faster than automatic transparent recalls caused by read or write systemcalls.

If no path arguments and no includefile are specified,nsrhsmrecall recalls files in the current directory.

OPTIONS−q Quiet. Turns off the verbose output.nsrhsmrecallnormally runs with verbose output.

−R Recurse subdirectories of directories specified on the command line, or in the input file. Bydefault,nsrhsmrecall does not recurse into subdirectories.

−u Stops when an error occurs during recall. Normally,nsrhsmrecall treats errors as warnings andtries to continue to recall the rest of the files requested. However, when this option is used,nsrhsmrecallwill stop recalling on the first error it encounters.

−v Enables more verbose output.

−I includefileTakes the paths to recall from the command line, and reads paths to recall from the file includefile.The paths must be listed one per line. If no paths are specified on the command line, only thosepaths specified in includefile will be recovered.

EXAMPLESTo recall all of the files in the current directory, you would enter the following command:

nsrhsmrecall

To recall all files named Index, you would enter the following command:

find . -name Index -print | nsrhsmrecall -I -

BUGSThe file name printed on recall is the file name at the time the file was premigrated.

Recover of HSM stub files to a different location will cause two HSM stubs to refer to the same data, how-ev er only one of the HSM stubs can be recalled at a time. Runnsrhsmrecall again to recall the files thatcould not be recalled.

SEE ALSOnsrhsmls(1m), nsrpmig(1m), nsrmig(1m).


NSRIB(1m) NSRIB(1m)

NAMEnsrib − NetWorker index browser daemon

nsriba − NetWorker index browser agent daemon

SYNOPSISnsrib [ −sserver] [ −t timeout] [ −v ] [ −M ] [ −i # ] [ −C # ] [ −D # ] [ −R # ] [ −T rdir ] [ dir ]

nsriba [ −sserver] [ −c client ] [ −p path] [ −v ] [ −t browse_date] [ −I index_type] [ −N session_name][ −i # ] [ −C # ] [ −D # ] [ −R # ] [ −T rdir ] [ dir ]

DESCRIPTIONThensrib (index browser) andnsriba (index browser agent) daemons provide a convenient NFS interfacein which to view NetWorker indexes. Usingnsrib is the preferred method as it will launch and manage theappropriatensriba processes as needed. Thensriba daemon gives you an NFS filesystem view of a partic-ular NetWorker client’s index as of a giv en time. Also, it can be used directly for situations where the flexi-bility provided bynsrib is not required. Thensrib andnsriba daemons appear to be an NFS server to thelocal kernel in a manner similar toautomount(1m).

The nsrib daemon will interpret names referenced indir without an ‘@’ as a NetWorker client index tobrowse. You can also construct names of the formclient@dateto browse a particular index as of a particu-lar time. You can also use the name of the form@dateto browse the index for the local machine. Thedateis interpreted as ansr_getdate(3) style string after replacing any underscores (_) with a space and alldashes (-) with a slash (/). Whennsrib gets such a name request, it will launch and manage the appropriatensriba process on a mount point that it builds indir automatically. If thensriba filesystem is not accessedwithin an appropriate interval,nsrib will attempt an umount of thensriba filesystem. If successful, thesymbolic link and mount directory created indir is removed.

Below thedir/client@datedirectory fornsrib (or within thedir directory fornsriba), a read-only filesys-tem consisting of the entire NetWorker index for the specified client can be seen. At times, a local machinemay not have NetWorkerrecover accessrights for the specified client. Seensr_client(5). There may be noentries in the NetWorker index for the specified client at the appropriate time. In either of those cases, thedirectory will be empty (nsrib) or the command will fail (nsriba). The files and directories with in thensriba filesystem will appear as normal UNIX files just as inrecover(1m), except that the access time(atime) of all files will be the "save time" of the file, not the access time of the file as stored in the index.Thus runningls -lu within annsriba directory will show all the file save times. If an file within annsribafilesystem is read, thennsriba will either recover the file and then return the resultant file as needed, orreturn an NFSERR_OPNOTSUPP ("Operation not supported") error. The actual behavior is dependent onthe −R and−C flags and whether the file appears to be currently "online" to NetWorker. If a file is to berecovered, the actual operation may take a long time depending primarily on the speed and location of theunderlying media that will be needed to recover the file. A separate process is used to do the actual filerecovery so that thensriba process can still respond to new NFS operations.

Within nsriba filesystems, hidden directories can be referenced for each file or directory to give informa-tion similar to therecover(1m) version command. These hidden directories are namedfile.V. Since thesehidden directories names are never returned for function calls such asreaddir (3), programs such asfind(1)that traverse the filesystem will never see these directories. The files and directories within these hiddendirectories are built up using the NetWorker file location information. The hidden directories for directoriescan either be named as ".V" within the directory or asdirname.V from above the directory. But when usingnsrib, you can only usedir/client@date/.Vto see all the versions for "/". Files within the hidden directoriescan be read (recovered) as any other file within thensriba filesystem.

nsrib and nsriba must not be terminated with theSIGKILL signal (kill -9). Without an opportunity tounmount itself and clean up properly, thensrib and nsriba mount points appear to the kernel as a non-responding NFS server. The recommended way to terminate annsrib or nsriba process is to send aSIGTERM(kill -15) signal to the daemon. Whennsriba receives aSIGTERMsignal, it attempts to unmountitself and exit if the unmount is successful (the filesystem is not currently busy). Whennsrib receives aSIGTERMsignal, it attempts to signal any childnsriba processes, that it started, to exit. If all childnsribaprocesses exit,nsrib then attempts to unmountdir itself and exits if the unmount is successful (the


NSRIB(1m) NSRIB(1m)

filesystem is not currently busy).

OPTIONSCommonnsrib andnsriba options:

−sserver Indicates the NetWorker server to use.

−i # Specifies "in place" mode if the corresponding file in the system is a symlink whose targetstring has the filename at the end.0 - nev er do "in place" recovers1 - do "in place" recovers only for exact matches with names of "file@date"2 - do "in place" recovers on any matching symlink targetDefault value is 1.

−v Runs in verbose mode. This should only be used for debugging purposes.

−C # Sets an upper limit on the number of concurrent file recovers. A value of 0 will disable allrecovers (independent of the−R value). Default value is 2.

−D # Specifies the debug level for messages. Using a number from 1 - 3 to get various (reason-able) levels of output. When running in a debugging mode,nsrib will not automatically runitself in the background. Default value is 0.

−R # Specifies recover mode on read.0 - nev er recover the file on NFS read.1 - recover the file on NFS read if "online".2 - always attempt a recovery of a file on NFS read.Default value is 2.

−T rdir Temporary directory to use to cache recovered files. Default value is"/usr/tmp/nsrib/Rtmp.client".

The−i, −s, −v, −C, −D, −R, and−T options tonsrib are passed through to eachnsriba program started.

The following options apply only tonsrib:

−t timeout Indicates the time in minutes to attempt umounts ofnsriba browsing directories. Default is30 minutes.

−M Indicates thatnsrib is being monitored by another process (such asnsrexecd(1m)), andshould not run in the background.

The following options apply only tonsriba:

−c client Indicates the NetWorker client index name to browse.

−p path Indicates the NetWorker index path to browse.

−t browse_dateIndicates ansr_getdate(3) string giving the "browse as of" time. Default value isnow.

−I index_typeIndicats the type of index that is being browsed. The default is a backup index.

−N session_nameIndicates the name to use to generate the NetWorker session name. Default value is themount directorydir .

EXAMPLESThese examples assume thatnsrib has been started on the/ib directory.

Finding filesTo find all versions of a file namedfoo that were owned byuser last week, use this command.Note that when usingfind(1), you shouldcd(2) to the directory first to avoid using the symboliclink instead of the resultant directory.


NSRIB(1m) NSRIB(1m)

cd /ib/@last_week; find . −namefoo−useruser−ls

Seeing saved versionsTo see all the saved versions of/var/adm/messagesfor a NetWorker clientclientname,use thiscommand. Note that by using the−u flag tols(1), the file save times will be displayed in thels datefield.

ls -lu /ib/clientname/var/adm/messages.V

Recovering filesTo recover/etc/fstabas of yesterday into the/tmp directory, use this command:

cp /ib/@yesterday/etc/fstab /tmp/fstab

FILES/etc/mtab This is the file that is updated on SunOS 4.1.x asnsrib andnsriba processes are mounted

and umounted.

/etc/mnttab This is the file that is updated on Solaris 2.x asnsrib andnsriba processes are mounted andumounted.

/ib This is the directory on whichnsrib will mount itself.

/usr/tmp/nsribThis is the default file cache directory tree.

LIMITATIONSThepwd(1) command fails from within a hidden ".V" directory.

The filesystem statistics returned to programs likedf(1) are of minimal use.

An unreliable heuristic is used to determine if a file is currently "online" for recovery if the NetWorkerserver version is 3.x or earlier. In particular, if a volume on a pre-4.0 NetWorker server is marked to be atsome location using themmlocate(1m) command,nsriba always performs as if the volume is "online".

SEE ALSOmount(2V), umount(2V), signal(3), nsr_getdate(3), nsr(5), nsr_client(5), automount(1m),nsrindexd(1m),nsrexecd(1m), recover(1m)


NSRIB(1m) NSRIB(1m)

NAMEnsrib − NetWorker index browser daemon

nsriba − NetWorker index browser agent daemon

SYNOPSISnsrib [ −sserver] [ −t timeout] [ −v ] [ −M ] [ −i # ] [ −C # ] [ −D # ] [ −R # ] [ −T rdir ] [ dir ]

nsriba [ −sserver] [ −c client ] [ −p path] [ −v ] [ −t browse_date] [ −I index_type] [ −N session_name][ −i # ] [ −C # ] [ −D # ] [ −R # ] [ −T rdir ] [ dir ]

DESCRIPTIONThensrib (index browser) andnsriba (index browser agent) daemons provide a convenient NFS interfacein which to view NetWorker indexes. Usingnsrib is the preferred method as it will launch and manage theappropriatensriba processes as needed. Thensriba daemon gives you an NFS filesystem view of a partic-ular NetWorker client’s index as of a giv en time. Also, it can be used directly for situations where the flexi-bility provided bynsrib is not required. Thensrib andnsriba daemons appear to be an NFS server to thelocal kernel in a manner similar toautomount(1m).

The nsrib daemon will interpret names referenced indir without an ‘@’ as a NetWorker client index tobrowse. You can also construct names of the formclient@dateto browse a particular index as of a particu-lar time. You can also use the name of the form@dateto browse the index for the local machine. Thedateis interpreted as ansr_getdate(3) style string after replacing any underscores (_) with a space and alldashes (-) with a slash (/). Whennsrib gets such a name request, it will launch and manage the appropriatensriba process on a mount point that it builds indir automatically. If thensriba filesystem is not accessedwithin an appropriate interval,nsrib will attempt an umount of thensriba filesystem. If successful, thesymbolic link and mount directory created indir is removed.

Below thedir/client@datedirectory fornsrib (or within thedir directory fornsriba), a read-only filesys-tem consisting of the entire NetWorker index for the specified client can be seen. At times, a local machinemay not have NetWorkerrecover accessrights for the specified client. Seensr_client(5). There may be noentries in the NetWorker index for the specified client at the appropriate time. In either of those cases, thedirectory will be empty (nsrib) or the command will fail (nsriba). The files and directories with in thensriba filesystem will appear as normal UNIX files just as inrecover(1m), except that the access time(atime) of all files will be the "save time" of the file, not the access time of the file as stored in the index.Thus runningls -lu within annsriba directory will show all the file save times. If an file within annsribafilesystem is read, thennsriba will either recover the file and then return the resultant file as needed, orreturn an NFSERR_OPNOTSUPP ("Operation not supported") error. The actual behavior is dependent onthe −R and−C flags and whether the file appears to be currently "online" to NetWorker. If a file is to berecovered, the actual operation may take a long time depending primarily on the speed and location of theunderlying media that will be needed to recover the file. A separate process is used to do the actual filerecovery so that thensriba process can still respond to new NFS operations.

Within nsriba filesystems, hidden directories can be referenced for each file or directory to give informa-tion similar to therecover(1m) version command. These hidden directories are namedfile.V. Since thesehidden directories names are never returned for function calls such asreaddir (3), programs such asfind(1)that traverse the filesystem will never see these directories. The files and directories within these hiddendirectories are built up using the NetWorker file location information. The hidden directories for directoriescan either be named as ".V" within the directory or asdirname.V from above the directory. But when usingnsrib, you can only usedir/client@date/.Vto see all the versions for "/". Files within the hidden directoriescan be read (recovered) as any other file within thensriba filesystem.

nsrib and nsriba must not be terminated with theSIGKILL signal (kill -9). Without an opportunity tounmount itself and clean up properly, thensrib and nsriba mount points appear to the kernel as a non-responding NFS server. The recommended way to terminate annsrib or nsriba process is to send aSIGTERM(kill -15) signal to the daemon. Whennsriba receives aSIGTERMsignal, it attempts to unmountitself and exit if the unmount is successful (the filesystem is not currently busy). Whennsrib receives aSIGTERMsignal, it attempts to signal any childnsriba processes, that it started, to exit. If all childnsribaprocesses exit,nsrib then attempts to unmountdir itself and exits if the unmount is successful (the


NSRIB(1m) NSRIB(1m)

filesystem is not currently busy).

OPTIONSCommonnsrib andnsriba options:

−sserver Indicates the NetWorker server to use.

−i # Specifies "in place" mode if the corresponding file in the system is a symlink whose targetstring has the filename at the end.0 - nev er do "in place" recovers1 - do "in place" recovers only for exact matches with names of "file@date"2 - do "in place" recovers on any matching symlink targetDefault value is 1.

−v Runs in verbose mode. This should only be used for debugging purposes.

−C # Sets an upper limit on the number of concurrent file recovers. A value of 0 will disable allrecovers (independent of the−R value). Default value is 2.

−D # Specifies the debug level for messages. Using a number from 1 - 3 to get various (reason-able) levels of output. When running in a debugging mode,nsrib will not automatically runitself in the background. Default value is 0.

−R # Specifies recover mode on read.0 - nev er recover the file on NFS read.1 - recover the file on NFS read if "online".2 - always attempt a recovery of a file on NFS read.Default value is 2.

−T rdir Temporary directory to use to cache recovered files. Default value is"/usr/tmp/nsrib/Rtmp.client".

The−i, −s, −v, −C, −D, −R, and−T options tonsrib are passed through to eachnsriba program started.

The following options apply only tonsrib:

−t timeout Indicates the time in minutes to attempt umounts ofnsriba browsing directories. Default is30 minutes.

−M Indicates thatnsrib is being monitored by another process (such asnsrexecd(1m)), andshould not run in the background.

The following options apply only tonsriba:

−c client Indicates the NetWorker client index name to browse.

−p path Indicates the NetWorker index path to browse.

−t browse_dateIndicates ansr_getdate(3) string giving the "browse as of" time. Default value isnow.

−I index_typeIndicats the type of index that is being browsed. The default is a backup index.

−N session_nameIndicates the name to use to generate the NetWorker session name. Default value is themount directorydir .

EXAMPLESThese examples assume thatnsrib has been started on the/ib directory.

Finding filesTo find all versions of a file namedfoo that were owned byuser last week, use this command.Note that when usingfind(1), you shouldcd(2) to the directory first to avoid using the symboliclink instead of the resultant directory.


NSRIB(1m) NSRIB(1m)

cd /ib/@last_week; find . −namefoo−useruser−ls

Seeing saved versionsTo see all the saved versions of/var/adm/messagesfor a NetWorker clientclientname,use thiscommand. Note that by using the−u flag tols(1), the file save times will be displayed in thels datefield.

ls -lu /ib/clientname/var/adm/messages.V

Recovering filesTo recover/etc/fstabas of yesterday into the/tmp directory, use this command:

cp /ib/@yesterday/etc/fstab /tmp/fstab

FILES/etc/mtab This is the file that is updated on SunOS 4.1.x asnsrib andnsriba processes are mounted

and umounted.

/etc/mnttab This is the file that is updated on Solaris 2.x asnsrib andnsriba processes are mounted andumounted.

/ib This is the directory on whichnsrib will mount itself.

/usr/tmp/nsribThis is the default file cache directory tree.

LIMITATIONSThepwd(1) command fails from within a hidden ".V" directory.

The filesystem statistics returned to programs likedf(1) are of minimal use.

An unreliable heuristic is used to determine if a file is currently "online" for recovery if the NetWorkerserver version is 3.x or earlier. In particular, if a volume on a pre-4.0 NetWorker server is marked to be atsome location using themmlocate(1m) command,nsriba always performs as if the volume is "online".

SEE ALSOmount(2V), umount(2V), signal(3), nsr_getdate(3), nsr(5), nsr_client(5), automount(1m),nsrindexd(1m),nsrexecd(1m), recover(1m)


NSRIM(1m) NSRIM(1m)

NAMEnsrim − NetWorker index management program

SYNOPSISnsrim [ −c client ] [ −N saveset] [ −x percent] [ −lnqvMX ]

DESCRIPTIONThe nsrim program is used to manage the NetWorker online file and media indexes. Normally,nsrim isinvoked bynsrmmdbd(1m) at startup, by thesavegrp(1m) command on completion, and bynsrd(1m)whenRemove oldest cycleis selected from the NetWorker Administrator program.nsrim is not normallyrun manually. Howev er, the command’s modes can be modified such that normal users may run the com-mand while retaining root privileges; seensr(1m) for more details.

nsrim usespoliciesto determine how to manage online entries. (Seensr_policy(5), nsr_client(5), and theLegato NetWorker Administrator’s Guidefor an explanation of index policies). Entries that have been in anonline file index longer than the period specified by the respective client’s browse policy are removed. Savesets that have existed longer than the period specified by a client’s retention policy are marked asrecyclablein the media index. When all of the save sets on a volume have been marked recyclable, then the volume isconsidered recyclable. Recyclable volumes may be selected by NetWorker (and automatically relabeled bya jukebox) when a writable volume is needed to hold new backups. When a recyclable volume is reused,the old data is erased and is no longer recoverable.

Unless the−q option is used,nsrim prints header and trailer information for each group of save sets. Theheader lists the save set type, the client name, the save set name, and the applicable browse and retentionpolicies that apply to the save set. (See the example in this man page). There are four types of save sets:

Normal All save sets backed up automatically usingsavegrpthat are associated with a schedule, a browsepolicy, and a retention policy.

Ad hocsUser-initiated save sets are designated by appendingad hocsto the header line.

ArchivesSave sets that never expire automatically are designated by appendingarchivesto the save set line.

MigrationsSave sets that never expire automatically, and were created by a file migration application, are des-ignated by appendingmigrationsto the save set line.

The trailer lists four utilization statistics of the save set afternsrim has applied the policies to it. The fourstatistics are the total number of browsable files remaining in the online index, the total of files currentlyassociated with the save set, and the amount of recoverable data out of the total of data associated with thesave set. For example,nsrim may print the following output for one save set name:

mars:/usr, retention policy: Year, browse policy: Month, ad hocs8481 browsable files of 16481 total, 89 MB recoverable of 179 MB total

mars:/usr, retention policy: Year, browse policy: Month, ad hocs0 browsable files of 13896 total, 163 MB recoverable of 163 MB total

mars:/usr, retention policy: Year, browse policy: Month 43835browsable files of 427566 total, 6946 MB recoverable of 7114 MB total

When the−v option is used, the following information is also printed for each save set: the save set id, cre-ation date, level, file count, size, and status. A sav e set’s status is one of the following:

browse The file entries for the save set are browsable (the save set files still exist in the online index).These files are easily restored using the NetWorker recover mechanisms.

recover The age of the save set does not exceed the retention policy for the save set, but its entries havebeen purged from the NetWorker online index. This means that save set is recoverable from the


NSRIM(1m) NSRIM(1m)

backup media usingrecover. (Seerecover(1m).) scanner(1m) may be also be used to recoverthe save set, but users should userecoverfirst.

recycle The save set is older than its associated retention policy and may be overwritten (deleted) once itsbackup media is recycled. Until the media is recycled, the save set is also recoverable from thebackup media.

delete The save set will be deleted from the media database.nsrim deletes only recyclable save sets thathave zero files.

The save set status may be followed by any of the following modifiers:

(archive)The save set never expires, and is exempt from any status change.

(migration)The save set was created by a file migration application and never expires, and is exempt from anystatus change.

(scanned in)The save set was restored using thescannercommand, and is exempt from any status change.

(aborted)A sav e set of questionable size, consuming backup media space.

If nsrim changes the status of a save set, then it prints the transition symbol−> followed by the new status.For example:

17221062 3/05/92 f 23115 files 158 MB recycle17212499 3/19/92 f 625 files 26 MB recover(aborted)->recycle17224025 5/23/92 i 0 files 0 KB recover->recycle->delete17226063 6/05/92 f 3115 files 58 MB recover17226963 6/09/92 f 3197 files 114 MB browse->recover17227141 6/10/92 f 3197 files 115 MB browse

Oncensrim has processed all of the save sets, it flags the file index for cross-checking innsrindexd(1m).If the −l flag is specified, the cross-check is attempted synchronously, otherwise, it is simply scheduled andnsrindexd performs the cross-check when the index is idle. At the same time,nsrim processes the statusof any affected NetWorker volumes. With the absence of the−q flag, a line is printed for each volume.The line includes the volume name, the amount of space used, the total number of save sets, and the status.The status will be one of the following:

appendableMore save sets may be appended to the volume. The status may also be modified with(currentlymounted)which signifies that the volume could transition to therecyclablestate if it was notmounted for writing.

read-only, fullNo more save sets can be appended to the volume, nor can the volume be reused since it containssome valuable save sets.

recyclableNo more save sets can be appended to the volume, and all save sets on the volume have expired.

In addition, the following modifier applies to all three of these states:

(manual-recyclable)The volume will not be automatically eligible for recycling when all of its save sets have expired.Instead, the volume may only be recycled by a manual relabel operation. Note that aread-onlyvolume can still be recycled unless themanual-recyclableflag is also set. The manual-recyclableflag can be set using the NetWorker Administrator GUI (nwadmin(1m)) or thensrmm(1m) andnsrjb (1m) commands when volumes are labeled or at any time thereafter. This flag is never set


NSRIM(1m) NSRIM(1m)

automatically.

If the volume status changes, thennsrim appends−>recyclable to the status. If the volume contains somebrowsable save sets, then this is noted; recoverable save sets are also noted. The odd case where anappendable volume has only recyclable save sets is also noted. For example:

jupiter.20: 3474 MB used, 398 save sets, full->recyclablejupiter.21: 4680 MB used, 440 save sets, full, 249 recoverablejupiter.22: 4689 MB used, 351 save sets, full, 351 browsablejupiter.24: 1488 MB used, 141 save sets, appendable, 141 browsable

RETENTION AND BROWSE POLICIESUnder normal circumstances, the association between browse or retention policies and client save sets isobvious. However, since a save set may be listed by more than one client resource with the same name, andeach client resource may specify different browse and retention policies, determining the policies applica-ble to a save set is not always straight forward.nsrim(1m), uses the following steps to select an instance ofa client resource with the client’s name. Once the client resource is selected, its browse or retention policyis used for managing information about the save set.

1) Locate all the client resources which belong to the same group that the save set belongs to. Withinthis set of client resources, apply the following rules to get the best match. If no client resourcebelongs to the save set’s group, or if the group no longer exists, or if the saveset is from a backup ear-lier than version 5 (when group information was not recorded in the save set), apply the followingrules to all the client resources to get the best match.

2) Locate a client resource explicitly listing the save set. If more than one client resource lists the saveset, choose the client resource with the longest policy.

3) Search for a client resource listing the save set "All". If more than one client resource lists the saveset "All", choose the client resource with the longest policy.

4) Find the client resource listing a save set with the most common prefix (longest) of the target saveset. If more than one client resource lists the save set with the most common prefix, choose the clientresource with the longest policy.

5) Among all of the client resources, choose the client resource with the longest policy.

Note that if two or more client resources with the same name exist, it is possible that the browse policyfrom one instance of the client resource and the retention policy of another instance may be used for man-aging save set information.

Save sets that have no corresponding NetWorker client resource use the NetWorker client resources of theserver to determine the browse or retention policies.

A sav e set cannot be purged from the index or marked for recycling until all of its dependent save sets arealso eligible for purging or recycling. See theLegato NetWorker Administrator’s Guidefor an explanationof dependent save sets.

The last (and only) Full save set will not be purged from the online index until it is also marked for recy-cling. In this case, the header line of the save set omits the browse policy and prints a message stating thatonly one browsable cycle exists.

With the exception of the−l option, manual ad hoc save sets are treated as full save sets that have no depen-dents. However, unlike true Full save sets, the last manual save set is not given any special considerationwith regard to index purging.

The retention time applied to save sets is rounded up to midnight when the elapsed time implied by thepolicies is greater than or equal to a day. Therefore,nsrim should produce the same results whether it isrun at 8 a.m. or 5 p.m. on the same day.


NSRIM(1m) NSRIM(1m)

OPTIONS−c client

Only process the online file index for the specified client. Normally, all client indexes are pro-cessed. This option may be repeated to process multiple clients.

−l Removes the oldest full save and all save sets dependant on it from the online index. Browse andretention policies are ignored. The save set header information will print the number of browsablefull cycles currently in the online index.Archiveandmigration save sets are ignored. With thisoption,manualsave sets are treated as normalincrementalsave sets. This option also sets the uti-lization threshold to 30 percent.

−M Master mode (not advised for manual operation). Advisesnsrim that it is being run bynsrd(1m)or another NetWorker daemon and that it should log messages with timestamps, and perform anyother behavior expected bynsrd.

−N save setProcess only save sets named; all others are skipped. This option can be repeated to process multi-ple save sets.

−n Do nothing. Instead, emulate the actions of this command without the index cross-check. Notethat trailer statistics reflect current (and not emulated) results.

−q Run quietly. This option will not generate header, trailer or save set messages.

−v Produce a more detailed report. This may produce a large amount of output. When both−v and−q are issued, they cancel each other.

−X Check the consistency of the data structures of the save set with the data structures of the volume.This is only required after a NetWorker crash. This option also sets the utilization threshold to 30percent.

−x percentSets the utilization threshold. If, after removing entries, the utilization of an online file index isless than the specified amount, the index is compressed automatically by passing this percentage tonsrindexd when requesting a cross-check. The default value is 50 (percent). Note that specifying−X or −l changes the default to 30 (percent).

FILES/nsr/tmp/.nsrim

nsrim locks this file, preventing more than one copy of itself from thrashing the media database.

SEE ALSOnsr_client(5), nsr_layout(5), nsr_policy(5), nsr(1m),nsrd(1m),nsrindexd(1m),nsrmm(1m),nsrm-mdbd(1m),nwadmin(1m), recover(1m),savegrp(1m),scanner(1m).

DIAGNOSTICSYou are not authorized to run this command

Only root or NetWorker administrators may runnsrim to modify the online indexes. However,any user may invoke the command with the−n option.

nsrim has finished (cross) checking the media dbThis notification message appears in the NetWorker messages window whennsrim completes andthe command was invoked with the−q option, but without the−c and−N options.


NSRINDEXASM(1m) NSRINDEXASM(1m)

NAMEnsrindexasm − NetWorker module for recovering indexes

SYNOPSISnsrindexasm[standard-asm-arguments]

DESCRIPTIONThe nsrindexasm is a standard, externalASM (Application Specific Module). It assists in the recovery ofNetWorker on-line save record index files that were saved with NetWorker versions earlier than version 6.

Seeuasm(1m) for a general description ofASM and the [standard-asm-arguments]. It is intended thatnsrindexasmbe invoked byuasmduringnsrck(1m) index recovery operations.

FILES/nsr/index/clientname/db6

This is directory whose data is recovered by thisASM.

SEE ALSOnsr_layout(5), nsrck(1m),nsrindexd(1m),mmrecov(1m),uasm(1m).


NSRINDEXD(1m) NSRINDEXD(1m)

NAMEnsrindexd − NetWorker file index daemon

SYNOPSISnsrindexd

DESCRIPTIONThe nsrindexd daemon is started by the servernsrd(1m) daemon. It should not be started by hand. Thedaemon provides an RPC-based service to the servernsrd(1m) daemon; direct network access to this ser-vice is not allowed. The RPC program and version numbers provided bynsrindexd are 390105 and 4,respectively.

The service provided to the NetWorker system is designed for high performance insertion and deletion ofsave records into indexes. This performance is obtained by keeping information cached in thensrindexdprocess address space. When the NetWorker system wishes to commit a save session’s records, it notifiesthensrindexd daemon (via a remote procedure call) to flush its volatile state to its file(s).

Since the daemon (or the server machine) may crash at any time, the index files may be left in an inconsis-tent state. Therefore, the maintenance program,nsrck(1m) is run automatically by thensrd daemon beforethe NetWorker service is started.

Sincensrindexd andnsrck are run at the same time, both programs use an advisory file locking mecha-nism on the filedb.SCAVENGE to synchronize their access to an index.

FILES/nsr/index/clientname/db Where the client’s index records are stored and accessed.

/nsr/index/clientname/db.SCAVENGEWhen this file exists andnsrindexd is not running, thensrck program must berun beforensrindexd is restarted.

SEE ALSOnsr_layout(5), nsr(1m), nsrck(1m), nsrd(1m), nsrim(1m), nsrindexasm(1m), nsrls(1m), nsrmm(1m),nwadmin(1m).

DIAGNOSTICSwaiting for lock on filename.

This message indicates that another program is accessing the same file that is required by this dae-mon. The daemon will wait for the advisory lock to be cleared.

lock on filenameacquired.Informative message that will eventually follow the previous message.


NSRINFO(1m) NSRINFO(1m)

NAMEnsrinfo − NetWorker file index reporting command

SYNOPSISnsrinfo [ −vV ] [ −sserver| −L ] [ −n namespace] [ −N filename] [ −t time] [ −X application] client

DESCRIPTIONThe nsrinfo command generates reports about the contents of a client file index. Given a required Net-Worker client name and no options,nsrinfo will produce a report of all files and objects, one per line, in thebackupname space for that client. It can also generate reports as follows: for a specific file index namespace, for all name spaces at once, or for a particular XBSA application. Reports can also be restricted to asingle time (the time at which the entry was entered into the file index, called thesavetime).

For example, to generate a report of all files backed up in the most recent backup of the/usr file system forthe clientmars, use the following sequence of commands (assuming the% character is the shell prompt):

% mminfo −r nsavetime −v −N /usr −c pegasus −ot | tail −1809753754% nsrinfo −t 809753754 mars

Note: The time used in the query is obtained by running themminfo(1m) command with a custom report toprint the save time for the most recent save set for/usr. The time printed is passed tonsrinfo along with thename of the client (mars).

Unless you use the−L option, you must have NetWorker Administrator privilege to use this command. Inthe case of−L , you must be a system administrator (that is, root on a UNIX system).

OPTIONS−v Verbose mode. In addition to the filename, it prints the type of the file, the internal file index iden-

tifier (if any), the size (if a UNIX file), and the savetime. This option may be combined with the−V option.

−V Alternate verbose mode. In addition to the filename, it prints the offset within the save set contain-ing the file, the size within the save set, the application name space (see the−n option for a list ofvalues), and the save time. This option may be combined with the−v option.

−sserverIndicates the name of the NetWorker system to be queried. By default, the server on the local sys-tem is queried.

−L Opens a file index directly without using the server. This option is used for debugging, or to querythe file index while NetWorker is not running.

−n namespaceIndicates the file index name space to query. By default thebackupname space is used. The otherrecognized values are:migrated, archive(reserved for future use),nsr (for internal use),informix(for INFORMIX data),sybase(for Sybase data),msexch(for Exchange data),mssql (for SQLServer data),notes(for Lotus Notes data),db2 (for DB/2 data),oracle (for Oracle data), andall .The name space field is case sensitive.

−N filenameIndicates an exact filename to look for in the file index. Only index entries matching this nameexactly print. Note that for some clients, such as NetWare, the name stored in the file index isoften not made up of printable ASCII characters, giving this option limited use.

−t time Restricts the query to a single, exact save time. The time can be in any of the NetWorkernsr_getdate(3) formats. Every save set created by NetWorker has a unique save time; these timescan be determined by using themminfo(1m) command.

−X applicationRestricts the query to list information for only a specific X/Open Backup Services (XBSA) appli-cation. Valid application types areAll, Informix, andNone. The application type is not case sen-sitive. See the APPLICATION TYPES section of this man page for more information.



FILE TYPESThe file index can store entries for all types of clients. Each index entry includes an index entry type. Ingeneral, only the client that created the index entry can decode the entry.

This section lists index entry types recognized bynsrinfo. Howev er, even though these types are recog-nized,nsrinfo can only completely decode one entry type: the UNIX version decodes UNIX entry types,and the NT version decodes NT entry types. For other recognized types, some information may be incom-plete.

old UNIX Clients running versions earlier than 3.0 of NetWorker for UNIX.

UNIX Clients running versions earlier than 4.0 of NetWorker for UNIX.

UNIX ASDF Index entries including extended ASM Structured Data Format (ASDF) informationfor clients running versions 4.1 and later of NetWorker for UNIX.

UNIX ASDF v2 Index entries from agentless saves for clients running versions 4.2 and later of Net-Worker for UNIX.

UNIX ASDF v3 Index entries for large files (files > 2 gigabytes) for clients running versions 5.1 forUNIX and later NetWorker for UNIX.

old DOS DOS clients running versions 2.0 and earlier of NetWorker for DOS.

DOS DOS, Windows, or OS/2 clients running version 2.0 of NetWorker for DOS, Win-dows, or OS/2.

DOS old ASDF DOS, Windows, or OS/2 clients running version 2.0 of NetWorker for DOS, Windowsor OS/2.

WIN ASDF Windows or NT clients running NetWorker for Windows NT 4.2 and above.

WIN ASDF v2 Windows or NT clients running NetWorker for Windows NT 4.2 and above, createdby using agentless saves.

old NetWare NetWare clients running version 3.0 and earlier of NetWorker for NetWare.

NetWare NetWare clients running version 3.0 and later of NetWorker for NetWare 3.0.

OSF 64bit A client running OSF/1 with 64bit file sizes and offsets.

continuation A special internal index entry, that is generated when a file crosses save set boundariesin a save set series.

APPLICATION TYPESAll This application type prints out all of the X/Open Backup Services API (XBSA) infor-

mation available for each object; only XBSA objects are printed. The-v and-V flagshave the same effect here as they do on files.

Informix This application type prints out only those objects recognized as Informix Databaseobjects (XBSA ObjectOwner.bsaObjectOwner is INFORMIX). The-v flag behaves asit does with files, while the-V flag prints out all the XBSA information about theobject (see All, above), including the normal-V information.

None This application type prints out objects that arenotXBSA objects, but match the givencriteria. For example, this option can be used to print a list of files backed up from aclient.

FILES/nsr/index/client/db6

SEE ALSOnsr_getdate(3), mminfo(1m),nsrck(1m),nsrindexd(1m).

DIAGNOSTICS



bad time value ‘time’The time value specified in the−t option is not in a validnsr_getdate(3) format.

cannot open index for clientclient: reasonThe file could not be opened using the−L option. The specific reason is printed, although theremay be several. The most likely reasons arepermission deniedif the user is not the superuser, andservice busy, try againif the file index is already locked (for example, bynsrindexd(1m)).

cannot create db scan onclientAn internal error occurred while attempting to query the file index. Contact Legato TechnicalSupport.

numberbad records for clientclientThis diagnostic prints at the end of a report if any bad index records were detected. This is a signthat the index is damaged, and may need to be recovered.

cannot connect to serverserverThe index server is not available for one of many reasons. For example, the NetWorker servermay be down, ornsrinfo may not be able to connect to a running server due to either a resourceshortage or a network problem.

cannot start session with serverserverThe index server is running, but refused the connection. The exact reason is printed on the subse-quent line of output. The most likely reasons arepermission deniedif the user is not a NetWorkeradministrator, andservice busy, try againif the file index is locked (for example, bynsrck(1m)).

lookup failed to serverserverThe index server is running, but was unable to process the query. The exact reason is printed onthe subsequent line of output.

LIMITATIONSThe command line options should be made as powerful as those ofmminfo(1m).

The−v and−V reports are not formatted into columns.

A query for a specific time can take a very long time due to the schema of the file index.

The queries are limited due to the lack of a cross-platform browser.


NSRJB(1m) NSRJB(1m)

NAMEnsrjb − NetWorker jukebox control command

SYNOPSISnsrjb [ −C ] [ −j name] [ −sserver] [ −v ] [ −f media device] [ −Sslots| −T Ta gs| volume names]

nsrjb−L [ −j name] [ −s server] [ −gimnqvMG ] [ −Y | −N ] [ −R | −B ] [ −b pool ] [ −f media device|−J hostname] [ −e forev er ] [ −c capacity] [ −o mode] [ −S slots | −T tags | −W current pool|volume names]

nsrjb−l [ −j name] [ −s server] [ −nvqrMG ] [ −R [ −b pool ] ] [ −f media device| −J hostname] { −Sslot | −T tags| −W current pool[ −D volume name] [ −e forev er] | volume names}

nsrjb−u [ −j name] [ −sserver] [ −qvM ] [ −f media device] [ −Sslot | −T tags| volume names]

nsrjb−I [ −j name] [ −sserver] [ −Evpq ] [ −f media device] [ −Sslots| −T tags]

nsrjb−p [ −j name] [ −sserver] [ −vq ] [ −f media device] −Sslot | −T tag | volume name

nsrjb−o mode[ −j name] [ −sserver] [ −Y ] { −Sslots| −T tags| volume names}

nsrjb−H [ −j name] [ −sserver] [ −EHvp ]

nsrjb−h [ −j name] [ −sserver] [ −v ]

nsrjb−U uses[ −j name] [ −sserver] [ −Sslots| −T tags]

nsrjb−V [ −j name] [ −sserver] [ −v ]

nsrjb−d [ −j name] [ −sserver] [ −v ] [ −N ] [ −P ports] [ −Sslots] [ −T tags] [ volume names]

nsrjb−w [ −j name] [ −sserver] [ −v ] [ −N ] [ −P ports] { −Sslots| −T tags| volume names}

nsrjb−a [ −j name] [ −sserver] [ −vd ] [ −A media type] { −T tags| [ −T tags] volume names}

nsrjb−x [ −j name] [ −sserver] [ −vwX ] −T tags

nsrjb−F [ −j name] [ −sserver] [ −v ] −f media device

DESCRIPTIONThe nsrjb program manages resources in two broad classes of jukeboxes,remotely managed jukeboxesand locally managed jukeboxes. Remotely managed jukeboxes are controlled through an external agent.nsrjb communicates with this agent to gain access to jukebox resources. The agent allows multiple appli-cations, including multiple NetWorker servers, to share resources in the jukebox. Examples of agents are


NSRJB(1m) NSRJB(1m)

SmartMedia and StorageTek’s ACSLS . nsrjbcommunicates directly with a locally managed jukebox,there is no intervening agent. Resources in a locally managed jukebox can be used by only one Networkerserver.

For a locally managed jukebox, the jukebox resource is used to track the state of the entire jukebox. Theresource records the number of drives and slots in the jukebox. It is also used to track whether devices areloaded, whether there is media residing the slots, and the name of any volume on the media, as well asother information. See nsr_jukebox(5).

The jukebox resource for a remotely managed jukebox does not reflect the current state of the entire juke-box, only NetWorker’s view. Media must be allocated, before NetWorker may use media in a remotelymanaged jukebox the media. For more details, see the description of the−a option. The number of slots ina remote jukebox resource increases as media is allocated for NetWorker’s use and decreases as media isdeallocated after NetWorker has no further use for the media. The order in which media is listed in thejukebox resource does not necessarily reflect physical location within the jukebox. The number of drivesin a remote jukebox is an upper bound on the number of volumes in the jukebox that NetWorker mayaccess simultaneously.

The nsrjb command is used to manage all jukeboxes for a NetWorker server. Use this command, ratherthannsrmm(1m), to label, load, and unload the volumes contained within a jukebox. Multiplensrjb com-mands may access a jukebox at any giv en time. Eachnsrjb program determines the resources in a jukeboxthat the command will require, and locks all needed resources before execution begins. For additionaldetails see the description of the jukebox attributes,slot/volume locks , drive locks ,andjukebox lock , innsr_jukebox(5).

Manually enterednsrjb commands which require use of jukebox resources do not directly perform therequested operation. Instead a manually run nsrjb makes a request of the NetWorker server process, whichassigns a lockid to the command and starts another instance ofnsrjb which performs the requested opera-tion. The server process starts the secondnsrjb with the−O option added to the original options to servicethe request. Users should never use the−O option on a manually entered command. This may result inlocks for resources in a jukebox resource not being released when the command terminates. The onlymethod for clearing such locks is to stop and start the NetWorker server process.

A volumeis a physical piece of media, for example, a tape cartridge or optical disk. Each volume within ajukebox and each jukebox has a name recognized by NetWorker. Avolume nameis specified when the vol-ume is first labeled by NetWorker. You can change the volume name when a volume is relabeled. Net-Worker refers to volumes by their volume names. For example, when requesting the mount of a volume,NetWorker asks for it byvolume name.The volume name ’cleaning tape’ may be used to load a cleaningcartridge to clean a tape device.

Before usingnsrjb , the jukebox and its device resources must be added to the NetWorker server. Usejbconfig to add the jukebox resource and its device resources to the NetWorker server. The jukeboxresource is described innsr_jukebox(5).

When a NetWorker server requires a volume for backup or recovery and an appropriate volume is notalready mounted, the server checks the media database to verify whether a jukebox contains a volume thatsatisfies the media request. If so,nsrjb is invoked to load the media into an idle device. TheAvailableSlotsattribute specifies the slots containing volumes available to automatically satisfy NetWorker requestsfor writable volumes. When automatically selecting a writable volume for backup, NetWorker only consid-ers volumes from the list of available slots. It is important to note that theAvailable Slotsattribute does notlimit what slots the user runningnsrjb can operate on.

nsrjb attempts to determine which jukebox to use based on the options−j , −f , or avolume name .If oneor more of these options do not uniquely identify a jukebox and one must be selected, thensrjb programprompts you to select a jukebox. You can set theNSR_JUKEBOX environment variable to the name ofthe jukebox you want thensrjb program to use by default.

OPTIONSOptions are separated into two groups. The first are the options which specify the operation to be per-formed, e.g. label or load media. The second group list the additional options which provide arguments for


NSRJB(1m) NSRJB(1m)

the operation, e.g. specifying the media to be labeled or loaded.

OPERATION OPTIONS−a This option is used in conjunction with the−T tagsoption, to allocate volumes in a remotely man-

aged jukebox. A volume must be allocated before it can be labeled and used by a NetWorkerserver.

For STL silos a−d option can be added for silos that support depositing (also known as importingor entering) tapes from their I/O ports. The−d must appearafter the −a on the command line.This function is usually handled by the silo management software, but is added here for ease ofuse. This option may not be supported on all silos supported by NetWorker.

There are two types of media which may be allocated or added to a SmartMedia jukebox resource,scratch or foreign. The term scratch, is used to indicate volumes currently not being used by Net-Worker. A foreign tape is one that was being used by NetWorker before being imported intoSmartMedia.

Use−a in conjunction with−T tagsoption to allocate scratch volumes for NetWorker’s use. Byspecifying the barcode or physical cartridge label with this option a volumes from specific mediacartridges may be allocated. This option can also be used to allocate a given number of volumesfrom unspecified media.

After importing a foreign volume (an existing NetWorker volume) into SmartMedia, this option isalso used to inform NetWorker that the imported volumes are available through SmartMedia. Use−a, in conjunction with−T tagsand volume namesto add foreign media to a SmartMedia juke-box resource. Thetag is the name given to the volume when it was imported into SmartMedia.The volume nameis the volume name recorded in NetWorker’s media database.

See−x for a description of how volumes are removed from a remote jukebox’s list of volumesavailable for use by a NetWorker server.

−C Displays the current volumes in the jukebox and the devices associated with the jukebox. This isthe default command option, used if no other command options are specified. It displays a list ofslot numbers, volume names, media pools, optional bar code information, volume ids and volumemodes. If the jukebox attributeBar Code Readeris enabled and there are bar code labels on themedia volumes, then the bar code label is included in the list. IfBar Code Readeris set and thevolume does not have a bar code label, a dash prints, indicating that there is no bar code label onthe media. By default the short volume id of a volume is displayed. Using the verbose option (-v)displays the long volume id along with other information described below. The -C option doesnotperform an actual jukebox inventory;nsrjb only reports on the volumes currently contained withinthe jukebox resource. Volumes may be succeeded by one of the following flags: an(R), to indi-cate the volume is read-only; or an(A), to indicate the volume is either an archive or a migrationvolume. When combined with the−v option, the capacity of the volumes that have been filled isalso displayed. Volumes that are not contained in the NetWorker media database are marked withan asterisk, "*".

The Mode column contains additional information about the mode of the volume. The Mode fieldcan have one of three values:manually recyclableto indicate that the volume will not be automati-cally recycled or relabeled;recyclable, to indicate that the volume is eligible for automatic recy-cling; or blank to indicate that neither of the other two values apply.

After the slot map prints, a line about each device is displayed. For each enabled device, the fol-lowing information is provided: drive number, device pathname, slot number and name of the cur-rently loaded volume, and an indication if NetWorker has the volume mounted. If the device isdisabled, only the drive number and pathname are displayed, along with the messagedisabled.When several device resources share a physical drive in the jukebox, via the samehardware id


NSRJB(1m) NSRJB(1m)

attribute value, the drive number is only displayed on the first device pathname sharing the drive.

−d Deposits (loads into the jukebox) one or more cartridges from the cartridge access ports (alsocalled import/export elements, mail slots, or I/E ports).

The number of cartridges to deposit is determined by the number of specified slots or tags. Allempty slots in the jukebox are deposited, if slots or tags are not specified. Multiple destination slotranges may be specified, full slots are skipped. If all available import ports are empty and thereare cartridges to deposit, the operator will be prompted to fill the import ports. When the−Noption is used in conjunction with the jukebox polling feature, the jukebox will poll for cartridgesin the import ports until all of the cartridges are deposited or an error occurs. Exceeding thepolling timeout waiting for additional cartridges is considered an error.

Specifying volume names on the command line is not recommended. The inventory commandshould be run to accurately determine the volume names.

If −d is used with a−T tagsoption, then the command is assumed to be running on a silo, and istreated internally as if it had been run with the−a and−d options. Specified volume tags (bar-codes) will be deposited into the silo and then NetWorker will attempts to allocate them for its use.Depending on the exact type of silo used, this allocation step may or may not succeed. You shouldverify the success of the allocation, and retry the command with just the−a option for all of thetag values specified. If the tags have already been allocated, you will see a message indicatingthis. This is not an error, and only means that the volumes had already been successfully allocatedfor use by NetWorker.

−F Releases a shared device contained within an STL silo. This option is only available for tapelibraries with device sharing. Seensr_jukebox(5).

−h Displays the actions and results of the past 120 jukebox commands issued. These include com-mands issued on the command line by the user, ornsrjb commands that were started automati-cally by NetWorker. Starting with Version 5.5, manynsrjb commands issued from the commandline will not actually perform the requested jukebox options. Instead, the manually runnsrjbsends a message to the NetWorker server process, which will then start the requirednsrjb com-mand(s). Therefore, it is not unusual to see two entries generated in the history for each commandissued on the command line. Instances ofnsrjb that are started by the server have an extra com-mand line option set (−O <instance number>). If you wish to change the number of commandlines saved in the history, you may set the environment variableNSRJB_HISTORY_COUNT toa value between20 and 2000. Values smaller than 20 will result in 20 being used, and valueslarger than 2000 will result in 2000 being used.

−H Resets the jukebox hardware (and the NetWorker database representing the jukebox) to a consis-tent state. The jukebox clears the transport and then unmounts and unloads volumes from thedrives to slots. An actual inventory is not performed; (see the−I option). If the jukebox sensesthat the inventory is out-of-date, it prints an appropriate message.

For silos, only devices which NetWorker thinks are loaded are unloaded. You can use the silo con-troller to empty other drives.

For SmartMedia jukeboxes, resets the jukebox devices and the NetWorker database representingthe jukebox to a consistent state. The operation synchronizes the state of the devices in the juke-box and the media in the jukebox resource with SmartMedia.nsrjb queries SmartMedia for infor-mation about volumes in the jukebox resource and which volumes are currently mounted. It usesthis information to synchronize the jukebox and device resources to be consistent with the infor-mation reported by SmartMedia. If the-p option is also specified a check operation will be per-formed on the loaded volumes.

nsrjb with −H option is executed for every SmartMedia jukebox, by NetWorker whenever the


NSRJB(1m) NSRJB(1m)

server is started.

−I Performs an inventory on the jukebox’s contents. Use this option to ensure that the mappingbetween slot number andvolume nameis correct. If necessary, the volumes in the specified slotsmay be loaded into a device, so their labels may be read. This option can take a long time to com-plete. For jukeboxes that have element status capability (for example, EXB-120, EXB-60, or HPoptical), you can use the−E option in conjunction with the−I option to reinitialize the jukebox’sinventory state. The-E option increases the amount of time it takes to inventory a jukebox,because the hardware must check every component, including all slots and drives, for the presenceof media. You should only use this option if you are manually swapping media in or out of a juke-box.

If a jukebox has a bar code label reader, and the jukebox resource attributeBar Code Readeris set,thenvolume nameassociated with a slot is derived from the media bar code label. If the bar codelabel is not unique or does not exist in the NetWorker media database, thevolume nameis readfrom the media. If a bar code label on the media has changed, then the NetWorker media databaseis updated with the new bar code label. Proper use of a jukebox’s bar code reader can minimizethe time it takes to perform an inventory.

For SmartMedia jukeboxes, this operation is used to synchronize NetWorker and SmartMediadatabase. It insures that SmartMedia and NetWorker agree to the state of all volumes allocated tothis NetWorker server and listed in this jukebox resource. If the−p option is also specifiednsrjbrequests the volumes be loaded so that labels on each volume may be verified.

To allocate slots in a jukebox for cleaning cartridges, set the jukebox resource attributeAuto Cleanto Yesand theCleaning Slotsattribute to a non-empty range of slots. For further information seensr_jukebox(1m). Volumes from slots that are reserved for cleaning cartridges are not loadedduring the inventory of a jukebox. For jukeboxes that do not support element status or have a barcode reader, the−U usesoption must be used to enter a cleaning cartridge into the jukebox’sinventory. For jukeboxes that support element status or have a bar code reader, cleaning cartridgeslots that were previously empty but now contain a cartridge have the number of uses for thecleaning cartridge is the value set in the jukebox attributeDefault Cleanings.

−l Loads and mounts specified volumes. Volumes are specified by name, by the slot in which the vol-ume resides, or for remote jukeboxes by thetag associated with the volume. The operation fails, ifthe number of volumes specified is greater than the number of available drives.

For SmartMedia jukeboxes, the command may only be used to mount volumes into devices acces-sible from the storage node upon whichnsrjb is running. You can use, the−W option to load andmount a volume in a SmartMedia jukebox which belongs to the specified pool.

This option can also be used to clean a device by loading a cleaning cartridge. To load a cleaningcartridge use the volume name ’cleaning tape’ or specify a slot that has been set aside for a clean-ing cartridge. When the volume name ’cleaning tape’ is used to load a cleaning cartridge, andmore than one cleaning cartridge in the jukebox has any uses left, the cleaning cartridge selected isthe one with the fewest remaining uses. See−I for a discussion of how the slots of a jukebox areset aside for cleaning cartridges.

The−f option can be used to specify a media devices into which volumes are loaded.

−L Labels the volumes in the specified slots, or for remotely managed jukeboxes, by specified tags.Names for the volumes labeled are derived from media bar code labels,volume namesspecified onthe command line, or generated by referencing thelabel templateresource for the givenpool. Ifyou do not specify any slots, the range of slots is as described in theNSR_jukebox resource forthe jukebox. Labeling a complete jukebox may take a long time.


NSRJB(1m) NSRJB(1m)

If the jukebox has a bar code label reader, and theNSR_jukebox resource attributesBar CodeReaderandMatch Bar Code Labelsare set, then the volume label is derived from the bar codelabel on the media. If the jukebox resource attributeMatch Bar Code Labelsis not set, or thejukebox does not have a bar code reader, then the volume label is derived fromvolume namesspecified on the command line. If more volumes are being labeled then volume names specifiedon the command line, then the volume label is derived from the label template. No matter how thevolume label is derived, if the media labeled has a media bar code label, the bar code is stored inthe NetWorker media database so that it can be used during inventory operations.

Volumes located in slots set aside for cleaning cartridges cannot be labeled. See−I for a discus-sion of how the slots of a jukebox are set aside for cleaning cartridges.

If an empty slot is encountered, an informational message is displayed and the operation contin-ues.

See the−m option if you want the volume to be automatically mounted after being labeled.

−o modeSets the mode of a volume or range of slots. The following mode values are available:[not]recyclable, [not]readonly, [not]full, or [not]manual. The [not]manual mode values areonly valid when used in conjunction with the−L option. If the−Y option is not used, you areprompted to confirm the operation for each volume. Seensrim(1m) for a discussion of the per-volume flags.

−p Verifies and prints a volume label. A slot or for remotely managed jukeboxes a tag may be speci-fied. The device used to read the volume may also may be specified. Seensrmm(1m).

−u Unloads a volume from a device. To unload a volume from a device, specify the name of the vol-ume, the device in which the volume is loaded, or the slot from which the volume was loaded. Ifno volume, device or slot is specified, media is unloaded from all loaded devices.

−U usesSets the number of times a cleaning cartridge can be used. Slots can also be specified. Any slotspecified must be in the range of slots set aside for cleaning cartridges in the jukebox. If a range ofslots is not specified, all slots set aside for cleaning cartridges are updated. For slots that are cur-rently empty in the jukebox’s inv entory, this option updates the inventory to indicate that the slot isoccupied by a cleaning cartridge. For a discussion of how slots of a jukebox are set aside forcleaning cartridges, see−I.

Usesmust be either a positive integer, or the reserved wordsremoveor default. The reserved wordremovecan be used (for example,-U remove) to delete the cleaning cartridge(s) from the Net-Worker inventory. Specifyingdefaultsets the number of times a cleaning cartridge may be used tothe value of thedefault cleaningsattribute for the jukebox. Seensr_jukebox(5).

You can use the−T option in conjunction with the−U option to add cleaning cartridges to a SiloTape Library (STL). This option sets aside a cleaning slot in the STL each time a cleaning car-tridge is added. For a description of how to remove cleaning cartridges from an STL, see−x. See−I for a discussion of how slots in a non-STL jukebox are set aside for cleaning cartridges.

−V Displays vendor-specific status information. When combined with the−v option, the configura-tion of the jukebox is displayed.

−w Withdraws (ejects media from the jukebox) one or more cartridges to the cartridge access ports.

Cartridges must be specified by slot, volume name or tag. Multiple slot ranges and volume namesmay be specified, empty and duplicate slots are ignored. If the available export ports are full andthere are cartridges to withdraw, the operator will be prompted to empty the export ports. Whenthe −N option is used in conjunction with the jukebox polling feature, the jukebox will poll for


NSRJB(1m) NSRJB(1m)

empty export ports until all cartridges are withdrawn or an error occurs. Exceeding the pollingtimeout waiting for empty ports is considered an error.

If −w is used with a−T tagsoption, then the command is assumed to be running on a silo, and istreated internally the same as if it had been run with the−x and−w options. Specified volume tags(barcodes) are withdrawn from the silo. Then NetWorker deallocates them from its list of volumesfor that silo. In general, you can only withdraw at most about 40 volumes from a silo at one time,although this limit differs on different silo models. If a given command does not cause any tapesto be withdrawn from the silo, try again using fewer tag values on the command line.

−x This option, when used in conjunction with the−T tagsoption, is used to remove volumes from aremote jukebox. The specified volumes are removed from the remote jukebox’s list of volumesavailable for use by a NetWorker server.

For STL silos, a−w option can be added to withdraw or eject tapes from the silo or tophysicallyremove the tapes from the silo. The−w must appearafter the−x on the command line. This func-tion is normally handled by the silo management software, but is added here for ease of use. Thisoption may not be supported on all silos supported by NetWorker.

See−a for a description of how volumes are allocated for use by a NetWorker server.

ADDITIONAL OPTIONS−A media type

This option may be used n conjunction with−a option to specify the type of media on which vol-umes allocated for use by a server may reside. Valid values for media type, include all choices forthe deviceresource attributemedia typeexceptlogical, seensr_device(5). This option is sup-ported only for SmartMedia jukeboxes and may appear multiple times on a command line.

−b poolSpecifies the media pool to which the volume should belong. The pool may be any pool currentlyregistered with the NetWorker server. The pool names can be viewed by selecting the Pools menuitem from the Media menu ofnwadmin(1m). The pool name is referenced by the NetWorkerserver when determining what save sets can reside on the volume. If you omit this option the vol-ume is automatically assigned to theDefault pool. If you specify a pool name without a volumename,nsrjb will use the next volume name associated with the specified pool’slabel templateresource. Seensr_label(5).

−c capacityOverrides the volume’s default capacity. Seensrmm(1m).

−B Verifies that the volume currently being labeled does not have a readable NetWorker label. Beforelabeling a volume, NetWorker attempts to read any existing labels written on the volume. If youspecify this option and the volume has a NetWorker label that is readable by the device currentlybeing used, the label operation is canceled and an error message is displayed. If the volume doesnot have a label, or has a label that is not readable by the current device, then the volume can belabeled. This option is used bynsrd(1m) to label volumes automatically whennsrmmd(1m)makes a request for a volume while saving data.

−D volume nameUsed with the-l, option when mounting a volume from a particular pool to exclude this volumefrom consideration. This option is supported for SmartMedia jukeboxes only and may appearmultiple times on a command line.

−e forev erSpecifies the volume to be an Archive volume. (seensrmm(1m)).

−E Initializes element status for jukeboxes that support this feature. You can use this option in con-junction with the−I or −H options. Some jukeboxes have the ability to keep track of whether or


NSRJB(1m) NSRJB(1m)

not there is media in a component in the jukebox. This feature is known as an "element status"capability. The−V option may be used to determine whether a jukebox has this capability. Whenswapping media into the jukebox where media was not previously loaded, it may be necessary toreinventory(−I) the jukebox with the−E option so the jukebox reinitializes its element status.

−f media deviceSpecifies a media device to be used for an operation. Use the pathname of the media device as it isconfigured in the jukebox resource. When more than a single media device has been configuredfor a jukebox,nsrjb selects available devices with the lowest value for the device resourceattributeaccesses.Seensr_device(5). When loading or verifying volumes, the number of devicesavailable must at least be greater than or equal to the number of volumes specified for the opera-tion. For other operations, the value of the jukebox attributemax parallelism is an upper boundon the number of devices that may be used by any nsrjb command. Seensr_jukebox(5). You canoverride the device selection by using the−f option. You can use this option multiple times, tospecify more than one media device.

For SmartMedia jukeboxes, the device resource is not tied to a physical device. It is a logicaldevice resource. An association between this logical device and the physical device lasts as long asmedia is loaded in a device. NetWorker never asks SmartMedia to load media into a particulardevice. It allows SmartMedia choose the device into which the media is loaded. Thennsrjb cre-ates an association between the actual device and NetWorker logical device resource by assigningvalues to the device’slogicalname,logicaltype, andlogicalfamily attributes. Seensr_device(5).SmartMedia and NetWorker have different names for device and media types.nsrjb maintains atable to map between SmartMedia and NetWorker names to be able to correctly set the values ofthese attributes. This table can be updated dynamically to support additional SmartMedia driveand/or media types. The file/nsr/res/smdevmap.txt is used to make additions to nsrjb’s maptable. Each line in this file contains four columns, SmartMedia cartridge type, SmartMedia bitfor-mat, NetWorker device resource media type, and NetWorker device resource family type. TheSmartMedia bitformat maybe a regular expression, all other values are strings. As an example theline,

DTL7000 DLT8000.* DLT8000 tape

maybe used for the DLT8000 device using SmartMedia DLT7000 cartridge type.

−g This option is kept for historical reasons only. It has no affect.

−G This option is used only by the server to have the autoloader mount or label a volume in a NetworkData Management Protocol (NDMP) device.

−i This option is kept for historical reasons only. It has no affect.

−j nameSpecifies a particular jukebox to use. The givennameis the one assigned by the user when thejukebox resource is created. This option overrides theNSR_JUKEBOX environmental variable.

−J hostnameSpecifies a particular hostname to use. Drive selection bynsrjb will be restricted to a drive on thegiven hostname.This option can be used with the−l (load) or−L (label) options, and cannot beused with the−f option.

−m Mount a volume after it has been labeled. There must be enough available drives to mount all vol-umes to be labeled.

−M Sends messages to the NetWorker daemon reporting progress and errors. This is used bynsrd(1m) when mounting, unmounting, and labeling volumes in response to requests made bynsrmmd(1m). This option is ignored ifnsrjb is run manually.

−n Loads, but does not mount, the volume when specified with the−l option. NetWorker is not noti-fied that this volume has been loaded.


NSRJB(1m) NSRJB(1m)

−N Tells nsrjb to skip the confirmation prompt when used in conjunction with the−LR options.When NetWorker recycles volumes, NetWorker prompts you to confirm that it is okay to overwriteany volumes considered to be nonrecyclable. Seensrim(1m) for a discussion of the per-volumeflags.

−O instanceIndicates the lock id used by the command when locking resources in a jukebox. Should only beused with a command run by the server. You should never specify this option on a command line.Doing so may result in resource locks never being released once the command exits. The only wayto clear such locks is to stop and start the NetWorker server.

−PportsSpecifies a cartridge access port or range of ports to deposit or withdraw volumes.

Ranges are specified aslow to high. Both low andhigh must be integers;low must be less than orequal tohigh. Both numbers are checked for validity against the resource describing the jukebox.You can specify only one port range for a command.

−q Runs thensrjb program in quiet mode. Turns off all of the messages normally produced whenverifying, labeling, loading, or unloading volumes, or inventorying a jukebox. You can use thisoption only with the−p −L, −l, −u or −I options.

−r Loads the volume as read-only. You can use this option only with the−l option. Seensrmm(1m).

−R Recycles the volume. If a volume is recyclable, you are not prompted for confirmation as towhether or not this volume may be overwritten. Seensrmm(1m) for a discussion of the per-volume flags.

−sserverSpecifies the controlling server whennsrjb is used on a storage node. To usensrjb on a storagenode, the command must be run on the storage node. Seensr_storage_node(5) for additionalinformation on storage nodes.

−SslotsSpecifies a slot or range of slots on which to operate. Specify the slot range from low to high inte-ger order. Bothlow andhigh must be integers;low must be less than or equal tohigh. Both num-bers are checked for validity against the resource describing the jukebox. You can specify multi-ple slot ranges for a command.

−T tagsSpecifies tags or barcodes of volumes in a remote jukebox. You can specify this option more thanonce for a command.

tags can specify a single volume tag or a volume tag template similar to a label template. Seensr_label(5). The volume tag Template is a list of template fields separated by slashes "/". A tem-plate field is a constant alphanumeric string or an alphabetic or numeric range represented by thelow and high value separated by "-".

This template differs from the templates used in NetWorker GUI. Each portion of the template isentered into a separate line in the GUI’s dialog box instead of using "/" as a separator.

The tag is used to identify the media when a request is made of the agent managing the remotejukebox. This identifier is determined by the remote agent. A tag often is a bar code label. Whenmaking a request to load media into a device, NetWorker sends thetag with the request to theagent to identify the media to be loaded. Volumes in a jukebox resource are listed in alpha-numeric order of their tags. Therefore, the order in the jukebox resource may change as media isallocated and deallocated, and has no relation to the slot in which the media may reside in a physi-cal library.


NSRJB(1m) NSRJB(1m)

−v Verbose. See other arguments for specific details.

−W current poolThis option is supported only for SmartMedia jukeboxes.

When labeling a volume, this options designates the pool from which a recyclable volume is to beselected. Only to be used in conjunction with the−L and−R options. When SmartMedia selectsthe volume to be labeled, it currently must be recyclable and in thecurrent pool . The volume willbe added to the pool specified by the−b option.

For the load operation, this option is used to mount an unspecified volume that belongs to thecur-rent pool . In such cases, SmartMedia selects which volume is to be loaded from all volumeswhich possess the required characteristics. This feature is used when mounting a volume for writ-ing data, e.g. saving or cloning data.

−X You can use this option in conjunction with−x to purge a volume from NetWorker’s mediadatabase when the volume is being deallocated. A prompt is displayed to confirm that the volumeis to be purged from the media database, unless−Y is also specified.

−Y Disables confirmation prompting. Rather than prompting for confirmation, ayes answer isassumed. Prompts are normally generated when a volume is being relabeled before its expirationdate, or when a volume is still registered in the NetWorker media database. If the operation is tolabel (−L ) a volume or to load (−l ) a volume, with the−R option also specified, and the volumeis recyclable, there is no prompt to confirm whether the volume may be overwritten.

volume nameSpecifies the name to be used when labeling a volume. After a volume has been labeled, the vol-ume name is used to select media for an operation. Multiple volumes names may be specified fora single command, and must come at the end of the command line.

EXAMPLESLabeling volumes:

To label all of the volumes in a jukebox, use the−L option:nsrjb −L

To specify a particular pool, use the−b option:nsrjb −L −bOffsite

Labeling the volumes in slots 5 through 19:To label the volumes in slots 5 through 19, use the−Soption:

nsrjb −L −S 5−19

Labeling a volume with a non-standard name:To label the volume in slot 20 with a name that does not match the label template associated with apool, specify the name along with the−L option:

nsrjb −L −S 20 mars.special

When more than one volume is to be labeled, the name must match the label template associatedwith the pool. This ensures thatnsrjb generates the subsequent names.

Mounting a volume after it has been labeled:To mount a volume after it has been labeled use the−m option:

nsrjb −L −S 20 −m

The command fails if there are not be enough drives to mount all volumes to be labeled.

Labeling volumes with a standard name:To label the volumes in slots 21 through 28, starting with a name different than that referenced bythe label template associated with the pool resource, specify the first name along with the−Loption. In order fornsrjb to generate the additional names, the specified name must match thelayout of the label template.


NSRJB(1m) NSRJB(1m)

nsrjb −L −bOffsite −S 21−28 Offsite.501

After labeling the volume in slot 21 with ‘Offsite.501’nsrjb uses the label template to generatenames for the volumes in slots 22 (‘Offsite.502’) through 28 (‘Offsite.508’). If the nextvolumenamein the sequence for a label template is already in use, the name is skipped.

Loading a volume:To load volumes, use the−l option.

nsrjb −l

nsrjb will select volumes to load into selected devices. It will continue loading volumes until allof the devices are loaded.

Loading specific volumes:To load a volume namedmars.001, specify thevolume namealong with the−l option:

nsrjb −l mars.001

To load the volume in slot 5, use the−Soption:nsrjb −l −S 5

To load the selected volume into device/dev/nrst1,include the−f option.nsrjb −l −f /dev/nrst1 mars.005

Load a cleaning cartridge to clean a deviceTo load the cleaning cartridge with fewest remaining uses into a device, use the volume name’cleaning tape’ along with the−l option. To clean device/dev/nrst1,include the−f option.

nsrjb −l −f /dev/nrst1 ’cleaning tape’

To clean a device by loading the cleaning cartridge in slot 6, use the−S option. Slot 6 must be aslot in the jukebox set aside for cleaning cartridges, and must contain a cleaning cartridge withuses remaining. To clean device/dev/nrst1,include the−f option.

nsrjb −l −S 6 −f /dev/nrst1

Unloading a volumeYou can unload a particular volume, slot, or device. To unload volumemars.0028, use the-uoption:

nsrjb −u mars.0028

To unload the volume in slot28, use the−Soption:nsrjb −u −S 28

To unload the volume in device/dev/nrst3, use the−f option.nsrjb −u −f /dev/nrst3

Displaying the jukebox’s current volumesTo display a list of slots and volumes, and which volumes are loaded in to a jukebox’s devices, usethe-C option:

nsrjb −C

The−C option is the default and is used when no other options are selected. A range of slots mayalso be specified. For example, to display the volumes in slots 10 through 23, use the-Soption:

nsrjb −S 10−23

Setting the number of uses for a cleaning cartridge:To set the number of times all cleaning cartridges in a jukebox may be used to 12, use the-Uoption:

nsrjb −U 12

To set the number of times the cleaning cartridge in slot 10 may be used, use the-S option:nsrjb −U 25 −S 10

Slot 10 must be a slot set aside for cleaning cartridges in the jukebox.


NSRJB(1m) NSRJB(1m)

Inventorying the volumes:To reconcile the actual volumes and the list of volumes produced bynsrjb, use the−I option.Each volume may be loaded into a device and examined for a NetWorker label (depending on barcode settings and other factors). The internal list is then updated with the new information. Afterall volumes have been examined, the new list is compared to the NetWorker media database, and amessage listing any volumes located in the jukebox but not in the database is produced. To inv en-tory the volumes in slots 17 through 43, use the-Soption:

nsrjb −I −S 17−43

Like labeling, volume inventory may take considerable time.

Using the NetWorker notification system:When NetWorker needs a volume, a "media event" is generated. To hav ensrjb automaticallyrespond to these events, the NetWorker notification system is used. This notification resource isautomatically generated.

Using the cartridge access port:To withdraw cartridges from jukebox slot 7 through 11 to the cartridge access port 5 through 10,use the-w option along with the-Sand-P options:

nsrjb −w −S 7−11 −P 5−10

To deposit cartridges into jukebox slot 8 through 10 from the cartridge access port 3 through 5, usethe-d option along with the-Sand-P options:

nsrjb −d −S 8−10 −P 3−5

Using barcode templates on tape libraries:To add volumes with barcodes D001A, D002A, ..., D100A to the volumes available for NetWorkerin the tape library, use the-a and-T options:

nsrjb −a −T D/001−100/A

For a SmartMedia jukebox, to allocate 3 volumes from any media and to make the volumes avail-able for NetWorker, use the-a and-a options:

nsrjb −a −T +3

To deposit tapes labeled with barcodes D001A, D002A, ..., D012A into the silo and also to makethe volumes available for NetWorker in the tape library, use the-a and-T options along with the-d option:

nsrjb −a −T D/001−012/A −d

To remove volume with barcode D055A from the volumes available for NetWorker in the tapelibrary, use the-x and-T options:

nsrjb −x −T D055A

To remove volume with barcode D055A from the volumes available for NetWorker in the tapelibrary, and to withdraw it from the tape library physically (for example, for off-site storage), usethe-x and-T options, along with the-w option:

nsrjb −x −T D055A −w

To label volumes with barcodes D010A, D011A, ... , D020A, use the-L and-T options:nsrjb −L −T D0/10−20/A

To add cleaning cartridge with barcodes C010A, that can be used the default number of time forthis jukebox, use the-U and-T options:

nsrjb −U default −T C010A


The NetWorker media database.

/nsr/res/nsrjb.res The jukebox resource descriptors.


NSRJB(1m) NSRJB(1m)

/nsr/res/smdevmap.txtThe file used to map from SmartMedia media and drive types to a NetWorker deviceresourcemedia type. jukebox.

SEE ALSOjbconfig(1m), jbexercise(1m),mminfo(1m),nwadmin(1m),nsr(1m),nsrd(1m),nsr_layout(5),nsr_device(5), nsr_jukebox(5), nsr_notification(5), nsr_storage_node(5), nsradmin(1m),nsrim(1m),nsrmm(1m),nsrmmd(1m),nsrwatch(1m)

DIAGNOSTICSSome errors have been classified and can be identified by the last three digits of the error number returnedby thensrjb command. Nonclassified errors are listed first.

must be run by rootA normal (non-super) user has attempted to use this command.

No drives are available for use (busy,secure, or disabled).This message is logged when the jukebox tries to acquire a drive to satisfy a backup or recovermedia request. If the drives are not actively saving or recovering, then the device is secured or dis-abled. Devices are secured in thepool resources. Devices are enabled or disabled in thedeviceresources.

All drives are busy or disabled.If the drives are not actively saving or recovering, then the device is disabled. Devices are enabledor disabled in the devices window.

/dev/nrst2: verifying label, error opening: waitingto become readySome tape drives take time to position to the beginning of the tape. While repositioning, thedevice cannot be accessed. After the tape reaches the correct position it is available for use andnsrjb resumes. If the device does not have a tape loaded, an I/O error message similar to the fol-lowing appears:read open error, I/O error (5).

All volume names for ‘xyz’ are in useAll the volume names for the given template have been used. The operator should change the tem-plate to accommodate more volume names.

No volumes found in the media database...continuing.The media database is empty. The user will typically see this message when the module has beennewly installed or all volumes have been deleted.

Another nsrjb is already running, please wait...Another nsrjb command is accessing a jukebox. The current command will keep attempting toaccess the device periodically. Once it has acquired the jukebox device, it will display the mes-sage ’Continuing,’.

slot ‘xyz’ does not have a bar code labelAn inventory operation was attempted with the jukebox resource attributematch bar code labelsenabled, while the media did not have a label. Either disable the attribute withnsradmin ornwadmin, or place a bar code label on the media.

slot ‘xyz’ has a duplicate bar code label ‘xyz‘Tw o or more media volumes have the same bar code label attached. Either disable the attributewith nsradmin or nwadmin, or place a unique bar code label on the media volume.

(001) Unknown jukebox modelThe model for this jukebox is not known to the NetWorker jukebox module.

(006) Unknown control portThere is no control port listed for this jukebox.


NSRJB(1m) NSRJB(1m)

(007) Invalid rangeThe given range could not be parsed bynsrjb .

(010) Source component emptyThe jukebox attempted to move media between components in the jukebox (for example, from aslot to a drive), but found nothing in the source component.

(011) Destination component fullThe jukebox attempted to move media between components in the jukebox (for example, from aslot to a drive), but found something already in the destination component.

(012) All slots fullThe jukebox attempted to unload a drive as part of a reset (−H) operation, with all slots containingmedia. Empty one of the slots or remove the media located in the drive from the jukebox.

(013) Slot xxx is empty.This is often seen during a label operation. The labeling process stops as soon as an empty slot isencountered. If you attempt to label a range of slots on jukeboxes that have the ability to sensewhether or not the slots are loaded, the error message is as follows:

Slot xxx is empty, attempted to label the slot range xxx-yyy. Specify a slot range which is full ofvolumes. No volumes were labeled.

(016) Slot emptyThe source slot did not have a volume in it.

(017) Unsupported operationThis jukebox does not have the functionality to support the requested operation.

(025) Vendor error occurredNormally you would not see the message Vendor error occurred. Instead, you would see an errorstring retrieved directly from the jukebox or device driver. The operator should consult the hard-ware or driver manual to determine the cause of the error.

(027) All drives full/busyAll drives are loaded and/or busy at the moment. Free up one of the drives by unloading thedevice. If all drives are in use, you must wait for a drive to become idle.

(029) Unable to retrieve any volume information fromthe media databaseIndicates thatnsrjbcould not access any volumes in the media database.

(036) All of the devices are in use by nsrmmdThe jukebox could not acquire a drive to use for a save or recover.

(038) All drives must be unloaded before jukeboxresource can be deletedYou cannot delete a jukebox resource if any volumes are loaded in the media drives. Unload allmedia drives before attempting to delete the jukebox resource. If no devices are loaded, issue thensrjb command with the−H option.

(039) This command only valid with a singleslot specifiedYou can only specify a single slot is, not a range of slots, for example, -S 4-6, on which to operate.

(040) The drive is loaded with a volume from adifferent slotThe user specified both a volume and the−f option, but the drive already has a volume loadedfrom a different slot.

(041) The drive is emptyNeed a volume loaded on which to operate.


NSRJB(1m) NSRJB(1m)

(042) Will not overwrite volume without confirmationNetWorker does not allow you to overwrite a volume with a valid NetWorker label without confir-mation.

(043) The volume name does not match what has beeninventoried. Please re-inventory the volume.The jukebox encountered a volume with a different label than expected. The operator should rein-ventory the jukebox.

(044) The volume from that slot is loaded in anotherdriveThe user specified both the−f and −s options, but the volume from the given slot is loaded inanother drive.

(045) The volume does not exist in the jukeboxThe named volume is not loaded in the jukebox.

(047) The alternate side of the media is busyThe other side of the optical media is in use. The side you are trying to access is unavailable untilthe alternate side is idle.

(048) Too many devicesThe user tried to add too many devices during the creation of the jukebox.

(049) Unlabeled volume, loaded but not mountedThe user tried to load a volume, but no label was found on the media.

(050) Drive door closedThe user was trying to perform an unload operation. When the jukebox went to move the mediafrom the drive to a slot, the transport found the media drive door closed.

(051) Unable to select a suitable volume in responseto media requestThe jukebox module could not find any volumes in the devices to respond to a media request.

(054) The drive is busy. Please try again later.An operation was attempted on a media device assigned a save or recover session. Try the opera-tion again later when the media drive is free.

(055) No element status capability for this jukebox.-E ignored.The jukebox does not have element status capability. The-E option is ignored.

(056) The drive is disabled. Enable the drive orchoose another.The media drive specified is disabled. If this media device is the only one in the jukebox, then itmust be enabled for use bynsrjb. If there are other media devices enabled, try selecting one ofthem.

(057) The media pool is not allowed on this device.The media drive specified is not allowed to mount volumes from the media pool specified. Eitherchange the media pool configuration to allow mounts of the pool on this device, or try usinganother device.

(058) All the media drives are disabled.All the media drives are disabled. Enable one or more devices, or select another jukebox or mediadevice outside the currently selected jukebox.

(059) The media pool is not allowed on any of the drives.None of the media drives in this jukebox are allowed to mount volumes from the media pool speci-fied. Either change the media pool configuration to allow mounts of the pool on these devices, ortry using another jukebox device or media device outside the currently selected jukebox.


NSRJB(1m) NSRJB(1m)

(060) All drives are busy, disabled, or do not allow mediafrom this pool.See error descriptions (027), (058), and (059). Some combination of these three errors is prevent-ing the requested operation.

(062) Can only reset jukebox when all drives are idle.When attempting to unload a media device, the device was found to be busy. Wait for the deviceto become idle and reattempt the reset operation.


NSRLIC(1m) NSRLIC(1m)

NAMEnsrlic − NetWorker license reporting command

SYNOPSISnsrlic [ −vi ] [ −sserver]

DESCRIPTIONThe nsrlic command generates reports about all license information currently active on a NetWorkerserver. This command queries the NetWorker resource database, and formats and displays the results tostandard output.

The nsrlic program reports: the number of server or universal client licenses, the number of server clientlicenses used, the number of server client licenses borrowed by workstation clients, the number of remain-ing server client licenses, the list of server clients connected to the named NetWorker server, the list ofserver clients defined in the specified NetWorker server, the number of workstation client licenses, the num-ber of workstation licenses needed or used, the number of remaining workstation client licenses, the list ofworkstation clients connected to the specified NetWorker server, the list of workstation clients defined onthe specified NetWorker server, the number of server clients listed by platform, the number of workstationclients listed by platform, the list of valid client paks installed on the system, and the list of client typesallowed by client paks and server enablers installed on the system.

When applications exist which require licensing,nsrlic also reports them in the same manner. In this case,however, the output will not contain any references unless either there are licenses available, or a connectedclient is utilizing a license count for such applications.

OPTIONS−i Selects interactive mode. In this mode, you can request different reports, refresh the information,

or switch to a different server. The information is requested once and cached until anothercon-nectcommand is issued.

−sserverSelects which NetWorker server to query. By default, the server on the local system is queried.

−v Selects verbose mode. In addition to the number of licenses or the number of clients, a list of con-nected and defined clients is gathered and displayed.

USAGEThe following commands are supported in the interactive mode:

connect[ server]Connects to the namedserver. By default, this is the server on the local system.

detailProduces a detailed report. Displays a list of connected clients (clients that saved to NetWorker)and a list of defined clients (clients that are defined in the NetWorker server, but not yet saved).

helpDisplays a list of available commands.

summaryDisplays a summary report.

?Is the same as online help.

quitPerforms an immediate exit fromnsrlic.

DIAGNOSTICSnsrlic displays a "usage" message describing the available options when characters are used that are notvalid for this command.


NSRLIC(1m) NSRLIC(1m)

command not foundIndicates that the command is not a supported command.

RPC error: Remote system error RPC error: Program not registeredIndicates that some problems were encountered while connecting to the NetWorker server on thespecified system. Thensrlic command requires that the NetWorker daemons be running. Startyour NetWorker daemons(nsrd) and rerunnsrlic. If nsrd is already running, you have probablyreached a resource limit on the server (for example, not enough memory or no more processes).

Serverxxxdoes NOT hav e Self-Identifying capabilityIndicates that the specified NetWorker server does not have the capability to report license infor-mation.

SEE ALSOnsrd(1m),nsradmin(1m),nwadmin(1m)


NSRLS(1m) NSRLS(1m)

NAMEnsrls − list statistics of NetWorker index files

SYNOPSISnsrls [ { clientname. . . |−m } ]

DESCRIPTIONWhennsrls is used without any specified options being specified, the number of records in an online indexand the usage of the online index with respect to the number of kilobytes allocated to its UNIX files isprinted. Administrators can use this command to establish how many files have been saved by a client.

OPTIONSWhen invoked with the−m option, nsrls prints the information for the media database. The mediadatabase has the following four statistics associated with it: an internal file ID (Fid), the size of the file(Size), the number of logical records in the file (Count), and a descriptive name for the internal file(Name).

The internal files are interpreted as follows:

saveset filesThese are the internal record files which store the actual data (for example, ss).

volume filesThese are the internal record files which store the volumes (for example, vol).

index filesThese internal b-tree index files hold the index records used for optimizing media databasequeries. The names of these files contain the extension "_i*" (for example, ss_i0 and vol_i1).

temporary filesThese files (beginning with the filename "temp_*") contain temporary records used during sorting.Temporary files are present only while a database is currently being modified.

The number, name, function, and interpretation of the internal files can change at any time.

An empty argument list prints the statistics for all known clients.

EXAMPLE% nsrls -m

Database id 0: /nsr/mm/mmvolumeFid | Size | Count | Name------------------------------------------0 | 16 KB | 6 | vol1 | 136 KB | 484 | ss2 | 16 KB | 6 | vol_i03 | 16 KB | 5 | vol_i14 | 16 KB | 5 | vol_i25 | 16 KB | 5 | vol_i36 | 16 KB | 0 | vol_i47 | 24 KB | 484 | ss_i08 | 24 KB | 484 | ss_i19 | 16 KB | 164 | ss_i210 | 24 KB | 483 | ss_i311 | 8 KB | 1 | temp_0

% nsrls jupiter

/space2/nsr/index/jupiter: 292170 records requiring 50 MB/space2/nsr/index/jupiter is currently 100% utilized


NSRLS(1m) NSRLS(1m)

SEE ALSOnsr_layout(5), nsrindexd(1m)

DIAGNOSTICS... is not a registered client

The client named is not a valid NetWorker client.


NSRMIG(1m) NSRMIG(1m)

NAMEnsrmig − migrate files for long-term storage with NetWorker HSM

SYNOPSISnsrmig [ −aMnvx ] [ −l percent] [ −sserver] [ −t savetime] [ −W width ] [ path]

DESCRIPTIONnsrmig migrates files to the NetWorker server. Migration means replacing a file with a "stub" that points toa copy of the file made from premigration. If the stub is accessed at some later time, the file is automati-cally recalled to disk by the NetWorker server.

Criteria specified in the NetWorker migration client resource are used to select files for migration. Cur-rently, only regular files are premigrated and migrated.

If no pathargument is specified, the current directory is migrated.nsrmig does not cross mount points, nordoes it follow symbolic links.

OPTIONS−a Sort files by access time. Valid for XDSM HSM only. Files selected for migration are sorted by

access time before being replaced by stubs. Files that have been least recently accessed will bemigrated first.

−M Use Networker’s migration index to find lists of files that have been premigrated. Valid for XDSMHSM only. Without this optionnsrmig walks the directory specified searching for files that areeligible for premigration. Using the index might be faster finding the files that have be premi-grated recently, but will not find files that have been recovered since premigration. Symbolic linkHSM always uses the migration index.

−l percentSpecify a goal percentage fornsrmig. Migration stops when the goal percentage is reached. Ifthe goal percentage is already reached whennsrmig is run, thennsrmig will do nothing and exit.If the −l option is not specified, then the goal percentage is read from the appropriate migrationclient resource.

−n Indicates no stub replacement. Estimates the total number of files and the total size that is freed bystub replacement, but does not actually replace qualified files with stubs.

−sserverUses the NetWorker server namedserver.

−t savetimeMigrates files that were premigrated atsavetime.

−v Verbose. Causes thensrmig program to tell you in detail what it is doing as it proceeds. If youspecify multiple−v options, then the verbosity level increases.

−W widthIndicates the width that is used when summary information output is formatted.

−x Cross mount points that are encountered.


0 Normal exit.−1 Abnormal exit.

SEE ALSOnsr_getdate(3), hosts(5), nsr(5), nsr(1m), nsr_client(5), nsr_device(5), nsr_group(5), nsr_service(5),nsrd(1m), nsrhsmck(1m), nsrindexd(1m), nsrpmig(1m), nsrmm(1m), nsrmmd(1m), nsrwatch(1m),savefs(1m),savegrp(1m)


NSRMM(1m) NSRMM(1m)

NAMEnsrmm − NetWorker media management command

SYNOPSISnsrmm

[ −C ] [ −v | −q ] [ −sserver] [ −f device]

nsrmm−m [ −v | −q ] [ −sserver] [ −f device] [ −r ] [ volume]

nsrmm−l [ −v | −q ] [ −sserver] [ −f device] [ −myB ] [ −e forever] [ −c capacity] [ −o mode] [ −b pool] [ −R | volume]

nsrmm{ −u | −j } [ −v | −q ] [ −sserver] [ −y ] [ −f device| volume.. ]

nsrmm −p [ −v | −q ] [ −sserver] [ −f device]

nsrmm{ −d | −o mode} [ −v | −q ] [ −sserver] [ −Py ] [ −Sssid[/cloneid]| −V volid | volume ...]

nsrmm−S ssid[ −w browse-time] [ −e retention-time]

DESCRIPTIONnsrmm a command line interface to manage the media and devices (tapes, disks, and files) used by Net-Worker servers and storage nodes.

A volumeis a physical piece of media, for example, a tape or disk cartridge. When dealing with file typedevices,volumerefers to a directory on a file system. NetWorker must have exclusive use of this directory,as files will be created and removed. The NetWorker system keeps track of which user files have beensaved on which volumes, so they can be more easily recovered. Every volume managed by NetWorker hasa volume name(also known as avolume label) selected by an operator. A volume name is specified whenthe volume is first introduced to the system. It can only be changed when a volume is relabeled. The vol-ume should have an external label displaying its volume name for future reference. NetWorker refers tovolumes by their volume names, for example, when requesting a volume for recovery.

The NetWorker system automatically manages an index that maps saved user files to volumes. NetWorkeralso keeps other attributes associated with a volume, including the expected capacity of the volume.

The NetWorker server requests that specific volumes be mounted by their name for recoveries, or anywritable volumes for saves. These requests are submitted through thensr_notification(5) mechanism. Thenwadmin(1m) console window or thensrwatch(1m) command can be used to monitor pending mountrequests. Typically, the requests will also be written to the system console, or logged in a file. The samerequests can be used as input for software that controls ajukebox(a device that automatically loads andunloads volumes).

Before thensrmm command can be used (that is, before any data can be saved or recovered), at least onedevice must be configured for the NetWorker server. The NetWorker configuration may be modified withthe nwadmin(1m) Administration menus or thensradmin(1m) command after NetWorker has beeninstalled.

OPTIONS−B Verifies that the volume you want to label does not have a readable NetWorker label. Before label-

ing the volume, an attempt is made to read any existing label the volume may already possess. Ifyou specify this option and the volume has a valid NetWorker label that is readable by the devicecurrently being used, the label operation is canceled and an error message is displayed. If the vol-ume does not contain a label that is readable by the current device, the volume may be labeled.This option is used bynsrd(1m) when automatically labeling volumes on behalf ofnsrmmd(1m)requests.


NSRMM(1m) NSRMM(1m)

−b poolSpecifies the pool to which the volume belongs.−b pool can name any pool currently registeredwith nsrd. The possible values can be viewed by selecting the Pools menu item from the Admin-istration menu ofnwadmin(1m) or using thensradmin(1m) command. The pool name is refer-enced bynsrd when determining which save sets can reside on the volume. If you omit thisoption, the volume is automatically assigned to theDefaultpool. If you specify a pool name with-out specifying a volume name, the next volume name associated with the pool’slabel templateresource is used.

−C Displays a list of NetWorker configured devices and the volumes currently mounted in them. Thislist displays only the devices and volumes assigned to the server, not the actual devices and vol-umes. The−p option verifies the volume label.−C is the default option.

−c capacityOverrides the default capacity of a volume. NetWorker normally uses built-in default capacitiesbased on the device type. This option overrides these defaults. The format of the specification isnumber multiplier. Multipliercan be one of ‘K’ (1024 bytes), ‘M’ (1000 KB), or ‘G’ (1000 MB).Lower case letters are also accepted, as are extra characters like spaces, or an extra ‘B’ after ‘K’,‘M’, or ‘G’. Numbermay be any value, including an integer or real number, with up to three deci-mal places.

−d Deletes the client file indexes and media database entries from the NetWorker databases. Theaction does not destroy the volume: instead, it removes all references used by NetWorker to thevolume and the user files contained on it. This option can be used to control the size of the Net-Worker databases.

−e time When used in conjunction with the−S option, it sets the retention time of the specified save set.The retention time should be specified in the format that is acceptable to the functionnsr_getdate(1m). Note that the save set retention time may not be set such that the save set wouldbecome recyclable while it is still browsable. Refer to the−w option for more details on browsetime. When used in conjunction with volumes, the volume labeled will be an Archive volume ifthe value oftime is fore ver (Archive volumes mean that the volume label never expires). Anyother value oftimeare not applicable to a volume.

−f deviceSpecifies a device explicitly. When more than one device has been configured,nsrmm will selectthe first device by default. This option overrides the selection made bynsrmm.

−j Ejects a volume from the device. This option is similar to performing an unmount operation,except that the volume is also physically ejected from the device, if possible. This feature is notsupported by some device types, disk devices, and tapes.

−l Labels (initializes) a volume for NetWorker to use and recognize. Labeling must be performedafter the desired volume is physically loaded into the device, either by an operator or a jukebox.

−m Mounts a volume into a device. Mounting is performedafter a volume is placed into a device andlabeled. You can mount only labeled volumes. The labeling and mounting operations can be com-bined into a single command line. See theEXAMPLES section.

−o modeSets the mode of a volume, save set, or save set instance (clone). Themodecan be one of the fol-lowing: [not]recyclable, [not]readonly, [not]full , [not]manual or [not]suspect. The[not]recyclable mode applies to both volumes or save sets, but not clones. The[not]readonly,[not]full and [not]manual modes apply only to volumes. The[not]manual mode is the onlyvalid mode when used with the−l option. The [not]suspect mode applies only to save setinstances, meaning it must be specified along with−S ssid/cloneid, not just −S ssid by itself.(Remember that every instance of a save set has a clone id, even the original.) Seensrim(1m) fora discussion of the per-volume flags. Thesuspectflag is set automatically when arecover(1m)encounters a media error recovering data from a particular save set clone.


NSRMM(1m) NSRMM(1m)

−P When used in conjunction with the−d option the corresponding file index entries are purged, with-out deleting the entries in the media database. Thescanner(1m) command can then be used torecover the file index entries.

−p Verifies and prints a volume’s label. To confirm that the external volume label matches the inter-nal label, load a volume into a drive and use this option to display the volume name in the label.Verifying a label unmounts mounted volumes.

−q Quiet mode. This option tellsnsrmm to print out as little information as possible while perform-ing the requested operation. Generally, only error messages are printed.

−R Relabels a volume. This option rewrites the volume label and purges the NetWorker indexes of alluser files previously saved on the volume. Some of the volume usage information is maintained.

−r Mounts a volume as read-only. To prevent NetWorker from writing to a volume, specify the read-only flag when mounting the volume. Volumes marked as full and those in the read-only mode(−o readonly) are automatically mounted read-only.

−sserverSpecifies the NetWorker server to perform thensrmm operation on. Seensr(1m) for a descriptionof server selection.

−Sssid Changes (−o) or removes (−d) a sav e set from the NetWorker databases, or used in changing thebrowse time (specified with−w) or the retention time (specified with−e) of the specified save setrecord. The save set is identified by a save set identifier,ssid. A save set instance, or clone, can bespecified using the formatssid/cloneid(but, it is ignored when used for the options−w and−e).Themminfo(1m) program may be used to determine save set and clone identifiers.

−u Unmounts a volume. A volume should always unmount a volume before you unload it from adevice.

−V volidRemoves a volume from the NetWorker databases when used in conjunction with the−d option.The volume is identified by a volume identifier, orvolid. Themminfo(1m) command can be usedto determine volume identifiers.

−v Verbose mode. This option polls the NetWorker server to print out more information as the opera-tion proceeds.

−w browse timeSpecifies the browse time for the specified save set (supplied with the−S option). The browsetime should be specified in the format that is acceptable to the functionnsr_getdate(1m). Thebrowse time has to be after the insert time in the save set record, but it cannot be after the retentiontime. If the option−e was not used, the existing retention time in the save set record is used forcomparing with the specified browse time. See under the option−e for more details on retentiontime.

−y Do not confirm (potentially destructive) operations before performing them. This option must beused with extreme care.

EXAMPLESLabeling new tapes:

To introduce a new tape, namedmars.001, to the NetWorker system, load the tape in an emptydrive, then use the command:

nsrmm −l mars.001

The tape is labeled withmars.001and an entry is made in the appropriate NetWorker indexes.The mminfo(1m) command may be used to inspect the volume database and display informationabout the volumes:

mminfo −m


NSRMM(1m) NSRMM(1m)

Mounting a tape:To mount a NetWorker volume, use the−m option. Note that the volume must have been labeledpreviously and loaded in the drive:

nsrmm −m

When mounting, a volume name can also be specified:nsrmm −m mars.001

The mount will fail unless the given volume name matches the one read from the media.

By mounting a volume, you make the volume available to NetWorker. Whennsrmmd(1m) needsthe volume, the label will be read again and confirmed, preventing accidental data loss. Volumesare also verified and mounted automatically if the server recovers after a crash.

Labeling and mounting a tape:A volume may be labeled and mounted with a singlensrmm command by combining the−m and−l options. The following example labels a volume asmars.003 and mounts it on device/dev/nrst0:

nsrmm −m −l −f /dev/nrst0 mars.003

Unmounting or ejecting a volume:When a volume needs to be unmounted, use either the−u or −j option, depending on whether ornot the device can physically eject a volume.

nsrmm −u

When more than one volume is mounted, you can specify either the volume name or device toselect the desired volume. The following example ejects the volume namedmars.003.

nsrmm −j mars.003

Displaying the current volumes:The −C option displays the configured devices and the mounted volumes. This is the defaultoption.

nsrmm −C

Deleting a volume:To remove references to a volume and the user files saved on it from the NetWorker indexes, usethe −d option. This option does not modify the physical volume, and should only be used whenthe physical volume is destroyed. By deleting a volume, you free up space in the NetWorker fileindex and the NetWorker media index, but not much more than if you had purged it. The amountof space released depends on the number of user files saved on the volume. The following exam-ple deletes the volumemars.003:

nsrmm −d mars.003

Thescanner(1m) command can be used to rebuild the database entries.

Purging file index entries:The file index contains information about each file saved by NetWorker. Due to size constraints, itmay be necessary to purge information from the file index. When a volume or save set is deleted,the corresponding file index entries are also removed. It is also possible to preserve the mediadatabase entries of a volume while purging the file index by specifying the−P option when delet-ing.

The following example purges all of the file index entries for volumemars.001:nsrmm −d −P mars.001

Thescanner(1m) command can be used to recover the file index.

SEE ALSOnsr(1m),nsr_getdate(3), nsr_layout(5), nsr_device(5), nsr_notification(5), mminfo(1m),nwadmin(1m),nsrmmd(1m),nsradmin(1m),nsrim(1m), recover(1m),scanner(1m).


NSRMM(1m) NSRMM(1m)

DIAGNOSTICStype family volumemounted ondevice, write enabled

Message indicating that the−m (mount) option was successfully performed on a device with thegiven mediatypeand mediafamily, for example, 8mm tape.

‘saveset’ is not a valid save set idThe given sav e set identifier is not in the valid format. The format is either a single number (forthe save set without reference to its instances), or two numbers separated by a slash (/) (represent-ing a save set and clone (instance) identifier pair).

duplicate name; pick new name or delete old oneIt is illegal to label two tapes with the same name. If you wish to reuse a name, remove that vol-ume from the index using the−d option.

Are you sure you want to over-writevolumewith anew label?An attempt is being made to relabel a volume. A positive confirmation will overwrite the existingdata on that tape.

Purge file index entries fortype family volume? ...After confirmation, the file index entries are removed.

volumenot in media indexThe media index has no entry associated withvolume,so the−m command cannot be used. Thisproblem may be caused by mistyping the volume name when the tape was originally labeled, ordeleting it.

No valid family labelThe tape or disk in the named device does not have a valid NetWorker label.


NSRMMD(1m) NSRMMD(1m)

NAMEnsrmmd − NetWorker media multiplexor daemon

SYNOPSISnsrmmd [−v] [−s server] [−r system] −nnumber

DESCRIPTIONThe nsrmmd daemon provides an RPC-based media multiplexing and demultiplexing service. The RPCprogram and version numbers provided bynsrmmd are 390104 and 5, respectively. Howev er, to supportmultiple instances of the protocol (if the Concurrent Device Support feature is enabled), the version num-bers used have the daemon number times one hundred added to them. The daemon numbers always start atone, so the first version registered will be 105, then 205, and so on. Onensrmmd per enabled device isstarted automatically bynsrd. Additionalnsrmmd daemons may be started when a mount request is pend-ing. To change the number of daemons, alter the number of enabled devices.

OPTIONS−n number

Specify the daemon number.

−sserverSpecify the controlling server. This option is used on a storage node (seensr_storage_node(5)).

−r systemSomensrmmd programs run on the server but are controlling a device attached to aNetworkerData Management Protocol (NDMP)system. Such instances ofnsrmmd have an optional−rargument specifying the system that is being controlled.

−v Verbose: Print out messages about what the daemon is doing.

SEE ALSOnsr(1m),nsr_layout(5), nsr_service(5), nsr_storage_node(5), nsrd(1m),nsrmm(1m),mm_data(5)


NSRMMDBASM(1m) NSRMMDBASM(1m)

NAMEnsrmmdbasm − NetWorker module for saving and recovering media databases

SYNOPSISnsrmmdbasm[ standard-asm-arguments]

DESCRIPTIONThensrmmdbasm is a standard, externalASM (Application Specific Module) that assists in the saving andrecovering of the NetWorker media multiplexor’s database files.

Seeuasm(1m) for a general description ofASM’s and the [standard-asm-arguments]. It is intended thatnsrmmdbasmonly be invoked bynsrmmdbd(1m) ormmrecov(1m) operations.

Actions performed bynsrmmdbasm, specific to the NetWorker application during asaveare:

Architecture independence:The high speed access methods and data structures implemented by the database code are machinedependent. ThisASM saves only the records (and not access indexes) in an architecture indepen-dent manner. Therefore, NetWorker media databases may be saved from one machine architectureand recovered to another.

Conservation:Since only changed records are saved, and not internal indexes, considerable network bandwidthand tape space are conserved.

Therecoveroperation of thisASM is the inverse of thesaveoperation.

FILES/nsr/mm/.nsr This directive file causes most files in the directory to be skipped during normal

save operations.nsrmmdbasm ignores this directive.

/nsr/mm/mmvolume6 The directory which is saved and recovered by thisASM.

/nsr/mm/mmvolume.r A temporary file that stores the contents of a recovered media database untilnsr-mmdbd(1m) has completed building a new media database.

/nsr/mm/mmvolume.s A temporary file that thisASM reads when backing up data.

/nsr/mm/volume.tmp A temporary file created when converting an older media database schema to thepresent schema during recovery.

SEE ALSOnsr(5), nsr_layout(5), mmrecov(1m), nsrmmd(1m), nsrmmdbd(1m), nsrindexasm(1m), recover(1m),savegrp(1m),uasm(1m)


NSRMMDBD(1m) NSRMMDBD(1m)

NAMEnsrmmdbd − NetWorker media (volume) management database daemon

SYNOPSISnsrmmdbd

DESCRIPTIONThensrmmdbd daemon provides an RPC-based database service to the localnsrd(1m) andnsrmmd(1m)daemons, and query-only network access to NetWorker clients. The RPC program number provided bynsrmmdbd is 390107. The RPC version numbers provided bynsrmmdbd are 3, 4, and 5.Nsrmmdbd isnormally started bynsrd(1m).

The daemon manages a ‘‘media and save set database’’ located in the file/nsr/mm/mmvolume. The pri-mary purpose of the database is to remember which save sets reside on which backup volumes. Numerousaccess methods are provided to both save set and volume records within the database.


File containing the volume data base.

/nsr/mm/cvt A temporary file created when converting an older media database schema to the presentschema.

/nsr/mm/.cmprssdFor performance and space reasons, the database is periodically rebuilt (or compressed).This file is created each time the database is rebuilt; its associated ctime is used to determinewhen to rebuild the database again. To forcibly compress the database, remove this file andrunnsrim.

/nsr/mm/mmvolume.sThis temporary file is created to hold the media database information that will be saved totape bynsrmmdbasm(1m).

/nsr/mm/mmvolume.rThe file (created bynsrmmdbasm) that is read when the media database information isbeing recovered.

SEE ALSOmmrecov(1m), nsr(1m), nsrd(1m), nsrim(1m), nsrmmd(1m), nsrmmdbasm(1m), nsrmm(1m),mminfo(1m)

DIAGNOSTICSThensrmmdbd diagnostic messages will normally be logged to the/nsr/logs/daemon.logfile.

Besides the messages listed below,nsrmmdbd may generate other diagnostics. Any diagnostics other thanthose listed below indicate a serious problem with the media database. It may be necessary to recover yourmedia database usingmmrecov(1m) if that occurs.

media db is convertingpathto version 5media converting to version 5

Any media databases created prior to the NetWorker 4.2 release have to be converted (once) to thenew database format. Plan on allowing one second for every 100 save sets.

media conversion donePrinted when the conversion is completed successfully.

media conversion failed!reasonPrinted when the conversion is terminates unsuccessfully. A more detailed reason may beappended to the message. NetWorker cannot work until the media database is converted success-fully.


NSRMMDBD(1m) NSRMMDBD(1m)

media db is convertingcountvolumesThis is printed after the volumes’ data has been dumped from the old database, but before it hasbeen loaded into the new database.

media db is convertingcountsave setsThis is printed after the save sets’ data has been dumped from the old database, but before it hasbeen loaded into the new database.

media db is saving its data, this may take a whilePrinted when the daemon is dumping its records to a temporary file when the database is beingbacked up. The service is unavailable while the database is dumping.

media db is recovering, this may take a whilePrinted when the daemon is reloading its database. The service is unavailable while the data isbeing reloaded.

media db is recovering old data, this may take a whileSimilar to the previous message, except that a pre-4.0 database is being recovered and it will haveto be converted before service resumes.

media db is cross checking the save setsPrinted each time the daemon is restarted. Upon start-up, the daemon sanity checks its recordsbefore providing its service.

media db is open for businessPrinted after any of the previous messages are printed to indicate that the service is once againavailable.

A copy of this process is already running!Another copy ofnsrmmdbd(1m) is currently running and has exclusive access to the mediadatabase. Only onensrmmdbd process should be running on a given machine at a time. This canhappen if the previousnsrmmdbd was not properly killed off. Usensr_shutdown(1m) orps(1)andkill (1) to identify and kill off all the NetWorker daemons before restartingnsrd(1m) again.

Cannot open lock fileAn internal error, check the permissions on the/nsr/tmp and/nsr/mm directories.


NSRMON(1m) NSRMON(1m)

NAMEnsrmon − command to remotely control NetWorker commands and daemons

SYNOPSISnsrmon

DESCRIPTIONThensrmon command is run only by NetWorker daemons.nsrd(1m) starts the command to remotely con-trol other commands and daemons on NetWorker storage nodes runningnsrexecd(1m). Commands anddaemons started remotely includensrjb (1m) andnsrmmd(1m). Seensr_storage_node(5) for additionaldetail on storage nodes.

SEE ALSOnsr(1m),nsr_storage_node(5), nsrd(1m),nsrexecd(1m),nsrjb (1m)nsrmmd(1m)


NSRNDMP_RECOVER(1m) NSRNDMP_RECOVER(1m)

NAMEnsrndmp_recover − use NetWorker and Network Data Management Protocol(NDMP) to recover data

SYNOPSISnsrndmp_recover[ −c client ] [ −sserver] { -r raw device -S ssid -m mount point | -F }

DESCRIPTIONnsrndmp_recover is used to coordinate recover operations with NetWorker and aNetwork Data Man-agement Protocol(NDMP) system. Only the super-user may run this command. There are two ways torecover data: destructive recovers and file-level recovers. Destructive recovers occur when a raw partitionis specifed by the−r option along with a save set id (−S) option. Only a single save set can be specified ata time since the targetraw devicepathname must be specified. Users may opt to use the administrative userinterface, nwadmin(1m), to perform the destructive recover operation through the save set recover window.Users can determine save set ids using the user interfaces or themminfo(1m) command. File level recoversare specified by the−F option in conjunction with the use of thenwrecover(1m) or recover(1m) com-mands. Users should not specify this option.

The status of a recover can be monitored using the X Window System-basednwadmin(1m) program or thecurses(3X) basednsrwatch(1m) program for other terminal types. Only volume information is available atthis time. The amount of data that has been recovered is not provided.

nsrndmp_recover is not responsible for moving data on the NDMP system. All such activity is handledby the NDMP system.nsrndmp_recover receives messages from the NDMP system and processes themappropriately. Such messages could request a new tape be mounted or to post a log message. Refer to theNDMP specification and documentation available atwww.ndmp.orgfor more details.

In order to recover data for another system, make sure the user performing thensrndmp_recoveroperationis on theremote accessattribute list of the client resource. Seensr_client(5).

OPTIONS−c client

Client is the name of the machine that saved the files.

−F Specifies that a file-level recovery is going to be performed. This option should only be specifiedby nwrecover(1m) orrecover(1m).

−m mount pointThe mount point of the raw device specified by the-r option. The filesystem will be unmountedfor the recover operation and mounted after the operation is complete.

−r raw deviceThis option specifies the pathname of the raw device the data is to be recovered to. You must usethis option for destructive recovers.

−sserverThis option selects which NetWorker server to use.

−Sssid This mandatory option is used to specify save set recover mode. This mode can be used to imple-ment fast batch file recovery without requiring the NetWorker file index entries.ssidspecifies thesave set id for the save set to be recovered. NDMP-generated save sets cannot be cloned, so nocloneid applies as it does for the standardrecover(1m) command.

DIAGNOSTICSSkipping file due to incomplete save set: /core

The user has marked a file that is associated with an incomplete save set. The user should runnsrim -X to resynchronize the file index and media database.

SEE ALSOnsr_client(5), nsrndmp_save(1m), recover(1m),nwrecover(1m)


NSRNDMP_SAVE(1m) NSRNDMP_SAVE(1m)

NAMEnsrndmp_save − use NetWorker and Network Data Management Protocol(NDMP) to save data

SYNOPSISnsrndmp_save −Tbackup-type[ −Lnq ] [ −c client-name] [ −g group ] [ −l level ] [ −m masquerade] [−N name] [ −s server] [ −t date] [ −e expiration] [ −w browse_time] [ −y retention_time] [ −W width ]path

DESCRIPTIONnsrndmp_savecoordinates the backup process with NetWorker and a targetNetwork Data ManagementProtocol(NDMP) system. Only the super-user may run this command. The user must specify abackuptype, client-name, server, andpath.

The behavior of the backup depends on the NDMP system being protected. Certain environment variablesmay be needed depending upon the target system. Documentation for such backups should be consultedfor more details.

The status of a backup can be monitored using the X Window System-basednwadmin(1m) program or thecurses(3X) basednsrwatch(1m) program for other terminal types.

nsrndmp_saveis not responsible for moving data on the NDMP system. All such activity is handled bythe NDMP system.nsrndmp_savereceives messages from the NDMP system and processes them appro-priately. Such messages could request a new tape be mounted or a file index entry to be created. Refer tothe NDMP specification and documentation available atwww.ndmp.orgfor more details.


In order to save data for another system, make sure the user performing thensrndmp_saveoperation is ontheremote accessattribute list of the client resource. Seensr_client(5).

OPTIONS−c client-name

Specifies the client name for starting the save session. This is useful on clients with multiple net-work interfaces, and hence multiple host names. It can be used to create multiple index databasesfor the same physical client. Note that this does not specify the network interface to use. This isspecified in theserver network interfaceattribute of the client resource. Seensr_client(5).

−g groupIs used bysavegrp(1m) andsavefs(1m) to denote the group of the save. Seensr_client(5) andnsr_group(5). It is also used by the NetWorker server to select the specific media pool.

−l level Indicates the level of the save. This option is used bysavegrp(1m) andsavefs(1m) to specify aparticular level for a scheduled save.

−L When two−L options are specified, this option causes an extra line to be printed at the end of theform ‘‘complete savetime=number’’, where number is the savetime of the save set created by thisbackup. Used bysavegrp(1m).

−m masqueradeSpecifies the tag to precede the summary line. This option is used bysavegrp(1m) andsavefs(1m) to aid in savegrp summary notifications.

−n Indicates no save. Not supported, but provided for compatibility.

−N nameIndicates the symbolic name of this save set. By default, the most common prefix of thepatharguments is used as the save set name.

−q Indicates quiet. Not supported, but provided for compatibility.

−sserverSpecifies which machine to use as the NetWorker server.

−t date Indicates the date (innsr_getdate(3) format) after which files must have been modified beforethey will be saved. This option is used bysavegrp(1m) andsavefs(1m) to perform scheduled


NSRNDMP_SAVE(1m) NSRNDMP_SAVE(1m)

saves by consulting with the media database to determine the appropriate time value based on theprevious saves for the save set and the level of the scheduled save.

−T backup-typeThe type of backup on the NDMP server, for examplecelestra.

−eexpirationSet the date (innsr_getdate(3) format) when the saved data will expire. When a sav e set has anexplicit expiration date, the save set remains both browsable and non-recyclable until it expires.After it expires and it has passed its browse time, its state will become non-browsable. If it hasexpired and it has passed its retention time, the save set will become recyclable. The special valuefore ver is used to indicate that a volume that never expires (i.e. an archive or a migration volume)must be used. By default, no explicit expiration date is used.

−w browseSets the date (innsr_getdate(3) format) after which the saved data will no longer be browsable.By default, the server determines the browse date for the save set based on the browse policies ineffect. This option allows overriding the existing policies on a save by sav e basis.

−y retentionSets the date (innsr_getdate(3) format) when the saved data will become recyclable. The specialvalue fore ver is used to indicate that a volume that never expires (i.e. an archive or a migrationvolume) must be used. By default, the server determines this date for the save set based on theretention policies in effect. This option allows overriding the existing policies on a save by sav ebasis.


SEE ALSOcurses(3X), nsr_getdate(3), nwadmin(1m), nsr_client(5), nsr_device(5), nsr_group(5), nsr_service(5),nsrd(1m), nsrim(1m), nsrindexd(1m), nsrmm(1m), nsrmmd(1m), nsrndmp_recover(1m), nsr-watch(1m), recover(1m),savefs(1m),savegrp(1m)


NSRPM(8) NSRPM(8)

NAMEnsrpm − NetWorker Power Monitor Service

SYNOPSISThis executable is run only as an MS Windows service.

DESCRIPTIONThe nsrpm program (NetWorker Power Monitor Service) supports Windows 2000 Microsoft AdvancedConfiguration and Power Interface (ACPI). The executable for this service is installed and configured forautomatic startup during NetWorker setup.

For successful scheduled backups of Windows 2000 clients, each NetWorker client must be able to respondwhen a NetWorker server contacts it for backup. Therefore, by default, the NetWorker Power Monitor ser-vice does not allow a NetWorker server or client host to enter the ACPI standby mode. This helps ensurethat scheduled backups may be performed as configured and unscheduled operations may be performed atany time.

WARNINGSDo not place a NetWorker server host or client host in standby mode during a time period when either com-puter is to participate in a scheduled backup.

Do not put a NetWorker client or server host in standby or hibernation mode while a NetWorker backup orrestore operation (involving that host) is in progress. Doing so will likely have sev ere consequences andyield unpredictable results. In such a case, the NetWorker operation might fail or hang, and there might beother severe consequences or unpredictable results. Moreover, depending on the devices and configuration,if a server is powered down while NetWorker operations are in progress, the server’s peripheral devicesmight be powered down as well. If this occurs, when the server’s power is restored, the tape devices willprobably rewind and NetWorker’s tape processes (which otherwise never would have stopped) will haveincorrect positioning information.

1

NSRPMIG(1m) NSRPMIG(1m)

NAMEnsrpmig − premigrate files for long-term storage with NetWorker HSM

SYNOPSISnsrpmig [ −BEiLnpqvx ] [ −LL ] [ −sserver] [ −N name] [ −f dirfile ] [ −b pool ] [ −g group ] [ −m mas-querade] [ −W width ] [ −I input file] path

DESCRIPTIONnsrpmig premigrates files to the NetWorker server. Premigration means making a copy of a file on Net-Worker storage in preparation for migration. When a file is later migrated, the on-disk copy of the file isreplaced with a reference to the premigrated copy in NetWorker.

Currently, only regular files are premigrated. Criteria specified in the NetWorker migration client resourceare used to select files for premigration. The progress of a nsrpmig session can be monitored using the XWindow System basednwadmin(1m) program or thecurses(3X) basednsrwatch(1m) program for otherterminal types.

The nsrpmig command will not cross mount points, nor will it follow symbolic links. If the path to besaved is mounted from a network file server,nsrpmig will instruct the user to run the save on the remotemachine or use the-L option.

The directive files, seensr(5), encountered in each directory will be read by default. They contain specialinstructions directing how particular files are to be saved, such as compressed or skipped. These files arenamed .nsrhsm. Note that the directive files used by NetWorker for save and recover named .nsr areignored bynsrpmig.

Each file in the subdirectory structures specified by thepatharguments will be encapsulated in a NetWorkersave stream. This stream of data is sent to a receiving process on the NetWorker server. Seensrd(1m). Theserver processes the data, adding entries to the online index for each file in the stream. Seensrindexd(1m).The data finally resides on some long term storage media. Seensrmmd(1m).


OPTIONS−E Estimates the amount of data which will be generated by the save, then perform the actual save.

Note that the estimate is generated from the inode information, and thus the data is only actuallyread once.

−i Ignores any.nsrhsm directive files as they are encountered in the subdirectory structures beingsaved.

−L Indicates local. Saves will be performed from the local NetWorker client, even when files arefrom a network file server. To recover these files, runrecover(1m) with the−c client arguments,whereclient is the name of the NetWorker client that did thesave.

−LL In addition to treating the backup as a local backup, this option causes an extra line to be printed atthe end of the form ‘‘complete savetime=number’’, where number is the savetime of the save setcreated by this backup. This option is meant to be used by thesavegrp(1m) command in perform-ing automatic cloning.

−m masqueradeSpecifies the tag to precede the summary line with. This option is used bysavegrp(1m) andsavefs(1m) to aid in savegrp summary notifications.

−n Indicates no save. Estimates the amount of data which will be generated by the save, but does notperform the actual save.

−v Indicates verbose. Causes thesaveprogram to report its progress.

−p Exits with status 0. Is used by server to determine if client installed properly.

−q Indicates quiet. Displays only summary information and error messages.


NSRPMIG(1m) NSRPMIG(1m)


−N nameIndicates the symbolic name of this save set. By default, thepathargument is used as the save setname.

−f dirfileIndicates the file from which to read prototype default directives. Seensr(5). A dirfile of - causesthe default directives to be read from standard input.

−b poolSpecifies a particular destination pool for the save.

−g groupIs used bysavegrp(1m) andsavefs(1m) to denote the group of the save. Seensr_client(5) andnsr_group(5). It is also used by the NetWorker server to select the specific media pool.

−I input_fileIn addition to taking the paths to save from the command line, this option reads paths to save fromthe named file. The paths must be listed one per line. If no paths are specified on the commandline, then only those paths specified in the file will be saved.

−W widthIndicates the width used when formatting summary information output.

−x Indicates that if a subdirectory is a mount point, scan it also. This option currently has no effect.

−B Forces a save of all connecting directory information from root (/) down to the point of invocation.


0 Normal exit.−1 Abnormal exit.

SEE ALSOcurses(3X), nsr_getdate(3), nsr(5), nsr(1m),nsr_client(5), nsr_device(5), nsr_group(5), nsr_service(5),nsrd(1m),nsrhsmck(1m),nsrindexd(1m),nsrmig(1m),nsrmm(1m),nsrmmd(1m),nsrwatch(1m),nwadmin(1m), recover(1m),save(1m),savefs(1m),savegrp(1m).


NSRPORTS(1m) NSRPORTS(1m)

NAMEnsrports − port configuration tool

SYNOPSISnsrports [ −sserver] [ −a auth_server] [ −S | −C ] [ range]

DESCRIPTIONThe nsrports command is used to display and set ranges of ports used by the NetWorker software. Theport ranges are stored bynsrexecd(1m) in the NSR system port rangesresource. Whennsrports isexecuted without any options, the program displays the configured ranges for the system on which the com-mand is being run.

Only users executing a tool on the system can change the system ports. These changes can be made onlyby usingnsradmin(1m) to modify the administrator attribute for the resource storing the ranges.

There are also two additional options for viewing and setting the port ranges. The first is by using thegraphical user interface,nwadmin(1m). The second is by usingnsradmin(1m). Execute the program asfollows:

# nsradmin -sserver-p nsrexec

whereserveris the system to display ports for.

OPTIONS−sserver

Specifies the system to contact.

−a auth_serverSpecifies a NetWorker server. This option is required ifnsrports is connecting to a remote systemthat is located on a platform different than the system on which the command is being executed.

−S Sets the system’s service ports range to the one specified.

−C Sets the system’s connection ports range to the one specified.

SEE ALSOnsrexecd(5), nsradmin(1m),nwadmin(1m)


NSRRETRIEVE(1m) NSRRETRIEVE(1m)

NAMEnsrretrieve − retrieve NetWorker archive sav e sets

SYNOPSISnsrretrieve [ −fnqu ] [ −i {nNyYrR}] [ −d destination] [ −sserver] { [ −Sssid[/cloneid]]. . . [ −A

annotation]. . . } [ path . . . ]

DESCRIPTIONnsrretrieve is used to restore archive sav e sets from a NetWorker server. No browsing is available viansr-retrieve. Use ofnsrretrieve is restricted to listed administrators and users of an archive client resource.See thensr_client(5) man page for further details. When not running as root, only the files that the userowns can be retrieved.

When nopatharguments are specified, the entire save set contents will be retrieved. To restrict the archivesave set retrieval to only particular directories or files matching a given path prefix, exact matchingpath’scan be specified to limit which directories and files are retrieved.

OPTIONS-A annotation

The annotationis a regular expression which uniquely identifies a single archive sav e set. Seensrarchive(1m). The regular expression is as used bygrep(1). At least one annotation or ssid(see below) must be specified.

-Sssid[/cloneid]Thessidspecifies the save set IDs for the save sets to be retrieved. When there are multiple cloneinstances for an archive sav e set, you can specify thecloneidto select the particular clone instanceto be retrieved. At least one annotation (see above) or ssid must be specified.

-d destinationSpecifies the destination directory to relocate retrieved files.

-s serverSelects which NetWorker server to use.

-q Thensrretrieve command normally runs with verbose output. This flag turns off the verbose out-put.

-f Indicates that retrieved files will overwrite existing files whenever a name conflict occurs.

-n Does not actually create any directories or files while retrieving.

-i {nNyYrR}Specifies the initial default overwrite response to use when retrieving files and the file alreadyexists. You may specify only one letter. This option is the same as theuasm -i option when run-ning in recover mode. See theuasm(1m) man page for a detailed explanation of this option.

-u Stop when an error occurs during retrieval. Normally, nsrretrieve treats errors as warnings andtries to continue to retrieve the rest of the files requested. However, when this option is used, nsr-retrieve will stop recovering on the first error it encounters.

SEE ALSOgrep(1), nsrarchive(1m),nsr_client(5), nsr_service(5), nsr(1m),nsrd(1m),uasm(1m).


0 Normal exit. This means that all of the requested data was successfully retrieved.<>0 Abnormal exit.

MessagesThe nsrretrieve command reports invalid options by printing a ‘‘usage’’ message describing the availableoptions.

Cannot contact media database onserverThis message indicates that some problem was encountered connecting to the NetWorker server on


NSRRETRIEVE(1m) NSRRETRIEVE(1m)

the named machine.

cannot retrieve backup save setsThensrretrieve command can only be used to restore archive sav e set data.

cannot retrieve migration save setsThensrretrieve command can only be used to restore archive sav e set data.

more than one saveset have the annotationThe specified annotation matched more than one archive sav e set. Usenwretrieve(1m) for retriev-ing save set that has non-unique annotation key.

cannot find saveset with unique annotationThe specified annotation matched no archive sav e set.


NSRSSC(1m) NSRSSC(1m)

NAMEnsrssc − NetWorker save set consolidation program

SYNOPSISnsrssc −cclient−N saveset[ −p pool ] [ -r ] [ −vq ]

DESCRIPTIONnsrsscconsolidates the most recent level 1 (partial) save set and its corresponding full-level sav e set into anew full-level sav e set. This consolidation process effectively achieves the same outcome as a full-levelbackup at the time partial backup was done.

Normally,nsrsscis invoked withinsavegrp(1m) as part of a consolidation-level backup. During the consol-idation-level backup,savegrp(1m) automatically generates a level 1 backup, then callsnsrsscto create aconsolidated backup, using the latest full-level sav e set.

Using nsrsscallows for greater flexibility in scheduling backups and save set consolidation. Unlike thesavegrp(1m)command, which completes a consolidation backup prompty after the level 1 backup is com-pleted,nsrsscallows you to schedule the consolidation at a different time. Scheduling a time between thefull backup and consolidation backup frees up NetWorker to complete other processes.

If you executensrsscmanually, the most recently backed-up save set must be a level 1 sav e set; otherwise,the consolidation will not be successful.

The nsrssccommand requires at least two active devices. The consolidation process uses simultaneousdevice reads and writes to create its consolidated save set. This mechanism creates a restriction upon thelocation of the newly created save set. The new sav e set cannot be created on the same volume where thepartial or full save set was derived. Also, volumes containing the previous full and level 1 sav e sets mustreside on the same storage node.

OPTIONS−c client

Indicates the name of the client whose save set should be included for the consolidation process.

−N savesetIndicates the name for the generated consolidated save set.

−p poolSpecifies the destination media pool to build the consolidated full save set. The pool may be anypool currently registered withnsrd(1m). The selected poolmustbe the same pool type as the pre-vious full-level sav e set. You can view pool values by selecting Pools from the Administrationmenu ofnetworker(1m). Pool values are also listed in theNSRpoolresource. Seensr_pool(5).

If this option is omitted, then the consolidated save sets are automatically built on volume(s)whose media pool is the same as that of the previous full-level sav e set.

−r Removes the level 1 sav e set. If the level 1 sav e set is on tape, then the save set will expire. If thelevel 1 sav e set is on diskfile type volume, then the save set (including its index entries, its mediadatabase entries, and the actual save set data on disk) is removed. Please note thatnsrsscwillneverattempt to remove the level 1 if consolidation fails.

−v Enables verbose operation. In this mode, additional messages may be generated during the con-solidation process.

−q Runs quietly. This is the default mode.

EXAMPLESThe following examples demonstrate how sav e set consolidation can be performed. In both examples, asave set defined in group nameelmancois consolidated for clientdelepanto.The save set data for groupelmancois /etcand/users .

Example:

To perform a save set consolidation, use the following commands:


NSRSSC(1m) NSRSSC(1m)

savegrp −l 1 −I −G elmanconsrssc −c delepanto −N /etcnsrssc −c delepanto −N /users

Note that this example is almost the same as doing asavegrp −G elmanco −l c.The only differences are:1) No index and bootstrap is backed up after data is consolidated.2) If there is a failure during the consolidated process, a full backup is not performed.

Example:

To direct level 1 data to a disk cache (file-type device) and have the level 1 sav e set removed after a full-level sav e set is built on tape, perform the following operations:

1. Set up a pool which only accepts level 1 data and its devices are only file type devices.2. Run the following commands:

savegrp −G elmanco −l 1 −Insrssc −c delepanto −N /etc −rnsrssc −c delepanto −N /users −r

This process removes the level 1 completely. Also, since fast media (disk-file type) is involved, this processmay be a much faster way of generating a full-level sav e set when compared to doing a regular full-levelbackup.

DIAGNOSTICSOn successful completion,nsrsscreturns zero; otherwise, a non-zero value is returned.

Some error codes are:

98Failed because the level 1 and previous full are not in the same storage node.

99Failed, most likely due to a renamed/deleted directory condition

You may also see one of the following messages:

You are not authorized to run this commandOnly root may runnsrssc

Cannot contact media databaseMost likely, nsrmmd(1m) is unavailable to answer queries, or an additional NetWorker daemonmay have completed processing. In either case, the system administrator needs to determine if theNetWorker services need to be restarted. Note, there may be a small interval during startup whenthe services may be unavailable to answer any queries.

SEE ALSOnsr_schedule(5), nsr_mminfo(1m),savegrp(1m)


NSRSTAGE(1m) NSRSTAGE(1m)

NAMEnsrstage − NetWorker save set staging command

SYNOPSISnsrstage

[ −v ] [ −d ] [ −sserver] [ −b pool ] −m −S{ −f file | ssid... }

nsrstage[ −v ] [ −sserver] −C −V volume

DESCRIPTIONThensrstageprogram is used to migrate existing save sets on a manual basis. Migration is the process ofmoving one or more save sets between volumes. The process begins with a clone of the specific save setsto the new volume specified, followed by the deletion of cloned save set entries from the media data base,and finally the possible removal of the save sets from the original source volumes. The second and the thirdoperations are triggered by the successful completion of the previous operation. The data is moved to newmedia volumes, making room for new data on the original volumes.

Migration can be onto any media type (for example, save sets on a file volume can be migrated to an opticaldisk). Thensrstageprogram does not perform simple volume migration; it migrates full save sets.

You can specify exactly which copy (clone) of a save set to use as the source. Ssee the−S option descrip-tion.

OPTIONS−b pool

Specifies the name of the media pool to which the data should migrate. The pool may be any poolcurrently registered withnsrd(1m). You can view values by selecting Pools from the Administra-tion menu ofnwadmin(1m). If you omit this option, the cloned save sets are automaticallyassigned to the Default Clone pool.

−m Performs the actual migration operation.

−sserverSpecifies a NetWorker server with save sets to migrate. Seensr(1m) for a description of serverselection. The default is the current system.

−v Enables verbose operation. In this mode, additional messages are displayed about the operation ofnsrstage, such as save sets that cross volumes, or save set series expansions.

−d Deletes the input file that specifies the save set identifiers to be staged. This option must always bespecified in conjunction with a-f option.

−C Instructsnsrstageto perform a volume cleaning operation. Scans a volume for save sets with noentries in the media data base and recovers their space. You can perform this operation on filetype volumes.

−Sssid Causesnsrstage to treat subsequent command line parameters as save set identifiers. Save setidentifiers are unsigned numbers. You can find out the save set identifier of a save set using themminfo -v command. Seemminfo(1m). The−S option is useful when you want to migrate indi-vidual save sets from a volume, or to migrate all save sets matching somemminfo query. Thesave set identifiers also specify exactly which copy of a sav e set to use as the source. To specifyexact copies, use thessid/cloneidformat for each save set identifier. In this case, thessidand thecloneidare unsigned numbers, separated by a single slash (/). You can find out thecloneid for aparticular copy by referring to themminfo -S report.

−f file Instructsnsrstageto read the save set identifiers from the file specified, instead of listing them onthe command line. The values must be listed one per line in the input file. Thefile may be -, inwhich case the values are read from standard input.

−V Specifies the name of the volume to be cleaned. This option cannot be used with−S or −moptions.


NSRSTAGE(1m) NSRSTAGE(1m)

EXAMPLESMigrate save sets1234 and 4568to a volume in theOffsite Clonepool:

nsrstage −b ’Offsite Clone’ -m -S 1234 4567

Migrate clone instance12345678 of save set 1234to a volume in theDefault Clonepool:nsrstage −m −S 1234/12345678

Migrate all save sets created since last Saturday to a volume in theDefault Clonepool:nsrstage −m −S ‘mminfo −r ssid \

-q ’savetime>last saturday’‘

Recover space from volumejupiter.013:nsrstage −C −V jupiter.013

Only complete save sets can be migrated bynsrstage(1m):

DIAGNOSTICSThe exit status is zero if all of the requested save sets migrated successfully; otherwise status is non-zero.

Several messages are printed denoting a temporary unavailability ofnsrd(1m) for migrating data. Theseare self-explanatory. In addition, you may see one of the following messages:

Adding save set series which includesssidIf running in verbose mode, this message prints whennsrstagenotices that a requested save set iscontinued and requires the entire series to be migrated (even if only part of the series was specifiedby the command line parameters).

Cannot contact media database onserverThe media database (and probably other NetWorker services as well) on the named server is notanswering queries. The server may need to be started. Or if it was just started, it needs to finish itsstartup checks before answering queries.

Cannot open nsrstage session withserverThis message prints when the server is not accepting migration sessions. A more detailed reasonprints on the previous line.

numberis not a valid save setThe given sav e set identifier is not valid. Two forms are understood: simple save set identifiersand those with a cloneid specified. Simple save set are unsigned numbers. You specify the saveset with the cloneid form as two unsigned numbers separated by a single slash (/).

save setnumberdoes not existThe given sav e set (from a−S save set list) does not exist. Verify your save set identifiers usingmminfo(1m).

save set clonenumber/cloneiddoes not existYou specified a specific clone of a save set, but that save has no clones with that clone identifier.Verify your save set identifiers usingmminfo(1m).

volumenamedoes not existThe given volume (if you specified the−V option) does not exist in the media database.

waiting 30 seconds then retryingA temporary error occur.nsrstageautomatically retries its request until the condition is cleared.For example, if all of the devices are busy saving or recovering,nsrstagecannot use these devicesand must wait for two of them to become free.

Space can only be recoverd from file type devices.The given volume (if you specified the−V option) is not a file type volume. This message alsoprints after a successfull migration of data from volumes of type other than file.

SEE ALSOnsrclone(1m) nsr_getdate(3), mminfo(1m),nsr(1m),nsr_pool(5), nsrd(1m),nsrmmd(1m),nwad-min(1m).


NSRTRAP(1m) NSRTRAP(1m)

NAMEnsrtrap − snmp notification scheme for NetWorker messages

SYNOPSISnsrtrap [ −c community] [ −t trap-type] [ −s specific-type] network_management_station

DESCRIPTIONnsrtrap is a mechanism to send NetWorker notifications using the Simple Network Management Protocol(SNMP) trap mechanism. A NetWorker administrator could create a custom NetWorker notificationscheme based on nsrtrap, by configuring the NetWorker events and priorities.

A NetWorker administrator could create notification schemes to receive messages on different networkmanagement consoles by configuring the events and priorities and specifying the desired network manage-ment station as the location to receive the trap messages.

To create a new SNMP notification, follow the steps below:

1. Open the Notifications window from the Customize menu.

2. Choose the Details option under the View menu.

3. Click on the Create button.

4. Enter the name of the new notification in the Name field.

5. In the Action field, enter the command nsrtrap along with the network management station name towhich the networker SNMP notification should be sent.

6. Set the events and priorities desired.

7. Click on the Apply button.

OPTIONS−c community the SNMP community string. This defaults to the community "public".

−sspecific-type the SMI (System Management Interface) Network Management Private Enterprise Code.

−t trap-type could be one of the SNMP trap types[0-6]. The default is 6, the "enterprise-specific traptype.

SEE ALSOnsr(1m)nsr_resource(5)


NSRWA TCH(1m) NSRWA TCH(1m)

NAMEnsrwatch − command for character-based display of NetWorker status

SYNOPSISnsrwatch [ −sserver] [ −p polltime]

DESCRIPTIONThe nsrwatch command displays a NetWorker server’s status. The server’s name is specified by theoptional−sserverargument. If no server is specified, it defaults to the same server that would be used by acommand, such asrecover(1m) in the current directory. If there is no NetWorker service on the selectedmachine, the command issues an error message. The polling interval is specified by the optional−p poll-timeargument (in seconds). The default is two seconds.

Users can runnsrwatch from any terminal that has enoughtermcap(5) capabilities for cursor positioning;it does not require any particular window system. Thensrwatch program gets its information via remoteprocedure calls to the specified server. This way it can be used from any machine that can access the serverthrough the network.

The nsrwatch display is divided into a header and several panels: theserverpanel, thedevicepanel, thesessionspanel, themessagespanel, and thepending messagepanel. The panel sizes are adjusted dependingon the size of the terminal or window being used.

The header contains the name of the server and the current time. Theserverpanel provides current statusof the server. The first line of the panel is reserved for error messages. This line is usually blank. The nextline tells how long the server has been up, and the server’s release version (which may not be the same asthe client’s release version). The following lines display how many sav es and recovers the current serverhas performed.

The devicepanel displays the devices known to the current server. For each device, the panel displays itsname, the device type, the name of the mounted volume, or(unmounted) if no volume is mounted, anddevice status. The name may be followed by(J) if the device is configured as part of a jukebox device.The sessionspanel provides current save set information for each active session (saving, recovering, orbrowsing). Themessagespanel displays a history of messages of general interest to the operator. Finally,thependingmessage panel displays messages that require operator intervention.

Thensrwatch program runs continuously until quit, stopped, or interrupted (Control-C, for example). Typ-ing theq character quits the program, the Control-L forces a screen clear and redraw, while any other char-acter forces the status to be updated.

Thensrwatch program checks for new devices at a slower rate than the polling rate, so it might take up to aminute after a new device is added before the device is noticed. To recognize the devices immediately,either restart the program or press Control-L. Deleted devices may cause a ‘‘resource does not exist’’ mes-sage temporarily, but otherwise they are noticed immediately.

The nsrwatch program adapts to changes in the screen size, if supported by the underlying environment.For example, if a window terminal emulator is resized, the size of each field may change to match the win-dow. If the window is too small, all the devices, sessions, and messages, might not be displayed. For bestresults, use a window of at least 30 lines.

OPTIONS−sserver

Sets the current NetWorker server toserver.

−p polltimeSets the polling interval to bepolltimeseconds.

SEE ALSOtermcap(5), nsr_notification(5), nsr_device(5), nsr_service(5), recover(1m),nsradmin(1m),nsr(1m),nsrd(1m),nwadmin(1m).


NWADMIN(1m) NWADMIN(1m)

NAMEnwadmin, networker − graphical administration interface to NetWorker

SYNOPSISnwadmin [ −sserver]networker [ −sserver]

DESCRIPTIONnwadmin is an X Window System application. It is used to administer and monitor NetWorker servers.

The server’s name may be specified with the−s serverargument. When no server is specified,nwadminuses the server selection rules found innsr(1m). When multiple NetWorker servers are accessible, theymay be selected from within thenwadmin command.

The nwadmin command is used to administer NetWorker servers. Clients may be added and deleted.Schedules controlling save lev els may be created and modified. Groups of clients may be formed and con-trolled together. Directives controlling how data is saved may be defined and changed. Cloning of savesets and recover by sav e sets may be specified. Cloning of entire backup volumes may also be specified.The notification messages may be displayed. In general, there is a panel for each component.

The current state of NetWorker servers may be monitored. The amount of data saved, the number of clientssaving, and requests for mounting and unmounting volumes, are among the items displayed. This informa-tion is displayed on the console panel.

The graphical interfaces for backup and recover are available throughnwbackup(1m) andnwrecover(1m),respectively.

A complete explanation of thenwadmin command may be found in theNetWorker Administrator’s Guide.

OPTIONS−sserver



The X11 resources fornwadmin.

NOTEnetworker is linked tonwadmin.

SEE ALSOnsr(1m),nsradmin(1m),nwbackup(1m),nwrecover(1m),The NetWorker Administrator’s Guide


NWARCHIVE(1m) NWARCHIVE(1m)

NAMEnwarchive − NetWorker graphical archive interface

SYNOPSISnwarchive [ −sserver]

DESCRIPTIONnwarchive is an X Window System application. It is a front end tonsrarchive(1m) and used to archivefiles to a NetWorker server on an ad hoc basis. Normally, files are archived automatically.

The server’s name may be specified with the−s serverargument. When no server is specified,nwarchiveuses the server selection rules found innsr(1m). When multiple NetWorker servers are accessible, theymay be selected from within thenwarchive command.

NetWorker supports both scheduled network-wide archives and manual archives of client system files anddirectories. To request an immediate manual archive, runnwarchive

Check that the correct NetWorker server is selected. The Server is identified in the Main window. You canchange servers using the Change Server command if necessary. This is the Server to which the client fileswill be backed up. The hostname of the current client is displayed in the Client field. The pathname of thecurrent directory is displayed in the Selection field.

Change directories by entering the full pathname in the Selection field or by highlighting the icon in theArchive window.

To perform a manual archive, first mark the files and directories that you want to back up by selecting theircheckboxes. Then select Start archive... from the File menu of the Archive window. You must enter anannotation for the archive.

Monitor the progress of the archive in the Archive Status window. Check to see that an archive volume ismounted in the Pending display of the Main window.

A complete explanation of the nwarchive command may be found in the NetWorker Administrator’s Guide,UNIX Version.

OPTIONS−sserver



The X11 resources fornwarchive.

SEE ALSOnsr(1m),nsradmin(1m),nsrarchive(1m),nsrretrieve(1m)

The NetWorker Administrator’s Guide, UNIX Version


NWBACKUP(1m) NWBACKUP(1m)

NAMEnwbackup − NetWorker graphical backup interface

SYNOPSISnwbackup [ −sserver] [ path]

DESCRIPTIONnwbackup is an X Window System application. It is a front end tosave(1m) and used to save files to aNetWorker server on an ad hoc basis. Normally, files are saved automatically.

The server’s name may be specified with the−s serverargument. When no server is specified,nwbackupuses the server selection rules found innsr(1m). When multiple NetWorker servers are accessible, theymay be selected from within thenwbackup command. Ifpath is specified,nwbackup will set initializethe current selection to the given path. The default selection ifpath is not specified is the current workingdirectory.

NetWorker supports both scheduled network-wide backups and manual backups of client system files anddirectories. To request an immediate manual backup, runnwbackup

Check that the correct NetWorker server is selected. The Server is identified in the Main window. You canchange servers using the Change Server command if necessary. This is the Server to which the client fileswill be backed up. The hostname of the current client is displayed in the Client field. The pathname of thecurrent directory is displayed in the Selection field.

Change directories by entering the full pathname in the Selection field or by highlighting the icon in theBackup window.

To perform a manual backup, first mark the files and directories that you want to back up by selecting theircheckboxes. Then select Start backup... from the File menu of the Backup window. You must selectwhether to compress files or exclude patterns in the Backup Options dialog box to continue the backup.

Monitor the progress of the backup in the Backup Status window. Check to see that a backup volume ismounted in the Pending display of the Main window.

A complete explanation of the nwbackup command may be found in the NetWorker User’s Guide.

OPTIONS−sserver



The X11 resources fornwbackup.

SEE ALSOnsr(1m),nsradmin(1m),nwbackup(1m),nwrecover(1m),save(1m)

The NetWorker User’s Guide


NWRECOVER(1m) NWRECOVER(1m)

NAMEnwrecover − NetWorker graphical recover interface

SYNOPSISnwrecover [ −sserver] [ −c client ] [ −T browse time] [ path ]

DESCRIPTIONnwrecover is an X Window System application. It is used to recover lost files that have been saved withNetWorker. If you are running in a non-X11 environment,recover(1m) may be used to recover files.

The server’s name may be specified with the−s serverargument. When no server is specified,nwrecoveruses the server selection rules found innsr(1m). When multiple NetWorker servers are accessible, theymay be selected from within thenwrecover command. Ifpath is specified,nwrecover will attempt to ini-tialize the current selection to the given path. The default attempted selection ifpath is not specified is thecurrent working directory.

If you are recovering files that were saved with Access Control Lists (ACLs), you need to be root or the fileowner to recover the file. Files with an ACL have a trailing ’+’ (e.g., -rw-r--r--+) after the mode bits whenviewing file details. Seerecover(1m) for more information about ACLs.

There are three basic steps to recover a lost file: (1) Browse NetWorker’s index in the Recover window tofind the lost file, (2) Mark the file for recovery by selecting its checkbox and (3) Start the recovery. In addi-tion, there are recover commands for relocating recovered files (Relocate), finding past versions of a file(Versions), changing the browse time (Change Browse Time), showing the files you have marked for recov-ery (Show Marked), and overwriting or renaming recovered files that are in conflict with existing files(Conflict Resolution).

Opening the Recover window connects the client to its indexes maintained on the server. The entries in theindex represent previously backed-up files and are organized exactly like the filesystem. To browse theindex for another filesystem, enter the pathname in the Selection field.

To browse the index: Use the Recover window View menu to select the browsing level of your directories.Use the mouse to open a directory and display its contents.

To mark files: After you have located your files by browsing the index, mark the files you want to recoverby selecting their checkboxes. Or, highlight a file and use the Mark command from the File menu to markfiles. You can list the files that you have marked for recovery with the Show Marked command from theView menu.

To start the recovery: Select the Start recover command from the File menu. The Conflict Resolution dialogbox appears, where you tell NetWorker what to do when a conflict occurs between a recovered file and anexisting file. You select whether to be prompted for each individual conflict or to select one global resolu-tion for all conflicts. Then you tell NetWorker whether to Rename the recover file with a .R extension topreserve both files, to Discard the recover file and preserve the existing file, or to Overwrite the existing fileto preserve the recover file as the only copy of the file.

After you press OK in the Conflict Resolution dialog box, the recover continues. NetWorker will then auto-matically determine the media needed to complete the recovery, prompt the operator to mount the media,and execute the recovery. You can monitor the status of the recovery in the Recover Status window.

Before starting the recovery, you have the option of relocating the recover files with the Relocate command.Enter the pathname of a new or existing directory in which to place your recovered files.

The Recover window also offers two commands for browsing the index in the past. Versions shows you theentire backup history for a file. Change Browse Time allows you to change the time at which you are view-ing the on-line index.

A complete explanation of thenwrecovercommand may be found in the NetWorker User’s Guide.

OPTIONS−sserver



NWRECOVER(1m) NWRECOVER(1m)

−c clientSet the current NetWorker client index to browse toclient.

−T browse timeSet the current index browse time tobrowse time(in nsr_getdate(3) format). Using this option isequivalent to using thechange browse timedialog withinnwrecover.


The X11 resources fornwrecover.

SEE ALSOnsr_getdate(3), nsr(1m),nsradmin(1m),nwbackup(1m), recover(1m)

The NetWorker User’s Guide


NWRETRIEVE(1m) NWRETRIEVE(1m)

NAMEnwretrieve − NetWorker graphical retrieve interface

SYNOPSISnwretrieve [ −sserver]

DESCRIPTIONnwretrieve is an X Window System application. It is used to retrieve files that have been archived withNetWorker. If you are running in a non-X11 environment,nsrretrieve(1m) may be used to retrieve files.

The server’s name may be specified with the−s serverargument. When no server is specified,nwretrieveuses the server selection rules found innsr(1m). When multiple NetWorker servers are accessible, theymay be selected from within thenwretrieve command.

If you are retrieving files that were archived with Access Control Lists (ACLs), you need to be in groupoperator or the file owner to retrieve the file. Seensrretrieve(1m) for more information about ACLs.

There are three basic steps to retrieve a lost file: (1) Browse NetWorker’s list of Archives in the Retrievewindow. (2) Select the Archive you wish to retrieve. (3) Start the retrieve.

Opening the Retrieve window connects with the NetWorker server indexes. Selecting the Query button dis-plays a list of archive available on the server. The entries in the list represent previously archived files.

To start the retrieve: Select the Start retrieve command from the File menu. The Retrieve Status dialog boxappears, and you may enter a path to relocate to and select if you want to overwrite existing files.

After you press OK in the Retrieve Status dialog box, the Retrieve will begin retrieving the selectedarchives and status will be displayed in the Status field. NetWorker will then automatically determine themedia needed to complete the retrieve, prompt the operator to mount the media, and execute the retrieve.

A complete explanation of the nwretrieve command may be found in the NetWorker Administrator’s Guide,UNIX Version.

OPTIONS−sserver


FILES

/usr/lib/X11/app-defaults/NetworkerThe X11 resources fornwretrieve.

SEE ALSOnsr(1m),nsradmin(1m),nwarchive(1m),nsrarchive(1m),nsrretrieve(1m)

The NetWorker Archive User’s Guide


ORAEMCASM(1m) ORAEMCASM(1m)

NAMEoraemcasm − A utility, which is part of the NetWorker Module for EMC Symmetirx for Oracle, used forsaving anOracle databasewhich resides on anEMC Symmetrixdisk array

SYNOPSISoraemcasm[ −avXZ ] [ −C | -F ] [ −h server] [ −c client-name] [ −Soracle-sid] [ −H oracle-home] [ −uresfile] [ −p directory] [ −g group] [ −b pool ] [ −T tablespace. . . ] [ −P pfile ]

DESCRIPTIONThe oraemcasm module is a standard external Application Specific Module (ASM) that uses EMCTimeFinder or SRDF technology to save an Oracle datebase that resides on an EMC Symmetrix unit. Seeuasm(1m) for a general description ofASM’s.

One or more tablespaces or an entire database can be processed by one backup operation. Theoraemcasmmodule provides the flexibility to perform archive log only backups, hot (online) backups, or cold (offline)backups.

Multiple tablespaces may be specified following a single−T option. To backup an entire database omit the−T option. Backups may also be performed by running thesave(1m) command and specifying a dummyfile. The dummy file must be in a directory which contains a directive file specifying oraemcasm should berun against the tablespace or database. Refer tosave(1m) for details.

Tw o databases sharing EMC Symmetrix devices cannot be backed up in parallel. After performing abackup the user should run and keep the output fromoraemcmap(1m) to help with recovery.

OPTIONSoraemcasmaccepts the options described below.

−a Archive log backups. With this option, only the backup control file and archive logs are saved.Default is to save the entire database.

−X Encryption. Causes thexlateasmto be called when saving tablespaces (seeuasm(1m) for detailson xlateasm). Default is to not encrypt files.

−Z Compression. Causes thecompressasmto be called when saving tablespaces (seeuasm(1m) fordetails on compressasm). Default is to not compress files.

−v Verbose. Cause the program to tell you in great detail what it is doing as it proceeds. Default isquiet (no verbosity).

−C Cold backup, normal. Used for cold backups. Database will be shutdown using the "shutdownnormal" command. Default is "hot" or online backups. If both -C and -F are specified, -F takesprecedence.

−F Cold backup, immediate. Used for cold backups. Database will be shutdown using the "shutdownimmediate" command. Default is "hot" or online back ups. If both -C and -F are specified, -Ftakes precedence.

−b poolSpecify which volume pool should be used for the save set.

−h serverSpecify which machine to use as the NetWorker server. If not specified, use local machine asserver.

−c client-nameSpecify the client name for starting the save session. This is useful on clients with multiple net-work interfaces, and hence multiple host names. It can be used to create multiple index databasesfor the same physical client. If not specified, use local machine as client-name.


ORAEMCASM(1m) ORAEMCASM(1m)

−Soracle-sidSpecify the Oracle SID of the database to be backed up. If not set, $ORACLE_SID from the envi-ronment is used.

−H oracle-homeSpecify the Oracle home directory of the database to be backed up. If not set, $ORACLE_HOMEfrom the environment is used.

−u resfileSpecify the resource file for the configuration information. If not set, /nsr/res/oraemcasm.res isused.

−p directorySpecify the location for the "backup" control file. If not set, the Oracle "log_archive_dest" direc-tory is used.

−g groupSpecify the save group that will be used to select the volume pool.

−T tablespaceSpecify the tablespace to be backed up. The user can enter multiple tablespaces following the −Toption. If this option is omitted, then all tablespaces (or the entire database) are backed up.

−P pfileSpecify the Oracle parameter file to be used to re-start the Oracle database after a cold backup. Ifnot specified, $ORACLE_HOME/dbs/init$ORACLE_SID.ora is used.

EXAMPLESBackup a Single Tablespace

In this example, the Oracle SID is ’EMC2’, the Oracle home directory is ’/usr/oracle’, and theresource file is /nsr/res/neptune.res. To backup tablespace ’Test1’ and ’Test2’, use:

oraemcasm -S EMC2 -H /usr/oracle -u neptune.res -T Test1 Test2

Backup Entire DatabaseTo backup the entire ’EMC2’ database using the resource file ’/nsr/res/pluto.res’ use:

oraemcasm -S EMC2 -H /usr/oracle -u pluto.res

FILES/nsr/res/oraemcasm.res

Sample oraemcasm resource file.

SEE ALSOemcdiscover(1m),oraemcmap(1m),nsr(5), nsr(1m),nsr_directive(5), nsrindexasm(1m),nsrmmdbasm(1m), recover(1m),save(1m),scanner(1m)


ORAEMCMAP(1m) ORAEMCMAP(1m)

NAMEoraemcmap − A utility, which is part of the NetWorker Module for EMC Symmetrix for Oracle, used forshowing the configuration of Oracle tablespaces and datafiles installed on anEMC Symmetrixdisk array.

SYNOPSISoraemcmap[ −Soraclesid] [ −H oraclehome] [ −M BCV | RDFBCV] [ −T tablespace. . . ]

DESCRIPTIONThe oraemcmaputility displays the current Oracle database and EMC Symmetrix TimeFinder or SRDFconfiguration. This program should be run after each successfuloraemcasm(1m) backup, and the outputpermanantely saved. Recovering an Oracle database which was saved from a BCV, or remote BCV device(across an SRDF link) usingoraemcasm(1m) requires knowing the original tablespace layout on the Sym-metrix devices and datafile names, which this utility will provide for you.

The oraemcmap(1m) utility queries the Symmetrix unit and your Oracle database to locate tablespacenames and how these tablespaces are mapped onto the Symmetrix devices.

When available,oraemcmap(1m) will display the following information for each tablespace in thedatabase:

Tablespace: tablesapce-nameDatafile: datafile-name

Disk Group: group-name Volume: volume-nameStd Dev: device-info paired dev: device-info

Note:On HP/UX, the "Disk Group" field label is called "Vol. Group".

The "device-info" is displayed asSymmID-SymmDev

The "paired dev" output varies depending upon which backup method is being employed. Possi-ble field label values are:

BCV - for BCV backupsRemote BCV - for remote BCV (or RDFBCV) backups

OPTIONS−Soraclesid

Specify which instance of the Oracle database should be queried. If not specified, $ORA-CLE_SID, from the environment, is used.

−H oraclehomeSpecify the Oracle home directory (for the instance being queried). If not specifed, $ORA-CLE_HOME, from the environment, is used.

−M BCV | RDFBCVSpecify the backup method that will be employed against this database. Backup methods corre-spond to values entered into your oraemcasm resource file. If no backup method is specified, BCVis used.

−T tablespace. . .Specify the tablespace (or tablespaces), you wish to display mapping information about. Thedefault is to list all tablespaces in the database.


ORAEMCMAP(1m) ORAEMCMAP(1m)

EXAMPLESDisplay Configuration Information

The following example displays remote BCV device information about the TOOLS tablespace:

oraemcmap -S EMCSOL2 -H /oracle/oracle_733 -M RDFBCV -T TOOLS

Tablespace: TOOLSDatafile: /lvms/lv01dg01/oradata/EMCSOL2/tools02.dbf

Disk Group: dg01 Volume: vol01Std Dev: 000183500313-0B9 Remote BCV: 000183500311-051Std Dev: 000183500313-0B8 Remote BCV: 000183500311-050

Datafile: /symm02/oradata/EMCSOL2/tools03.dbfDevice: /dev/rdsk/c1t1d1s1Std Dev: 000183500313-0B1 Remote BCV: 000183500311-049

Datafile: /dev/rdsk/c1t1d3s4Std Dev: 000183500313-0B3 Remote BCV: 000183500311-04B

In this example, notice that the TOOLS tablespace is comprised of three datafiles. Also notice, the Sym-metrix ID is 000183500313 for the standard devices and 000183500311 is the Symmetrix ID where remoteBCVs are located.

This example displays BCV device information (default) for the USERS tablespace. Also, $ORA-CLE_HOME and $ORACLE_SID have both been set.

oraemcmap -T USERS

Tablespace: USERSDatafile: /symmvxfs01/oradata/EMCHP1/users01.dbf

Vol. Group: vg01 Volume: lvol2Std Dev: 000183500311-0C7 BCV: 0EFStd Dev: 000183500311-0C6 BCV: 0EE

Notice that this was run from an HP/UX machine (i.e. "Vol. Group" instead of "Disk Group"). Also noticethat the Symmetrix ID is not displayed for the BCV device.

An appropriate message will be displayed if the device is not a Symmetrix device.

Send oraemcmap output to the printerAssuming $ORACLE_SID contains the name of your Oracle instance, and $ORACLE_HOMEpoints to the Oracle home directory, you can run oraemcmap and send the output to your defaultprinter by running:

oraemcmap | lpr

SEE ALSOoraemcasm(1m),emcdiscover(1m)


pathownerignore(5) pathownerignore(5)

NAMEpathownerignore − ignore path-ownership rules during scheduled saves

SYNOPSIS<nsr_bin>/pathownerignore

DESCRIPTIONIn a clustered environment, the NetWorker software must distinguish between filesystems associated withthe physical client, and those that are managed by a resource group (a virtual client). These criteria arereferred to as the path-ownership rules. These rules determine which client index a sav e set is written to.

If a filesystem owned by a virtual client is defined in the save set list for a physical client resource, bydefault the filesystemwill not be backed up during a scheduled save. The same is true for a filesystemowned by a physical client defined in the save set list for a virtual client resource. In both cases, the filesys-tem is omitted. This occurs because the NetWorker software views the client (which owns the filesystem) asnot having matched the client of the current scheduled save.

To check the NetWorker path-ownership rule:

1. Run the following command on the NetWorker server:# sav e grp -p -cclient_name

2. Review which filesystems are owned by client_name. This procedure is part of the normal cluster instal-lation setup. For detailed instructions, refer to the appropriate Legato NetWorker Installation Guide.

To test for the existence of misappropriated save sets, run a test probe with the verbose option set. The com-mand output will warn you to which client indexes a sav e set will be saved. For example:

# sav e grp -pv -cclient_name group_name

To ignore NetWorker default path-ownership rules, you can create the <nsr_bin>/pathownerignore file. Thisfile causes the NetWorker software to back up the filesystem in question; however, the filesystem will besaved under the index of its correct owner. Creating the <nsr_bin>/pathownerignore file is not recom-mended, but it might be required under special circumstances. The <nsr_bin>/pathownerignore file doesnot override the default path-ownership rules. It causes the path-ownership rules to be ignored when deter-mining if a filesystem should be backed up during a scheduled save.

SEE ALSOsave(1m),savegrp(1m),savefs(1m)

NOTESTo override the path-ownership rules and have a sav e set written to an index other than its default owner,one must use the "save -cclient_name" command. Refer tosave (1m)for more information.


PMODE(1m) PMODE(1m)

NAMEpmode − print mode sense data

SYNOPSISpmode[ -f filename]

DESCRIPTIONThe pmode program will parse the data output by themsense(1m) program and print in technologicalEnglish. (C-style variables with hexadecimal numbers)

OPTIONS−f filenameSpecifies input; otherwise standard input is assumed.

EXAMPLESample output might look like:

viper# msense -a 0.0.0 -p 0x03 | pmodeMode Header: mdl=35 mtype=0x0 dparm=0x10 bdlen=8Block Desc[0]: dens=0x0 nblks=3933040 blklen=512

Fixed Page, code 0x03 (Format Device):tracks_per_zone: 0xf

alt_sectors_per_zone: 0x22alt_tracks_per_zone: 0x0alt_tracks_per_vol: 0x0sectors_per_track: 0x5e

data_bytes_per_sect: 0x200interleave: 0x1

track_skew_factor: 0x8cylinder_skew_factor: 0x11

SSEC: 0x0HSEC: 0x1RMB: 0x0SURF: 0x0

SEE ALSOmsense(1m)


PRECLNTSAVE(1m) PRECLNTSAVE(1m)

NAMEpreclntsave − Child process to run pre-processing commands for Networker savepnpc.

SYNOPSISpreclntsave−s server −c client −g group[−D debuglevel]

DESCRIPTIONThe preclntsaveprocess verifies if there is an existing /nsr/tmp/<grpname>.tmp file, which indicates thatthe pre-processing commands had been run. If so, it exits with status 0 to let savepnpc resume its normalsave task. Otherwise, it locks the /nsr/res/<grpname>.res.lck file, invokes all the pre-processing commandsspecified in the /nsr/res/<grpname>.res file, then creates /nsr/tmp/<grpname>.tmp file, spawns thepst-clntsaveprocess, and exits with status 0.

Note: This is to be invoked bysavepnpcprogram only. It is not meant for users to use.

OPTIONS−sserver

Specifies the controlling server.

−c clientThe name of the client where the pre-processing commands will be performed on.

−g groupSpecifies the group name that is being run.

−D debuglevelFor debugging purpose, the debug level could be 1, 2 or 3.

SEE ALSOsavepnpc(1m),pstclntsave(1m),save(1m)


PSTCLNTSAVE(1m) PSTCLNTSAVE(1m)

NAMEpstclntsave − Child process of preclntsave to run post-processing commands for Networker savepnpc.

SYNOPSISpstclntsave−s server −c client −g group[−p pollinterval] [−t timeout] [−D debuglevel]

DESCRIPTIONThe pstclntsaveprocess checks the WORKLIST attribute of the CLIENT resource from the server everynumber of seconds specified in the poll interval. Whenever the time_out condition or the WORKLIST isNIL (whichever comes first)pstclntsave performs all the post-processing commands specified in/nsr/res/<grpname>.res file, unlinks /nsr/tmp/<grpname>.tmp and /nsr/res/<grpname>.lck, then records theresults (success or failure) in /nsr/log/savepnpc.log file.

Note: This is to be invoked bypreclntsaveprogram only. It is not meant for users to use.

OPTIONS−sserver

Specifies the controlling server.

−c clientThe name of the client where the pre-processing commands will be performed on.

−g groupSpecifies the group name that is being run.

−p pollintervalSpecifies how often (in seconds) to poll the server.

−t timeoutThe timeout condition in nsr_getdate() format string to start the post-processing commands. Thiscan also be specified in the /nsr/res/<grpname>.res file.

−D debuglevelFor debugging purpose, the debug level could be 1, 2 or 3.

SEE ALSOpreclntsave(1m),savepnpc(1m),save(1m)


RECGEMS(1m) RECGEMS(1m)

NAMErecgems − recover a GEMStation, migrate to a new version of GEMS

SYNOPSISrecgems[ −nispmglNh ] [ −E directory] [ −L directory]

DESCRIPTIONrecgemsrecovers the information in a GEMStation previously backed-up withsavegems(1m). recgemswill restore a GEMStation by bringing down the GEMS server software, then importing the configurationdata for each GEMS subsystem. Once all imports have completed, the GEMStation is brought up.

OPTIONS−n Do not restart the GEMStation after recovery.

−i Do not initialize the RCSM database. The default is to initialize the RCSM database.

−s Do not initialize the JMAPI database. The default is to initialize the JMAPI database.

−p Do not import the policy engine database. The default is to import the policy enginedatabase.

−m Do not import the managed object database. The default is to import the managed objectdatabase.

−g Do not import the GEMS software manager database. The default is to import the soft-ware manager database.

−l Do not import the GEMS license manager database. The default is to import the licensemanager database.

−N Do not remove the recovered directory when done. The recovered directory is whererecgemsexpects files to be located. The default location for this directory is listed in theFILES section below. The−E option used for upgrading from a previous version ofGEMS changes where the files are located. The default behavior is to remove the recov-ered directory.

−h Prints the usage forrecgems, then exits.

−E directory Specify the directory where import files will be read. See the−N option. Using thisoption implies thatrecgemsis being used to upgrade a GEMStation from a previous ver-sion of GEMS.

−L directory Enables log files and places them in the specified directory. See theFILES sectionbelow.

FILESrecgems.out.log

The standard output of the import processes. This file contains information useful for tracking theprogress of a recovery. This file is only created whenrecgemsis executed with the−L option.

recgems.err.logThe error output of the import processes. This file contains information useful for determining thecause of a failed recovery. This file is only created whenrecgemsis executed with the−L option.

/gems/tmp/backupThe directory where import data files are read. This directory is usually recovered from a Net-Worker save set prior to recovering a GEMStation withrecgems. If recgemsis executed with the−E option, thenrecgemswill reference the directory specified instead of this directory.

SEE ALSOnewgems(1m),nwrecover(1m), recover(1m),savegems(1m)

1.3.Build.149a May 04, 2001 1

RECGEMS(1m) RECGEMS(1m)


0 Normal exit. This means that the recover executed normally.<>0 Abnormal exit. This means thatrecgemswas not executed with the correct arguments, or that one

of the GEMS subsystems responsible for importing could not be found.

Log MessagesSeveral messages may be present in the log files listed above.

NOTESThe log files generated (seeFILES above) should be checked to insure that the recovery process was suc-cessfully completed by each subsystem responsible for importing data into the GEMStation.

1.3.Build.149a May 04, 2001 2

RECOVER(1m) RECOVER(1m)

NAMErecover − browse and recover NetWorker files

SYNOPSISrecover [-f] [-n] [-q] [-u] [-i {nNyYrR}] [-d destination] [-c client] [-t date] [ -s server] [ dir ]recover [-f] [-n] [-u] [-q] [-i {nNyYrR}] [-I input file] [-d destination] [-c client] [-t date] [ -s server] -apath. . .recover [-f] [-n] [-u] [-q] [-i {nNyYrR}] [-d destination] -s server-S ssid[/cloneid][-S ssid[/cloneid]]. . . [path]. . .recover [-f] [-n] [-q] [-i {NYR}] -R recover-target-c client [-d destination] [-t date] [ -s server] [ dir ]recover [-f] [-n] [-q] [-i {nNyYrR}] [-t date] [-s server] [-N system save set]

DESCRIPTIONrecover browses the saved file index and recovers selected files from the NetWorker system. The file indexis created when files are saved withsave(1m). When in interactive mode (the default), the user is presentedwith a view of the index similar to a UNIX filesystem, and may move through the index to select andrecover files or entire directories. In automatic mode (−a option), the files specified on the command lineare recovered immediately without browsing. While in save set recover mode (−S option), the save set(s)specified are retrieved directly without browsing the NetWorker file index. Use of save set recover mode isrestricted to root.

When using recover without the −S option, users who are root may recover any file. The remaining permis-sion checking rules described in the paragraph apply to users who are not root. For files that don’t hav e anAccess Control List (ACL), the normal Unix mode bits must allow you to read the file in order to recover it.Files with an ACL can only be recovered by their owner or by root.

OPTIONS−a Specifies automatic file recovery with no interactive browsing.Pathspecifies one or more files or

directories to be recovered. Symbolic links are not followed, though the link file itself will berecovered. Mount points are also not followed unless the most recentsave(1m) was performedwith the ’-x’ option.

−Sssid[/cloneid]Specifies save set recover mode and can only be used by root. This mode can be used to imple-ment fast batch file recovery without requiring the NetWorker file index entries.ssidspecifies thesave set id’s for the save set(s) to be recovered. When there are multiple clone instances for a saveset, thecloneidcan also be specified to select the particular clone instance to be recovered from.When nopatharguments are specified, the entire save set contents will be recovered. One more ormorepath’s can be specified to limit which directories and files are actually recovered. Ifpath’sare supplied, then the beginning of each path name as it exists in the save set must exactly matchone of thepath’s before it will be recovered. Shell like file name matching using meta characterslike ‘* ’, ‘ ?’, and ‘[...]’ is not done. You can use apath that ends in with a slash (‘/’) to force adirectory only match (e.g., use apath of /etc/fs/ instead of/etc/fs to prevent files like/etc/fsckfrom being recovered as well).

−d destinationSpecifies the destination directory to relocate recovered files. Using this option is equivalent tousing therelocatecommand when in interactive mode (see usage). Relative paths are interpretedrelative to the current working directory.

−sserverSelects which NetWorker server to use.

−c clientClient is the name of the machine that saved the files. When browsing a directory that was savedby another client, the pathnames will reflect the file tree of the client that saved the files. Bydefaultsaveandrecover determine the client name from the filesystem table. This option mightbe necessary if the−L option was used on thesavecommand. This option cannot be used in con-junction with the−Sssidoption (save set recover mode).



−t date Display/recover files as of the specifieddate (in nsr_getdate(3) format). Using this option isequivalent to using thechangetimecommand with the givendatewhen in interactive mode (seeusage). This option cannot be used in conjunction with the−Sssidoption (save set recover mode).

−q Turns off the verbose output. Therecovercommand normally runs with verbose output.

−f Forces recovered files to overwrite any existing files whenever a name conflict occurs. This is thesame as specifying−iY .

−n Does not write or create any files or directories when recovering.

−i {nNyYrR}Specifies the initial default overwrite response to use when recovering existing files. Only one let-ter may be specified. This option is the same as theuasm −i option when running in recovermode. See theuasm(1m) man page for a detailed explanation of this option. For directed recov-ers (see the -R option), only ’N’, ’Y’, and ’R’ are valid values.

−I input fileTakes the paths to recover from the command line, and read paths to recover from the named file.The paths must be listed one per line. If no paths are specified on the command line, then onlythose paths specified in the file will be recovered. To be used in conjunction with -a option.

−R recover-targetSpecifies the name of the remote machine to direct the recovery. This is used in conjunction withthe -c option to specify browsing of another client’s index. When the -R option is used, either the-f or the -i option must also be specified in order to instruct the recover target what to do when it isrecovering existing files. Note that the values ’N’, ’Y’, and ’R’ are the only valid ones to use withthe -i option for directed recovers. Note also that the -a option is not supported with the -R option.

−N system save setUsed to recover the following system save sets: SYSTEM DB, SYSTEM FILES, or SYSTEMSTATE. (Windows Only)

−u Stops when an error occurs during recovery. Normally, recover treats errors as warnings and triesto continue to recover the rest of the files requested. However, when this option is used, recoverwill stop recovering on the first error it encounters. This option is not valid for directed recovers.

USAGEWhen using recover in the interactive mode, an image of the filesystem at a particular time is presented.Using commands similar to the shell, you can change the view and traverse the filesystem. Files may beselected for recovering, and the actual recover command issued.

The following commands manipulate the view of the filesystem and build the list of files to recover. In allof the commands that take anameargument pattern matching characters can be used. The pattern matchingcharacters and regular expression format are the same as for the UNIX shellsh(1).

ls [ options] [ name ...]List information about the given files and directories. When nonamearguments are given,ls liststhe contents of the current directory. When anameis given andnameis a directory, its contentsare displayed. Ifnameis a file, then just that file is displayed. The current directory is representedby a ‘.’ (period). The options to this command correspond to those of the UNIX command,ls(1).An additional recover specific−S option can be used to select the save time instead of the lastmodified time for sorting (with the−t option) and/or printing (with the−l option). Files that havebeen added to the recover list are preceded by a ‘+’. Files that have an ACL have a trailing ’+’(e.g. -rw-r--r--+) after the mode bits when viewing file details.

lf [ name ...]is the same asls −F. Directories are marked with a trailing ‘/’, symbolic links with a trailing ‘@’,sockets with a trailing ‘=’, FIFO special files with a trailing ‘|’, and executable files with a trailing‘* ’.



ll [ name ...]is the same asls −lgsF. Generates a long format listing of files and directories. This commandcan be used to find the value of a symbolic link.

cd [ directory]Change the current working directory to [directory ]. The default directory is the directoryrecover was executed in. Ifdirectory is a simple symbolic link,cd will follow the symbolic link.However, ifdirectory is a path containing symbolic links anywhere but at the end of the path, thecd command will fail; you shouldcd a component of the path at a time instead.

pwd Print the full pathname of the current working directory.

add [ name ...]Add the current directory, or the named file(s) or directory(s) to the recover list. If a directory isspecified, it and all of its descendent files are added to the recover list. Symbolic links are not fol-lowed, though the link file itself will be recovered. Mount points are also not followed unless themost recentsave(1m) was performed with the ’-x’ option.

delete[ name ...]Delete the current directory, or the named file(s) or directory(s) from the recover list. If a directoryis specified, that directory and all its descendents are deleted from the list. The most expedientway to recover a majority of files from a directory is to add the directory to the recover list, andthen delete the unwanted files.

list [ −l ] | [ −c ]Display the files on the recover list. With no arguments the recover list is displayed as a list of fullpath names, one per line, followed but a total count of the files to be recovered. The-c argumentprints just the total count of files to be recovered. The-l argument prints the files in the same for-mat as thell command with the−dSoptions.

volumesPrints a list of the volumes needed to recover the current set of files on the recover list.

recoverRecover all of the files on the recover list from the NetWorker server. Upon completion therecover list is empty.

verboseToggle the status of the ‘‘verbose’’ option. When verbose mode is on,recover displays informa-tion about each file as it is recovered. When verbose mode is off,recover only prints informationwhen a problem occurs. The default is verbose mode on.

force If name conflicts exist, overwrite any existing files with recovered files.

noforceCancel theforce option. When in ‘noforce’ mode, a prompt is issued each time a naming conflictarises between a file being recovered and an existing file. At each prompt, six choices are pre-sented: ‘y’, ‘Y’, ‘n’, ‘N’, ‘r’ and ‘R’. To overwrite the existing file, select ‘y’. To rename the fileto an automatically generated alternative name, select ‘r’. Selecting ‘n’ causes the recovered fileto be discarded. The capital letters invoke the same action for all subsequent conflicts without fur-ther prompting. Hence, selecting ‘Y’ will cause all existing conflicting files to be overwritten, ‘N’will cause all conflicting recovered files to be discarded, and ‘R’ will automatically rename allconflicting recovered files (except when an external ASM has a conflicting file name that alreadyends in the rename suffix).

relocate[ directory]Change the target recover location todirectory. If directory is not specified then the user will beprompted for a destination directory. Relative paths are interpreted relative to the current workingdirectory within therecover program. The recovered files will be placed into this directory, whichwill be created if necessary. When files from multiple directories are being recovered, they will beplaced below this directory with a path relative to the first common parent of all the files to be



recovered. For example, if/usr/include/sys/errno.hand /usr/include/stdio.hare being recovered,and the relocation directory is set to/tmp, then the first common parent of these two files isinclude, so the recovered files will be named/tmp/sys/errno.h, and/tmp/stdio.h.

destinationPrint destination location for recovered file.

exit Immediately exit fromrecover.

help Display a summary of the available commands.

? Same ashelp.

quit Immediately exit fromrecover. Files on the recover list are not recovered.

changetime[ time]Display the filesystem as it existed at a different time. If notime is specified the ‘current’ time isdisplayed, and a prompt is issued for a ‘new’ time. The new time is given innsr_getdate(3) for-mat. This format is very flexible. It accepts absolute dates, such asMarch 17, 1997, and relativedates, such aslast Tuesday. Absolute dates can be given in two formats:MM/DD[/YY], andMonth DD[, YYYY]. Times can also be specified as either absolute or relative, with absolute timesin the format:HH[[:MM][:SS]] [am|pm] [time zone]. For example, 12:30 am, 14:21, and 10 pmPST. The current time is used to calculate unspecified parts of a relative date (e.g. 2 days agomeans 2 days ago at the current time), and the end of the day is assumed for unspecified times onan absolute date (e.g. July 2 means July 2 at 11:59:59 PM). By default, the present is used as thecurrent time. The resolution of the filesystem image at a time in the past depends on how oftensavewas run and how far back the NetWorker file index information goes.

versions[ name]All instances of the current directory, ifnameis not specified, or the named file or directory, foundin the NetWorker file index are listed. For each instance, three lines of data are displayed. Thefirst line is similar to thell output. The second line lists the instance’s sav e time. The third linespecifies which tape(s) this instance may be recovered from. With appropriate use of thechange-time command, any one of the entries may be added to the recover list. As withls, lf , andll , filesthat have been added to the recover list are preceded by a ‘+’.

SEE ALSOls(1), nsr_getdate(3), nsr_service(5), nsr(1m),nsrd(1m),nsrindexd(1m),nwrecover(1m),save(1m)

DIAGNOSTICSRecover complains about bad option characters by printing a ‘‘usage’’ message describing the availableoptions.

Message from server: other clones exist for failed save setThe request failed on a save set that had multiple clones. The server automatically picks a differ-ent clone on each attempt. Recover automatically re-submits its recover request to the server, ifany files remain to be recovered.

Path nameis within machine:export-pointAn informative message that lets you know that the given path name is mounted from a networkfile server and that the recovery will use the index for the named file server. If themachineis not aNetWorker client, then the−c option may be necessary.

Browsing machine’s on-line file indexAn informative message that explicitly states which NetWorker client’s index is being browsed forinteractive recovers that resolve to another machine.

Usingserveras server forclientAn informative message that lets you know which NetWorker server was selected for client’sindex.



Cannot open recover session withserverThis message indicates that some problem was encountered connecting to the NetWorker server onthe named machine.

error, nameis not on client listThis message indicates that the client invoking therecover command is not in the server’s clientlist. Seensr_service(5) for details.

path: Permission deniedThe filenamecannot be recovered because you are not root, and you don’t hav e read permissionfor the file.

path: Permission denied (has acl)The filenamecannot be recovered because you are not root, the file has an ACL (Access ControlList), and you are not the owner of the file.


REGCNSRD(8) REGCNSRD(8)

NAMEregcnsrd <option> − used to create or delete a NetWorker server resource, or register and unregister aresource extension.

SYNOPSISregcnsrd [ options ]

options:[ −c | −C ] [ −d | −D ] [ −r | −R ] [ −u | −U ]

DESCRIPTIONThe regcnsrd command must be run in an MS Windows NT MSCS cluster environment.

This command is used for the following:

1.) To create and register the resource-type "NetWorker Server" in a cluster where this resource-typedoes not exist.

Usage Tips:

a) Use the option −c or −C to create and register the resource-type NetWorker Server.

b) You need to use the command regcnsrd only once (from any ONE node) in the clusterto create the resource-type NetWorker Server (assuming the resource-type NetWorkerServer does not already exist in the cluster).

Once this resource-type is created, you can run the command with −r option from all othernodes in the cluster to register (See below for details on −r option).

c) Using the option −c will fail if the resource-type NetWorker Server already exists in thecluster.

d) Use the −C option when you want to create the resource-type NetWorker Server, so that theresource of this type can be managed by MSCS cluster.

IMPORTANT: You must ensure that there is only one resource of the resource-type Net-Worker Server created.

2.) Register the resource-type NetWorker Server in any node(s) in the cluster where the resource-typeNetWorker Server was already created (using the −c option as explained above). Registering theresource-type NetWorker Server in a node allows the resource of this resource-type (NetWorkerServer) to be managed using Cluster Administrator in this particular node.

Usage tips:

a) Use the option −r or −R to register the resource-type NetWorker Server.

b) It is better to run this command with the −r option in all nodes except in the node that wasused to create the resource-type NetWorker Server (using −c option). So that, the resource ofresource-type can be managed from all nodes in the cluster.

It is not needed to run the command with the -r option in the node that was used to create theresource-type NetWorker Server because the −c option does both the creating and register-ing of this resource.

1


c) When you use the −r option, the following output might be displayed. Select the appropriateoption and continue:

Is this machine a member of the cluster on which you want to register Resource Extensionfor NetWorker Server resource?

Enter ’y’ if the command was run from any node in the cluster where the resource-type Net-Worker Server exists.

Enter ’n’ if the NT machine on which this command was run is not part of the cluster (wherethe resource-type NetWorker Server exists).

Choosing the second option (’n’) allows you to manage the resource-type NetWorker Serverin a remote cluster. To use this second option, the NT host must already have the clustersoftware, or cluster administrator software installed.

d) Use −r to manage the NetWorker Server resource from a node other than the node used tocreate the resource-type NetWorker (with the−c option).

3.) Unregister the resource-type NetWorker Server in any nodes in the cluster where the resource-typeNetWorker Server exists. Once unregistered, the resource-type NetWorker Server cannot be man-aged or might not function properly in the node where the command was run.

Usage tips:

a) Use the option −u or −U to unregister.

b) Do not unregsiter the resource-type NetWorker Server from a node if this is the only nodefrom which the resource-type NetWorker Server can be managed. That is, the resource-typehas already been unregistered from all other nodes in the cluster (or, not registered at all inthem).

Once you finished running the command with this option ( −u ) in all nodes except the lastone, you can run the command with the option −d in the last node to unregister (from the lastnode) and delete the resource-type NetWorker Server (from cluster). Further information onthe−d option appears later in this document.

c) When you use the −u option, the following output might be displayed. Choose the appropri-ate option and continue:

Is this machine a member of the cluster on which you want to unregister Resource Exten-sion for NetWorker Server resource?

Enter ’y’ if the command was run from any node in the cluster where the resource-type Net-Worker Server exists.

Enter ’n’ if the NT machine on which the command was run is not part of the cluster (wherethe resource-type NetWorker Server exists).

With the second option (’n’), you give up the ability to manage the resource-type NetWorkerServer installed on a remote cluster.

To use this second option, the NT host must have either the cluster software, or the cluster

2


administrator software installed.

4.) Delete the resource-type NetWorker Server from the cluster so that it is no longer managed as a cluster resource in the cluster.

Usage tips:

a) Use the option −d to unregister the resource-type NetWorker Server from a node, anddelete this resource-type from the cluster.

b) Before you run this command, do the following if a resource-type NetWorkerServer exists in the cluster:

- Bring the the resource of resource-type NetWorker Server offline.

- Delete this resource

You may use the Cluster Administrator to do both of the above steps.

Refer to the NetWorker Administrator’s Guide for information explaining how tobring NetWorker server off-line and then delete it. You may also refer to the HELPthat comes with Cluster Administrator.

c) Before running this command from the last node, make sure the resource-type Net-Worker Server has been already unregistered (using −r option) from all other nodes inthe cluster. You can designate any node as the last node in the cluster to run the com-mand with −d option.

IMPORTANT NOTE BEFORE YOU UNINSTALL:

Do the following before uninstalling NetWorker Server software from the MSCS cluster if Net-Worker Server is configured to run as cluster managed resource with failover capability:

1) Bring the resource of resource-type NetWorker Server offline.

2) Delete the resource of resource-type NetWorker.

3) Unregister the resource-type NetWorker Server from all nodes except one node (des-ignate as last node and it can be any node). Use the command regcnsrd -u to unres-gister. See above for the details.

4) Unregister the resource-type (NetWorker Server) from the last node, and delete theresource-type from the cluster. Use the command regcnsrd -d to unresgister, anddelete. See above for the details.

OPTIONS−c | −C To Create NetWorker Server resource and Register Resource Extension.

−d | −D To Delete NetWorker Server resource and Un-register Resource Extension.

−r | −R To Register Resource Extension.

−u | −U To Unregister Resource Extension.

3

RELEM(1m) RELEM(1m)

NAMErelem − read element status

SYNOPSISrelem [ -a b.t.l ] [ -fvtb ] [ -m {0|1|2} ] [ -r eladdr.nel] [ -l ]

DESCRIPTIONTherelem program will send a READ ELEMENT STATUS command to all changers, or to the (optionally,with the-a option) named device.

SIMPLE OPTIONS−a b.t.l Selects a specific ordinal SCSI address, whereb is the logical SCSI bus,t is the SCSI target, andl

is the SCSI logical unit number (LUN) on that target. Seelibscsi(1m).

−f generates full output (somewhat verbose).

−v generates very verbose output.

−t causes volume tags, if present, to be printed.

−b causes the returned element status data to be dumped as hexadecimal codes rather than interpreted.

−l Performs a complete LUN search for all SCSI adapters in the system when performing Autodetec-tion. This argument is accepted on all systems, but does not have any effect on HP-UX systems.Due to the method used to scan for available devices on HP-UX systems, all accessible devices arealways shown, and the−l option has no additional effect. On all other systems, the normal behav-ior is to start checking at LUN 0 for SCSI deivces. The first empty LUN found will end the searchfor a given target ID. With the−l option, all LUNS present on all target IDs for all SCSI busses inthe system will be checked for jukeboxes. This can take avery long time and should thereforeonly be used when necessary. For example, a Fibre Channel adapter can support 126 target IDs,each of which may have 80 or more LUNs. Checking all LUNs on this single adapter may takeover 10 minutes.

METHOD OPTIONS-m 0 all element status data is fetched in one call

-m 1 element status data is fetched per element type (e.g., all drive elements are read at once, then allslot elements, etc.)

-m 2 element data is fetched per element (which is the default method)

One trivial reason to specify these different ways of fetching element data is that the SCSI specificationallows each one of these ways. A more important real reason is that some changers have bugs with respectto one way or another. For example, one changer may be accurate in reporting all elements in a class, butreturn all zeros if asked for all elements at once.

RANGE OPTIONS−r eladdr.nelis used to read a range of addresses, whereeladdr is the starting decimal address (in the

particular changer’s numbering scheme) of the element to start from, andnel is the number of ele-ments of status to read. Usechangersto get the particular addresses used by your changer.

SEE ALSOlibscsi (1m)


RESOURCE(5) RESOURCE(5)

NAMEresource descriptors − RAP resource file format

SYNOPSISresource::= attribute list <blank line>attribute list::= attribute[ ; attribute]*attribute::= name[ : value[ , value]* ]name, value::= <printable string>

DESCRIPTIONFiles with the.res suffix use a common format to describeresources.Generally, a resource representssomething that a system administrator might want to manage (for example, devices, backup schedules, filesystems), or that a user might want to locate. The encoding of the information describing a resource iscalled theresource descriptor. Resource description files are are accessed by applications and services thatuse the Resource Administration Platform(RAP),but they can also be viewed with a normal text editor.

Each resource descriptor is made up of a list of attributes, and ends in a blank line. Each attribute in theattribute list has a name and an optional list of values. The attribute name is separated from the attributevalues by a colon (:), attribute values are separated by commas (,), and attributes are separated by semi-colons (;). A comma at the end of a line continues the line, as does a back-slash (\) character. The back-slash character can also be used to escape the special meaning of a single character (such as comma, semi-colon, double quote, and back-slash), or the string can be included in quotes. A line beginning with apound-sign (#) is a comment and the rest of the line is ignored. The end of a resource attribute list ismarked with a blank line.

The attribute name and values can contain any printable character. Upper and lower case is ignored oncomparisons, and extra white space is ignored on both ends but not in the middle of names and values. Forexample,

Name: testing 1 2;will match

name : Testing 1 2 ;but is different than

Name: testing 1 2;

Below is an example which includes two resources. The first resource has eight attributes:type, name,server, schedule, directive, group, save set, andremote access. Thegroup attribute has two values:mar-keting andsales. Theremote accessattribute has no value. The second example includes an attribute thatneeds quotes because it contains a colon.


server: mars;schedule: Default;directive: custom;

group: marketing, sales;save set: /, /usr;

remote access: ;

type: NSR group;name: engineering servers;


Each resource includes the special attributetype. The type attribute defines which other attributes aresource can contain. For example, a resource with type printer might include an attributepaper size, whilein a resource of typeNFS filesystemthis attribute makes no sense.

Thenameattribute is a descriptive name of the object that a resource represents. In the example above, the


RESOURCE(5) RESOURCE(5)

name of the second resource isengineering servers, which describes a group of machines to be savedtogether.

Theadministrator attribute is the list of users that have permission to modify this resource. This attributeis inherited from the server resource when a new resource is created. The administrator in the serverresource also controls who has permission to create new resources and delete old ones.

The resource identifier is set and used internally by theRAP system. It provides a unique identification ofeach resource, and although it is sometimes printed like an attribute, it is stored differently. When newresources are created theresource identifierattribute should be left off. This signals the system that this isa new resource and a new identifier will be assigned.

TYPESThere are special resources that define the attributes found in a given type. They are called resource typedescriptors. Type descriptors have the same syntax as other resources except that they hav e a type attributewith the valuetype and atype nameattribute with the value of the type they describe. For example, theresource type descriptor for type NFS filesystem would have, among its other attributes:

type:type; type name:NFS filesystem

Type descriptors are used internally, and should normally never be stored in files or seen by administrators.For each of the other attributes in a type descriptor, there are three or more values. The first value gives thebase type, the second value gives a list of flags separated by spaces, the third value is a string for on-linehelp, and any subsequent strings are default values. This type information is used by system administrationtools to improve the user interface.

FILES*.res Files that contain resource descriptors.

SEE ALSOrap(1m).


SAVE(1m) SAVE(1m)

NAMEsave − sav e files to long term storage with NetWorker

savepnpc − save files to long term storage with NetWorker and performs pre and post processing commandson a NetWorker client.

SYNOPSISsave[ −BEiKLnquSVvx ] [ −s server] [ −c client-name] [ −N name] [ −e expiration] [ −f dirfile ] [ −bpool ] [ −F file ] [ −I input_file] [ −g group] [ −l level] [ −t date] [ −m masquerade] [ −w browse_time] [−y retention_time] [ −W width ] [ path . . . ]

savepnpc[ −BEiKLnquSVvx ] [ −s server] [ −c client-name] [ −N name] [ −eexpiration] [ −f dirfile ] [−b pool ] [ −F file ] [ −I input_file] [ −g group ] [ −l level ] [ −t date] [ −m masquerade] [ −W width ] [−w browse_time] [ −y retention_time] [ path . . . ]

DESCRIPTIONsavesaves files, including directories or entire filesystems, to the NetWorker server (seensr(1m)). Theprogress of a save can be monitored using the X Window System basednwadmin(1m) program or thecurses(3X) basednsrwatch(1m) program for other terminal types.

The user of this command may retain root privileges if the command’s modes are properly set as describedin nsr(1m).

If no path arguments are specified on the command line or via the−I option, the current directory will besaved. savewill save a directory by saving all the files and subdirectories it contains, but it will not crossmount points, or follow symbolic links. If the paths to be saved are mounted from a network file server,saveinstructs the user to run the save on the remote machine or use the-L option.

The directive files (seensr(5)) encountered in each directory are read by default, and they contain specialinstructions directing how particular files are to be saved (i.e. compressed, skipped, etc.). These files arenamed ’.nsr’.

Each file in the subdirectory structures specified by thepatharguments is encapsulated in a NetWorker savestream. This stream of data is sent to a receiving process (seensrd(1m)) on the NetWorker server, whichprocesses the data, adding entries to the on-line index (seensrindexd(1m)) for each file in the stream, withthe data finally ending up on a long term storage media (seensrmmd(1m)).


savepnpcconsists of the same command options assaveand behaves just likesave. In addition, prior tothe actual save on a Networker client,savepnpcperforms pre-processing commands if any exists in the/nsr/res/<grpname>.res file, and at the end of the save of the last save set on the client, the post-processingcommands (if any) will be invoked. In the condition of failure to run the pre-processing commands,savep-npc aborts itself. All results are logged in /nsr/logs/savepnpc.log file on the client. A timeout condition canbe set by the user to indicate at which point in time the post-processing commands need to be run withoutwaiting for all the save sets to be backed up. This timeout attribute resides in the /nsr/res/<grpname>.resfile. The timeout should be specified in such a format thatnsr_getdate() can understand (seensr_getdate(3)).

An example of /nsr/res/<grpname>.res can be described as:type: savepnpc;precmd: /bin/true;pstcmd: /bin/true, "/bin/sleep 5";timeout: "12:00pm";

Theprecmd field can be manually modified to contain any number of commands that are needed to be runat the beginning of the save of the 1st save set. Thepstcmd is to hold any commands that are needed to berun at the end of the save of the last save set. The post-processing commands are run after the save of thelast save set or the timeout condition, whichever comes first.


SAVE(1m) SAVE(1m)

OPTIONS−b pool

Specifies a particular destination pool for the save.

−c client-nameSpecifies the client name for starting the save session. This is useful on clients with multiple net-work interfaces, and multiple host names. It can be used to create multiple index databases for thesame physical client. This does not specify the network interface to use. This is specified in theserver network interfaceattribute of the client resource (seensr_client(5)). This option can alsobe used on a cluster when performing manual saves, or in specifying a non-defaultbackup com-mand for scheduled saves. This option directs NetWorker to override the cluster path-ownershiprules, saving the path argument(s) as belonging toclient-nameand making index entries in theindex forclient-nameinstead of using the name of the physical host or virtual host which owns thepath, according to the cluster management software. Refer topathownerignore(5) for more infor-mation about path-ownership rules.


−f dirfileThe file from which to read prototype default directives (seensr(5)). A dirfile of - causes thedefault directives to be read from standard input.

−g groupThis option is used bysavegrp(1m) and savefs(1m) to denote the group of the save (seensr_client(5) andnsr_group(5)) and is used by the NetWorker server to select the specific mediapool.

−i Ignores any .nsr directive files as they are encountered in the subdirectory structures being saved.

−l level The level of the save. This option is used bysavegrp(1m) andsavefs(1m) to specify a particularlevel for a scheduled save.

−m masqueradeSpecifies the tag to precede the summary line. This option is used bysavegrp(1m) andsavefs(1m) to aid in savegrp summary notifications.savepnpc(1m) also uses this tag to identifyclient operations on the savegrp’s work list that should complete beforepstclntsave(1m) will trig-ger its post-processing.

−n No save. Estimate the amount of data which will be generated by the save, but do not perform theactual save.

−q Quiet. Displays only summary information and error messages.


−t date The date (innsr_getdate(3) format) by which files must have been modified for them to be saved.This option is used bysavegrp(1m) andsavefs(1m) to perform scheduled saves by consulting withthe media database to determine the appropriate time value based on the previous saves for thesave set and the level of the scheduled save.

−u Stop the save if an error occurs. Thesaveprogram normally treats errors as warnings and contin-ues to save the rest of the files in the backup. When this option is set, errors will causesaveto exitand abort the save. This option is not recommended for general use, although it can be usefulwhen a group of files needs to be backed up as a set.


SAVE(1m) SAVE(1m)

−v Verbose. Causes thesaveprogram to provide great detail about thesaveas it proceeds.


−w browse_timeSets the date (innsr_getdate(3) format) after which this save set will no longer be browsable. Bydefault, the server determines the browse date for the save set based on the browse policies ineffect. This option allows overriding the existing policies on a save by sav e basis.


−B Force save of all connecting directory information from root (‘‘/’’) down to the point of invocation.

−E Estimate the amount of data which will be generated by the save, then perform the actual save.Note that the estimate is generated from the inode information; thus, the data is only read once.

−F file Only save files whose change time is newer than the file modification date offile.

−I input_fileIn addition to taking the paths to save from the command line, read paths to save from the namedfile. The paths must be listed one per line. If no paths are specified on the command line, thenonly those paths specified in the file will be saved.

−K Does not build connecting directory index entries.

−L Local. Saves will be performed from the local NetWorker client, even when files are from a net-work file server. To recover these files, runrecover(1m) with the−c client arguments, whereclient is the name of the NetWorker client that did thesave.

−LL In additional to treating the backup as a local backup, causes an extra line to be printed at the endof the completion output of the form ‘‘complete savetime=number’’, where number is the savetimeof the save set created by this backup. This option is meant to be used by thesavegrp(1m) com-mand in performing automatic cloning.

−N nameThe symbolic name of this save set. By default, the most common prefix of thepatharguments isused as the save set name. If the -N option is used when saving any of the SYSTEM save sets(SYSTEM STATE, SYSTEM FILES, and SYSTEM DB), the path must also be specified and mustmatch the name value assigned with the -N option.

−S Allows only save set recovery. This performs the save without creating any index entries. Thismeans that the save set will not be browsable, although save set recovery may be used to recoverthe data.

−V Prevent the OFC mechanism from creating a point-in-time copy of the source volume. (Includedfor compatibility with NT NetWorker servers.)


SEE ALSOcurses(3X), nsr_getdate(3), nwadmin(1m), nsr(5), nsr(1m), nsr_client(5), nsr_device(5), nsr_group(5),nsr_service(5), nsrd(1m), nsrim(1m), nsrindexd(1m), nsrmm(1m), nsrmmd(1m), nsrwatch(1m),recover(1m),savefs(1m),savegrp(1m),pathownerignore(5).


0 Normal exit. This means that a save set was correctly created on the server. Messages about indi-vidual file backup failures arewarnings, and do not cause abnormal exit.


SAVE(1m) SAVE(1m)

<>0 Abnormal exit. A save set was not correctly created on the server.

Messageshost: saveset level=level, size time countfiles.

This message (with the appropriate client host name, saveset name, level, total save set size,elapsed time, and file count) is printed wheneversaveis run bysavegrp(1m) and exits normally.

host: filename: warningMessages of this form are warnings about difficulties backing up individual files. Such messagesdo not normally cause thesaveto fail, and therefore may appear in thesaveoutput found in the‘‘Savegroup Completion’’ message’s Successful section.


SAVEFS(1m) SAVEFS(1m)

NAMEsavefs − save filesystem to a NetWorker server

SYNOPSISsavefs [ options] filesystem

savefs −p[ options] [ filesystem... ] [ −M filesystem... ]

options: [ −BEFnpqRv ] [ −sserver] [ −N name] [ −g group] [ −c client ] [ −l level | −C schedule] [ −eexpiration] [ −w browse] [ −y retention] [ −f filename] [ −W width ] [ −t date] [ −T seconds]

DESCRIPTIONThe savefscommand saves a filesystem (usingsave(1m)) to a NetWorker server. Mount points are notcrossed, and symbolic links are not followed.NOTE: running savefsdirectly is not recommended; usesavegrp(1m) instead.

A level-based system (similar todump(1m)) is used to save only those files which have been modifiedsince some previous save (apartial save).

The nsr_schedule(5) for the local NetWorker client is examined to determine the proper level of sav e forthe current date.

The set of files saved depends on when, and at what level, previous saves hav e been performed, in additionto the effects of the default directives (seensr_directive(5)), and the various directive files (seensr(5))which are encountered while processing the filesystem.

FILESYSTEM PROBESThe savefscommand may also be used to probe a client for its filesystems and recent save times. Whenprobing,savefsdoes not save data, but instead produces a machine-parsable report describing the layout ofthe client’s filesystems. When used with the−p probe option, the local NetWorker client’snsr_client(5)resources are examined, and the filesystems listed in thesave setattribute are probed (if no filesystems arelisted on the command line). If the save set list consists of the keywordAll , then the/etc/fstab file(/etc/vfstabon Solaris,/etc/mnttab on SCO, and a kernel table on AIX) are examined to determine whichfilesystems should be saved, making sure to save only local, mounted filesystems.

Note that metadevices within the Sun Solaris Online DiskSuite and Logical Volumes within the HP-UXLogical Volume Manager are treated like independent disks. This approach allows each to be saved in itsown session, assuming sufficient parallelism.

Care should be taken when theNSR client resource explicitly lists the save sets, for two primary reasons.First, this list must be manually updated when new filesystems are added which need saving. Second, sincesavefsonly stops at the end of a path or a mount point, if you list two sav e sets in the same filesystem, andone is a subdirectory of the other, the subdirectory will be saved twice.

Filesystemarguments can be specified to limit the filesystem saves to only those specified, but the specifiedfilesystems must appear on someSave Setlist for this client (see the−F option).

Probes are also useful when testing how NetWorker will behave in a clustered environment. In this setupownership of shared filesystems must be determined, and performing a probe with the verbose option setallows one to examine the default ownership rules. Refer topathownerignore(5) for a description of path-ownership rules.

OPTIONS−B Force save of all connecting directory information from root (‘‘/’’) down to the point of invocation.

This option is used bysavegrp(1m), for example, when saving the server’s bootstrap information.

−c clientThe name of the client whose filesystem needs to be saved. This option is especially needed in acluster environment where a physical host can represent its own hostname as well as hostnames ofany virtual (also known as "logical") hosts that exist in this physical host. Without this option, thehostname of the physical host is assumed by default. This option is required if a filesystem thatbelongs to any of the virtual hosts needs to be saved.



−C scheduleThe name of the schedule (seensr_schedule(5)) to use when automatically determining the savelevel. If this option is not specified,savefsuses the schedule named by the NSR client resourcefor the specified filesystem.


−w browseSets the date (innsr_getdate(3) format) after which this save set will no longer be browsable. Bydefault, the server determines the browse date for the save set based on the browse policies ineffect. This option allows overriding the existing policies on a save by sav e basis.

−y retentionSets the date (innsr_getdate(3) format) when the saved data will become recyclable. By default,the server determines this date for the save set based on the retention policies in effect.

−E Estimate. Before saving any data, browse the filesystem trees to be saved and accurately estimatethe amount of data that will be generated. Without this flag, the estimate size is zero. This flagconsumes an amount of time proportional to the number of files in each filesystem. This isbecause the entire directory is browsed before any saving begins and browsed again when actuallysaving the directory, but the file data is only read from the disk the last time. In many cases, theoverhead for using this flag is small and is well-justified.

−f filenameThe file from which application specific modules (or ASMs) should take their directives (seensr(5)). By default, these are taken from theNSR directive resource named by thedirectiveattribute in theNSR client resource for each client (seensr_directive(5)).

−F Force. Save every argument like a filesystem, even if it is not listed infstab(5)or nsr_client(5).

−M As part of a probe, signifies that all subsequent filesystems should be probed for their ability to bemigrated. This option is quietly ignored on systems that do not support file migration.

−g groupRestrict the scope of the client to a particular group. If this option is not specified, save sets fromall instances of the NSR client resource for this client will be used, regardless of the group. Thisvalue is also passed on tosave(1m), which uses it to select a specific media pool.

−l level The level of sav e to perform. There are 12 levels:full , lev els1 though9, incr , andskip. Fullspecifies that all files are to be saved. It is analogous to a level 0 dump indump(1m). Incr speci-fies incremental saves in which only those files that have been modified since the most recent save,at any lev el, are saved. This level has no exact analogue indump(1m) since the last save at anylevel, including previous incremental saves, are considered when determining what to save.Skipcauses no files to be saved. The levels1 though9 cause all files to be saved which have been mod-ified since anylower level sav e was performed. As an example, if you did a full save on Monday,followed by a level 3 sav e on Tuesday, a subsequent level 3 sav e on Wednesday would contain allfiles modified or added since the Monday full save. By default, the save lev el is determined auto-matically from the NetWorker client’s schedule (seensr_schedule(5)). By using the history ofprevious saves maintained bynsrmmd(1m) on the NetWorker server, the needed time for thegiven lev el can correctly be computed. By using media information on the server, times computedfor saves that are based on previous save lev els will automatically be adjusted as required whentapes are deleted.



−n No save. Accurately estimates the amount of data that would be generated (as described for−E,but doesn’t sav e any data.

−N nameThe symbolic name this set of saves is to be known by. By default, the firstfilesystemargument isused as the name.

−p List the name of the filesystems, the level of sav e that would be performed, and the time sincewhich files must have been modified to be saved, but don’t actually do the save. This informationis gleaned from the/etc/fstab file (or another operating system specific file, as described above)and thensr_schedule(5).


−qq Really quiet. Display only error messages.

−R Causesavefsto report on its success or failure, by echoing a simple "succeeded" or "failed" mes-sage. This is used bysavegrp(1m) when it is runningsavefs.

−sserverSpecifies which machine to use as the NetWorker server. Seensr(1m) for the algorithm Net-Worker uses to choose a server when none is specified.

−t date The date (innsr_getdate(3) format) from which to base schedule level calculations. If not speci-fied, the current time is used.

−T secondsSpecifies the ‘‘inactivity timeout’’ in seconds. Ifsavefsdetects that the (local) server has made noprogress in the specified time, then it concludes that thesavecommand is hung. A message isprinted to stderr andsavefsexits normally. This option should only be used on NetWorker servermachines.

−v Verbose. Causes lots of debugging style output. This option is also used bysavegrp(1m) when itis probing for the capabilities of the client’ssavefs, for supporting multiple versions.

−W widthThe width used when formatting output or notification messages. By default, this is 80.

RESOURCE TYPESNSR client

These resources specify the client’s sav e sets, default schedule, and directives to use when savingthem.

NSR directiveA resource of this type is named by thedirectiveattribute in eachNSR client resource. These arethe directives used for the save sets specified in the associatedNSR client resource.

NSR scheduleA resource of this type is named by thescheduleattribute in eachNSR client resource. This is theschedule used for the save sets specified in the associatedNSR client resource.

FILES/etc/fstab

If All is specified in thesave setattribute for aNSR client resource, then the list of local filesys-tems is taken from this file.

/etc/vfstabSolaris only. The same as /etc/fstab on other operating systems.

/etc/mnttabSCO only. The same as /etc/fstab on other operating systems.

SEE ALSOnsr_getdate(3), fstab(5), mnttab(F) (SCO only),vfstab(5) (Solaris only),nsr(5), nsr_service(5),nsr_schedule(5), dump(1m),nsr(1m),nsrd(1m),nsrindexd(1m),nsrmmd(1m), recover(1m),save(1m),



savegrp(1m),pathownerignore(5).


0 Normal exit.

255 Abnormal exit.










OPTIONSsavegems









1.3.Build.149a May 04, 2001 1











1.3.Build.149a May 04, 2001 2

SAVEGRP(1m) SAVEGRP(1m)

NAMEsavegrp − start a group of NetWorker clients saving their filesystems

SYNOPSISsavegrp [ options] [ −R | −G ] [ groupname]

options:[ −EIOFXmnpv ] [ −l level | −C schedule] [ −N parallelism] [ −eexpiration] [ −w browse] [ −yretention] [ −t date] [ −r retries] [ −P printer ] [ −W width ] [ −c client [ −c client ... ] ]

DESCRIPTIONThe savegrpcommand runs a group of NetWorker clients through the process of saving their filesystems(usingsave(1m)). The group of clients is selected by naming a NetWorker group (seensr_group(5)), fromwhich individual clients can be selected by using one or more−c options. If no group name is specified,the NetWorker groupDefault is used. If a NetWorker group is named, clients whosensr_client(5)resources specify the named group in theirgroup attribute will be saved. If an explicit client list is alsospecified,savegrpwill only back up those clients, with respect to the named group. Thesavegrpcommandwill automatically make a clone of the newly saved data when the appropriate attributes are set on theNSRgroup resource (see below).

The savegrp command is normally run automatically bynsrd(1m), as specified by each group’snsr_group(5) resource.

Thesavegrpcommand will set up an RPC connection tonsrexecd(1m) to runsave(1m) on each client (andwill fall back on using thercmd(3) protocol and the client-sidershd(1m) if nsrexecdis unavailable on theclient) for each filesystem listed in thensr_client(5) resourcesave setattribute. If a sav e set ofAll is speci-fied for a client,savegrp will request from the client a list of filesystems to be saved (this is called theprobeoperation). The probe expandsAll into a list by looking for filesystems that are both local and auto-matically mounted on that client machine (e.g. NFS mount points and filesystems mounted manually aregenerally ignored). The exact determination of which filesystems to save varies between different operatingsystems. Seesavefs(1m) for additional details on the probe operation. To see which filesystems a clientsaves, run a savegrp preview,savegrp -c client -p(assuming the client is in the Default group). Eachfilesystem saved is called asave set.

The savegrp command attempts to keep multiple clients busy by individually scheduling the client savesets. As save sets complete, the output is collected and another save set will be started bysavegrp.

Theparallelism attribute in thensr_service(5) resource is the maximum number of save sets to run simul-taneously. Modifications to this parameter will take effect as save sets complete − if the value is reduced,no new sav e set will be started until the number of active sav e sets running drops below the new value.

When all of the save sets are completed on a client, the client’s index on the NetWorker server are saved. Ifthe NetWorker server is one of the machines being saved, its index is sav ed after all the other clients arecompletely done. When the server’s index is sav ed, the bootstrap save set information is printed to thedefault printer (or another specified printer). Ifsavegrpdetects that the NetWorker server is not listed inany active group (a group with its autostart attribute set), then the server’s bootstrap is saved with everygroup.

Thesavegrpcommand detects other active inv ocations of the same group, and exits with an error message.If two different NetWorker groups are running simultaneously, they each will run up toparallelismsavesessions simultaneously; however, the NetWorker server will only allowparallelism of these sessions towrite to the backup devices at a time. Note that running multiplesavegrpcommands simultaneously canuse up significant server resources, due to the number of pending saves.

The progress of the actively saving clients can be monitored using the X11 basednwadmin(1m) programor thecurses(3X) basednsrwatch(1m) program. Thensradmin(1m) browser may also be used to exam-ine thecompletionstatus andwork list of eachNSR groupresource, although thehiddenattribute displayoption must be selected (seensradmin(1m)). These two attributes allow you to track the progress of eachsavegrp. Seensr_group(5) for more details.

Whensavegrpstarts, it sends anNSR notification (seensr_notification(5)) with an event ofsavegrpand



priority of info to the NSR notification system. This event is normally logged in themessagesattribute ofthensr_service(5) resource, and in the log file specified in theLog defaultNSR notification resource.

The save sets are automatically cloned, when all the save sets have finished, and theNSR group resourcehas theclonesattribute enabled. The client save sets and their indexes are cloned before the bootstrap saveset is generated so the bootstrap information can track both the original set of save sets and their clones.The bootstrap save set is also cloned. Clones will be sent to the pool named in theclone poolattribute.Changing the values of these attributes whilesavegrp is running has no effect; they must be set beforesavegrp starts. Thensrclone(1m) command is used to clone the save sets.savegrp uses a heuristic todetermine which save sets were generated as part of the group; it may occasionally clone more save setsthan expected, if a client has its filesystems separated into multiple groups that run at the same time. Notethat at least two enabled devices are required to clone save sets.

When the save sets are all complete and cloned (if cloning is enabled), anNSR notification with an eventof savegrpand priority ofnotice is sent to the NSR notification system. This is generally set up to cause e-mail to be sent to theroot user specifying the list of clients who failed (if any), and all the output collectedfrom all clients. The format and common error messages included in the savegrp notification are explainedin theSAVEGRP COMPLETION NOTIFICATION MESSAGE section.

OPTIONS−E Causessave(1m) on each client to estimate the amount of data which will be generated by each

save set before performing it. This will result in the filesystem trees being walked twice − once togenerate a estimate of how much data would be generated, and again to generate a save stream tothe NetWorker server. Note that the data is only read from the disk on the final filesystem walk, asthe estimate is performed by using inode information.

−I Disables the saving of each client’s index.

−O Saves only each client’s index (the bootstrap is also saved).

−m Disable monitor status reporting, including allNSR notification actions.

−n No save. Causesaveto perform an estimate as described for−E, but not to perform any actualsaves. This option also sets−m.

−p Runs the probe step on each client, so you can see which filesystem would be saved and at whatlevel, but do not actually save any data. This option also sets−m. The output generated by the−poption may show sev eral save lev els for each save set at different points in the output, assavegrplearns the correct level. This is the expected behavior, and can be useful for debugging. Theactual level the savegrp uses is shown the last time each save set is displayed in the output. Themedia pool the save set would be directed to is also listed in the preview output.

−v Verbose. Prints extra information about whatsavegrp is doing. The−q flag is also not passed tothesavecommand.

−G Run the group; apply no restart semantics. This is the default mode of operation; the option is pro-vided for compatibility with other versions ofsavegrp.

−R Restart. This option is used to restart a group that was stopped or if savesets failed and they needto be retried.

−l level The level of sav e (seensr_schedule(5)) to perform on each client. This overrides the save lev elwhichsavegrpwould normally automaticly determine.−l and−C cannot be specified together.

−C scheduleThe name of theNSR schedule(seensr_schedule(5)) to be used in the automatic save lev el selec-tion process whichsavegrpnormally performs. This overrides the save schedule whichsavegrpwould normally use for a given client.−l and−C cannot be specified together.

−eexpirationSet the date (innsr_getdate(3) format) when the saved data will expire. When a sav e set has anexplicit expiration date, the save set remains both browsable and non-recyclable until it expires.



After it expires and it has passed its browse time, its state will become non-browsable. If it hasexpired and it has passed its retention time, the save set will become recyclable. The special valuefore ver is used to indicate that a volume that never expires (i.e. an archive or a migration volume)must be used. By default, no explicit expiration date is used.

−w browseSets the date (innsr_getdate(3) format) after which the saved data will no longer be browsable.By default, the server determines the browse date for the save set based on the browse policies ineffect. This option allows overriding the existing policies on a save by sav e basis.


−t date The time to use instead of the current time for determining which level to use for this savegrp (innsr_getdate(3) format). By default, the current time is used.

−F Automatically perform a full level backup if save set consolidation fails. This option is ignored ifthe backup level is not "c".

−X Automatically remove the level 1 sav e set after save set consolidation builds a full level sav e set.This option is ignored if the backup level is not "c". It is also ignored if the backup level is "c" butthe save set consolidation process fails.

−r retriesThe number of times failed clients should be retried beforesavegrpgives up and declares themfailed. The default is taken from the group resource. Abandoned saves are not retried, becausethey may eventually complete. Retries are not attempted if−p is specified.

−P printerThe printer whichsavegrpshould use for printing bootstrap information.

−W widthThe width used when formatting output or notification messages. By default, this is 80.

group Specifies the NetWorker group of clients that should be started, rather than the default NSR group(which has thenameattribute ofdefault). Seensr_group(5) for more details.

−c clientThe name of a client on which to save filesystems. There can be multiple−c client specifications.When −c options are specified, only the named clients from the specified group (which is"Default" if no group is specified) will be run.

−N parallelismThe parallelism value overrides any other parallelism considerations thatsavegrp may use toavoid over-utilizing the system’s resources.

RESOURCE TYPESNSR Use theparallelism attribute for the maximum number of saves to start simultaneously.NSR group The attributework list contains values in groups of 3, specifying the client name, level of

save, and path to save, for each save set not yet completed. The attributecompletioncon-tains values in groups of 4, specifying the client name, path saved, status, and the output,for each save set completed.

NSR scheduleUsed by thesavegrp command with each client’snsr_client(5) resource to determinewhich level of sav e to perform for each specified save set.

NSR client Each client resource names the groups it should be saved by, the names of the save setswhich should be saved, the name of the schedule to use (seensr_schedule(5)) and thename of the directives to use (seensr_directive(5)).



NSR notificationThree kinds of notices are sent to the NSR notification system, both with theev entattributeof savegrp. While a savegrp is in progress, status notices are sent with thepriority of info.At completion of a savegrp, a notice is sent containing the collected output of all saves, andthe name of clients which had a save which failed (if any). This notice will have anev enttype of savegrp, and apriority of notice. If savegrp is interrupted, a notice stating thegroup was terminated, with anev ent type ofsavegrp, and apriority of alert will be sent.These last two typically will result in the notice being encapsulated in a mail message toroot.

SAVEGROUP COMPLETION NOTIFICATION MESSAGEThe savegroup completion notification message contains 5 parts: the header, the Never Started Save Sets,the Unsuccessful Save Sets, the Successful Save Sets, and the Cloned Save Sets. Each client in the groupwill be listed in one or more of sections categories (more than one if some save sets are in one category, andother save sets in another category). The clients are listed in alphanumeric order, with the server listed last.

The header shows the name of the group and lists which clients failed. If the group was aborted, the headerincludes an indicator of this as well. The header also shows the time the group was started (or restarted, ifthe −R option was used), and the time the savegrp completed. The failed clients list in the header showsonly those clients for which saves were attempted, not those for which saves nev er started.

The Never Started Save Sets section is optional and is included only if there are some save sets of someclients in the group that were never started. This should only occur when a savegrp is aborted, either bykilling the master savegrp daemon or by selecting theStopfunction in theGroup Controlwindow or theStop Nowattribute in theGroup window of nwadmin(1m). Each entry listed in this section shows theclient and save set that was never started (orAll if no save sets were saved for that client). No other errormessages should appear in this section.

The Unsuccessful Save Sets section shows all of the saves that were attempted but failed. This section willonly be present if at least one save set failed. There are many reasons for a save to fail. The most commonare listed below. More reasons will be listed in the future. It is important to differentiate between the manyreasons for a save to fail, so that the administrator can quickly determine the cause and fix it.

Each entry in the Unsuccessful Save Sets section lists the client and save set that failed, along with one ormore lines of error and information messages. Each client is separated by a blank line, and all the failedsave sets for a client a listed together. Typical error or information messages are listed at the end of thissection, (without the client:saveset prefix), with the necessary action(s) to take to correct the problem.

Each entry in the Successful Save Sets section lists the client and save set that succeeded, along with levelof the save, the amount of data saved, the time to run the save set, and the number of files saved. Eachentry may also be preceded by one or more warning or informational messages, the most common of whichare listed below. These warning or informational messages are usually (but not always) prefixed by ‘‘* ’’.A sav e set’s output may include warnings; these do not necessarily mean the save set was unsuccessful.

The Cloned Save Sets section refers to the save sets cloned, and not the clients that originated those savesets. The output shown in this section is the output of thensrclonecommand. See thensrclone(1m) manpage for information on the output ofnsrclone.

The following is a list of common informational, warning and error messages found in the completion noti-fication. This list is not complete. The messages you see may vary slightly from those shown here due todifferences in the operating system vendor-supplied error messages. Since many messages include client orserver names, it is most efficient to look for a keyword in the error message. The messages are listed belowin alphabetical order, by the first non-variable word in the message. (Note: initial words like "save", "asm"and "savefs" may or may not vary, and initial pathnames are always assumed to vary).

abortedThis informational message only occurs when you abort a running savegrp, generally by selectingStopfrom the Group Control Window of thenwadmin(1m) interface. It means that the specifiedsave set had started saving, but had not completed when the savegrp was aborted. The session (inthe Sessions display ofnwadmin(1m)) for this save set may not disappear immediately, especially



if savegrp’s attempt to kill the save session fails. The save set will be retried if and when youRestartthe savegrp (e.g. from the Group Control Window).

Access violation fromclient - insecure portNThis message, generated by thesavecommand on client, means thatsaveis not setuid root. Makesure that thesavecommand on the client is owned by root and has its setuid bit set. Ifsaveis onan NFS mounted filesystem, make sure the filesystem was not mounted on that client using the"-nosuid" option.

Access violation − unknown host:clientThis message is caused when then the client’s hostname and IP address are not correctly listed inone or more of /etc/hosts, NIS or DNS on the server. You must change the appropriate host table(depending on which one(s) are in use on your server) to list the client’s name as it is known toNetWorker (client’s primary name), or you must add the name listed at the end of the error mes-sage to thealiasesattribute of the client’s Client resource(s).

asm: cannot openpath: I/O errorThis message generally means that there are bad blocks on the disk(s) containing the specified fileor directory. You should immediately run a filesystem check on the named client filesystem andcheck your client’s system error log. If there are bad block, repair them if possible, or move thefilesystem to a different disk.

asm: cannot statpath: Stale NFS file handleasm: cannot statpath: Missing file or filesystem

These informational messages (or variants of them for other operating systems) mean that thewhensaveattempted to test the named directory (to determine if it was a different filesystem fromthe one currently being saved), the filesystem was NFS mounted, but the mount point was bad.While this message does not affect the saved data, it does mean you have a network or NFS prob-lem between the specified client and one or more of its fileservers. You may need to remountfilesystems on the client, or perhaps reboot it to correct the problem.

/path/nsrexecd: Can’t make pipe/path/nsrexecd: Can’t forkfork: No more processes

The specified client-side resource has been exceeded. There are too many other services runningon the client while savegrp is running. Inspect the client and determine why it has run out ofresources. The client may need to be rebooted. You should also consider re-scheduling any jobsautomatically started on the client (e.g. viacron(1m)) that run while savegrp is running.

asm: chdir failed path: Permission deniedThis message means that while backing up the specified save set,savewas unable to enter thenamed directory. This may mean thatsave is not setuid root on the specified client, or that thedirectory is an NFS mount point for which root is not allowed access. Check the permissions onsaveon the specified client (usingls(1)) and make sure thatsave is owned by root and that thesetuid bit is set.

connect to address AA.BB.CC.DD:messageTrying AA.BB.CC.DD...

These informational messages are displayed only when the−v option is used. They mean that theconnection to the client failed on the address specified in the first line of the message. If the clienthas more than one IP address, savegrp has attempted the address listed in the second line. Subse-quent lines of the completion mail show if this second address succeeded. You may want to checkand change your network routing tables to avoid getting these messages.

Connection refusedThis means the client machine is up, but it is not accepting new network connections fornsrexecd(or rshd). This could mean the client was in the process of booting when the savegrp attempted toconnect, or that the client had exceeded some resource limit, and was not accepting any new con-nections. You should attempt to log into the client and verify that it is accepting remote



connections. If the client is a non-Unix machine, you may need to start the NetWorker client onthat machine. Refer to your ClientPak installation for more information.

Connection timed outThis usually means the client has crashed or is hung. Make sure the client has rebooted, and thatnsrexecdis running on it (if you are usingnsrexecd). If the client is a non-Unix machine, youmay need to ensure that the network protocols are loaded, and that the NetWorker client is runningon that machine. Refer to your ClientPak installation for more information.

asm: external ASM ‘asm2’ exited with code 1This message generally accompanies another message reporting a specific problem while saving afile or directory on the named save set. The backup will attempt to continue and attempt to saveother data. Generally, the backup will not be listed in the failed save sets section of the completionmail if any files on the save set are saved successfully, even if it only saves the top directory of thesave set.

save:pathfile size changed!This informational message is often generated when NetWorker backs up log files. It may alsooccur for other files. For files that you expect to grow whilesavegrp is running, you can use adirective specifying that thelogasm(1m) should be used to back up the file. See alsonsr(5) andnsr_directive(5).

asm: getwd failedThis message means that while backing up the specified save set, an attempt to determine the cur-rent directory’s name failed. This occurs on clients, generally running older versions of the Net-Worker ClientPak, on which thegetwd(3) library call is broken. You may want to contact LegatoTech Support to find out if there is a patch available for your client platform to work around thisvendor-specific bug, or contact your operating system vendor to see if a more recent OS versionaddresses this problem.

Group groupnameaborted, savegroup is already runningThis message is only delivered by itself. It occurs when the named group has already been startedor restarted (eg after a reboot, or when requested via the Group Control Window ofnwad-min(1m)), either automatically bynsrd(1m) or manually, from the command line. You can useps(1) to find out the process id of a running savegrp. The existance of a running group is deter-mined by looking for a file named/nsr/tmp/sg.group which, if existing and locked, means asavegrp is running.

nsrexec: Attempting a kill on remote saveclient:savesetaborted due to inactivity

The client has not sent any data to the server for the specified inactivity timeout.savegrp willattempt to kill the backup in progress so that the hung client will not impede other backups orcloning operations.

has been inactive forN minutes sincetime.client:savesetis being abandoned by savegrp.

A backup of the specified save set started, but afterN minutes of no activity,savegrpgave up onthe save set. Generally, this means that the client is hung waiting for an NFS partition. Unfortu-nately, NetWorker (or any other program) has no way of reliably telling if an NFS partition willhang until after it tries to access the partition. When the partition comes back on line, the save willcomplete, despite thatsavegrpabandoned it. You should check the client, since you sometimesmay need to reboot the client to unhang NFS partitions. Non-Unix clients also hang for other rea-sons such as defects in the operating system implementation of their network protocols.

Host is unreachableThe NetWorker server cannot make TCP/IP connections to the client. This generally means thenetwork itself is not configured correctly; most commonly, one or more gateways or routers aredown, or the network routes were not set up correctly. You should verify that the server can con-nect to the client, and if not, check and, if necessary, reconfigure your routers, gateways or routing



tables.

Login incorrectThis message is generated when theremote userattribute for the client is not set to a valid login onthe client. Verify that theremote userattribute for the client is set to the correct login name. Youmay see this message even when runningnsrexecdif nsrexecdhas not been started (or was killed)on the client.

asm: missing hard links not found:This message is generated when a backed-up file had one or more hard links that were not found.The message is followed by a list of one or more file names which were backed up minus somelinks. The message means that the files were either created (with multiple hard links) while thebackup was occurring, so some of the links were missed due to the order of filesystem tree walk-ing, or the file (or some links) were removed while the backup was occurring. Only those linksthat were found can be recovered; additional links will have been lost. One can do an additionalincremental backup of the affected filesystem if a consistent state for the affected file is essential.

lost connection to server, exitingsave: network error, server may be down

The backup of the named filesystem was begun, but the connection to the NetWorker server closedpart way through. This typically means that the server machine rebooted, one or more NetWorkerserver daemon processes were killed by the system administrator or by the system itself (e.g. dueto overwriting the binary or a disk error in swap space), or there was some transport problem thatcaused the network connection to dropped by the operating system. Restart the save at a latertime.

No save sets with this name were found in the media database;performing a full backup

This informational message is added bysavegrp to any sav e set that is saved at the levelfullinstead of the level found in the client’s schedule. Due to timing problems, you can occasionallysee this message when the clocks on the client and server are out of sync, or whensavegrpstartsbefore midnight and ends after midnight. You may also get spurious messages of this type fromsome versions of NetWorker client software backing up a NetWare BINDERY, which ignore theschedule and perform a full save. In both these cases, the client re-checks the level, and overridesthe server’s requested level.

No more processesSee "Can’t make pipe" message information.

No ’NSR client’ resource for clientclienthostnamesavefs: cannot retrieve client resources

This pair of messages occurs if the the client’s hostname changed (in /etc/hosts, NIS or DNS).You may also have deleted the client’sClient resource while savegrp was running. In the formercase, you will need to add the client’s new name to thealiasesattribute of the client (this is a hid-den attribute) usingnsradmin(1m) (selecting theHidden display option) ornwadmin(1m)(selecting theDetails Viewoption for theClient window). In the latter case, no additional action isrequired if this deletion was intentional (the next run ofsavegrp will not attempt to save theclient). If it was accidental, and you did not want to delete the client, you should add the clientback again and add the client back into the appropriate group(s). The next timesavegrp runs, itwill back up the client, just as if the client had been down the previous day.

no outputThe save set completed, but returned no status output. The most common reasons are that theclient crashed or lost its network connection (i.e.. a router between the client and server crashed)while the client was being backed up. Another is that the disk on which the client status was beinglogged filled up (perform adf /nsr/tmp to see if this was the case). To determine if the save setwas sav ed, you can usemminfo(1m). For example, runmminfo -v -c clientname-t ’1 day ago’and look at theflagscolumn for the completion status. An ’a’ flag means it aborted. Use a moredistant time (the−t option) to look further back in time.



filesystem: No such file or directoryAn explicit save set was named in theClient resource for the specified client, and that save setdoes not exist (or is not currently mounted) on the client. Make sure you spelled the save set namecorrectly (and that it is capitalized correctly), and log into the client and verify that the save set ismounted.

/path/nsrexecd: Couldn’t look up address for your host/path/nsrexecd: Host address mismatch forserver

Thensrexecddaemon on the client managed to look up the server in the client’s host table, but theaddress listed there did not match the address of the server. Every interface of the server musthave a unique name listed in the host table (possibly with non-unique aliases or CNAME’s), andeach unique name must be listed as a valid server tonsrexecd.

/path/nsrexecd: Hostservercannot request command execution/path/nsrexecd: Your host cannot request command execution

The server is not listed innsrexecd’s list of valid servers on the specified client. The list of validservers is either on thensrexecdcommand line (with one or more−s serveroptions tonsrexecd),or in a file (with the−f file option tonsrexecd). If neither is specified,nsrexecdwill look for afile namedservers in the same directory that contains thensrdb configuration database (e.g./nsr/res/nsrdbon a typical Unix server). Also the server may not be listed in one or more of/etc/hosts, NIS, or DNS, on the client, in which casensrexecdcannot validate the server until theclient’s host naming configuration is fixed.

/path/nsrexecd: Invalid authenticator/path/nsrexecd: Invalid command

These two messages should never occur in a savegroup completion message. They mean thatsavegrpdid not follow its protocol correctly.

/path/nsrexecd: Permission deniedPermission denied

These similar messages are generated bynsrexecd and rshd, respectively. In either case, theserver does not have permission to execute commands on the client. In the case of the first mes-sage, make sure that the server is listed as a valid server on the client (see "Hostservercannotrequest command execution", above, for details). In the case of the second message, which doesnot mentionnsrexecd, make sure that "servername" is listed in the client’s/.rhostsfile (or, if youhave set theremote userattribute for this client, the.rhostsfile in the home directory for that useron the client).

/path/savegrp: printing bootstrap information failedSee "unknown printer" message information.

reading log file failedAfter the specified save set completed, savegrp was unable to read the log file of the output statusfrom the save set. This generally means that someone, or an automated non-NetWorker adminis-trative program or script, removed the log file. This message can also occur if the filesystem onwhich the client logs are stored has run out of space (usedf /nsr/tmp to determine if this is thecase). Verify that no scripts remove files from/nsr/tmp(which is where savegrp stores the save setlog files).

request from machineserverrejectedThe server is not listed in the PC (NetWare or DOS) client’s list of acceptable servers. See yourClientPak installation guide for instructions on adding the server to the client-side list.

N retries attempted1 retry attempted

One of these informational messages is prepended to a save set’s output ifsavegrp is unable tobackup the data on the first try and if theclient retriesattribute for the group has a value greaterthan zero. In this case, the specified number of retries was performed before the backup of thesave set succeeded or was finally marked as failed.



RPC error: details...Cannot open save session with ‘server’

Thesavecommand generates this message if it is unable to back up data to the NetWorker server.There are several possibledetails. The most likely causes are: resources are exceeded on theserver so nsrd cannot accept new sav e sessions, nsrd actually died since savegrp started (however,this is unlikely, since you cannot normally receive a sav e grp completion message after nsrd dies,but you can see this when using the−p option), there are numerous network errors occurring andsave cannot open a session to save its data (check this by runningnetstat -sand see how many net-work errors are occurring; you may need to do this several times a few minutes apart to get thechange in errors). Save cannot tell which of these three causes are the real cause. If you see theseerrors frequently, and it looks like a server resource problem, you might consider increasing thevalue of theclient retriesattribute of thegroup resource having these problems. This won’tdecrease the resource utilization, but will makesavegrp more robust. (The trade-off is thatincreasingclient retrieswill increase the load on the server even more).

nsrexecd onclient is unavailable. Using rsh instead.This informational message is only displayed when the−v flag has been used for verbose informa-tion. This message means thatnsrexecdis not running on the client, and that savegrp is attempt-ing to use thershd service instead for backward compatibility with older versions ofsavegrp.

save:clientname2is not onclient’s access listThis error occurs when the named client has more than one name, for example, a short name,client, and a fully-qualified domain name,client.legato.com. When the client attempts to connectback to the NetWorker server to start a save, that client is calling itself by the nameclient, whichmatches the client resource name, but when the server looks up the client’s network address, it isgetting back the nameclientname2. If this is correct, add the nameclientname2to the client’saliasesattribute and re-run the save.

save: path length ofxxxxtoo long, directory not savedThis message can occur if you have a directory tree that is very deep, or directory names that arevery long. This message can also occur if there are bad blocks in the specified filesystem, or if thefilesystem is corrupt. NetWorker limits the full pathname to 1024 characters which is the systemimposed maximum on most systems. To sav e such directories, you need to rename or move thedirectories so that the full pathname is shorter than 1024 characters. If the filesystem appears to becorrupted (for example, a very long pathname that looks like it has a loop in the name), perform afilesystem check on the specified client.

/path/save: Command not found/path/savefs: Command not found/path/save: Not found/path/savefs: Not found

Thesaveor savefscommand could not be found in the specified path. If you are usingnsrexecd,this probably means that thesaveor savefscommand is not in the same directory in whichnsrex-ecd is installed (or thatsaveor savefswas removed). If you are usingrshd for remote execution,then you need to set theexecutable pathattribute in theClient resource for this client to be thedirectory in which the NetWorker executables are installed on the client.

savefs: error starting save offilesystemThis informational message accompanies several othersaveor asm messages listed here. Thismessage means that savefs has detected the failed save and has marked the save set as failed.

save: unknown host name:serversavefs: unknown host name:server

The host table on the specified client (either /etc/hosts, NIS or DNS, depending on that client’sconfiguration) does not include the server’s name. Add the server’s hostname to the specifiedclient’s host table. If you use DNS but the server’s Client resource name (i.e. the client resourcefor the server itself) is not fully qualified (i.e. it looks like "server", not "server.dom.ain"), and theserver is in a different domain from the client, add the nameserver to the domain table for the



domain containing the client. If you use NIS, this error means that either the NIS hosts map doesnot contain the server, the/etc/hostsfile does not list the server, or the NIS master for the specifiedclient is otherwise mis-configured (the server is a secondary server and there is noyppush(1m)from the primary; runypwhich -m on the client to find which NIS server is providing mastertranslation).

savegrp: client rcmd(3) problem for command ’command’This error message normally accompanies another, more specific, error message. It is generatedwhen the attempt to run the specified command (usuallysaveor savefswith several command lineparameters) failed on the specified save set. The previous line of error output should include themore specific error message (look for that message elsewhere in this section). Generally, the prob-lem is a bad hosttable configuration, or various permissions denied errors (server not specifiedwhen starting nsrexecd, or missing permissions in .rhosts if not using nsrexecd). If not, log intothe NetWorker server as root and run the commandsavegrp -p -v -cclientname groupnamegivingthe appropriate client forclientnameandgroupname .This verbose output should include the nec-essary additional information needed for fixing the problem.

savegrp: suppressedN lines of verbose outputSometimes a backup will generate a huge amount of output, such as when one runssavegrp -v.Whensavegrpupdates the group completion attribute in the server, it may suppress some initiallines of of this output, since logging all of the output to the completion attribute can causensrd touse an unexpectedly large amount of memory. The entire output ofsavegrp -vcan be found in thelog files or the savegrp completion mail, rather than the completion attribute.

socket: All ports in useThe NetWorker server has run out of socket descriptors. This means that you have exceeded thesocket resource limit on your server. To avoid such future messages, you should determine whatother network services are running while savegrp is running, and consider re-scheduling eithersavegrp or the other service(s). You can also reduce the parallelism in thensr_service(5)resource, to reduce the resource utilization.

socket: protocol failure in circuit setup.The client does not seem to support the TCP/IP protocol stack, or has not used a privileged port forsetting up its connection. The latter could occur if you usensrexecdbut did not start it as root onthe specified client. Thensrexecddaemon must run as root on each client.

path: This data set is in use and cannot be accessed at this timeThis message is generated by save sets on PC clients running DOS or NetWare. The NetWorkerclient software on these systems cannot back up files open for writing, due to the interface pro-vided by the operating system. This message actually comes from Novell’s TSA and is notchangeable.

unknown hostThe specified client is not listed in the host table on the server (note: a similar "save" or "savefs"specific message is described above). Depending on your host configuration, this means the clientis not listed in one (or more) of /etc/hosts, NIS, or the Domain Name Service. If you use fullyqualified domain names, you may need to make a new client resource for this client, using thatfully qualified domain name (i.e. name the client resource "mars.legato.com", not "mars").

printer: unknown printerpath/savegrp: printing bootstrap information failed(reproduced below)

This message, or similar messages, accompanies the bootstrap information whensavegrp wasunable to print the bootstrap on the printer. You need to either specify a different printer in theprinter attribute for the group, or configure your print server to recognize the printer (by default,your system’s default printer is used). The bootstrap information is listed as part of the savegrpcompletion mail. You should print out this information immediately, in case your server has a dis-aster and loses a disk, and fix the printer name used bysavegrp.

Warning − file ‘path’ changed during saveThis warning message is generated whensavenotices that the file’s modification time changedwhile the file was being backed up. NetWorker does not attempt to lock files before saving them,



as this would make backups run extremely slowly. You may wish to backup files which generatethis message manually, to ensure that a consistent copy is sav ed. NetWorker does not attempt thisautomatically, to avoid trying forever on the same file.

Warning: ‘ client’ is not in the hosts table!This message is generated by asaveor savefscommand run on the specified client to save thatclient’s filesystems. The client’s hostname is not listed in the host table on the client (either/etc/hosts, NIS or DNS, depending on that client’s configuration). This almost always results in afailed save. Fix the client’s host table and re-run the save.

asm: pathwas not successfully savedThis message generally accompanies one or more other more-specific messages for the save set.The specified path within the current save set was not saved successfully. The backup will con-tinue trying to back up other files and directories on the save set.

asm: xdr _opfailed for pathThis error can be caused by several possible conditions (for example, out of memory, defectivenetworking software in the operating system, an external ASM unexpectedly exiting, a lost net-work connection). If it was due to a lost network connection, then the NetWorker server mostlikely exited (due tonsr_shutdown). After restarting the server, rerun the group. If due to anASM exiting unexpectedly (in this case, the message should be accompanied by a messagedescribing which ASM exited unexpectedly), you may have found a bad block on the disk, or per-haps a defect. Check if the client ran out of memory (there may be console messages), and verifythat there are no bad blocks on the save set’s disk. If there were network errors, there may alsohave been messages logged by other programs on the system console (client or server), or to sys-tem log files.

FILES/nsr/tmp/sg.group A lock file to keep multiple savegrps of the same group from running

simultaneously./nsr/tmp/sg.group.client.* Temporary files used to log the output of individual save sets for the

named group and client./nsr/tmp/ggroup* On filesystems with short names (less than 64 characters), the temporary

files used to log the output of individual save sets for the named group.

SEE ALSOls(1), ps(1), nsr_getdate(3), rcmd(3), fstab(5), nsr(5), nsr_directive(5), nsr_notification(5),nsr_service(5), nsr_group(5), nsr_schedule(5), nsr_resource(5), mminfo(1m),nsrssc(1m),netstat(1m),nsr(1m),nsradmin(1m),nsrexec(1m),nsrexecd(1m),nsrwatch(1m),nwadmin(1m), rshd(1m),save(1m),savefs(1m),pathownerignore(5), yppush(1m).


SAVE(1m) SAVE(1m)

NAMEsave − sav e files to long term storage with NetWorker

savepnpc − save files to long term storage with NetWorker and performs pre and post processing commandson a NetWorker client.

SYNOPSISsave[ −BEiKLnquSVvx ] [ −s server] [ −c client-name] [ −N name] [ −e expiration] [ −f dirfile ] [ −bpool ] [ −F file ] [ −I input_file] [ −g group] [ −l level] [ −t date] [ −m masquerade] [ −w browse_time] [−y retention_time] [ −W width ] [ path . . . ]

savepnpc[ −BEiKLnquSVvx ] [ −s server] [ −c client-name] [ −N name] [ −eexpiration] [ −f dirfile ] [−b pool ] [ −F file ] [ −I input_file] [ −g group ] [ −l level ] [ −t date] [ −m masquerade] [ −W width ] [−w browse_time] [ −y retention_time] [ path . . . ]

DESCRIPTIONsavesaves files, including directories or entire filesystems, to the NetWorker server (seensr(1m)). Theprogress of a save can be monitored using the X Window System basednwadmin(1m) program or thecurses(3X) basednsrwatch(1m) program for other terminal types.

The user of this command may retain root privileges if the command’s modes are properly set as describedin nsr(1m).

If no path arguments are specified on the command line or via the−I option, the current directory will besaved. savewill save a directory by saving all the files and subdirectories it contains, but it will not crossmount points, or follow symbolic links. If the paths to be saved are mounted from a network file server,saveinstructs the user to run the save on the remote machine or use the-L option.

The directive files (seensr(5)) encountered in each directory are read by default, and they contain specialinstructions directing how particular files are to be saved (i.e. compressed, skipped, etc.). These files arenamed ’.nsr’.

Each file in the subdirectory structures specified by thepatharguments is encapsulated in a NetWorker savestream. This stream of data is sent to a receiving process (seensrd(1m)) on the NetWorker server, whichprocesses the data, adding entries to the on-line index (seensrindexd(1m)) for each file in the stream, withthe data finally ending up on a long term storage media (seensrmmd(1m)).


savepnpcconsists of the same command options assaveand behaves just likesave. In addition, prior tothe actual save on a Networker client,savepnpcperforms pre-processing commands if any exists in the/nsr/res/<grpname>.res file, and at the end of the save of the last save set on the client, the post-processingcommands (if any) will be invoked. In the condition of failure to run the pre-processing commands,savep-npc aborts itself. All results are logged in /nsr/logs/savepnpc.log file on the client. A timeout condition canbe set by the user to indicate at which point in time the post-processing commands need to be run withoutwaiting for all the save sets to be backed up. This timeout attribute resides in the /nsr/res/<grpname>.resfile. The timeout should be specified in such a format thatnsr_getdate() can understand (seensr_getdate(3)).

An example of /nsr/res/<grpname>.res can be described as:type: savepnpc;precmd: /bin/true;pstcmd: /bin/true, "/bin/sleep 5";timeout: "12:00pm";

Theprecmd field can be manually modified to contain any number of commands that are needed to be runat the beginning of the save of the 1st save set. Thepstcmd is to hold any commands that are needed to berun at the end of the save of the last save set. The post-processing commands are run after the save of thelast save set or the timeout condition, whichever comes first.


SAVE(1m) SAVE(1m)

OPTIONS−b pool

Specifies a particular destination pool for the save.

−c client-nameSpecifies the client name for starting the save session. This is useful on clients with multiple net-work interfaces, and multiple host names. It can be used to create multiple index databases for thesame physical client. This does not specify the network interface to use. This is specified in theserver network interfaceattribute of the client resource (seensr_client(5)). This option can alsobe used on a cluster when performing manual saves, or in specifying a non-defaultbackup com-mand for scheduled saves. This option directs NetWorker to override the cluster path-ownershiprules, saving the path argument(s) as belonging toclient-nameand making index entries in theindex forclient-nameinstead of using the name of the physical host or virtual host which owns thepath, according to the cluster management software. Refer topathownerignore(5) for more infor-mation about path-ownership rules.


−f dirfileThe file from which to read prototype default directives (seensr(5)). A dirfile of - causes thedefault directives to be read from standard input.

−g groupThis option is used bysavegrp(1m) and savefs(1m) to denote the group of the save (seensr_client(5) andnsr_group(5)) and is used by the NetWorker server to select the specific mediapool.

−i Ignores any .nsr directive files as they are encountered in the subdirectory structures being saved.

−l level The level of the save. This option is used bysavegrp(1m) andsavefs(1m) to specify a particularlevel for a scheduled save.

−m masqueradeSpecifies the tag to precede the summary line. This option is used bysavegrp(1m) andsavefs(1m) to aid in savegrp summary notifications.savepnpc(1m) also uses this tag to identifyclient operations on the savegrp’s work list that should complete beforepstclntsave(1m) will trig-ger its post-processing.

−n No save. Estimate the amount of data which will be generated by the save, but do not perform theactual save.

−q Quiet. Displays only summary information and error messages.


−t date The date (innsr_getdate(3) format) by which files must have been modified for them to be saved.This option is used bysavegrp(1m) andsavefs(1m) to perform scheduled saves by consulting withthe media database to determine the appropriate time value based on the previous saves for thesave set and the level of the scheduled save.

−u Stop the save if an error occurs. Thesaveprogram normally treats errors as warnings and contin-ues to save the rest of the files in the backup. When this option is set, errors will causesaveto exitand abort the save. This option is not recommended for general use, although it can be usefulwhen a group of files needs to be backed up as a set.


SAVE(1m) SAVE(1m)

−v Verbose. Causes thesaveprogram to provide great detail about thesaveas it proceeds.


−w browse_timeSets the date (innsr_getdate(3) format) after which this save set will no longer be browsable. Bydefault, the server determines the browse date for the save set based on the browse policies ineffect. This option allows overriding the existing policies on a save by sav e basis.


−B Force save of all connecting directory information from root (‘‘/’’) down to the point of invocation.

−E Estimate the amount of data which will be generated by the save, then perform the actual save.Note that the estimate is generated from the inode information; thus, the data is only read once.

−F file Only save files whose change time is newer than the file modification date offile.

−I input_fileIn addition to taking the paths to save from the command line, read paths to save from the namedfile. The paths must be listed one per line. If no paths are specified on the command line, thenonly those paths specified in the file will be saved.

−K Does not build connecting directory index entries.

−L Local. Saves will be performed from the local NetWorker client, even when files are from a net-work file server. To recover these files, runrecover(1m) with the−c client arguments, whereclient is the name of the NetWorker client that did thesave.

−LL In additional to treating the backup as a local backup, causes an extra line to be printed at the endof the completion output of the form ‘‘complete savetime=number’’, where number is the savetimeof the save set created by this backup. This option is meant to be used by thesavegrp(1m) com-mand in performing automatic cloning.

−N nameThe symbolic name of this save set. By default, the most common prefix of thepatharguments isused as the save set name. If the -N option is used when saving any of the SYSTEM save sets(SYSTEM STATE, SYSTEM FILES, and SYSTEM DB), the path must also be specified and mustmatch the name value assigned with the -N option.

−S Allows only save set recovery. This performs the save without creating any index entries. Thismeans that the save set will not be browsable, although save set recovery may be used to recoverthe data.

−V Prevent the OFC mechanism from creating a point-in-time copy of the source volume. (Includedfor compatibility with NT NetWorker servers.)


SEE ALSOcurses(3X), nsr_getdate(3), nwadmin(1m), nsr(5), nsr(1m), nsr_client(5), nsr_device(5), nsr_group(5),nsr_service(5), nsrd(1m), nsrim(1m), nsrindexd(1m), nsrmm(1m), nsrmmd(1m), nsrwatch(1m),recover(1m),savefs(1m),savegrp(1m),pathownerignore(5).


0 Normal exit. This means that a save set was correctly created on the server. Messages about indi-vidual file backup failures arewarnings, and do not cause abnormal exit.


SAVE(1m) SAVE(1m)

<>0 Abnormal exit. A save set was not correctly created on the server.

Messageshost: saveset level=level, size time countfiles.

This message (with the appropriate client host name, saveset name, level, total save set size,elapsed time, and file count) is printed wheneversaveis run bysavegrp(1m) and exits normally.

host: filename: warningMessages of this form are warnings about difficulties backing up individual files. Such messagesdo not normally cause thesaveto fail, and therefore may appear in thesaveoutput found in the‘‘Savegroup Completion’’ message’s Successful section.


SCANNER(1m) SCANNER(1m)

NAMEscanner − NetWorker media verifier and index rebuilder

SYNOPSISscanner[ options] −B device

scanner[ options] −B −Sssid[ −im ] device

scanner[ options] −i [ -Sssid] [ -c client ] [ -N name] device

scanner[ options] −m [ -Sssid] device

scanner[ options] [ −S ssid] [-c client] [-N name] device <command>

options: [ −npqv ] [ −f file ] [ −r record] [ −sserver] [ −t type] [ −b pool ]

command: −x command[ arg ... ]

DESCRIPTIONThescannercommand reads NetWorker media, such as backup tapes or disks, to confirm the contents of avolume, to extract a save set from a volume, or to rebuild the NetWorker online indexes. As installed, onlythe super-user may run this command. However, the command’s modes can be modified such that normalusers may run the command while retaining root privileges; seensr(1m) for more details. Thedevicemustalways be specified, and is usually one of the device names used by the NetWorker server. For tape drives,it must be the name of a ‘‘no-rewind on close’’ device.

Whenscanner is invoked with either no options or−v, the volume on the indicateddeviceis opened forreading, scanned, and a table of contents is generated. The table of contents contains information abouteach save set found on the volume. By default, one line of information is written to standard output foreach save set found containing the client name, save set name, save time, level, size, files, ssid and a flag.The client name is the name of the system that created this save set. Thename is the label given to thissave set bysave(1m), usually the path name of a file system. Thesave timeis the date and time the saveset was created. Thelevel values are one-letter abbreviated versions offull, incremental, levels0 through9,or blank for manual saves. Thesizeis the number of bytes in the save set. Thefiles labeled by column pro-vide the number of client files contained in the save set. Thessid (save set identifier) is an identifier usedinternally to reference and locate this save set. This same identifier may be specified explicitly with the−Soption to extract a particular save set.

The table of contents is based on synchronization (sometimes called ‘‘note’’)chunks(seemm_data(5))interspersed with the actual save set data. There are four types of note chunks: Begin, Continue, Synchro-nize, and End, symbolized by aflagof B, C, S or E respectively. TheBegin note is used to mark the start ofa sav e set. When a beginning chunk is written, the save set size and number of files are not known. TheContinue note is used to indicate that this save set started on a different volume. TheSynchronize notemarks locations in the save set where you may resume extracting data in the event of previous media dam-age (a client file boundary). TheEnd note marks the end of the save set, and causes the table of contentsline to be printed. The other notes are displayed only when the−v option is selected.

OPTIONS−b pool

Specifies which pool the volume should belong to. This option only applies for versions of Net-Worker that do not store the pool information on the media. For these versions, you might need tospecify the media pool the volume should belong to if the user does not want the volume to be amember of theDefaultpool. For volumes where the pool information is stored on the media, themedia must be relabeled (destroying all data on the media) to assign the media to a different pool.

−B When used in absence of the−S option,scannerquickly scans the tape for the start of bootstrapsave sets. The program only reads the first record at each file mark on the media to see if a saveset with the name ‘‘bootstrap’’ exists. When the entire media has been exhausted, the save set idand file location of the most recent bootstrap save set is printed. When used in conjunction withthe−Soption, the save set id specified is flagged as that of a bootstrap.



−c clientProcess only save sets that come from the specified NetWorker client machine. This option can beused multiple times and in conjunction with the−N option, but only in presence of the−i or −xoption.

−f file Starts the scan at the specific media file number. This option is not useful on media such as opticaldisks and file device types, for example.

−i Rebuilds both the media and the online file indexes from the volumes read. If you specify a singlesave set with the−S ssidoption, only entries for the specified save set are copied to the online fileindex. Note that for version 6.0 and later, if you have the tape that contain the index backups thatgo along with the data backups, the recommended way of restoring your indexes is to runscanner−m to reload the media database entries for the index and data backups. Once that is done, youshould runnsrck −L7 −t date <clientname>to recover the index for the client as of the time ofthe backups on the tape. This will roll the index entries for that time back into the index. How-ev er, if you have tapes for which there are no index backups, then you will need to use the−ioption to reconstruct the index entries.

−m Rebuild the media indexes for the volumes read. If you specify a single save set with the−S ssidoption, only entries for the specified save set are copied to the online file index.

−n Checks all media without rebuilding the media or index databases. When used with the−i option,this option provides the most complete media checking available, while not modifying thedatabases at all.

−N nameOnly processes save sets specified byname(a literal string only). This option can be used multi-ple times and in conjunction with the−c option, but only in presence of the−i or −x option.

−p Prints out information save set notes as they are processed.

−q Displays only errors or important messages.

−r recordStarts the scan at the specific media record number.

−sserverSpecifies the controlling server when usingscanneron a storage node. Seensr_storage_node(5)for additional detail on storage nodes.

−Sssid Extracts the specified save set(s). When used with the−i or −x option, this option can be usedmultiple times and is in addition to any sav e sets selected using the−c and−N options. Otherwise,the volume is scanned for save setssid,which will be written to the standard output. Most oftenthis is piped to auasm(1m) program running in recover mode to process the save set (potentiallywith a directory list to limit the files to be recovered and potentially using a−m argument to mapthe file location). When using−S without −i or −m, scanner prompts for the volume block size ifthe volume label is not readable. If the volume information is still in the media database, the userhas the option of running recover by sav e set (seerecover(1m)). When−B is also specified,ssidis taken to be that of a bootstrap. Only one ssid is allowed in this case.

−t type Specifies the type of media, for example,optical for an optical disk, or8mm 5GBfor an 8mm 5GBtape). Normally the media type is obtained from the NetWorker server, if a known device is beingused (seensr_device(5)).

−v Displays more verbose messages, such as a log of each note chunk, and a message after every 100media records. When the−i option is used, a line is printed for each client file (an enormousamount of output can be produced).

−x command arg ...Specifies an arbitrary UNIX command to process each new selected save set. This argument canonly occur once at the end of the argument list (afterdevice). The save stream for each save set isconnected to the stdin of a new instance of the command. Most often this command isuasm(1m)



running in recover mode to process each save set (potentially using a−m argument to map the filelocation). If the volume information is still in the media database, the user has the option of run-ning recover by sav e set (seerecover(1m)). Do not attempt console I/O by specified UNIX com-mand. Instead specify conflict resolution parameters as arguments passed to the command (e.g.:scanner -S <ssid> -x uasm -iR -rv). If console interaction is required, pipe scanner output to thedesired Unix command instead of invoking the command using the -x option.

EXAMPLESVerifying a tape:

scanner /dev/nrst0

scanner: scanning 8mm tape mars.001 on /dev/nrst0

client name save set save time level size files ssid Sspace /export 10/07/94 12:38 f 100762460 10035 16983 Espace /usr 10/07/94 13:14 f 27185116 3185 16984 Espace /nsr 10/07/94 12:40 f 77292280 8436 16980 Sspace / 10/07/94 13:22 f 1693192 518 16985 Sscanner: reached end of 8mm tape mars.001

Rebuilding the online file index for a client from a tape:

scanner -m /dev/nrst8

scanner: scanning 4mm tape monday.fulls on /dev/nrst8scanner: ssid 17458697: scan completescanner: ssid 17458694: scan completescanner: ssid 17458698: scan completescanner: ssid 17458693: NOT completescanner: reached end of 4mm tape monday.fulls

scanner: when next tape is ready, enter device name [/dev/nrst8]?

nsrck -L7 -t "06/07/99" supernova

nsrck: checking index for ’supernova’nsrck: The file index for client ’supernova’ will be recovered.nsrck: Recovering index sav esets of ’supernova’ from ’quasar’Recover completion time: Fri Jun 16 14:03:16 2000nsrck: completed recovery of index for client ’supernova’

nsrck: /disk1/nsr/index/supernova contains 85782 records occupying 14 MBnsrck: Completed checking 1 client(s)

Extracting a save set for /usr and relocating to /mnt:

scanner -S 637475597 /dev/nrst8 | uasm -rv -m /usr=/mntor

scanner -S 637475597 /dev/nrst8 -x uasm -rv -m /usr=/mnt

Extracting all save sets from client mars and relocating to /a:

scanner -c mars /dev/nrst8 -x uasm -rv -m/=/a

SEE ALSOmm_data(5), mminfo(1m), nsrmmdbasm(1m), nsr(1m), nsrck(1m), nsrindexasm(1m), nsrmmd(1m),nsr_device(5), nsr_storage_node(5), uasm(1m).

DIAGNOSTICSxdr conversion error, fn %d, rn %d, chunk %d out of %d



unexpected file number, wanted %d got %dunexpected record number, wanted %d got %d

All three preceding messages are indicative of media errors (tape blocks are either lost or dam-aged). In the case of an xdr conversion error, a non-zero ‘‘chunk’’ number means that the blockmay be partially salvageable. Unexpected file numbers are normal whenscannerreaches the logi-cal end of the media that has been recycled.

continuation of data in nsrscan.NNNNN.MMMMMMAfter an XDR decode error (an error denoted by one or more of the messages described above),scannerattempts to re-synchronize and send the rest of the stream. However, because programslike uasm(1m) are unable to handle decoding streams with parts missing in the middle, scannersends the remainder of the stream to a file. You can decode this stream manually. For example, ifyour original command was:

scanner -S ssid | uasm -rand a synchronization error occurs, you can decode the rest of the stream with the following com-mand:

uasm -r < nsrscan.NNNNN.MMMMMMwhere the file name you enter corresponds to the name printed in the diagnostic message.

unexpected volume id, wantedvolid1got volid2This message normally appears when running in verbose mode on a tape or disk that has beenrecycled. It does not indicate an error condition, but details the conditions normally treated as theend of the volume.

ssid %d: finished, but incompleteScanner has detected the end of a save stream, but the stream was aborted, and is of dubiousvalue. If online indexes are being rebuilt, the end of the aborted stream may precipitate the nextmessage.

(ssid %d): error decoding save streamAs indexes are being rebuilt,scannerdetected that the bytes in the save stream are invalid. This isusually caused by processing an aborted save stream. Other causes may include a damaged tape.Once this condition is detected, the process of rebuilding the indexes for the particular save streamexits. This may precipitate the next message.

write failed, Broken pipePrinted byscannerwhen a process rebuilding a save stream’s indexes exits before consuming theentire stream.

You are not authorized to run this commandA normal (non-root) user invoked this command.

could not convert ‘arg’ to a file numberThe −f and −r options require a numeric argument for the starting file or record number of themedia.

already exists in the media indexThe−i or −m option was specified and the volume was already listed in the media database. Thismessage is purely informational, and means that the volume is not being added to the mediadatabase because it is already listed there.

fn %d rn 0 read error I/O errordone withtape_typetape volid volume_name

These messages, when occurring together, are a consequence of scanner encountering consecutivefilemarks at end of the media. They do not indicate an error condition and can be ignored.

LIMITATIONSscannercan run without the NetWorker services (for example,nsrd(1m) andnsrmmdbd(1m)) when notreconstructing the media or the online file indexes with most device types. For logical and NDMP devices,the NetWorker services have to be running in order to query these device configurations.



File index backups imported from volumes from other Networker servers cannot be recovered bynsrck-L7. You must usemmrecov to recover the Bootstrap of that NetWorker server before the file indexes canbe recovered.

When scanning a relabeled optical volume (that is, a re-writable optical volume that had been written once,then re-labeled and used again),scannermay read off the end of the new data, and attempt to read the olddata from the previous version of the volume, terminating with an ‘‘unexpected volume id’’ error. Thiserror occurs after all the good data has been read, and can be ignored.


SJIDOPEN(1m) SJIDOPEN(1m)

NAMEsjidopen − test the SJI Jukebox Interface SJIDOOROPEN command

SYNOPSISsjidopendevname

DESCRIPTIONThe sjidopen program tests the SJIDOOROPEN command on SJI compliant Jukeboxes (seelibsji(1m)).This command tests the open/close capability of the main door to a Jukebox. If a Jukebox doesn’t supportthis feature, an appropriate error message will be printed out.

The devnameargument should be any device name that can be used to reach a SJI compliant Jukeboxdriven by the system. Typical usage is in constructing the name that contains the SCSI Bus, the SCSI TargetID and the SCSI Lun on that target, e.g.,[email protected].

SEE ALSOlibsji(1m)


SJIIELM(1m) SJIIELM(1m)

NAMEsjiielm − test the SJI Jukebox Interface SJIIELEM command

SYNOPSISsjiielm devname[ { drive | slot | inlt | mt } address nelements]

DESCRIPTIONThe sjiielm program tests the SJIIELEM command on SJI compliant Jukeboxes (seelibsji(1m)). Thiscommand tests the Initialize Element Status interface for a Jukebox. If a Jukebox doesn’t support this fea-ture, an appropriate error message will be printed out.


The additional optional arguments are for Jukeboxes that support the initialization of a specific range of ele-ments. In this case, you select one of element types ofdrive, or slot, or inlt (Import/Export element), ormt(Media Transport, or the "Robot Arm"), and give an SJI normalized (i.e., starting from 1) address and num-ber of elements to initialize.

SEE ALSOlibsji(1m)


SJIINQ(1m) SJIINQ(1m)

NAMEsjiinq − test the SJI Jukebox Interface SJIINQ command

SYNOPSISsjiinq devname

DESCRIPTIONThe sjiinq program tests the SJIINQ command on SJI compliant Jukeboxes (seelibsji(1m)). This com-mand returns a string that identifies a Jukebox. If a Jukebox doesn’t support this feature, an appropriateerror message will be printed out.


SEE ALSOlibsji(1m)


SJIRDP(1m) SJIRDP(1m)

NAMEsjirdp − test the SJI Jukebox Interface SJIRDP command

SYNOPSISsjirdp devname

DESCRIPTIONThe sjirdp program tests the SJIRDP command on SJI compliant Jukeboxes (seelibsji(1m)). This com-mand reads SJI ordinal device positions from a Jukebox.

Typical output for this command might look like:

[email protected] has 2 DAT A TRANSPORT Elements starting at address [email protected] has 1 MEDIA TRANSPORT Element starting at address [email protected] has 25 STORAGE Elements starting at address [email protected] has 1 IMPORT/EXPORT Element starting at address 1


SEE ALSOlibsji(1m)


SJIRDTAG(1m) SJIRDTAG(1m)

NAMEsjirdtag − test the SJI Jukebox Interface SJIRTA G command

SYNOPSISsjirdtag devname

DESCRIPTIONThe sjirdtag program tests the SJIRTA G command on SJI compliant Jukeboxes (seelibsji(1m)). Thiscommand reads media presence and tag data from a Jukebox.

Example output for this command:

Tag Data for 0.2.1, Element Type DAT A TRANSPORT:Elem[001]: tag_val=0 pres_val=1 med_pres=1 med_side=0

Tag Data for 0.2.1, Element Type STORAGE:Elem[001]: tag_val=0 pres_val=1 med_pres=1 med_side=0Elem[002]: tag_val=0 pres_val=1 med_pres=1 med_side=0Elem[003]: tag_val=0 pres_val=1 med_pres=1 med_side=0Elem[004]: tag_val=0 pres_val=1 med_pres=1 med_side=0Elem[005]: tag_val=0 pres_val=1 med_pres=0 med_side=0Elem[006]: tag_val=0 pres_val=1 med_pres=1 med_side=0Elem[007]: tag_val=1 pres_val=1 med_pres=1 med_side=0

VolumeTag=<00000098>Tag Data for 0.2.1, Element Type MEDIA TRANSPORT:

Elem[001]: tag_val=0 pres_val=1 med_pres=0 med_side=0


The output is sorted by types available, and then by ascending order.

The boolean tokentag_val states whether the tag data is available (valid). In the example above, only theseventh storage element had valid tag data (which was the bar code numbered ’00000098’).

The boolean tokenmed_presstates whether the Jukebox believes that there is media present at this loca-tion.

The boolean tokenpres_val is a subtle and overloaded indicator that states (if set to true, or 1), that the thetokenmed_presshould be believed absolutely. Ifpres_val is not true (set to zero), ormed_presis true (setto one), it is probable that thereis media at that location, but that there is some exception associated with it.A possible exception is that a tape has a missing, upsidedown, or unreadable bar code label. Also, a Juke-box cannot determine whether media that existed at one point (for example, prior to a power failure) hasbeen removed from a Data Transport element (tape drive). If the tokenpres_val is not true (set to zero),andmed_presis also not true (set to zero), there probably isn’t media in that location. This uncertainty iseliminated the next time an INITIALIZE ELEMENT STATUS is done (seesjiielm(1m)).

The tokenmed_sizeis for two sided media (e.g., optical disks), where the Jukebox has kept track of whichside is up.

SEE ALSOlibsji(1m)sjiielem(1m)


SJIRELEM(1m) SJIRELEM(1m)

NAMEsjirelem − test the SJI Jukebox Interface SJIRELEM command

SYNOPSISsjirelem devname

DESCRIPTIONThe sjirelem program tests the SJIRELEM command on SJI compliant Jukeboxes (seelibsji (1m)). Thiscommand reads media presence and origin data from a Jukebox.


Element Data for 0.2.1, Element Type DAT A TRANSPORT:Elem[001]: pres_val=1 med_pres=1 med_side=0

Origin: type STORAGE, address 5Element Data for 0.2.1, Element Type STORAGE:

Elem[001]: pres_val=1 med_pres=1 med_side=0Elem[002]: pres_val=1 med_pres=1 med_side=0Elem[003]: pres_val=1 med_pres=1 med_side=0Elem[004]: pres_val=1 med_pres=1 med_side=0Elem[005]: pres_val=1 med_pres=0 med_side=0Elem[006]: pres_val=1 med_pres=1 med_side=0Elem[007]: pres_val=1 med_pres=1 med_side=0

Element Data for 0.2.1, Element Type MEDIA TRANSPORT:Elem[001]: pres_val=1 med_pres=0 med_side=0


The output is sorted by types available, and then by ascending order.

The boolean tokenmed_presstates whether the Jukebox believes that there is media present at this loca-tion.

The boolean tokenpres_val is a subtle and overloaded indicator that states (if set to true, or 1), that the thetokenmed_presshould be believed absolutely. Ifpres_val is not true (i.e., set to zero), andmed_presistrue (i.e., set to one), it is probable that thereis media at that location, but that there is some exception asso-ciated with it. This may mean (in some cases and for some Jukeboxes) that a tape has a missing, upsidedown, or unreadable bar code label. It may also mean, hoever, in the case of media located inside a DAT ATRANSPORT element (tape drive) that the Jukebox cannot really tell, but that as far as it knew, therewasmedia there at some point in the past (say, prior to a power failure). If the tokenpres_val is not true (i.e.,set to zero), andmed_presis also not true (i.e., set to zero), it is pretty likely that there isn’t media in thatlocation. Typically, most of this uncertainty gets eliminated the next time an INITIALIZE ELEMENTSTATUS is done (seesjiielm(1m)).

The tokenmed_sizeis for two sided media (e.g., optical disks), where the Jukebox has kept track of whichside is up.

When the Jukebox can, it might also say where the last place a piece of media had been prior to its currentlocation, in which casesjirelemwill print that out.

SEE ALSOlibsji(1m)sjiielem(1m)


SJIRJC(1m) SJIRJC(1m)

NAMEsjirjc − test the SJI Jukebox Interface SJIRJC command

SYNOPSISsjirjc devname

DESCRIPTIONThe sjirjc program tests the SJIRJC command on SJI compliant Jukeboxes (seelibsji (1m)). This com-mand reads internal configuration information and options about a Jukebox and prints it out.


Device: [email protected]

Number of Drives: 1Number Drive Pairs: 1

Number of Import/Export Elements: 0Number of Import/Export Pairs: 1

Number of Slots: 7Number of Slot Pairs: 1

Number of Transport Elements: 1Number of Transport Pairs: 1

Initialize Element Status SupportedAuto Eject Supported


Since the SJI specification allows for multiple disjoint sets of elements, you will always see a count of’Pairs’ (but the author has never run across a value other than ’1’). The output tells you how man drives,slots, transports (gripper arm), and import/export elements there are.

In addition, various internal library options are printed out.

SEE ALSOlibsji(1m)


ssi(8) ssi(8)





(Unix only)









ssi(8) ssi(8)



(NT only)




(All platforms)


OPTIONSmini_el




PARAMETERSssi








ssi(8) ssi(8)













stk_eject(1m) stk_eject(1m)

NAMEstk_eject − eject tapes from a StorageTek silo

SYNOPSISstk_eject

acsls_hostnamecap_idmodevolser0 ... volsern

DESCRIPTIONThe stk_eject command allows you to eject one to 42 tapes at a time from a StorageTek silo. By dealingdirectly with ACSLS (through ssi on the NetWorker system), this utility originally was intended to avoidnsrjb’s limitation of ejecting only one tape per command. The limitation in nsrjb was removed in Net-Worker version 5.5 Patch 3. However,stk_eject is being maintained for compatability and for use in cus-tomer developed scripts.

Starting with version 1.1 ofstk_eject,you will be able to eject volumes from devices controlled by morethan one ACSLS server, as long as you are also usingssiversion 1.06 or greater. The correctssi to use willbe automatically detected bystk_ejectbased on the ACSLS hostname passed in on the command line.

NOTE: stk_eject doesNOT deallocate volumes for NetWorker. YouMUST perform the deallocation sep-arately using thensrjb -x -Txxx command to properly maintain NetWorker’s idea of what is and what isnot present in the silo.

All parameters are required, and the list of volsers can be from 1 to 42 elements long. Volsers beyond thelimit of 42 will be ignored.

acsls_hostname− 1 word host ID of system running ACSLS or Library Station - e.g.expo1

cap_id− STK cap name, no internal spaces - e.g.0,0,0

mode− should this command wait until the eject is completed?values areWAIT or NOWAIT for QUIETmode

− or −WAITV or NOWAITV for VERBOSEmode

volser− 6 character STK volser list - not checked for size or form - only first 6 characters are used. Volsersshould be separated by a single space character.

*** You may eject up to 42 VOLSERS per command ***

NOTE: inNOWAIT or NOWAITV mode, there is no confirmation of ejection on this system.

If you require confirmation, you should use theWAIT or WAITV modes. This program will then receiveconfirmation fromACSLS, but it will not return to the caller until all ejected tapes have been removed fromthe CAP.

You will not receive any messages regarding emptying the CAP when it fills, or when the eject is done -youmust use the ACSLS console to see what the CAP status is.

This utility usesssi to communicate withACSLS, sossimust be properly configured and running on thesystem where this command is used. Note that this utility does not depend on any other NetWorker files, soit can be run on any system that is running ssi which can communicate with the desiredACSLS system.

SEE ALSOSTK_silo(1m)


STK_silo(8) STK_silo(8)





(Unix only)












(NT only)




(All platforms)


OPTIONSmini_el




PARAMETERSssi








STLI(1m) STLI(1m)

NAMEstli − Standard Tape Library Interface

DESCRIPTIONSTLI is the documented interface for connecting tape libraries to NetWorker.Tape libraries are often referred to as silos. Examples of tape libraries are StorageTek ACSLS silos,ABBA systems from Grau or the IBM 3494. Tape libraries differ from other jukeboxes supported by Net-Worker in two major points:

o The tape libraries have their own volume and slot management. The volumes are accessed byNetWorker via their barcode labels. The exact location of a given tape is unknown to NetWorker.

o The interface to the library for issuing requests like mount or unmount of particular volumes differs fromthat of jukeboxes, which are normally connected via a SCSI or V24 interface. Standard Tape Librarycommands are transmitted to a Standard Tape Library server, which receives these commands andinvokes the appropriate library actions. Usually, the connection between NetWorker and the Library con-troller is over the network, although serial (RS-232) connections are also used. Note that the actual con-nection is hidden from NetWorker by the STL library. Accordingly NetWorker and other applicationscalling the STL server are called library clients.

To keep applications independent of differing interfaces to various tape library systems, an API called STLI(Standard Tape Library Interface) has been defined, which is used by NetWorker to invoke library requests.The STLI specifies a shared library with well defined functions, which is dynamically linked to NetWorker.These STLI interface libraries, which transform the STLI function calls to library-specific calls to the pro-prietary tape library server, may be provided by Legato or the manufacturers of the tape library.

Not all functions specified in this paper must be implemented in the STLI library. These functions are theminimum necessary for a functional library:

stl_open()stl_close()stl_mount()stl_unmount()

Implementation of stl_error() is recommended for easier use of the library and better troubleshooting.

If you wish to support dynamic device reservation these are the relevant functions:

stl_reserve_dev()stl_release_dev()stl_dev_reservation()

Optional fucntions for added features:

stl_query_volume()stl_withdraw_volume()stl_withdraw_volumes()stl_deposit_volume()stl_deposit_volumes()stl_version()

stl_close()Declaration:

int stl_close(char* stl);

Description:

Close the connection to the tape library.

<stl> is the handle returned by stl_open().


STLI(1m) STLI(1m)

The return value on success is STL_ERR_NOERR. For error return values see appendix.

stl_deposit_volumeDeclaration:

int stl_deposit_volume(char *stl, char *volume, char *capname)

Description:

Causes the specified volume to be inserted into the library from the specified cartridge access port(inport/export facility, mailslot....)


<volume> is the barcode of the volume to be deposited into the library.

<capname> specifies the cartridge access port/inport-export area/mailslot to be used to insert thevolume into the library. It is a character string, which is understood by the tape library as a namefor that device. This argument can be NULL, in which case the default CAP will be used.

This function returns STL_ERR_NOERR if the volume is successfully inserted.STL_ERR_NOVOL is returned if the volume is not present and was not inserted. Other returnvalues are possible if errors occur. See appendix for possible values.

NOTE: on some libraries, this function may not be needed.

o The IBM 3494 will automatically import any tapes placed into its in/out area. There is no ’deposit’ func-tion in the 3494’s API.

o StorageTek libraries with the CAP set to automatic mode behave in the same manner as the 3494. How-ev er, if their CAP is set to manual, then a deposit call is required.

o On EMASS/Grau libraries, a single call to deposit one volume from the EIF will in fact deposit all avail-able volumes. However, subsequent deposit calls will return quickly and the error can be ignored.

stl_deposit_volumesDeclaration:

int stl_deposit_volumes(char *stl, char *volumes, char *capname)

Description:

Causes the specified volumes to be inserted into the library from the specified cartridge access port(inport/export facility, mailslot....). This function will be used instead of stl_deposit_volume() if itis defined and stl_version returns 1.3 or greater. Therefore, it should be capable of functioningwith either a single volume specified or with a comma separated list.


<volumes> is a comma separated list of barcodes of the volumes to be deposited into the library.There should be no extraneous spaces added between the individual barcodes since the space char-acter (ASCII 32) is a valid barcode character itself.

<capname> specifies the cartridge access port/inport-export area/mailslot to be used to insert thevolume into the library. It is a character string, which is understood by the tape library as a namefor that device. This argument can be NULL, in which case the default CAP will be used.

This function returns STL_ERR_NOERR if the volume is successfully inserted.STL_ERR_NOVOL is returned if the volume is not present and was not inserted. Other returnvalues are possible if errors occur. See appendix for possible values.

NOTE: on some libraries, this function may not be needed.

o The IBM 3494 will automatically import any tapes placed into its in/out area. There is no ’deposit’ func-tion in the 3494’s API.

o StorageTek libraries with the CAP set to automatic mode behave in the same manner as the 3494. How-ev er, if their CAP is set to manual, then a deposit call is required.


STLI(1m) STLI(1m)

o On EMASS/Grau libraries, a single call to deposit one volume from the EIF will in fact deposit all avail-able volumes. However, subsequent deposit calls will return quickly and the error can be ignored.

stl_dev_reservation()Declaration:

int stl_dev_reservation(char *stl, char *device, int *state)

Description:

Get the reservation state of device <device>.


<device> specifies the device to get the reservation state of. It is a character string, which isunderstood by the tape library as a name for that device.

*<state> returns the reservation state:

STL_DEV_FREE: Free

STL_DEV_RESERVED: Reserved for NetWorker’s use

STL_DEV_OCCUPIED: Occupied by another host


stl_error()Declaration:

char *stl_error(void)

Description:

Gives a printable error message belonging to the preceding STLI function call. The functionreturns the address of a buffer which contains a message describing the status of the last STLIfunction call. These messages can be constant strings, for instance the messages contained in stl.hfor error codes less than 100. But these messages can also be built up with actual parameters,which describe more exactly the error situation.

Error and other status information should be maintained in global static variables in the library,since this call is made without any parameters.

The function returns NULL if no message available. See appendix.

stl_mount()Declaration:

int stl_mount(char* stl, char* volume, char* device);

Description:

Move volume <volume> into drive <drive>.


<volume> is the barcode of the volume to be mounted.

<device> specifies the device, on which the volume shall be mounted. It is a character string,which is understood by the tape library as a name for that device.

This call may not return until the volume is loaded into the drive and the drive has come ready.The exact sequence is dependent on the library. It can therefore take sev eral minutes to complete.


stl_open()Declaration:


STLI(1m) STLI(1m)

int stl_open(char* server, char** stl);

Description:

Connect to the tape library.

<server> is a character string which contains all information necessary to establish a connection tothe tape library. The information in this string is proprietary to the special type of tape library.Generally it should be of the form:

[<host>] [<par1>=<val1> [<par2>=<val2>]...]

In most cases the string contains merely <host>, the nodename of a library server, which receivesand serves the STLI requests over the network.

*<stl> is a handle returned for use by the other STLI functions. This handle can be used to storeinternal information between subsequent function calls. For some libraries, the parameter to thiscall may or may not be used, as environment variables may be used to hold the required configura-tion information.

The return value on success is STL_ERR_NOERR. For error return values see appendix

stl_query_volumeDeclaration:

int stl_query_volume(char *stl, char *volume)

Description:

Queries a silo to establish the presence of a volume. This function is currently used to verify thepresence of a volume before allocating that volume for use with NetWorker.


<volume> is the barcode of the volume to be mounted.

This function returns STL_ERR_NOERR if the volume is present, or STL_ERR_NOVOL if thevolume is not present.

Other return values are possible if errors occur. See appendix for possible values.

stl_release_dev()Declaration:

int stl_release_dev(char *stl, char *device);

Description:

Release device <device>, which has previously been reserved by stl_reserve_dev().


<device> specifies the device to be released. It is a character string, which is understood by thetape library as a name for that device.


stl_reserve_dev()Declaration:

int stl_reserve_dev(char *stl, char *device);

Description:

Reserves device <device> for NetWorker’s use.


<device> specifies the device to be reserved. It is a character string, which is understood by thetape library as a name for that device.


STLI(1m) STLI(1m)


stl_version()Declaration:

int stl_version(void)

Description:

Returns STLI version information for the STL library

This function returns the version of the STL library * 10. I.e., it returns a value of 12 for an STLlibrary that supports the functions for STLI version 1.2.

STLI version 1.0 specified the following calls:

stl_close()stl_dev_reservation()stl_mount()stl_open()stl_release_dev()stl_reserve_dev()stl_unmount()

STLI version 1.1 added the following calls:

stl_query_volume()stl_version()


stl_deposit_volume()stl_withdraw_volume()


stl_deposit_volumes()stl_withdraw_volumes()

Note that stl_version does not return a value that can be interpreted as an ’STL_’ error. Attempt-ing to do so will result in unpredictable results.

stl_unmount()Declaration:

int stl_unmount(char* stl, char* volume, char* device);

Description:

Remove volume <volume> from drive <drive>.


<volume> is the barcode of the volume to be removed.

<device> specifies the device, from which the volume shall be removed. It is a character string,which is understood by the tape library as a name for that device.

Either <volume> or <device> can be NULL. If both values are specified, they must be consistent.

This call will not return until the volume is ejected from the drive and returned to its slot by thelibrary. It can therefore take sev eral minutes to complete.


stl_withdraw_volumeDeclaration:

int stl_withdraw_volume(char *stl, char *volume, char *capname)

Description:


STLI(1m) STLI(1m)

Causes the specified volume to be ejected from the library through the specified cartridge accessport (inport/export facility, mailslot....)


<volume> is the barcode of the volume to be withdrawn.

<capname> specifies the cartridge access port/inport-export area/mailslot to be used to remove thevolume from the silo. It is a character string, which is understood by the tape library as a name forthat device. This argument can be NULL, in which case the default CAP will be used.

This function returns STL_ERR_NOERR if the volume is successfully withdrawn from thelibrary. STL_ERR_NOVOL is returned if the volume is not present and STL_ERR_VOLBUSY isreturned if the volume is currently in use and cannot be withdrawn.


stl_withdraw_volumesDeclaration:

int stl_withdraw_volumes(char *stl, char *volumes, char *capname)

Description:

Causes the specified volume to be ejected from the library through the specified cartridge accessport (inport/export facility, mailslot....) This function will be used instead ofstl_withdraw_volume() if it is defined and stl_version returns 1.3 or greater. Therefore, it shouldbe capable of functioning with either a single volume specified or with a comma separated list.


<volumes> is a comma separated list of barcodes of the volumes to be deposited into the library.There should be no extraneous spaces added between the individual barcodes since the space char-acter (ASCII 32) is a valid barcode character itself.

<capname> specifies the cartridge access port/inport-export area/mailslot to be used to remove thevolume from the silo. It is a character string, which is understood by the tape library as a name forthat device. This argument can be NULL, in which case the default CAP will be used.

This function returns STL_ERR_NOERR if the volume is successfully withdrawn from thelibrary. STL_ERR_NOVOL is returned if the volume is not present and STL_ERR_VOLBUSY isreturned if the volume is currently in use and cannot be withdrawn.


Appendix: Return ValuesReturn values 0 - 99 are reserved for common, library type independent error codes. The header file stl.hdefines the common return values together with a short error message.

Return values greater 100 can be used by each STLI implementation for proprietary error codes.

It is recommended, that a STLI implementation should map all error situations to the common STLI errorcodes and should provide the function stl_error() for more detailed error messages. This allows NetWorkerto react on known error codes, but also to forward the more detailed error messages via the user interface.

No proprietary error codes are allowed in situations, where the common error codesSTL_ERR_DEVEMPTY, STL_ERR_DEVFULL or STL_ERR_ALRDYMNTED apply.

The currently defined return codes are:

Error value Meaning-------------------------------------------------------

STL_ERR_NOERRSuccessful call return - no error


STLI(1m) STLI(1m)

STL_ERR_UNKNOWNError, no details available

STL_ERR_CONNECTCannot connect to tape library

STL_ERR_BUSYTape library busy, try later

STL_ERR_ACCESSPermission denied (to access the library, the requested device, volume or operation)

STL_ERR_NODEVDevice not known to the tape library or physically not available

STL_ERR_NOVOLVolume not known to the tape library or physically not available

STL_ERR_DEVFULLDevice already loaded with another volume

STL_ERR_DEVEMPTYDevice empty

STL_ERR_DEVBUSYDevice busy

STL_ERR_ERRNOLocal UNIX error, see errno

STL_ERR_INVALIDInvalid parameter

STL_ERR_VOLBUSYVolume already loaded in another drive or is otherwise occupied

STL_ERR_LIBRARYTape library internal error

STL_ERR_CONFIGRequest doesn’t comply with tape library configuration

STL_ERR_DEVOCCDevice reserved by another host

STL_ERR_DEVRESDevice already reserved

STL_ERR_DEVNOTRESDevice not reserved

STL_ERR_NOTINSTSTLI-Library is a dummy library

STL_ERR_NOTSUPPDummy function return

STL_ERR_ALRDYMNTEDRequested volume already mounted in requested device

SEE ALSOnsrjb (1m),nsr_jukebox(5), IBM_silo (1m),EMASS_silo(1m),STK_silo(1m).


SYM2XDM(1m) SYM2XDM(1m)

NAMEsym2xdm − convert symbolic link HSM files to XDSM HSM files

SYNOPSISsym2xdm[ −v ] [ −s <NWserver> ] [<path> ]

DESCRIPTIONsym2xdm is a NetWorker XDSM HSM-specific command to convert HSM symbolic links used by previ-ous versions of NetWorker HSM.sym2xdm searches for HSM symbolic links in the directory tree speci-fied by path.

If no path is specified on the command line,sym2xdmuses the current directory.

OPTIONS−v Displays more detailed information.

−s <NWserver>Specifies which machine to use as the NetWorker server (hostname or IP address).

SEE ALSOnsrpmig(1m), nsrmig(1m).


TAPEEXERCISE(1m) TAPEEXERCISE(1m)

NAMEtapeexercise − exercise a tape drive

SYNOPSIStapeexercise[ −vBEFP ] devname

DESCRIPTIONThe tapeexerciseprogram writes sample data to a tape, and tests to see if positioning and read operationsperform as expected. It needs a write-enabled tape on the indicated (no-rewind) drive. Tapeexercise shouldbe used with extreme caution. Successful completion is indicated by a ‘‘<test name>: test begin,’’‘‘<test name>: test ok’’ pair for each test that is run. An example is as follows:

BasicTest: test beginBasicTest: test ok

OPTIONS−v Operate in verbose mode.

−B Perform only the basic test.

−E Perform only the EOT test.

−F Perform only the File Space Forward test.

−P Perform only the ’SCO’ positioning test.

If none of the optionsBEFP are set, then all of the tests are performed.

devnameThe device name of the tape device under test. This should be a non-rewind device, following thelocal operating system convention.

SEE ALSOnsrmmd(1m),Hardware Compatibility Guide.

LIMITATIONSThe tapeexerciseprogram will generally fail for QIC drives, because these devices do not support all of thefunctionality assumed bytapeexercise, most importantly, they do not support back-skip-file. Such devicesmay work withnsrmmd(1m); consult theHardware Compatibility Guidefor a complete list of supporteddevices.


TUR(1m) TUR(1m)

NAMEtur − test unit ready

SYNOPSIStur [ -a b.t.l ] [ -l ]

DESCRIPTIONThetur program will send a TEST UNIT READY command to all SCSI devices attached to the system.

−a b.t.l Selects a specific ordinal SCSI address, whereb is the logical SCSI bus,t is the SCSI target, andlis the SCSI logical unit number (LUN) on that target. Seelibscsi(1m).

−l Performs a complete LUN search for all SCSI adapters in the system. This argument is acceptedon all systems, but does not have any effect on HP-UX systems. Due to the method used to scanfor available devices on HP-UX systems, all accessible devices are always shown, and the−loption has no additional effect. On all other systems, the normal behavior is to start checking atLUN 0 for SCSI deivces. The first empty LUN found will end the search for a given target ID.With the −l option, all LUNS present on all target IDs for all SCSI busses in the system will bechecked for devices. This can take avery long time and should therefore only be used when nec-essary. For example, a Fibre Channel adapter can support 126 target IDs, each of which may have80 or more LUNs. Checking all LUNs on this single adapter may take over 10 minutes.

SEE ALSOlibscsi(1m)


UASM(1m) UASM(1m)

NAMEuasm − NetWorker module for saving and recoveringUNIX filesystem data

SYNOPSISuasm −s[ -benouv] [ -ix ] [ −t time] [ −f proto ][ −p path] path...uasm −r [ −dnuv ] [ −i { nNyYrR} ] [ −m <src>=<dst> ] −z suffix] [ path]...uasm −c[−nv] [ path]...

DESCRIPTIONTheuasmcommand is the default filesystemASM (Application Specific Module). It is built intosave(1m)and recover(1m). uasm may also be called directly in a manner similar totar (1). This description ofuasm applies to allASMs. For clarity, onlyuasm is mentioned in many of the descriptions in this manpage.

uasmhas three basic modes: saving, recovering, and comparing. When saving,uasmwill browse directorytrees and generate a save stream (seensr_data(5)), to the associatedstdoutfile representing the file anddirectory organization. When recovering,uasm reads a save stream from the associatedstdinfile and cre-ates the corresponding directories and files. When comparing,uasm reads a save stream from itsstdinfileand compares the save stream with the files already located in the file system.

During backup sessions, the behavior ofuasm can be controlled bydirectives. Directives control howdescendent directories are searched, which files are ignored, how the save stream is generated, and howsubsequent directive files are processed. (Seensr(5)). When browsing a directory tree, symbolic links arenever followed, except in the case ofrawasm.

ASMs can recover sav e streams from current or earlier versions of NetWorker. Note: olderASMs may notbe able to recover files generated by newerASMs.

The following list provides a brief description of theASMs supplied with NetWorker:

always ThealwaysASM always performs a back up of a file, independent of the change time of the file.

atimeasmTheatimeasmis used to backup files without changing the access time of the file. This function-ality is a subset ofmailasm. On most systems,atimeasmuses the filemtimefor selection andthen resets the fileatimeafter the backup (which changes the filectime). On systems that supportinterfaces for maintaining the fileatimewithout changing the filectime,atimeasmhas no effect,since the fileatimeis normally preserved.

compressasmThe compressasmuses a software compression algorithm to compress file data. This ASM doesnot compress directories. The amount of compression achieved is data-dependent.compressasmuses considerable amounts of CPU resources, so its benefits may be limited on low-powered sys-tems.

holey Theholey ASM handles holes or blocks of zeros when backing up files and preserves these holesduring recovery. On some filesystems interfaces can be used to find out the location of file holeinformation. Otherwise, blocks of zeros that are read from the file are skipped. This ASM is nor-mally applied automatically and does not need not be specified.

logasm The logasmenables file changes during backup sessions.logasmcan be used for “log” files andother similar files where a file changes during a backup operation is not worth noting.

mailasmThemailasm uses mail-style file locking and maintains the access time of a file, preserving “newmail has arrived” flag on most mail handlers.

mtimeasmThe mtimeasm is used to backup files using the filemtime for file selection instead of the filectime.


UASM(1m) UASM(1m)

nsrindexasmThe nsrindexasm is used to recover from NetWorker file index backups performed prior to Ver-sion 6. During recovery from these older index backups,nsrindexasm is invoked automaticallyby nsrck andmmrecov.

nsrmmdbasmThe nsrmmdbasm is used to process NetWorker’s media index. Normally,nsrmmdbasm isinvoked automatically bysavegrpandmmrecov, and should not be used in NetWorker directives.

null The null ASM does not back up the specified files and directories, but keeps the file name in theonline index of the parent directory.

nullasmnullasm is an alternate name for thenull ASM, named for backward compatibility with earlierreleases wherenullasm was a separate executable program instead of aninternal ASM.

posixcrcasmTheposixcrcasmis used to calculate a 32-bit CRC for a file during a backup. This CRC is storedalong with the file and is verified when the file is restored; no verification occurs during thebackup itself. Using thisASM it is possible to validate a file at restore time, but it does not providea way to correct any detected errors.

rawasmThe rawasm is used to back up/deventries (for example, block− and character−special files) andtheir associated raw disk partition data. On some systems,/deventries are actually symbolic linksto device specific names. Unlike otherASMs, rawasm follows symlinks, allowing the shorter/devname to be configured. When recovering,rawasm requires that the filesystem node for the rawdevice exist prior to the recovery. This protects against the recovery of a/deventry and the over-writing of data on a reconfigured disk. You can create the/deventry, having it refer to a differentraw partition, and force an overwrite if desired. If you create the/deventry as a symbolic link, thedata is recovered to the target of the symbolic link. Precautions should be taken when usingrawasm, see the CAVEATS section.

skip Theskip ASM does not back up the specified files and directories, and does not place the filenamein the online index of the parent directory.

swapasmTheswapasmdoes not backup actual file data, but recreates a zero-filled file of the correct size onrecovery. This ASM is used on systems where the swapping device is a swap file that must berecovered with the correct size, but the contents of the swap file are not important and do not needto be backed up.

xlateasmThexlateasmtranslates file data so that data backed up is not immediately recognizable.

Internal ASMs are not separate programs, but are contained within allASMs. ExternalASMs are separateprograms, and are invoked as needed. ExternalASMs provided with NetWorker arensrmmdbasm andnsrindexasm. All other ASMs previously listed are internal.

For security reasons, externalASM names must end inasmand be located in theorigin directory, which isthe same directory as the originally invoked program (typicallysaveor recover). In some system architec-tures, other directories relative to theorigin will be searched if anASM cannot be located in theorigindirectory.

WalkingASMs traverse directory trees. Theskip, null , andnullasm ASMs do not walk.

The internalASMs described here are modes, and a number of different internalASMs may be applied at thesame time. When an externalASM is needed to process a file, the newASM is invoked and generates thesave stream. When afiltering ASM is traversing a directory tree and invokes anotherASM, thatASMs sav estream is processed by thefiltering ASM. Hence, while usingcompressasmto backup a directory, themailasm can still be used to process the mail files correctly. Note that once different modes are set, theonly way to turn them off is to explicitly match anASM directive foruasm.


UASM(1m) UASM(1m)

Auto-appliedASMs are used under certain conditions, and do not need to be specifically mentioned in adirective file. For example, when a large file only has a small number of disk blocks allocated, theholeyASM is automatically invoked to process the file. Auto-appliedASMs are not used when a file namematches an explicit directive.

When used in conjunction withrecover, all standardASMs support security at recovery time. If a file issaved with an access control list (ACL), then only the owner of the file, root or Administrator may recoverthe file. For files that do not contain an ACL, the standard mode bits are used to determine who mayrecover a file. The file’s owner, root and Administrator may always recover the file. Note that whenASMsare invoked by hand, these security checking rules do not apply.

OPTIONSAll ASMs accept the options described in this section. These options are generally referred to as thestan-dard-asm-arguments. ASMs may also have additional options, which must be capital letters.

Either −s (saving),−r (recovering) or-c (comparing) mode must be specified and must precede any otheroptions. When saving, at least onepathargument must be specified.Pathcan be either a directory or filename.

The following options are valid for all modes:

−n Performs a dry run. When backing up, browse the file system, create the save stream, but do notattempt to open any files. When recovering or comparing, consume the input save stream and per-form basic sanity checks, but do not create any directories or files when recovering or comparingfile data.

−u This option makes the ASM stop when an error that would normally cause a warning occurs. Thiscan be useful if you are recovering to a file system that may not have enough disk space or you areperforming a save and you want any warnings to stop the save. If you use this option withuasmon recovery, it will stop if it runs out of disk space. Without this option,uasmwill continue to tryto recover each file until it has processed the entire save stream.

−v Turns on verbose mode. The currentASM, its arguments, and the file being processed are dis-played. When a filteringASM operating in filtering mode (processing the save stream of anotherASM) modifies the stream, its name, arguments and the current file are displayed within squarebrackets.

When saving, the following options may also be used:

−b Produces a byte count. This option is similar to the−n option, but byte count mode will estimatethe amount of data that would be produced instead of actually reading file data. (It is faster butless accurate than the−n option.) Byte count mode produces three numbers: the number ofrecords (for example, files and directories), the number of bytes of header information, and theapproximate number of bytes of file data. Byte count mode does not produce a save stream; itsoutput cannot be used as input to another ASM in recover mode.

−e Do not generate the final "end of save stream" boolean string. This flag should only be used whenan ASM invokes an externalASM and as an optimization chooses not to consume the generatedsave stream itself.

−f protoSpecifies the location of a.nsr directive file to interpret before processing any files (seensr(5)).Within the directive file specified byproto, <<path>> directives must resolve to files within thedirectory tree being processed, otherwise their subsequent directives will be ignored.

−i Ignores all save directives from.nsrdirective files found in the directory tree.

−o Produces a (seensr_data(5)) save stream that can be handled by older NetWorker servers.


UASM(1m) UASM(1m)

−p pathThis string is prepended to the name of each file as it is output. This argument is used internallywhen one ASM executes another external ASM.Ppathmust be a properly formatted path that iseither the current working directory or a trailing component of the current working directory.

−t date The date (innsr_getdate(3) format) after which files were modified will be backed up.

−x Cross filesystem boundaries. Normally, filesystem boundaries are not crossed during walking.Symbolic links are never followed, except in the case ofrawasm. 1.0v

When recovering, the following options may also be used:

−i {nNyYrR}Specifies the initial default overwrite response. Only one letter can be used. When the name ofthe file being recovered conflicts with an existing file, the user is prompted for overwrite permis-sion. The default response, selected by pressing [Return], is displayed within square brackets.Unless otherwise specified with the−i option, "n" is the initial default overwrite response. Eachtime a response other than the default is selected, the new response becomes the default. WheneitherN, R, or Y is specified, there is no prompting (except when auto-renaming files that alreadyend with the rename suffix) and each subsequent conflict is resolved as if the corresponding lowercase letter had been selected.

The valid overwrite responses and their meanings are:

n Do not recover the current file.

N Do not recover any files with conflicting names.

y Overwrite the existing file with the recovered file.

Y Overwrite files with conflicting names.

r Rename the conflicting file. A dot, “.”, and a suffix are appended to the name ofthe recovered file. If a conflict still exists, the user will be prompted again.

R Automatically renames conflicting files by appending a dot, (“.”), and a suffix.If a conflicting file name already ends in a “.” suffix, the user is prompted toavoid potential auto rename looping condition.

−m src=dstThis option maps the file names that are created. Any files that start exactly withsrc will bemapped to have the path ofdst, replacing the leadingsrc component of the path name. This optionis useful for the relocation of recovered files that were backed up using absolute pathnames into analternate directory (for example,−m /usr/etc=.).

−z suffixSpecifies the suffix to append when renaming conflicting files. The default suffix is “R”.

path Used to restrict the files being recovered. Only files with prefixes matchingpath will be recov-ered. This checking is performed before any potential name mapping is done using the−m speci-fication. Whenpath is not specified, no checking is done.

CAVEATSRaw partitions are often used to store active DBMS data. If your raw partition contains data managed andupdated by an active DBMS product,rawasm alone will not give a consistent backup. The database mustnot be updating the data in an uncontrolled fashion whilerawasm saves or recovers data on the partition.The partition must be offline, the database manager shutdown, or the partition placed in an appropriate statefor backup. Legato has products to assist with online database backup. Similarly ifrawasm is used to savea partition containing a Unix filesystem, the filesystem must be unmounted or mounted read-only to obtaina consistent backup.

Ideally, recovery of a raw partition should take place on a system configured with the same disk environ-ment and same size partitions as the system which performed the backup. If the new partition is smallerthan the original partition, the recovery will not complete successfully. If the new partition is larger than


UASM(1m) UASM(1m)

the original partition, only the amount of data originally saved will be recovered.

If the partition backed up includes the disk label − the label often contains the disk geometry − recoveringthis partition to a new disk also recovers the label, changing the new disks geometry to match the originaldisk. Similarly, if a Unix filesystem partition is backed up usingrawasm, recovering the partition resets allinformation on the partition, including timestamps concerning mount times (if applicable).

Sincerawasm does not discover the size of the partition it backs up until the backup iscompleted, theestimated size reported on recovery is not accurate.

EXAMPLESCopying files

To copy all of the files in the current directory totarget_dir, use:uasm −s . | (cdtarget_dir; uasm −rv)

This preserves ownership, time, and the other Unix attributes. Only the data in holey files iscopied; the holes are not copied.

Copying a file tree to an archive directoryTo copy the file tree under the directoryhere to archiveand overwrite any files with conflictingnames, use:cd hereuasm −s . | (cdarchive; uasm −r −iY)

Change directory (cd) to herefirst and give the firstuasm determining the save a relative path so that theseconduasmperforming the recover will recreate the file tree underarchive.

Another way to achieve the same result is to use the−m option on the seconduasmperforming the recoverto explicitly map the path names.

uasm −shere| uasm −r −iY −mhere=archive

FILES.nsr Save directive files located throughout the filesystem.

SEE ALSOnsr(5), nsr_directive(5), nsrmmdbasm(1m),nsrindexasm(1m),nsrck(1m),nsr_data(5), recover(1m),save(1m),scanner(1m),XDR(3N).


WRITEBUF(1m) WRITEBUF(1m)

NAMEwritebuf − write device buffer

SYNOPSISwritebuf -a b.t.l [ -m mode] [ -b buffer-id] [ -o buffer-offset] [ -p plist-len] -f filename

DESCRIPTIONThe writebuf program will send a WRITE BUFFER command to the named SCSI device. It is typicallyused to download new microcode to SCSI devices.

The required-a argument must be used to select a specific ordinal SCSI address (seelibscsi(1m)).

The required argument set of-f filenamemust refer to a file containing the data to write.

OPTIONS−b buffer-idSpecifies the identity of the buffer to be written.

−o buffer-offsetSpecifies the offset from the beginning of the buffer to begin writing.

−p plist-lenSpecifies the parameter list length.

SEE ALSOThe ANSI SCSI-2 specification for more in-depth explanation of the arguments.

BUGS AND WARNINGSSome microcode files are large. Be sure that the platform that you run this command on, and the hostadapter that attaches the device to which you are directing this command, can support sending all of themicrocode in one command. Loading a fraction of the microcode and failing can be disastrous and candamage a device requiring it to be sent back to the factory. Use thelusbinfo(1m) command and find the I/Otransfer limit size for these constraining factors. Exercise caution using this command.

SEE ALSOlibuscsi(1m)lusbinfo(1m)


NRM(8) NRM(8)

NAMEnrm - Introduction and overview of NetWorker Recovery Manager

DESCRIPTIONNetWorker Recovery Manager facilitates the protection and recovery of network computer systems. Net-Worker Recovery Manager depends upon Networker for file backup. The machine being protected must beregularly backed up using Networker. (seensr(8)).

NetWorker Recovery Manager uses a client-server model to provide the disaster protection service. At leastone machine on the network is designated as the Recovery Manager server and the machines with SCSIdisks to protect are Recovery Manager clients. Currently it protects only SCSI disks attached to SparcSolaris machine.

The Legato NetWorker Recovery Manager Administrator’s Guide provides information on configuring andadministering a Recovery Manager system. It includes examples and rationales for setting up and running asuccessful disaster protection operation.

INSTALLATIONSee the NetWorker Recovery Manager Installation Guide for detailed installation instructions.

SERVER COMMANDSNetWorker Recovery Manager uses a client-server model to provide disaster recovery services. The follow-ing commands encompass the server side of Recovery Manager.

nrmplan (8) - The NetWorker Recovery Manager planning tool. The nrmplan command prepares a machinefor disaster recovery.

nrmgatherdata (8) - The NetWorker Recovery Manager information gathering tool. The nrmgatherdatacommand is run on a regular basis. It protects a computer from disaster.

nrminfo (8) - The NetWorker Recovery Manager information display tool. The nrminfo command displaysinformation about a specified client.

nrmrestore (8) - The NetWorker Recovery Manager restore command. The nrmrestore command is used torecover SCSI disks attached to a Sparc Solaris computer.

CLIENT COMMANDSThe following command encompasses the client side of Recovery Manager.

nrmlcient (8) - The NetWorker Recovery Manager client daemon. During the planning phase, this nrmclientdaemon collects information about disk partitions, filesystems, etc. This information is sent to nrmgather-data, which stores it in the NetWorker Recovery Manager repository.

SEE ALSOnrmplan (8), nrmsbsplan (8), nrmgatherdtata (8), nrminfo (8), nrmrestore (8), nrmclient (8)

NetWorker Recovery Manager 6.1.Build.149a May 04, 2001 1

NRMCLIENT(8) NRMCLIENT(8)

NAMEnrmclient − NetWorker Recovery Manager daemon.

SYNOPSISnrmclient [ −p port ] [ −v ]

DESCRIPTIONThenrmclient daemon runs on both NetWorker Recovery Manager Server and NetWorker Recovery Man-ager client. Various programs on server connect to this daemon and send it requests.

During the planning phase this daemon collects the information about disk partitions, file systems etc. Thisinformation is sent tonrmgatherdata , which stores it in the NetWorker Recovery Manager repository.This information is then used bynrmrestore during the client recovery.

During the recovery process,nrmrestore sends the necessary information to this daemon, which performsthe actual recovery operations like, partitioning the disk, making file systems, installing the boot block etc.

OPTIONS−p port The port number on which the daemon will run.

By defaultnrmclient runs on port 999. Port number can be specified through either command line optionor the /etc/services file. The service called ’nrm’ for tcp is used to get the port number. If the port number isspecified through command line option, the port number in /etc/services is ignored.

−v Turn the verbose mode on.

SEE ALSOnrmrestore(8),nrmgatherdata(8),nrminfo (8),nrmplan (8)


NRMGATHERDAT A(8) NRMGATHERDAT A(8)

NAMEnrmgatherdata − Gathers information about NetWorker Recovery Manager client.

SYNOPSISnrmgatherdata −c client [ −p port ] [ −v ]

nrmgatherdata −l [ −p port ] [ −v ]

DESCRIPTIONThe nrmgatherdata command is run from NetWorker Recovery Manager server on regular basis. It pro-tects the machine from the disaster. It contacts thenrmclient daemon running on the protected machine andcollects the necessary information to automate the recovery.

Thenrmclient daemon collects the information about the partitions, file systems etc, and sends it tonrm-gatherdata . The nrmgatherdata stores this information in the NetWorker Recovery Manager repository.This information is then used bynrmrestore during a client recovery.

OPTIONS−c client

Name of the client which is to be protected from disaster.

−p port The port for client-server communications (default is 999).

−l Use the list of clients from /nrm/hostlist file. This file lists all the clients, one on each line, thatwill be protected from disaster. This option is used to specify multiple hosts.

−v turn the verbose output mode on.

NOTEThe -l and -c options are mutually exclusive.

SEE ALSOnrmplan (8),nrmrestore(8),nrmclient(8),nrminfo(8)


NRMINFO(8) NRMINFO(8)

NAMEnrminfo − Displays the information about a NetWorker Recovery Manager client.

SYNOPSISnrminfo [ −B ] −c client

DESCRIPTIONThe nrminfo command displays the information about the specified client. It displays the informationabout the SCSI disks attached to the client. The information displayed includes which controller the disk isattached to, and disk’s SCSI-ID. It also shows the partitioning information and file systems on each parti-tion.

If the specified client is a Networker server, thenrminfo command can also display the information aboutthe server’s bootstrap save sets.

OPTIONS−B Use this option if the client is a Networker Server. This option displays the information about the

save sets of a server’s bootstrap backups.

−c clientUse this option to specify the machine to display information about.

SEE ALSOnrmgatherdata(8)


NRMPLAN(8) NRMPLAN(8)

NAMEnrmplan − Prepares a NetWorker Recovery Manager client for remote boot procedure.

SYNOPSISnrmplan −c client−n boot_image_dir[ −N Subnet_Boot_Server:/boot_server_dir]

DESCRIPTIONnrmplan prepares the specified machine, for remote boot procedure used during disaster recovery. see (nrmrestore(8)). This command also copies appropriate Networker binaries to the boot image.

This command must be executed from NetWorker Recovery Manager server, once for each client, aftercompleting installation of NRMclnt on NetWorker Recovery Manager client. After executing nrmplan foreach clientnrmgatherdata can be run on a regular basis.

OPTIONS−c client


−n boot_image_dirThe name of the directory containing the installable boot image created bysetup_install_serversee (install_scripts(1M))

−N Subnet_Boot_Server:/boot_server_dirThe complete path of the Subnet Boot Server image directory. Required only if the client is on adifferent subnet than the NetWorker Recovery Manager Server.

SEE ALSOnrmsbsplan(8),nrmgatherdata(8),install_scripts(1M)


NRMRESTORE(8) NRMRESTORE(8)

NAMEnrmrestore − NetWorker Recovery Manager recover command.

SYNOPSISnrmrestore −c client [ −p port ] [ −v ]

DESCRIPTIONnrmrestore is used to recover the SCSI disk(s) attached to Sparc Solaris machine. When the command isrun, it displays a list of all the SCSI disks protected on a client and allows you to choose which disks torecover.

Before recovering a disk, it may be neccessary to replace a damaged disk. Thenrminfo command can beused to display information needed to replace the disk. The replacement disk should be as large as the dam-aged disk. Additionally the replacement disk should be connected to same cotroller and have same SCSI-ID, as the damaged disk.

After replacing the disk, boot the machine using ’boot net’ command from the openboot prompt. Once theboot process is complete, startnrmclient from the command shell. Next, startnrmrestore from the Net-Worker Recovery Manager server and select which disk to recover. On the machine being recovered,answer the questions asked and wait for recovery to finish.

OPTIONS−c client

Name of the client that needs to be recovered from a disaster.

−p port The port for client-server communications (default is 999).

−v Turn the verbose mode on.

SEE ALSOnrmgatherdata(8),nrmclient (8),nrminfo (8)


NRMSBSPLAN(8) NRMSBSPLAN(8)

NAMEnrmsbsplan − Prepares a NetWorker Recovery Manager client for remote boot procedure from a differentsubnet.

SYNOPSISnrmsbsplan −cclient−n boot_server_dir−N NRM_Server:/boot_image_dir

DESCRIPTIONnrmsbsplan prepares the specified machine, for remote boot procedure used during disaster recovery. see (nrmrestore(8)).

This command must be executed from the Solaris Subnet Boot Server, once for each client on that subnet,after completing installation of NRMclnt on NetWorker Recovery Manager client. After executing nrmsbs-plan for each clientnrmgatherdata can be run on a regular basis.

OPTIONS−c client


−n boot_server_dirThe name of the directory containing the boot server image created bysetup_install_server -bsee( install_scripts(1M))

−N NRM_Server:/boot_image_dirThe complete path (along with the NetWorker Recovery Manager Server name) of the directorycontaining the complete boot image created bysetup_install_serversee (install_scripts(1M))

SEE ALSOnrmplan (8),nrmgatherdata(8),install_scripts(1M)


DB2UEXT2 ( 8 ) Maintenance Procedures DB2UEXT2 ( 8 )

NAMEdb2uext2 − User Exit program that archives and retrieves db2 database transactional log files from the Net-Worker server

SYNOPSISdb2uext2 −OSos−RLrelease−RQrequest−DBdbname−NNnodenumber−LPlogpath−LN logname

DESCRIPTIONThedb2uext2program saves and retrieves db2 transactional log files from the NetWorker server.

If a db2 database is enabled for rollforward recovery (by having the USEREXIT database configurationparameter set to ON) db2 database manager invokes thedb2uext2 program whenever a transactional logrequires archiving. It will be saved for the long term storage using NetWorker. During db2 database roll-forward recovery, db2 database manager will invokedb2uext2 to retrieve the logs from the NetWorkerserver. For more information on thedb2uext2 user exit program refer to the DB2 UDB AdministationGuide, Legato NetWorker Module for DB2 UDB Administrator’s Guide and Release Supplement.

OPTIONS−OSos operating system. For example, it can be Solaris or AIX.

−RLreleaseDB2 UDB release. For example, for DB2 UDB V6.1 Enterprise Edition you should specifySQL06010.

−RQrequestrequest. Only ARCHIVE or RETRIEVE are valid options.

−DBdbnamedatabase name. Specify the database name in capital letters.

−NNnodenumbernodegroup. This is the node partition that contains the database. For example, for the first nodeuse NODE0000.

−LPlogpathlog file path. During the retrieval of a transactional log, specify the database log path directory thatthe log was backed up.

−LN lognamelog file name.


0 The transactional log was successfully archived or retrieved. During log retrieval, if the log backupis not found on NetWorker serverdb2uext2returns 0.

12 hardware error16 software error20 invalid parameters28 unknown error32 user terminated

LOGSSet NSR_DB2UEXT2_DEBUG_FILE environment variable to a path of a file. When the user exit isinvoked, debugging information aboutdb2uext2execution will be directed to it. If a file cannot be openedby db2uext2, the default /nsr/nsrdb2uext2.log will be used. If you executedb2uext2manually, you shouldset NSR_DB2UEXT2_DEBUG_FILE in the environment, ifdb2uext2 is called by db2 database manager,NSR_DB2UEXT2_DEBUG_FILE environment variable should be set in db2 vendor configuration file. Formore information refer to your Legato NetWorker Module for DB2 UDB Administrator’s Guide andRelease Supplement.

NetWorker rt_2001_1Q Last change: March 22, 2001 1

DB2UEXT2 ( 8 ) Maintenance Procedures DB2UEXT2 ( 8 )

NetWorker rt_2001_1Q Last change: March 22, 2001 2

NMI_CONFIG(8) NMI_CONFIG(8)

NAMEnmi_config − Legato NetWorker Module for Informix configuration tool

SYNOPSISnmi_config [ −sserver]

DESCRIPTIONnmi_config is used during the Legato NetWorker Module for Informix installation process to connect to aNetWorker server and create the appropriate resources. Most users need to separate their data by storingtheir database backups in one pool and their transaction log backups in another. This script creates mediapool (nsr_pool(5)) and label template (nsr_label(5)) resources named "DBMIData" and "DBMILogs".

nmi_config usesnsradmin(8) to create the resources. Tw o messages are displayed if this script succeedsin connecting to a NetWorker server and creating the resources. Typical output will look like the following:

# ./nmi_configCreating appropriate resources.Done creating resources.

See the documentation accompanying the product for more information.

OPTIONS−sserver

Connect to the specified NetWorkerserver.

FILESNone.

SEE ALSOnsr(8), nsr_pool(5), nsr_label(5), nsradmin(8).

NetWorker 6.1 May 04, 2001 1

nsrdb2(8) nsrdb2(8)

NAMEnsrdb2 − Legato NetWorker Module for DB2 command

SYNOPSISnsrdb2 <save options>

DESCRIPTIONThis shell script is used bysavegrp(8) to trigger backups on a DB2 database server in conjunction with theuse of db2 backup command, the backup interface provided by IBM.nsrdb2 is normally only run bysavegrp and not run manually. To use this command for scheduled backups, the client resource (seensr_client(5)) should listnsrdb2 in thebackup commandattribute.

If you want to modify this command, use the nsrdb2 file, which is a copy of the original template nsrdb2.sh.So that you always have a reference to the original behavior, do not modify the template file nsrdb2.sh.

There are several configurable variables that must be set in the script to control backups to the NetWorkerserver. For more information refer to the latest version of the Legato NetWorker Module for DB2 Adminis-trator’s Guide or Release Supplement.

OPTIONSRefer to the thesave(8) man page for a description of options supported bynsrdb2.

FILESnsrdb2 A template of this command is provided in nsrdb2.sh.

SEE ALSOmminfo(8), nsr_client(5), nsr_pool(5), nwadmin(8), save(8), savegrp(8).


NSRDB2SV(8) NSRDB2SV(8)

NAMEnsrdb2sv − save DB2 databases to long term storage with NetWorker

SYNOPSISnsrdb2sv <save options>

DESCRIPTIONnsrdb2sv saves DB2 databases to the NetWorker server (seensr(8)). The progress of a nsrdb2sv can bemonitored using the X Window System basednwadmin(8) program or thecurses(3X) basednsrwatch(8)program for other terminal types.

nsrdb2svshould not be evoked manually.nsrdb2svexecutes as part of a scheduled DB2 database backupand is called bynsrdb2 script with appropriate arguments. Environment variables innsrdb2 script and forDB2 instance must be set correctly. For more information refer to the latest version of the Legato Net-Worker Module for DB2 Administrator’s Guide or Release Supplement.

OPTIONSRefer to the thesave(8) man page for a description of options supported bynsrdb2sv.

SEE ALSOnwadmin(8), nsrwatch(8), save(8), savegrp(8), savefs(8), nsrdb2(8), nsr(5), nsr(8), nsr_client(5),nsr_device(5), nsr_group(5), nsr_service(5), nsr_pool(5), nsrd(8), nsrim(8), nsrindexd(8), nsrmm(8),nsrmmd(8)


nsrdbmi(8) nsrdbmi(8)

NAMEnsrdbmi − Legato NetWorker Module for Informix command

SYNOPSISnsrdbmi [ −v ] [ −sserver] [ −c client ] [ −N saveset_name] [ −eexpiration] [ −g group] [ −l level] [path . . . ]

DESCRIPTIONThis shell script is used bysavegrp(8) to trigger backups on an Informix database server in conjunctionwith the use ofON-Bar, the backup interface provided by Informix.nsrdbmi is normally only run bysavegrp and not run manually. To use this command for scheduled backups, the client resource (seensr_client(5)) should listnsrdbmi in thebackup commandattribute.

Users wishing to modify this command should use the template provided in /etc/nsrdbmi.sh. Do not mod-ify the template file so you always have a reference to the original behavior.

There are several configurable variables users can set in the command to control backups to the NetWorkerserver.

Variable: PRECMDDefault value: NONEDescription: This variable can be used to run a command before dbspace and log file backups. If this com-mand does not run successfully (in other words, if it returns a non-zero status),nsrdbmi will exit.

Variable: POSTCMDDefault value: NONEDescription: This variable can be used to run a command after dbspace and log file backups have com-pleted. This command is not run ifonbar returns a non-zero exit code.

Variable: NSR_DAT A_VOLUME_POOLDefault value: DBMIDataDescription: If this variable is set to name a media pool, then dbspace backups will be directed to the givenmedia pool.

Variable: DO_LOGFILE_BACKUPSDefault value: YESDescription: If this variable is set to YES, then logfile backups will be executed after dbspace backups. Ifset to NO, no logfile backups will occur.

Variable: NSR_LOG_VOLUME_POOLDefault value: DBMILogsDescription: If this variable is set to name a media pool, then logfile backups will be directed to the namedmedia pool.

Variable: DO_BOOTFILE_BACKUPSDefault value: YESDescription: If this variable is set to YES, then Informix critical file backups will be executed afterdbspace backups.

Variable: BOOTFILE_LISTDefault value: NONEDescription: This environment variable specifies the file name that lists the Informix critical files tobackup. If not specified and DO_BOOTFILE_BACKUPS is set to YES, a default list of files will be backedup. Note that the default list might not include all the necessary files for XPS.

Variable: NSR_COMPRESSION


nsrdbmi(8) nsrdbmi(8)

Default value: FALSEDescription: This environment variable determines if compression is performed on the data. The defaultvalue is "FALSE" indicating no compression should take place. Set this value to "TRUE" to turn on client-side compression.

Variable: PATHDefault value: $INFORMIXDIR/bin:/bin:/usr/sbinDescription: Set up PATH environment variable. This must be configured to include the path to theonbarexecutable and the NetWorker programmminfo(8).

Variable: INFORMIXDIRDefault value: /usr/informixDescription: Set up INFORMIXDIR environment variable which lists the directory where the InformixRDBMS is installed.

Variable: ONCONFIGDefault value: onconfig.stdDescription: Lists the name of the Informix RDBMS configuration file being used.

Variable: INFORMIXSQLHOSTSDefault value: $INFORMIXDIR/etc/sqlhostsDescription: Describes where thesqlhostsfile is located.

The nsrdbmi command also backs up the emergency boot file and $ONCONFIG file using thesavecom-mand. These files will be backed up to the same pool as the dbspace backups. If you wish to have thesefiles sent to another pool, the shell script should be modified appropriately.

OPTIONSPlease see thesave(8) man page for a description of options supported bynsrdbmi.

Note that each optional path argument is typically of the formINFORMIX:/<IDS_server_name>/<dbspace_name>.

FILES/etc/nsrdbmi.sh A template of this command is provided in /etc.

SEE ALSOmminfo(8), nsr_client(5), nsr_pool(5), nwadmin(8), save(8), savegrp(8).


NSRDOCRC(8) NSRDOCRC(8)

NAMEnsrdocrc − NetWorker Module for Lotus Notes deleted document(s) recovery command

SYNOPSISnsrdocrc

[ −c client ] [ −D debug_level] { −p database} [ −s server] [ −t date] [ −T temporary_directory]

DESCRIPTIONThensrdocrc command restores all deleted data documents in a Notes database since a specified backup ofthe database. The database to be restored must be backed up to a NetWorker server’s online client index.

nsrdocrc works by first recovering the specified version of the database to a temporary directory and thencopying the deleted documents from this database to the existing database. Once the deleted documents arecopied to the existing database, the database recovered to the temporary directory is deleted.

OPTIONS−c client

Specifies the name of the client where the files originated. Use this option when restoring to aclient other than the one from where the files originated.

−D debug_levelSpecifies the level of debugging to use. 1-9 where level 9 is the highest.

−p databaseSpecifies the database name with the full path containing deleted documents to be restored.

−sserverSpecifies the NetWorker server to which the data was backed up.

−t Specifies the date innsr_getdate(3)format to recover data that was backed up previous to themost recent back up.

−T temporary_directorySpecifies the the temporary directory to use. The default directory is /nsr/applogs.

EXAMPLEAn example of the command to recover the deleted data documents since the last full backup of thedatabase to NetWorker server neptune:

nsrdocrc -s neptune -T /tmp −p /usr/lotusdata/mail/jsmith.nsf

SEE ALSOnsrnotesrc(8)

DIAGNOSTICS" nsrdocrc does not load "

The value forLD_LIBRARY_PATH (Solaris) / LIBPATH (AIX) variable in the environmentdoes not point to the libnotes.so file (Solaris) or libnotes_r.a file (AIX).

1

nsrnmo(8) nsrnmo(8)

NAMEnsrnmo − NetWorker Module for Oracle scheduled backup command script

SYNOPSISnsrnmo <save options>

DESCRIPTIONThensrnmo shell script is used by the NetWorker server programsavegrp(8) to trigger scheduled backupson an Oracle8 or Oracle8i database server. Thensrnmo script sets specific environment variables for thescheduled Oracle backup and invokes theOracle Recovery Manager (RMAN), the backup interface pro-vided by Oracle8 and Oracle8i.

The nsrnmo script is usually run by savegrp only. It is not run manually. To use thensrnmo script for ascheduled Oracle backup, you must specifynsrnmo in theBackup commandattribute of the client resource(seensr_client(5)) for the scheduled Oracle backup.

To modify thensrnmo command script, you should copy the template file/etc/nsrnmo.shto thensrnmofile and modify thensrnmo file. Do not modify the template file/etc/nsrnmo.sh.Keep the template fileunchanged, as a reference to the originalnsrnmo script.

For information on how to create a separatensrnmo script file for each Oracle instance, refer to Chapter 5on scheduled Oracle backups in the latestNetWorker Module for Oracle Administrator’s Guide.

There are several environment variables in thensrnmo script that you can customize for a particular sched-uled Oracle backup. The variables ORACLE_HOME and PATH aremandatoryfor each scheduled Oraclebackup. The other variables areoptional,and you can leave them undefined in the script, if desired. Notethat the variables in thensrnmo file are all initially undefined. Please refer to the/etc/nsrnmo.shtemplate,your site-specificnsrnmo script, and the latestNetWorker Module for Oracle Administrator’s Guidefordescriptions of the various environment variables.

OPTIONSPlease see thesave(8) man page for a description of options supported bynsrnmo.

FILES/etc/nsrnmo.sh A template of this command is provided in /etc.

SEE ALSOnsrnmostart(8), mminfo(8), nsr_client(5), nsr_pool(5), nwadmin(8), save(8), savegrp(8).


nsrnmostart(8) nsrnmostart(8)

NAMEnsrnmostart − NetWorker Module for Oracle scheduled backup executable program

SYNOPSISnsrnmostart <backup options>

DESCRIPTIONThensrnmostart executable program is used by thensrnmo script to invoke scheduled backups on an Ora-cle8 or Oracle8i database server. Thensrnmostart program launchesOracle Recovery Manager(RMAN) by passing the appropriate arguments.

Thensrnmo script runs thensrnmostart program. Do not run it manually.nsrnmostart uses environmentvariables set in thensrnmo script and backup options passed bynsrnmo (seensrnmo (8)). If the manda-tory environment variables are not set in thensrnmo script,nsrnmostart returns a nonzero status code. Fordetails on the environment variables to set in thensrnmo script, see Chapter 5 in theNetWorker Module forOracle Administrator’s Guide.

The directory wherensrnmostart is located must be included in the PATH environment variable in thensrnmo script.

The nsrnmostart program writes debugging information into the debug file specified by theNSR_SB_DEBUG_FILE environment variable in thensrnmo script.

Thensrnmostart program runs a pre-command script specified by the PRECMD environment variable--ifthe variable is set in thensrnmo script. If the pre-command does not run successfully,nsrnmostart returnsa nonzero status code tonsrnmo without performing the scheduled Oracle backup.

Thensrnmostart program invokes RMAN for Oracle database backups based on the inputs fromnsrnmoandsavegrp(seensrnmo (8) andsavegrp(8)).

When RMAN execution is finished, thensrnmostart program runs a post-command script specified by thePOSTCMD environment variable--if the variable is set in thensrnmo script. If the post-command does notrun successfullynsrnmostart returns a nonzero status code tonsrnmo. Even if RMAN fails, the post-command is executed.

SEE ALSOnsrnmo(8), save(8), savegrp(8)


NSRNOTESRC(8) NSRNOTESRC(8)

NAMEnsrnotesrc − NetWorker Module for Lotus Notes recover command

SYNOPSISnsrnotesrc [ −qNXZ ] [ −s server] [ −c client ] [ −d destination_path] [ −D debug_level] [ −t date] [ −Ttemporary_directory] { NOTES| filename[ filename[ ... ] ] }nsrnotesrc [ −N ] [ −s server ] { −l number_of_logs} { −d destination_path} [ −c client ] [ −Ddebug_level]

DESCRIPTIONThensrnotesrccommand is used restore Notes data backed up to a NetWorker server’s online client index.The information contained in the client and media indexes is used for restoring data from NetWorker back-ups.

The NetWorker Module for Lotus Notes recover command allows you to identify the scope of restore andrestore to different clients and directories when needed.

OPTIONS−q Specify this option to set verbosity to zero.

−N Specify this option during disaster recovery to skip initialization of Notes API.Please see adminguide for more details on disaster recovery.

−X Specify this option if you do not wish to apply the transaction logs to the recovered file.

−Z Specify this option to assign a new DBIID to the recovered database. Use this option when you arerecovering a database to different directory and you already have a database with same name inDomino server’s data directory.Specify -ZZ, to assign a new DBID as well as a new ReplicaID to the database being recovered.

−sserverSpecifies NetWorker server to which the data was backed up.

−c clientSpecifies name of the client where the files originated. Use this option when restoring files to aclient other than the one where the files originated.

−d destination_pathSpecifies a destination path to restore the data to. If this path is not specified then the data will berestored to the same directory it was backed up from.

−D debug_levelSpecifies level of debugging to use. 1-9 where level 9 is the highest.

−l number_of_logsSpecifies number of transaction logs to be recovered. This option is used only during disasterrecovery. Please see admin guide for more details on disaster recovery.

−t Specifies date innsr_getdate(3)format to restore data that was backed up previous to the mostrecent back up.

−T temporary_directorySpecifies temporary directory to be used. The default is /nsr/applogs.

NOTESSpecify this option to restore all the databases backed up by nsrnotesv command.

EXAMPLEAn example of a recover command for restore two Notes databases from the NetWorker server, mars at aspecified point in time:

nsrnotesrc -s mars -t 01/02/00 /usr/lotusdata/accounting.nsf /usr/lotusdata/payroll.nsf

1

NSRNOTESRC(8) NSRNOTESRC(8)

An example of a recover command to restore a database file from last full backup from the NetWorkerserver neptune to a different directory:

nsrnotesrc -s neptune -d /usr/lotusdata/accounting /usr/lotusdata/payroll97.nsf

SEE ALSOnsrnotesv(8)

2

NSRNOTESV(8) NSRNOTESV(8)

NAMEnsrnotesv − NetWorker Module for Lotus Notes backup command

SYNOPSISnsrnotesv

[ -CIZ ] [ −sserver] [ −a comfort_span] [ −b pool ] [ −c client ] [ −d temporary_directory] [ −Ddebug_level] [ −E exclude_list_file] [ −g group] { −R|-S|-T| filename[ filename[ ... ] ] }

DESCRIPTIONThe nsrnotesvcommand provides effective backups of both online and offline Lotus Notes database andsystem files. This is important as database files, such as the Notes public mailbox, are rarely offline. Lotusfiles with extensions .nsf, .ntf, .njf, .ncf, .box, .id, .dsk, .dic and notes.ini are backed up by nsrnotesv.

nsrnotesvoffers many backup options, such as compression and encryption, which allow the users to iden-tify the scope and type of backup to perform.

To perform scheduled backups use the options identified below, within the backup script, to specify thebackup requirements for each client’s system. The default setting is myArgs=-R, which performs a backupof all Notes data residing in the notes recommended directory. The specification of an incremental or full istaken from the client’s resource on the NetWorker server.

Thensrnote backup script is installed on each client and coordinates scheduled backup processes betweenthe NetWorker server and the NetWorker Module for Lotus Notes.nsrnote calls upon nsrnotesv to performbackup of Notes data.

Use thensrnotesvcommand on a Notes/Domino server or client to perform manual backups of the system.

Backup processes can be monitored from either the NetWorker Administrator or the NetWorker Module forLotus Notes. Upon backup completion NetWorker provides reports in the NetWorker messages log and theNetWorker server’s bootstrap file.

OPTIONS−C Specifies data compression during backup before the data is moved over the network or written to

tape.

Compressing data may speed up the backup process, as long as the Notes/Domino database system is ableto send data to the NetWorker server fast enough to keep the tape drive streaming. Also, data compressionwill increase CPU usage but will reduce the amount of data sent to the NetWorker server.

−I Specifies this option to take an incremental backup.

−Z Specifies data encryption during backup.

−sserverSpecifies the NetWorker server to backup client data to.

−a comfort_spanSpecifies the Comfort span in kilo bytes for incremental backup. When this option is specified,nsrnotesv determines whether to perform a full backup or incremental backup, based on the argu-ment. If 4000 is specified as comfort span, then nsrnotesv will perform full backup if the amountof data changed since the last backup, for this file, is more than 4000KB. Using this optionreduces the number of incremental backups required for recovery.

−b poolSpecifies a particular destination pool for a save set.

1

NSRNOTESV(8) NSRNOTESV(8)

−c clientSpecifies a client name to be used for client indexing.

−d temporary_directorySpecifies the temporary directory to use. The default directory is /nsr/applogs.

−D debug_levelSpecifies the level of debugging to use. 1-9 where level 9 is the highest.

−E exclude_list_fileSpecifies the full path of file containing list of filenames or regular expressions to be excluded dur-ing backup. Each entry in the list should be separated by newline.

−g Specifies a backup group to be used bysavegrpto denote the group used by the NetWorker serverto select a specific media pool.

−R Specifies that the database files being backed up reside in the Lotus Notes recommended directory.Lotus Notes recommended directory is the directory where the file notes.ini is present. nsrnotesvidentifies this directory by going through each component in the PATH environment variable andlooking for the notes.ini file in that path component.

−S Specifies that the database files being backed up reside outside the recommended Lotus Notesdirectory.

−T Specifies that the database files being backed up may reside within or outside of the recommendedLotus Notes directory.

EXAMPLEAn example of a backup command for backing up notes databases to the default NetWorker server, mars:

nsrnotesv -s mars -R

An example of a backup command to perform a backup with compression and encryption of a Lotus Notesdata file to the NetWorker server neptune:

nsrnotesv -s neptune -g mygroup -C −Z /usr/lotusdata/log.nsf

SEE ALSOnsrnotesrc(8)

DIAGNOSTICS" Scheduled backups are not functional"

The value forLD_LIBRARY_PATH (Solaris)or LIBPATH (AIX) environment variable speci-fied in the nsrnote script does not point to the libnotes.so file (Solaris)or the libnotes_r.a (AIX).Another reason could be that the name of the backup command specified, in the client’s resourceon the NetWorker server, is not the same as the name of the backup script nsrnote.

2

NSRSAPSV(8) NSRSAPSV(8)

NAMEnsrsapsv − NetWorker Module for SAP R/3 on Oracle scheduled backup command

SYNOPSISnsrsapsv -ffilename

DESCRIPTIONThensrsapsvcommand provides scheduled backups of SAP R/3 on Oracle databases.

The required -ffilenameoption is used to identify the absolute path and filename to the NetWorker Modulefor SAP R/3 on Oracle scheduled backup configuration file. This file contains information about your SAPR/3 on Oracle environment. For a complete description of the scheduled backup configuration file, see theLegato NetWorker Module for SAP R/3 on Oracle Administrator’s Guide.

The nsrsapsvcommand should not normally be launched from the command line. Rather, this commandshould be entered in the Backup Command attribute field for this client’s Client Resource in the NetWorkerAdministrator program. For more information about configuring your Client Resource to perform sched-uled backups of your SAP R/3 on Oralce database environment, see theLegato NetWorker Module for SAPR/3 on Oracle Administrator’s Guide.

OPTIONS−f filename

Specifies the absolute path and filename of the NetWorker Module for SAP R/3 on Oracle sched-uled backup configuration file. This file contains information that is required for the NetWorkerModule to find the appropriate SAP R/3 on Oracle environment. For a complete description of thescheduled backup configuration file, see theLegato NetWorker Module for SAP R/3 on OracleAdministrator’s Guide.

The default location and filename for this file is /nsr/res/nsrsapsv.res.

SEE ALSONone

1

nsrsyb(8) nsrsyb(8)

NAMEnsrsyb − BusinesSuite Module for Sybase backup command

SYNOPSISnsrsyb <save options>

DESCRIPTIONThis shell script is used bysavegrp(8) to trigger backups on a Sybase SQL Server. This command setsenvironment variables, constructs a database consistency check PRECMD by default, and calls nsrsybsv toperform the backups.

nsrsyb is only run by savegrp and is not run manually. To use this command for scheduled backups, theclient resource (seensr_client(5)) should listnsrsyb in the backup commandattribute, the Sybase SQLServer userid in theusernameattribute, and the SQL Server password in thepasswordattribute.

There are several configurable variables users can set in the command to control backups to the NetWorkerserver.

Variable: USE_CONSISTENCY_CHECKDefault value: TRUEDescription: This variable specifies whether a database consistency check should be constructed as thePRECMD and issued before the scheduled backup. The database consistency check ensures that thedatabase is not corrupt before it is backed up. There are cases where the underlying dbcc commands canincorrectly report errors when the-o ckal option is set (which it is by default for Systems prior to 11.5.)See thensrsybcc(8) man page for more information.

Note that if the consistency check is to be performed, the Sybaseusernameandpasswordspecified in theclient resource must have sufficient permissions to run the dbcc command. Although an account with theSybaseoper role has sufficient permissions for doing backups, it does not have sufficient permissions forperforming a database consistency check.

Variable: DBCCOPTDefault value: NONEDescription: This specifies thensrsybccoptions that are to be used for the database consistency check.The default value for this is none, which means that the defaults from thensrsybcccommand are used. Seethe man page fornsrsybcc(8) for more information on the default consistency check options.

Variable: PRECMDDefault value: NONEDescription: This variable can be used to run a command before database backups. If this command doesnot run successfully (in other words, if it returns a non-zero status),nsrsybwill not backup the database(s).ThePOSTCMDwill be executed, however. If theUSE_CONSISTENCY_CHECKoption is TRUE and thePRECMDis set, a database consistency check will not be performed. IfUSE_CONSISTENCY_CHECKisTRUE and thePRECMDis not set, the a database consistency check will be constructed as thePRECMD.

Variable: POSTCMDDefault value: NONEDescription: This variable can be used to run a command after database backups have completed. Thiscommand is always run, even ifnsrsybsvreturns a non-zero exit code.

Variable: NSR_DAT A_VOLUME_POOLDefault value: NONEDescription: If this variable is set to name a media pool, then database (full) backups will be directed to thegiven media pool. If this is not set, then backups will be directed to pools based on the server policy. If nopolicy is set up at the server, backups will be directed into the default pool.


nsrsyb(8) nsrsyb(8)

Variable: NSR_LOG_VOLUME_POOLDefault value: NONEDescription: If this variable is set to name a media pool, then transaction log (incremental) backups will bedirected to the given media pool. If this is not set, then backups will be directory to pools based on theserver policy. If no policy is set up at the server, backups will be directed into the default pool.

Variable: NSR_COMPRESSIONDefault value: FALSEDescription: This environment variable determines if compression is performed on the data. The defaultvalue isFALSEindicating no compression should take place. Set this value toTRUEto turn on client-sidecompression.

Variable: PATHDefault value:/bin:/usr/sbin:/opt/networker/bin:/usr/bin:/usr/sbin/nsrDescription: Set up PATH environment variable. This must be configured to include the path to the Net-Worker programsnsrsybsv(8), nsrsybcc(8),and mminfo(8).

Variable: SYBASEDefault value: NONEDescription: Set this variable to the directory where the Sybase "interfaces" file exists. This is the top leveldirectory in which Sybase SQL Server was installed. If this is not set, then Sybase will look for the inter-faces file in the user’s home directory. Because nsrsyb is run as root, it is extremely unlikely that theSybase interfaces file will be found in the user’s home directory.

OPTIONSPlease see thesave(8) man page for a description of options supported bynsrsyb. There are a number ofoptions supported bysave(8) that are accepted bynsrsybsv(8) but are not used. See the man page fornsrsybsv(8) for a description of which options are used bynsrsybsv(8).

SEE ALSOmminfo(8), nsr_client(5), nsrsybsv(8), nsrsybcc(8), nsr_pool(5), nwadmin(8), save(8), savegrp(8).


0 Normal exit. This means that the save set was correctly created on the server.1 Abnormal exit. A save set was not correctly created on the server.

MessagesThis is a listing of messages thatnsrsyb can generate. It may also produce any errors thatnsrsybsvandnsrsybcccan generate.

nsrsybsv returned status ofnumbernsrsyb exiting.

This indicates that nsrsybsv returned a non-zero status and that the save set was not correctly cre-ated on the server.


NSRSYBCC(8) NSRSYBCC(8)

NAMEnsrsybcc − consistency check Sybase database(s)

SYNOPSISnsrsybcc[ −hqv? ] [ −o ckdb] [ −o ckdbnoidx] [ −o ckal ] [ −o ckcat] [ −o ckstor] [ −U user-name] [ −Ppassword] [ −g group ] [ −c client-name] [ −s server ] [ −D debug-level] SYBASE:/sql-server-name[/database-name]

DESCRIPTIONnsrsybcc consistency checks Sybase SQL Server databases by running SQL Server "dbcc" consistencychecks on the database.

The database(s) to be checked are specified by the SYBASE:/sql-server-name[/database-name] value. Thedatabase-nameis optional; if omitted, all databases on the SQL Server except the tempdb database will bechecked.

The user-nameandpasswordprovide the information necessary to log in to the SQL Server to be backedup. The user-name must have sufficient permissions to run the dbcc command on the databases to bechecked. If the user-name and password are not supplied, the SQL Server name and database name,client-name, group,andserverare used to query the client resource defined on the server for the user-name andpassword.

If no checking options are supplied, the default for Sybase Adaptive Server 11.5 and later is the-o ckstoroption. For earlier versions of Sybase SQL Server, the default is a combination of the-o ckdbnoidx, -ockal, and-o ckcatoptions.

The -o ckstoroption is available only with Sybase Adaptive Server version 11.5 or later. The-o ckstoroption is preferable to the other options because it provides higher concurrency. The-o ckal option mayoccasionally cause errors 2540 or 2546 to occur when no real error conditions exist. Running SQL Serverin single user mode will prevent these condition from being incorrectly reported. See the SQL Server(TM)Troubleshooting and Error Messages Guide for more information on interpreting these and other consis-tency check errors.

The results of the consistency check are written to stdout. There may be a large quantity of informationalmessages, and these informational messages may be suppressed by using the−q flag.

OPTIONS−h Display usage.

−v Verbose. Cause thensrsybccprogram to tell in great detail what it is doing as it proceeds.

−q Quiet. Display only error messages.

−o ckdbThis performs a dbcc checkdb check on the given database(s). This checks index and data pages,the sorting of indices, pointers, and data rows for each table in the database.

−o ckdbnoidxThis performs a dbcc checkdb check on the given database(s) with the skip non clustered indexoption. Non clustered indices may be easily (although not necessarily speedily) recreated, and thisoption only checks clustered indices. It causes a faster consistency check at the cost of possiblegreater recovery time if one of the non clustered indices were corrupt.

−o ckal This performs a dbcc checkalloc check on the given database(s). The checkalloc option checks thepage allocations of the database for inconsistency.

−o ckcatThis performs a dbcc checkcatalog check on the given database(s). This checks the system tablesin the database for consistency.

−o ckstorThis performs a dbcc checkstorage check on the given database(s). This option is available forSybase Adaptive Server version 11.5 and later. It performs a faster consistency check than the



other options, and it has greater concurrency capabilities than the−o ckal option.

−U user-nameSpecify the SQL Server user that will check the databases.

−P passwordSpecifies the password for the SQL Server user that will perform the consistency checks.


−D debug-levelSet the NSR_DEBUG_LEVEL for this command. Valid values range from 0 (no information) to10 (much information.) The default value is 2.

−c client-nameSpecify the client name for querying the user-name and password. This is useful on clients withmultiple network interfaces, and hence multiple host names. It can be used to create multipleindex databases for the same physical client. Note that this does not specify the network interfaceto use. This is specified in theserver network interface attribute of the client resource (seensr_client(5)).

−g groupThis option is used bysavegrp(8) andsavefs(8) to denote the group to query for the username andpassword (seensr_client(5) andnsr_group(5)).

EXAMPLESConsistency checking a database:

To consistency check database "accounting" on SQL Server "production", enter the followingcommand:

nsrsybcc −U sa −P password SYBASE:productionaccounting

This will issue the default consistency checks, which will be a-o ckal, -o ckdbnoidx, and-o ckcatfor Sybase SQL Servers prior to 11.5. For 11.5 systems, this will perform a-o ckstorcheck.

Consistency checking all database in a SQL Server:To consistency check all databases on SQL Server "production", with the-o ckdband -o ckaloptions enter the following command:

nsrsybcc −U sa −P password −o ckdb −o ckal SYBASE:production

SEE ALSOcurses(3X), nsr_getdate(3), nwadmin(8), nsr(5), nsr(8), nsr_client(5), nsr_device(5), nsr_group(5),nsr_service(5), nsrd(8), nsrim(8), nsrindexd(8), nsrmm(8), nsrmmd(8), nsrsyb(8), nsrsybcc(8), nsrsy-brc(8), nsrwatch(8), recover(8), savefs(8), savegrp(8)


0 Normal exit. This means the consistency check completed without error.1 Abnormal exit. There may be consistency errors in the checked databases.

MessagesThis is a partial listing of common error messages. Refer to the Administrator’s Guide for more completelisting of error messages and how to resolve them.

The context allocation routine failed...Cannot access file/sybase/config/objectid.dat

This message indicates that the $SYBASE environment variable was not set. This environmentvariable needs to be set so thatnsrsybcccan find the Sybase localization files required at runtime.The $SYBASE environment variable should be set to the directory where Sybase was installed(the one where theinterfaces file is.) This error may also appear if the Sybase Open Client



software is not installed. Sybase Open Client must be installed fornsrsybccto run.

Error from server: Msg number, Lev elnumber, StatenumberError text

This message indicates that there was an error message returned from Sybase. Refer to the SybaseTroubleshooting and Errors Guide for an explanation of the error and its resolution.


NSRSYBRC(8) NSRSYBRC(8)

NAMEnsrsybrc − recover Sybase database(s) from NetWorker.

SYNOPSISnsrsybrc [ −hkqv? ] [ −U user-name] [ −P password] [ −d SYBASE:/sql-server-name[ /database-name ]] [ −s server ] [ −c client-name ] [ −t date ] [ −D debug-level] { SYBASE:/sql-server-name |SYBASE:/sql-server-name/database-name[ SYBASE:/sql-server-name/database-name ... ] }

DESCRIPTIONnsrsybrc restores Sybase SQL Server 11.0 and later databases from NetWorker. The progress of a nsrsybrccan be monitored using the X Window System basednwadmin(8) program or thecurses(3X) basednsr-watch(8) program for other terminal types.

The database(s) to be restored are specified by the SYBASE:/sql-server-name[/database-name] value. Thedatabase-nameis optional; if omitted, all databases on the SQL Server will be restored except the masterdatabase.

This command should be issued from the same operating system userid that launched the Sybase BackupServer. That is the userid that owns the backups, and it should be used to restore them. Other operatingsystem userids may be unable to see the savesets that the Sybase Backup Server created.

Note that the master database must be restored before the other system and user databases. The order forrestoring a complete SQL Server system is to first restore the master database and then to restore theremainder of the databases. In order to restore the master database, you must restart the SQL Server inmaster recovery mode (the -m option of startserver.) Using nsrsybrc, you would first issue the command:nsrsybrc -U sa -P password SYBASE:/sql-server-name/master Once this successfully restores the masterdatabase, the SQL Server will automatically shut itself down. You will need to restart the SQL server andrestore the rest of the databases with thensrsybrc -U sa -P password SYBASE:/sql-server-name command. This will restore the rest of the systemdatabases and the user databases.

Exercise extreme caution when restoring the master database. If you are restoring it to a new database,remember that it has path names to the device files. If you restore the master database to another SQLServer on the same machine, the new SQL Server will attempt to use the same device files as the old SQLServer.

Theuser-nameandpasswordprovide the information necessary to log in to the SQL Server to be restored.The user-name must have sufficient permissions to perform database loads on the databases to be restored.

OPTIONS−k Perform a database consistency check on each database once it is restored. The results of the

database consistency check are written to stdout. The default is not to perform a database consis-tency check, and if this option is supplied, the consistency check is done with the default optionsdescribed in the man page fornsrsybcc(8).

−h Display usage.

−v Verbose. Cause thensrsybrc program to tell you in great detail what it is doing as it proceeds.

−? Display usage.


−U user-nameSpecify the SQL Server user that will perform the backups.

−P passwordSpecifies the password for the SQL Server user that will perform the backups.

−d SYBASE:/sql-server-name[ /database-name ]Specify a new destination for this database. If this option is not set, the database is restored to itsoriginal SQL Server and database name. If this option is specified, the database is restored to thisnew destination. This can be used to restore a database to a new SQL Server and database name or



a new database name in the same SQL Server. If a new destination is supplied, only one databasename (or sql-server-name) can be specified on the command line for this restore. The destinationdatabase(s) must already exist; nsrsybrc does not create them automatically. For the fastest load-ing, new databases created to be restored to should be created with the "for load" option.


−c client-nameSpecify the client name for starting the restore session. This is useful on clients with multiple net-work interfaces, and hence multiple host names. It can be used to use multiple index databases forthe same physical client. Note that this does not specify the network interface to use. This isspecified in theserver network interfaceattribute of the client resource (seensr_client(5)).

−D debug-levelSet the NSR_DEBUG_LEVEL for this command. Valid values range from 0 (no information) to10 (much information.) The default value is 2.

−t date This specifies the date and time to restore to. The default for this is the current date and time. Ifthe SQL Server is Sybase Adaptive Server version 11.5 or later, and the database supports transac-tion log backups, this time will be used as the "until_time" that the database will be restored to. InSybase SQL Server 11.x Systems earlier than 11.5, the database will be restored to the latest timepossible that is not after the date supplied.

EXAMPLESRestoring a database:

To restore the model database in SQL Server "production" to the current time from the NetWorkerServer called "networker", use the following command:

nsrsybrc −U sa −P password −s networker SYBASE:productionmodel

Restoring the master database:To restore the master database in server "production" to the current time, restart "production" inmaster recover mode (with the "-m" option) and issue the command

nsrsybrc −U sa −P password −s networker SYBASE:productionmaster

Once the master database is restored, the SQL Server will shut itself down. To restore the rest ofthe databases, restart the SQL Server and issue the command

nsrsybrc −U sa −P password −s networker SYBASE:production

This will restore the rest of the databases on the production SQL Server.

Restoring a database to a new server:To restore a database to a new server, use the "-d" option to specify the new server name. Torestore the database "accounting" from "production" to "test", issue the command

nsrsybrc −U sa −P password −s networker −d SYBASE:testaccountingSYBASE:productionaccounting

The destination database must already exist in the server, and the user name and password are onesthat apply to the test server.

SEE ALSOcurses(3X), nsr_getdate(3), nwadmin(8), nsr(5), nsr(8), nsr_client(5), nsr_device(5), nsr_group(5),nsr_service(5), nsrd(8), nsrim(8), nsrindexd(8), nsrmm(8), nsrmmd(8), nsrsyb(8), nsrsybcc(8),nsrsybsv(8), nsrwatch(8), recover(8), savefs(8), savegrp(8)


0 Normal exit. This means that the database(s) were correctly restored.1 Abnormal exit. The database(s) were not all restored.





This message indicates that the $SYBASE environment variable was not set. This environmentvariable needs to be set so thatnsrsybrc can find the Sybase localization files required at runtime.The $SYBASE environment variable should be set to the directory where Sybase was installed(the one where theinterfacesfile is.) This error may also appear if the Sybase Open Client soft-ware is not installed. Sybase Open Client must be installed fornsrsybrc to run.

Archive API error...Unable to open API library for device...Library path is /sybase/lib/libbms .ext

This message indicates that the symbolic link to the libbms.ext library is missing from the$SYBASE/lib directory. Make sure the link exists. The libbms.ext file to link to is in the samedirectory as the NetWorker binary files.

Archive API error...Attempting to open byte stream device...network error (Severity 5 Number 12): Business Suite Module for Sybase has not been prop-erly enabled.

This message indicates that the server from which you are restoring is not enabled for Sybase.Install an enabler for the Sybase module on the server and retry the operation.

Archive API error...Attempting to open byte stream device...network error (Severity 5 Number 13):client is not a registered client

This message indicates that the client on whichnsrsybrc is being run is not a registered client ofthe server from which the backup is to be restored. Create a client resource for this client on theserver and retry the operation.

Error: no backup was found for databasedatabase-name

This message indicates that there are no backups found for the database requested. Make sure thatthe Operating System user used to restore the backups is the same as the one that created the back-ups. Usensrinfo to see the objectowner field for each backup. The objectowner is the OperatingSystem user that must restore the data.

Error from server: Msg 3108, Level 16, State 1LOAD DAT ABASE must be used in single user mode if trying to restore the Masterdatabase.

In order to restore the master database, the SQL Server needs to be restarted in single user mode.Shut down the SQL Server and restart it with the−m option.

CT_LIBRARY error: ...operation terminated due to disconnectCT_LIBRARY error: ...The connection has been marked dead.

These messages are expected when the master database is recovered. They indicate that the SQLServer has shut down on the successful restoration of the master database. The SQL Server willneed to be restarted in order to restore the user databases or in order to make it available to users.

Not restoring database master. It needs to be loadedseparately before the rest of the instance is loaded.

This message indicates that the entire SQL server is being restored (because nodatabase-name



was specified on the command line), and that the master database will not be restored. The masterdatabase must be restored before all of the other databases, and it will shut down the SQL Serveronce it is restored. Therefore, no other databases can be restored at the same time as master.




NSRSYBSV(8) NSRSYBSV(8)

NAMEnsrsybsv − save Sybase databases to long term storage with NetWorker

SYNOPSISnsrsybsv[ −CEGRThvq? ] [ −U user-name] [ −P password] [ −s server] [ −S stripe-count] [ −c client-name] [ −N SYBASE:/sql-server-name[ /database-name ]] [ −b pool ] [ −g group ] [ −l level ] [ −W width] [ −D debug-level] SYBASE:/sql-server-name[ /database-name ]

DESCRIPTIONnsrsybsv saves Sybase SQL Server databases to the NetWorker server (seensr(8)). The progress of ansrsybsv can be monitored using the X Window System basednwadmin(8) program or thecurses(3X)basednsrwatch(8) program for other terminal types.

The database(s) to be saved are specified by the SYBASE:/sql-server-name[/database-name] value. Thedatabase-nameis optional; if omitted, all databases on the SQL Server will be saved except the tempdbdatabase.

The user-nameandpasswordprovide the information necessary to log in to the SQL Server to be backedup. The user-name must have sufficient permissions to perform database dumps on the databases to besaved. If the user-name and password are not supplied, thename, client-name, group,andserverare usedto query the client resource defined on the server for the user-name and password.

Each database saved is created in its own save stream. If all databases on the SQL Server are saved, therewill be a save stream for each database and a save stream for the SQL Server. This stream of data is sent toa receiving process (seensrd(8)) on the NetWorker server, which will process the data, adding entries tothe on-line index (seensrindexd(8)) for each database, with the data finally ending up on some long termstorage media (seensrmmd(8)).

Thestripe-countoption allows striping each database to multiple save streams.

When command line backups are run, the client index (and bootstrap for the server) are not automaticallysaved. Scheduled backups save these automatically, which ensures that disaster recovery takes the mini-mum time possible. If command line backups are run, thesavegrp(8) command should be used with the"-O" option to save the client index and server bootstrap. This will ensure that recovery from a catastrophicfailure can occur in the most expedient fashion.

Details about handling media are discussed innsrmm(8) andnsr_device(5).

OPTIONS−C Use client-side compression for this save. Compression is not used by default.

−E Use encryption for this save. Encryption is not used by default.

−G Perform a dump with the NO_LOG options. This option is not used for normal backups, but maybe necessary in cases where there is a device failure or full log device.

−R Do not truncate the transaction log for the database after this backup. The transaction log is nor-mally truncated after each backup for all databases. When this option is used, the transaction logis left intact. This is useful for backing up a database when there has been a device failure.

−T Truncate the transaction log but do not perform a backup. This is not normally used, but it can beuseful once the transaction log has filled. This command should be followed by a full backup inorder to ensure the database can be restored.

−h Display usage.

−v Verbose. Cause thensrsybsvprogram to tell you in great detail what it is doing as it proceeds.


−U user-nameSpecify the SQL Server user that will perform the backups.



−P passwordSpecifies the password for the SQL Server user that will perform the backups.


−Sstripe-countSpecify the number of stripes to use in backing up each of the databases in this set. The defaultvalue for the stripe-count is 1.

−c client-nameSpecify the client name for starting the save session. This is useful on clients with multiple net-work interfaces, and hence multiple host names. It can be used to create multiple index databasesfor the same physical client. Note that this does not specify the network interface to use. This isspecified in theserver network interfaceattribute of the client resource (seensr_client(5)).

−N SYBASE:/sql-server-name[ /database-name ]The symbolic name of this save set. By default, the name of the database is used for the name(SYBASE:/sql-server-name/database-name.)

−b poolSpecifies a particular destination pool for the databases and transaction logs.

−g groupThis option is used bysavegrp(8) andsavefs(8) to denote the group of the save (seensr_client(5)andnsr_group(5)) and is used by the NetWorker server to select the specific media pool.

−l level The level of the save. This option is used bysavegrp(8) andsavefs(8) to specify a particular levelfor a scheduled save. Valid levels are "incr" and "full". The default level is "full". A full backupis a backup of the entire database. An incremental backup is a backup of the database’s transac-tion log and is only available when the transaction log is on its own device.


ENVIRONMENT VARIABLES−D debug-level

Set the NSR_DEBUG_LEVEL for this command. Valid values range from 0 (no information) to10 (much information.) The default value is 2.

SYBASEThe SYBASE environment variable must have the path to the Sybase SQL Server interfaces file.This has no default value, but if if is not set, Sybase attempts to find the interfaces file in the user’shome directory.

PRECMDThe PRECMD environment variable contains a command to execute before the database is backedup. If the PRECMD exits with a non zero return code, the database will not be backed up. Thedefault value for the PRECMD is no command if nsrsybsv is executed from the command line. Ifa scheduled backup initiates nsrsybsv, then the default PRECMD is a consistency check of thedatabases to be backed up.

POSTCMDThe POSTCMD environment variable contains a command to execute after the database is backedup. The POSTCMD is always executed, even if the PRECMD returns a result of 1. The defaultvalue for the POSTCMD is no command.

EXAMPLESBacking up a database:

To back up database model in SQL Server "production" to NetWorker server networker, enter thefollowing command:

nsrsybsv −U sa −P password −s networker SYBASE:productionmodel



Backing up an entire SQL Server:To back up all databases (except tempdb) in SQL Server "production" to NetWorker server net-worker, enter the following command:

nsrsybsv −U sa −P password −s networker −l full SYBASE:production

Backing up an entire SQL Server to multiple stripes:To back up all databases (except tempdb) in SQL Server "production" to NetWorker server net-worker using three stripes, enter the following command:

nsrsybsv −U sa −P password −s networker −l full −S3 SYBASE:production

Backing up a transaction log:To back up the transaction log for database "orders" in SQL Server "production" to NetWorkerserver networker, enter the following command:

nsrsybsv −U sa −P password −s networker −l incr SYBASE:productionorders

SEE ALSOcurses(3X), nsr_getdate(3), nwadmin(8), nsr(5), nsr(8), nsr_client(5), nsr_device(5), nsr_group(5),nsr_service(5), nsrd(8), nsrim(8), nsrindexd(8), nsrmm(8), nsrmmd(8), nsrsyb(8), nsrsybcc(8), nsrsy-brc(8), nsrwatch(8), recover(8), savefs(8), savegrp(8)


0 Normal exit. This means that a save set was correctly created on the server.1 Abnormal exit. A save set was not correctly created on the server.


host: saveset level=level, size time countfiles.This message (with the appropriate client host name, saveset name, level, total save set size,elapsed time, and file count) is printed whenevernsrsybsv is run bysavegrp(8) and exits nor-mally.


This message indicates that the $SYBASE environment variable was not set. This environmentvariable needs to be set so thatnsrsybsvcan find the Sybase localization files required at runtime.The $SYBASE environment variable should be set to the directory where Sybase was installed(the one where theinterfacesfile is.) This error may also appear if the Sybase Open Client soft-ware is not installed. Sybase Open Client must be installed fornsrsybsvto run.

Archive API error...Unable to open API library for device...Library path is /sybase/lib/libbms .ext

This message indicates that the symbolic link to the libbms.ext library is missing from the$SYBASE/lib directory. Make sure the link exists. The libbms.ext file to link to is in the samedirectory as the NetWorker binary files.

Archive API error...Attempting to open byte stream device...network error (Severity 5 Number 22): volume pool not found

This message indicates that the backup attempted to write to a volume pool that does not exist onthe NetWorker server. Check the server to see which pools are defined and try to backup to anexisting pool.



Archive API error...Attempting to open byte stream device...network error (Severity 5 Number 12): Business Suite Module for Sybase has not been prop-erly enabled.

This message indicates that the server to which your backups are directed is not enabled forSybase backups. Install an enabler for the Sybase backup product on the server and retry the oper-ation.

Archive API error...Attempting to open byte stream device...network error (Severity 5 Number 13):client is not a registered client

This message indicates that the client on which the backup is being run is not a registered client ofthe server to which the backup is directed. Create a client resource for this client on the server andretry the operation.

Archive API error...Attempting to open byte stream device...network error (Severity 5 Number 13): no groupgroupconfigured on server

This message indicates that the group supplied on the command line is not defined in the server towhich the backup is directed. Check the group name supplied to make sure it exists on the server.nsrsybsv does not require a group name, so the command may be retried without specifying agroup name.












OPTIONSsavegems









1.3.Build.149a May 04, 2001 1











1.3.Build.149a May 04, 2001 2

SAPCLONE(8) SAPCLONE(8)

NAMEsapclone − Clone all or selected SAP R/3 savesets

SYNOPSISsapclone[ −a ] [ −p pool ] [ −n ] [ −sserver] −u user

DESCRIPTIONsapcloneis available for cloning savesets created by the SAP Interface. For instance, you may want toclone your SAP backup data to an "offsite" media pool.

The sapclonecommand is executed by aNetWorkerserver system. If periodic execution is desired, thesapclonecommand can be scheduled by (cron(1m)) or another scheduling mechanism.

OPTIONS−a Clone all SAP saveset, not just those from the last 24 hours.

−n Trial run. Don’t actually initiate the cloning operation.

−p poolClone to this pool instead of the default clone pool.

−sserverOperate against the designated NetWorker server.

SEE ALSOcron(1m),nsr(8), nsr_pool(5), nsradmin(8)

SAP R/3 Backup 1.0.1 $Date: 2000/06/29 13:04:35 $ 1

nsrd(1m) nsrd(1m) · option has no additional effect. on all other platforms, the normal behavior...

Documents