user guide (bms cluster deployment) - huawei cloud...sap hana user guide (bms cluster deployment)...
Post on 09-Apr-2020
13 Views
Preview:
TRANSCRIPT
SAP HANA
User Guide (BMS Cluster Deployment)
Issue 03
Date 2018-02-12
HUAWEI TECHNOLOGIES CO., LTD.
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
i
Copyright © Huawei Technologies Co., Ltd. 2018. All rights reserved.
No part of this document may be reproduced or transmitted in any form or by any means without prior
written consent of Huawei Technologies Co., Ltd.
Trademarks and Permissions
and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.
All other trademarks and trade names mentioned in this document are the property of their respective
holders.
Notice
The purchased products, services and features are stipulated by the contract made between Huawei and
the customer. All or part of the products, services and features described in this document may not be
within the purchase scope or the usage scope. Unless otherwise specified in the contract, all statements,
information, and recommendations in this document are provided "AS IS" without warranties, guarantees or
representations of any kind, either express or implied.
The information in this document is subject to change without notice. Every effort has been made in the
preparation of this document to ensure accuracy of the contents, but all statements, information, and
recommendations in this document do not constitute a warranty of any kind, express or implied.
Huawei Technologies Co., Ltd.
SAP HANA
User Guide (BMS Cluster Deployment) Contents
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
ii
Contents
1 Overview ......................................................................................................................................... 1
1.1 Common Concepts ........................................................................................................................................................ 1
1.1.1 Node and Role ........................................................................................................................................................... 1
1.1.2 Scale-up and Scale-out............................................................................................................................................... 3
2 Deployment .................................................................................................................................... 4
2.1 Deployment Modes ....................................................................................................................................................... 4
2.1.1 Cluster Scenario (Single NFS Server Deployed) ....................................................................................................... 4
2.1.2 Cluster Scenario (Two NFS Servers Deployed to Work in Active/Standby Mode) ................................................... 5
2.2 Data Planning................................................................................................................................................................ 7
2.2.1 SAP HANA Nodes ..................................................................................................................................................... 7
2.2.2 Other Nodes ............................................................................................................................................................... 7
2.2.3 Network Planning ...................................................................................................................................................... 8
2.3 Preparing for Resources .............................................................................................................................................. 14
2.3.1 Software and Tools .................................................................................................................................................. 14
2.3.2 License ..................................................................................................................................................................... 15
2.3.3 Creating a VPC ........................................................................................................................................................ 15
2.3.4 Applying for a Subnet and Configuring a Security Group ....................................................................................... 16
2.4 Creating Servers.......................................................................................................................................................... 17
2.4.1 Creating an SAP HANA BMS ................................................................................................................................. 17
2.4.2 Creating an SAP HANA Studio ECS ....................................................................................................................... 19
2.4.3 Creating a NAT Server ............................................................................................................................................. 21
2.4.4 Creating an NFS Server ........................................................................................................................................... 23
2.4.5 Configuring SSH Switching Permissions ................................................................................................................ 26
2.4.6 Configuring an IP Address for a BMS NIC ............................................................................................................. 27
2.4.7 Configuring the Mapping Between SAP HANA Host Name and IP Addresses ...................................................... 34
2.4.8 Modifying OS Configurations ................................................................................................................................. 35
2.5 Installing SAP HANA (Cluster Deployment and Single NFS Server Deployed) ....................................................... 36
2.5.1 Formatting the Disks Attached to an NFS Server .................................................................................................... 36
2.5.2 Formatting an SAP HANA Disk .............................................................................................................................. 37
2.5.3 Installing the SAP HANA Software ......................................................................................................................... 40
2.5.4 Installing the SAP HANA Studio on a Windows ECS ............................................................................................. 45
2.5.5 Installing the SAP HANA Studio on a Linux ECS .................................................................................................. 46
SAP HANA
User Guide (BMS Cluster Deployment) Contents
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
iii
2.5.6 Connecting SAP HANA Nodes to the SAP HANA Studio ...................................................................................... 49
2.5.7 Configuring the Backup Path ................................................................................................................................... 53
2.5.8 Configuring SAP HANA Storage Parameters .......................................................................................................... 55
2.6 Installing SAP HANA (Cluster Deployment and Two NFS Servers Deployed) ......................................................... 56
2.6.1 Configuring NFS Servers to Be Trusty to Each Other ............................................................................................. 56
2.6.2 Formatting the Disks Attached to an NFS Server .................................................................................................... 57
2.6.3 Configuring HA on NFS Servers (SLES 12 SP1) .................................................................................................... 59
2.6.4 Formatting an SAP HANA Disk .............................................................................................................................. 71
2.6.5 Installing the SAP HANA Software ......................................................................................................................... 74
2.6.6 Installing the SAP HANA Studio on a Windows ECS ............................................................................................. 79
2.6.7 Installing the SAP HANA Studio on a Linux ECS .................................................................................................. 80
2.6.8 Connecting SAP HANA Nodes to the SAP HANA Studio ...................................................................................... 83
2.6.9 Configuring the Backup Path ................................................................................................................................... 87
2.6.10 Configuring SAP HANA Storage Parameters ........................................................................................................ 89
A Appendix ..................................................................................................................................... 91
A.1 Logging In to a Linux ECS Using an SSH Key ......................................................................................................... 91
A.2 Obtaining the Password for Logging In to a Windows ECS ...................................................................................... 92
A.3 Querying the NIC IP Address of a BMS .................................................................................................................... 93
A.4 Querying the NIC IP Address of an ECS ................................................................................................................... 93
B Change History ........................................................................................................................... 95
SAP HANA
User Guide (BMS Cluster Deployment) 1 Overview
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
1
1 Overview
1.1 Common Concepts
1.1.1 Node and Role
SAP HANA Node
An SAP HANA system consists of one or multiple SAP HANA nodes.
SAP HANA nodes are logical units that constitute an SAP HANA system. An SAP HANA
node contains the CPU, memory, and storage (such as log, data, shared, and backup disks)
resources with specified specifications, as shown in Figure 1-1.
Figure 1-1 SAP HANA node
SAP HANA Studio
The SAP HANA Studio provides management, monitoring, and information modeling of the
SAP HANA system. It can also function as a client and provides capabilities to access user
data. The information that the SAP HANA Studio provides includes the system information
(such as software version), alarm information (generated by Statistics Server), and statistics of
key system resources.
NAT Server
Provides the capability to switch to the HANA ECS using SSH. It allows you to switch to an
SAP HANA node from the NAT server using Secure Shell (SSH).
SAP HANA
User Guide (BMS Cluster Deployment) 1 Overview
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
2
NFS Server
Provides the file sharing capability. You can bind the shared and backup volumes to the NFS
server and share them with the SAP HANA node.
SFS
Scalable File Service (SFS) provides the file sharing service. Create a file system to provide
the shared volumes, backup volumes, and the shared path to SAP HANA nodes.
SAP HANA Roles
In SAP HANA cluster deployment mode, the roles assigned to SAP HANA nodes are shown
in Figure 1-2.
Figure 1-2 Roles assigned to SAP HANA nodes
Each SAP HANA node has the NameServer and IndexServer processes. Table 1-1 provides
the roles of the processes.
Table 1-1 SAP HANA roles
Process Type Description
NameServer Configured
Role
Initially configured roles
Master: functions as a global transaction coordinator,
which coordinates global transactions and stores the
global metadata of the information about the
computing node cluster. A cluster has three master
nodes configured, but only one of them is activated.
Slave: An SAP HANA cluster can have one or
multiple slave nodes configured in a distributed
architecture to cache metadata and perform database
operations assigned by the master node. A cluster has
multiple slave nodes.
Actual Role Activated roles due to the election mechanism
Master: activated master node elected from the
configured master nodes
Slave: nodes except master nodes
SAP HANA
User Guide (BMS Cluster Deployment) 1 Overview
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
3
Process Type Description
IndexServer Configured
Role
Initially configured roles
Worker: nodes that are running and performing
database operations
Standby: takes over services of a faulty node. A
cluster can have no or any number of standby nodes,
and one standby node by default. In normal cases,
software on the node is running, but the node's
memory database has no data and cannot process
services. The standby node takes over services of a
faulty node automatically.
Actual Role Activated roles due to the election mechanism
Master: node elected from worker nodes. It is the
same as the master node in Name Server.
Slave: worker nodes except the master nodes
Standby: same as parameter Config Role in Index
Server. Any node can be the standby node due to
multiple times of service switchovers.
1.1.2 Scale-up and Scale-out
SAP HANA nodes can be expanded in scale-up or scale-out mode.
Scale-up mode
Also called single-node system mode, in which an SAP HANA system contains only one
valid node. If high availability (HA) is required, construct such architecture using two
single nodes through system replication. This system architecture supports only scale-up
expansion. In this mode, add CPUs, memory capacity, and hard disks to the node.
The system does not support expansion for a node on which SAP HANA is running.
Scale-out mode
Also called cluster system mode, in which an SAP HANA system contains multiple
nodes. In scale-out mode, when the system requires expansion, add more nodes to the
system.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
4
2 Deployment
2.1 Deployment Modes
2.1.1 Cluster Scenario (Single NFS Server Deployed)
Figure 2-1 shows the cluster scenario where a single NFS server is deployed.
Figure 2-1 Cluster scenario where a single NFS server is deployed
The concepts involved in the preceding figure are as follows:
VPC network: All SAP HANA nodes are deployed within a VPC, and all ECSs and
HANA ECSs in the SAP HANA system belong to the same AZ to ensure network
security.
Public subnet:
− Network Address Translation (NAT) server: allows you to access SAP HANA ECSs
using Secure Shell (SSH).
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
5
− SAP HANA Studio ECS: runs the SAP HANA Studio software. You can use RDP or
SSH to access the SAP HANA Studio ECS and manage the SAP HANA system.
Private subnet:
SAP HANA server: used for deploying the SAP HANA software. The server has the
following disks attached:
− OS disk: provides the directory for installing the OS.
− Data volume: periodically stores the data transmitted from the SAP HANA IMDB
(a database running in high-performance memory). The period is 5 minutes by
default.
− Log volume: stores the data triggered by an event. When an event, for example, a
record or a batch of records are updated, is triggered for the server IMDB, the
system will write the latest IMDB data into the log volume.
− Shared volume: stores the SAP HANA installation software and SAP HANA
database log files.
− Backup volume: stores the SAP HANA database backup files.
Cluster scenario: Three to eight servers are required to constitute a cluster. Among these
servers, one functions as the STANDBY node, and the others are WORKER nodes.
Network File System (NFS) server: One HANA ECS is required to provide the file
sharing service. The shared and backup volumes are provided by the NFS server for
sharing them with SAP HANA ECSs.
Figure 2-2 shows the deployment flowchart in the cluster scenario where one NFS server is
required.
Figure 2-2 Deployment flowchart in the cluster scenario where one NFS server is required
2.1.2 Cluster Scenario (Two NFS Servers Deployed to Work in Active/Standby Mode)
Figure 2-3 shows the cluster scenario where two NFS servers are deployed to work in
active/standby mode.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
6
Figure 2-3 Cluster scenario where two NFS servers are deployed to work in active/standby mode
The concepts involved in this scenario are the same as those involved in 2.1.1 Cluster
Scenario (Single NFS Server Deployed) in addition to the following remarks:
The two NFS servers constitute a cluster, which provides the file sharing service for the SAP
HANA cluster.
Figure 2-4 shows the deployment flowchart in the cluster scenario where two NFS servers are
deployed to work in active/standby mode.
Figure 2-4 Deployment flowchart in the cluster scenario where two NFS servers are deployed to
work in active/standby mode
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
7
2.2 Data Planning
2.2.1 SAP HANA Nodes
SAP poses specific requirements on SAP HANA nodes depending on application scenarios.
The time zone of the server on which SAP NetWeaver is deployed must be the same as that of SAP
HANA nodes, excepting the specifications described in this section.
Specifications in BWoH Scenario
In BWoH scenario, SAP HANA works with SAP Business Warehouse. In this scenario, SAP
HANA provides OLAP functions and supports rapid computing and analyzing on massive
data. The SAP HANA processing performance and the network bandwidths between SAP
HANA nodes are the key concern.
Table 2-1 lists the requirements on BMSs in BWoH scenario.
The following volume sizes are recommended. For more configuration information, see the official SAP
documents.
Table 2-1 BMS specifications in BWoH scenario
CPU Number and Model Memory Disk
4 x Broadwell EX 8890 v4 2 TB Local disks:
OS disk: 2 x 600 GB SAS disk
EVS disks:
Data volume: 6 TB, ultra-high I/O,
shared disk
Log volume: 512 GB, ultra-high I/O,
shared disk
Table 2-2 BMS OS specifications
Item Specifications
OS SUSE Linux Enterprise Server 12 SP1 for SAP
SUSE Linux Enterprise Server 12 SP2 for SAP
2.2.2 Other Nodes
Other nodes include the NAT server, SAP HANA Studio, and NFS server nodes. Table 2-3
lists the requirements on these nodes.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
8
Table 2-3 Data planning for other nodes
Node Specifications
SAP HANA Studio OS:
NOTE
Based on service requirements, use a Windows or Linux ECS to deploy the
SAP HANA Studio.
− Windows: Windows Server 2008
− Linux: SUSE Linux Enterprise Server 12 SP1
Flavor: s1.xlarge (4 vCPUs and 16 GB memory capacity)
Disk: 80 GB system disk
NAT server OS:
SUSE Linux Enterprise Server (SLES) 12 SP1
Flavor: s1.medium (1 vCPU and 4 GB memory capacity) or
higher
Disk: 40 GB system disk
NFS server The NFS server is required only in cluster scenarios.
OS:
SUSE Linux Enterprise Server 12 SP1 for SAP
Flavor: s1.xlarge (4 vCPUs and 16 GB memory capacity)
Disk:
NOTE
If two NFS servers are required to work in active/standby mode, each type of
the shared, backup, and SBD volumes requires one for the two NFS servers.
− System disk: 100 GB non-shared disk
− Shared volume: ultra-high I/O, SCSI, shared disk.
Number of worker nodes/4 x Memory size. The value of
number of worker nodes/4 is rounded up for decimal places.
− Backup volume: 6144 GB, SCSI, shared disk, ultra-high I/O
− SBD volume, 10 GB, high I/O, shared disk This volume is
required only when NFS servers are deployed to work in
cluster or active/standby mode.
2.2.3 Network Planning
Network Plane Planning in the Cluster Scenario Where a Single NFS Server Is Deployed
Figure 2-5 shows the network plane planning in the cluster scenario where a single NFS
server is deployed.
The network segments and IP addresses are for reference only.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
9
Figure 2-5 Network plane planning in the cluster scenario where a single NFS server is deployed
Table 2-4 lists the entire network planning.
Table 2-4 Network planning in the cluster scenario where a single NFS server is deployed
Parameter Description Example Value
IP address of the
server/client plane
Specifies the IP address of the
primary NIC plane. Allows an
SAP HANA node to
communicate with service
software (such as ERP) or SAP
HANA Studio client software.
NFS Server uses the primary NIC
plane to synchronize time with
the NTP server.
SAP HANA node: 10.0.3.2
to 10.0.3.6 (taking five SAP
HANA nodes as an
example)
NFS server: 10.0.3.101
NAT server: 10.0.0.202
EIP Allows you to access the SAP
HANA Studio and NAT server.
Automatically allocated
IP address of the
internal communication
plane
Allows SAP HANA nodes, each
of which is equipped with a
single-root I/O virtualization
(SR-IOV) NIC, to exchange data.
SAP HANA node: 10.0.2.2
to 10.0.2.6 (taking five SAP
HANA nodes as an
example)
IP address of the cloud
management or backup
plane
Allows the shared and backup
volumes on the NFS server to be
attached to all HANA ECSs.
SAP HANA node: 10.0.1.2
to 10.0.1.6 (taking five SAP
HANA nodes as an
example)
NFS server: 10.0.1.101
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
10
Network Plane Planning in the Cluster Scenario Where Two NFS Servers Are Deployed to Work in Active/Standby Mode
Figure 2-6 shows the network plane planning in the cluster scenario where two NFS servers
are deployed to work in active/standby mode.
The network segments and IP addresses are for reference only.
Figure 2-6 Network plane planning in the cluster scenario where two NFS servers are deployed to
work in active/standby mode
Figure 2-6 lists the entire network planning.
Table 2-5 Network planning in the cluster scenario where two NFS servers are deployed to work
in active/standby mode
Parameter Description Example Value
IP address of the
server/client plane
Specifies the IP address of the
primary NIC plane. Allows an
SAP HANA node to
communicate with service
software (such as ERP) or SAP
SAP HANA node: 10.0.3.2
to 10.0.3.6 (taking five SAP
HANA nodes as an
example)
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
11
Parameter Description Example Value
HANA Studio client software.
NFS Server uses the primary
NIC plane to synchronize time
with the NTP server.
NFS server 1: 10.0.3.101
NFS server 2: 10.0.3.102
NAT server: 10.0.0.202
EIP Allows you to access the SAP
HANA Studio, NAT Server,
NFS Server, and SAP HANA
node.
Automatically allocated
IP address of the internal
communication plane
Allows SAP HANA nodes,
each of which is equipped with
an SR-IOV NIC, to exchange
data.
SAP HANA node: 10.0.2.2
to 10.0.2.6 (taking five SAP
HANA nodes as an
example)
NFS server 1: 10.0.2.101
NFS server 2: 10.0.2.102
IP address of the cloud
management or backup
plane
Allows the shared and backup
volumes on the NFS server to
be attached to all HANA ECSs.
SAP HANA node: 10.0.1.2
to 10.0.1.6 (taking five SAP
HANA nodes as an
example)
NFS server 1: 10.0.1.101
NFS server 2: 10.0.1.102
NFS server floating IP
address: 10.0.1.103
Security Group Rules
The network segments and IP addresses are for reference only. The following security group rules
are recommended practices. You can configure your own security group rules as you need.
In the following table, ## stands for the SAP HANA instance ID, such as 00. Ensure that this ID is
the same as the instance ID specified when you install the SAP HANA software.
For more information about specific ports and security group rules to be accessed by SAP, see
https://help.sap.com/viewer/575a9f0e56f34c6e8138439eefc32b16/2.0/en-US/616a3c0b1cc748238de
9c0341b15c63c.html.
Table 2-6 Security group rules (SAP HANA)
Source Protocol Port Range Description
Inbound
10.0.0.0/24 TCP 5##13 to 5##14 Allows the SAP HANA Studio
to access SAP HANA.
10.0.0.0/24 TCP 3##15 Provides ports for the service
plane.
10.0.0.0/24 TCP 3##17 Provides ports for the service
plane.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
12
Source Protocol Port Range Description
10.0.2.0/24 TCP 3##00 to 3##10 Provides ports for internal
communication between SAP
HANA nodes.
10.0.0.0/24 TCP 22 Allows SAP HANA to be
accessed using SSH.
10.0.2.0/24 TCP 22 Allows NAT servers to be
accessed using SSH.
10.0.0.0/24 TCP 43## Allows access to XS Engine
from the 10.0.0.0/24 subnet
using HTTPS.
10.0.0.0/24 TCP 80## Allows access to XS Engine
from the 10.0.0.0/24 subnet
using HTTP.
10.0.0.0/24 TCP 8080 (HTTP) Allows Software Update
Manager (SUM) to access SAP
HANA using HTTP.
10.0.0.0/24 TCP 8443 (HTTPS) Allows Software Update
Manager (SUM) to access SAP
HANA using HTTPS.
10.0.0.0/24 TCP 1128-1129 Allows access to SAP Host
Agent using SOAP/HTTP.
Automatically
specified by the
system
ANY ANY Security group rule created by
the system by default
It enables ECSs in the same
security group to communicate
with each other.
Outbound
ANY ANY ANY Security group rule created by
the system by default
Allows all peers to access SAP
HANA.
Table 2-7 Security group rules (SAP HANA Studio)
Source Protocol Port Range Description
Inbound
0.0.0.0/0 TCP 3389 Allows users to access the SAP
HANA Studio using RDP.
This rule is required only when
the SAP HANA Studio is deployed on a Windows ECS.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
13
Source Protocol Port Range Description
0.0.0.0/0 TCP 22 Allows users to access the SAP
HANA Studio using SSH.
This rule is required only when
the SAP HANA Studio is
deployed on a Linux ECS.
Automatically
specified by the
system
ANY ANY Security group rule created by
the system by default
It enables ECSs in the same
security group to communicate
with each other.
Outbound
ANY ANY ANY Security group rule created by
the system by default
Allows all peers to access the
SAP HANA Studio.
Table 2-8 Security group rules (NAT server)
Source Protocol Port Range Description
Inbound
0.0.0.0/0 TCP 22 Allows users to access the
NAT server using SSH.
10.0.3.0/24 TCP 80 (HTTP) Allows access to instances in
the same VPC using HTTP.
10.0.3.0/24 TCP 443 (HTTPS) Allows access to instances in
the same VPC using HTTPS.
Automatically
specified by the
system
ANY ANY Security group rule created by
the system by default
It enables ECSs in the same
security group to communicate
with each other.
Outbound
10.0.3.0/24 TCP 22 (SSH) Allows the NAT server to
access the 10.0.3.0 subnet
using SSH.
0.0.0.0/0 TCP 80 (HTTP) Allows instances in a VPC to
access any network.
0.0.0.0/0 TCP 443 (HTTPS) Allows instances in a VPC to
access any network.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
14
Table 2-9 Security group rules (NFS server)
Source Protocol Port Range Description
Inbound
10.0.1.0/24 TCP&UDP 111, 2049,
4000~4002
Allows SAP HANA nodes to
access the NFS server using
TCP or UDP.
10.0.0.0/24 TCP 22 Allows NFS servers to be
accessed using SSH.
Automatically
specified by the
system
ANY ANY Security group rule created by
the system by default
It enables ECSs in the same
security group to communicate
with each other.
Outbound
ANY ANY ANY Security group rule created by
the system by default
Allows all peers to access the
NFS server.
2.3 Preparing for Resources
2.3.1 Software and Tools
Table 2-10 lists the software and tools to be obtained.
Download the readme.txt file at https://obs-sap.obs.myhwclouds.com/readme.txt on a local computer
to obtain the location where the software and configuration file are stored.
Table 2-10 Required software and tools
Item Description How to Obtain
Local computer Runs a Windows OS
which is Windows 7 or
later.
N/A
WinSCP Uploads key files to
HANA ECSs.
https://www.winscp.net
PuTTY and PuTTYgen Used for logging in to a
HANA ECS and running
commands.
http://www.chiark.greenend.org.uk/
~sgtatham/putty/download.html
SAP HANA 1.0
SAP HANA 2.0
SAP HANA installation
package. You can visit
For detailed paths where the files
can be downloaded, see the sections
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
15
Item Description How to Obtain
NOTE
Install the required SAP
HANA version based on
version mapping
relationship between SAP
HANA software version
and SUSE OS version on
the SAP official website.
the SAP official website
or access the OBS
bucket to download the
installation package.
describing how to install SAP
HANA.
SAP HANA Studio
NOTE
Install the required SAP
HANA Studio version
based on version mapping.
SAP HANA Studio
installation package.
Configuration script Contains the script and
configuration file used
for installing and
configuring SAP
HANA.
Access the following URLs:
https://obs-sap.obs.myhwclouds.c
om/hana/cfgandscript/scale_out_s
cript.zip and
https://obs-sap.obs.myhwclouds.c
om/hana/cfgandscript/ha_auto_sc
ript.zip.
2.3.2 License
SAP HANA is authorized in Bring Your Own License (BYOL) mode. In this mode, you must
log in to the SAP technical support website and apply for a license.
In addition to applying for a license, you must purchase public cloud-related resources.
2.3.3 Creating a VPC
Scenarios
All ECSs of an SAP HANA system must be in the same VPC. Therefore, you must apply a
VPC for an SAP HANA system and specify the subnet segment for the VPC.
Procedure
Step 1 Log in to the management console.
Step 2 On the homepage, choose Network > Virtual Private Cloud.
Step 3 Click Create VPC on the right of the page.
Step 4 On the Create VPC page, configure VPC parameters.
The parameters are as follows:
Name: specifies the VPC name.
VPC CIDR: specifies the IP address segment for the VPC. The subnet segment in the
VPC must be within this IP address segment. Therefore, you must configure this
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
16
parameter according to the deployment plan described in section 2.2.3 Network Planning.
For example, set this parameter value to 10.0.0.0/8.
AZ: specifies the available zone (AZ) of the VPC subnet.
Subnet Name: specifies the name of the default VPC subnet. You are recommended to
set this parameter value to the name of the service or client plane subnet, for example,
10.0.3.x.
CIDR: specifies the IP address segment for the default VPC subnet. Ensure that the IP
address segment for the default VPC subnet is within the IP address segment for the VPC
subnet. Configure this parameter based on the deployment plan described in section 2.2.3
Network Planning.
Gateway: specifies the gateway IP address of the subnet.
DNS Server Address: The external DNS server address is used by default. If you need
to change the DNS server address, ensure that the DNS server addresses are available.
Step 5 Click Create Now to complete the VPC creation.
----End
2.3.4 Applying for a Subnet and Configuring a Security Group
Scenarios
To ensure proper communication between the servers in an SAP HANA system, apply a
subnet for the ECSs and configure a proper security group.
Procedure
Step 1 Apply for a subnet.
1. Log in to the management console.
2. On the homepage, choose Network > Virtual Private Cloud.
3. In the VPC navigation tree, choose the VPC name and switch to the VPC to which SAP
HANA belongs.
4. Click Subnets on the left and then Create Subnet on the Subnets page in the right pane.
5. In the Create Subnet pane, configure parameters as prompted.
− AZ: specifies the available zone (AZ) of the VPC subnet.
− Name: Configure the subnet name that is easy to identify, for example,
service_subnet.
− CIDR: Configure this parameter according to the deployment plan described in
section 2.2.3 Network Planning.
− Gateway: Use the default setting for this parameter.
6. Click OK to complete the subnet configuration.
7. Repeat Step 1.4 to Step 1.6 to create all required subnets according to the requirements
specified in section 2.2.3 Network Planning.
Step 2 Set security groups.
You need to create a security group for all nodes in the SAP HANA system.
1. Click Security Group on the left and then Create Security Group on the right side of
the page. The Create Security Group page is displayed.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
17
2. Enter the security group name and click OK to complete the security group
configuration.
Name the security group that is easy to identify, for example, studio_security_group.
3. Repeat Step 2.1 to Step 2.2 to create other security groups.
4. In the security group list on the right of the page, expand the security group for which a
rule is to be added.
5. Click Add Rule.
6. On the Add Rule page, add the rule according to the requirements specified in section
2.2.3 Network Planning.
The default security group rules cannot be deleted.
7. Repeat Step 2.4 to Step 2.6 to configure all security groups.
----End
2.4 Creating Servers
2.4.1 Creating an SAP HANA BMS
Scenarios
SAP HANA software runs on BMSs. Depending on deployment scenarios, you need to
allocate one or multiple BMSs for deploying the SAP HANA software.
For details about the number of BMSs to be applied for and their planned data, see sections
2.1 Deployment Modes and 2.2 Data Planning.
Procedure
Step 1 Log in to the management console.
Step 2 On the system homepage, choose Computing > Bare Metal Server to switch to the Bare
Metal Server page.
Step 3 On the right side of the page, click Buy BMS to switch to the Buy BMS page.
Step 4 Configure the parameters listed in Table 2-11.
Table 2-11 BMS parameter description
Parameter Description
Name Specifies the BMS name. The name contains a maximum of 13
characters.
When you allocate BMSs in batches, the Name is generated
automatically in ascending order based on the Quality value that you
filled in. For example, if you fill hana in Name, the first BMS is
hana-0001, and the second BMS is hana-0002.
AZ Specify the AZ in which HANA BMSs locate. The AZ must support
SAP HANA. Choose the AZ as required.
All nodes in the SAP HANA system must be within the same AZ.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
18
Parameter Description
Flavor Set the flavor as required.
Image Select the image based on the BMS OS.
Disk You do not need to create a local disk. To create an EVS disk, click
Add Data Disk. For details, see section 2.2.1 SAP HANA Nodes.
VPC and Security
Group
Use the VPC and security group that are created in section 2.3.4
Applying for a Subnet and Configuring a Security Group.
NIC Determine the number of NICs based on the information provided in
section 2.2.3 Network Planning and click Add NIC to configure the
NICs as required.
You do not need to create a NIC for the internal communication
plane because it will be created in section 2.4.6 Configuring an IP
Address for a BMS NIC using a configuration file.
EIP Do not bind an EIP to the BMS.
Key Pair In this mode, the Secure Shell (SSH) key certificate is used for
authenticating users who attempt to log in to servers. To create a key
pair, click View Key Pair and then click Create Key Pair on the
Key Pair page.
Ensure that the HANA ECSs/ECSs where SAP HANA, SAP HANA
Studio, NFS servers, and NAT servers are to be deployed use the
same key. Otherwise, SAP HANA installation will fail.
NOTE
If you choose an existing SSH key certificate from the drop-down list, make
sure that you have saved the certificate locally. Otherwise, you may fail to log
in to the BMS.
To create a key, do as follows:
Click View Key Pair. On the Key Pair page that is displayed, click Create
Key Pair, specify the key pair name, and click OK. In the Information dialog
box that is displayed, click OK. Then, you can query and save the private key
as prompted.
User Data
Injection
These parameters are displayed only after you click Show behind
Advanced Settings.
Select Text and enter the following content in the text box to assign
permissions to user root so that the user can log in to a BMS using a
private key through SSH:
#cloud-config
disable_root: false
runcmd:
- sed -i 's/^PermitRootLogin.*$/PermitRootLogin
without-password/' /etc/ssh/sshd_config
- sed -i '/^KexAlgorithms.*$/d' /etc/ssh/sshd_config
- service sshd restart
Quantity Set this parameter as required.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
19
Step 5 Click Buy Now. On the page that is displayed, click Submit Application.
Step 6 The system returns to the Bare Metal Server page. Check the status of the created task in
Task Status on the right of the page.
After the BMS is allocated, you can view the BMS from the BMS list on the right of the page.
Step 7 Allocate other BMSs as required.
Step 8 Change the root password for logging in to all BMSs.
Properly keep the root password. In addition, ensure that all ECSs and BMSs use the same
root password.
1. Log in to a BMS.
2. Run the following command to switch to user root:
sudo su -
3. Run the following command to change the password for user root:
passwd
Enter the password as prompted for confirmation.
----End
2.4.2 Creating an SAP HANA Studio ECS
Scenarios
An SAP HANA system requires an ECS for deploying the SAP HANA Studio software.
Procedure
Step 1 Log in to the public cloud management console and choose Computing > Elastic Cloud
Server to switch to the Elastic Cloud Server page.
Step 2 On the right side of the page, click Buy ECS to switch to the Create ECS page.
Step 3 Configure the parameters listed in Table 2-12.
Table 2-12 ECS parameters
Parameter Description
ECS Name Specifies the ECS name. The name contains a maximum
of 13 characters.
AZ Specifies the AZ where the ECS is located. Select an AZ
as required.
All ECSs and HANA ECSs in the SAP HANA system
must be within the same AZ.
ECS Type Select General-purpose.
vCPU and Memory s1.xlarge (4 vCPUs and 16 GB memory capacity)
Image Types Select Public Image and an image as required.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
20
Parameter Description
Disk 80 GB system disk
For details about disk requirements, see section 2.2.2
Other Nodes.
VPC and Security Group Select the VPC and security group created in Applying for
a Subnet and Configuring a Security Group.
NIC Select the correct NIC based on 2.2.3 Network Planning.
Elastic IP Address Set these parameters as required.
Bandwidth This parameter is configured as required only when
Elastic IP Address is set to Buy Now.
Key Pair This parameter is available only when the Login Mode is
set to Key pair.
An SSH key certificate is used for authenticating users
who attempt to log in to an ECS. To create a key pair,
click View Key Pair and then click Create Key Pair on
the Key Pair page.
Ensure that the HANA ECSs/ECSs where SAP HANA,
SAP HANA Studio, NFS servers, and NAT servers are to
be deployed use the same key. Otherwise, SAP HANA
installation will fail.
NOTE
If you choose an existing SSH key certificate from the drop-down
list, make sure that you have saved the certificate locally.
Otherwise, you may fail to log in to the HANA ECS or ECS.
If you want to create an SSH key pair, do as follows:
Click View Key Pair. On the Key Pair page that is displayed,
click Create Key Pair, specify the key pair name, and click OK.
In the Information dialog box that is displayed, click OK. Then,
you can query and save the private key as prompted.
File Injection This parameter is displayed only after you click Show
behind Advanced Settings.
Do not configure them in this scenario.
User Data Injection This parameter is displayed only after you click Show
behind Advanced Settings.
When creating a Windows ECS, you do not need to
configure this parameter.
When creating a Linux ECS, assign permissions to user
root in user data injection mode so that the user can log
in to an ECS using a private key through SSH.
To do so, select Text and enter the following content in
the text box:
#cloud-config
disable_root: false
runcmd:
- sed -i 's/^PermitRootLogin.*$/PermitRootLogin
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
21
Parameter Description
without-password/' /etc/ssh/sshd_config
- sed -i '/^KexAlgorithms.*$/d' /etc/ssh/sshd_config
- service sshd restart
Quantity 1
Step 4 Click Buy Now. On the page that is displayed, click Submit Application.
Step 5 The system returns to the Elastic Cloud Server page. Check the status of the created task in
Task Status on the right of the page.
After the ECS is created, you can view the ECS from the ECS list on the right of the page.
----End
2.4.3 Creating a NAT Server
Scenarios
In SAP HANA systems, you must create an ECS for deploying the NAT server. You can visit
the NAT server and then switch to an SAP HANA node using SSH for fault diagnosing and
locating.
Procedure
Step 1 Log in to the management console.
Step 2 On the system homepage, choose Computing > Elastic Cloud Server to switch to the
Elastic Cloud Server page.
Step 3 On the right side of the page, click Buy ECS to switch to the Create ECS page.
Step 4 Configure the parameters listed in Table 2-13.
Table 2-13 ECS parameters
Parameter Description
ECS Name Specifies the ECS name. The name contains a maximum
of 13 characters.
AZ Specifies the AZ where the ECS is located. Select an AZ
as required.
All ECSs and HANA ECSs in the SAP HANA system
must be within the same AZ.
ECS Type Select General-purpose.
vCPU and Memory s1.medium (1 vCPU and 4 GB memory capacity) or higher
Image Types Select Public Image and an image as required.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
22
Parameter Description
Disk 40 GB system disk
For details about disk requirements, see section 2.2.2
Other Nodes.
VPC and Security Group Select the VPC and security group created in Applying for
a Subnet and Configuring a Security Group.
NIC Select the correct NIC based on 2.2.3 Network Planning.
Elastic IP Address Set these parameters as required.
Bandwidth This parameter is configured only after an elastic IP
address is bound to an ECS. Specifies the communication
channel bandwidth when using EIP. Set the parameter as
required, for example, 30 Mbit/s.
This parameter is configured as required only when
Elastic IP Address is set to Buy Now.
Key Pair This parameter is available only when the Login Mode is
set to Key pair.
An SSH key certificate is used for authenticating users
who attempt to log in to an ECS. To create a key pair,
click View Key Pair and then click Create Key Pair on
the Key Pair page.
Ensure that the HANA ECSs/ECSs where SAP HANA,
SAP HANA Studio, NFS servers, and NAT servers are to
be deployed use the same key. Otherwise, SAP HANA
installation will fail.
NOTE
If you choose an existing SSH key certificate from the drop-down
list, make sure that you have saved the certificate locally.
Otherwise, you may fail to log in to the HANA ECS or ECS.
Click View Key Pair. On the Key Pair page that is displayed,
click Create Key Pair, specify the key pair name, and click OK.
In the Information dialog box that is displayed, click OK. Then,
you can query and save the private key as prompted.
File Injection This parameter is displayed only after you click Show
behind Advanced Settings.
Do not configure them in this scenario.
User Data Injection Assign permissions to user root in user data injection
mode so that the user can log in to an ECS using a private
key through SSH.
This parameter is displayed only after you click Show
behind Advanced Settings.
Select Text and enter the following content in the text box
to assign permissions to user root so that the user can log
in to a HANA ECS using a private key through SSH:
#cloud-config
disable_root: false
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
23
Parameter Description
runcmd:
- sed -i 's/^PermitRootLogin.*$/PermitRootLogin
without-password/' /etc/ssh/sshd_config
- sed -i '/^KexAlgorithms.*$/d' /etc/ssh/sshd_config
- service sshd restart
Quantity
Quantity
1
Step 5 Click Buy Now. On the page that is displayed, click Submit Application.
Step 6 The system returns to the Elastic Cloud Server page. Check the status of the created task in
Task Status on the right of the page.
After the ECS is created, you can view the ECS from the ECS list on the right of the page.
----End
2.4.4 Creating an NFS Server
Scenarios
In SAP HANA systems, you must create one ECS for deploying the NFS server in
single-node deployment mode or two ECSs for deploying the NFS servers in active/standby
deployment mode. The NFS servers provide disks that can be shared with SAP HANA nodes.
Procedure
Step 1 Log in to the public cloud management console and choose Computing > Elastic Cloud
Server to switch to the Elastic Cloud Server page.
Step 2 On the right side of the page, click Buy ECS to switch to Create ECS page.
Step 3 Configure the parameters listed in Table 2-14.
Table 2-14 Parameters of the ECS where the NFS server is deployed
Parameter Description
ECS Name Specifies the ECS name. The name contains a maximum
of 13 characters.
AZ Specify the AZ in which an ECS locates. Select an AZ as
required.
All ECSs and HANA ECSs in the SAP HANA system
must be within the same AZ.
ECS Type Select General-purpose.
vCPU and Memory s1.xlarge (4 vCPUs and 16 GB memory capacity)
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
24
Parameter Description
Image Types Select Public Image and an image as required.
Disk For details about disk requirements, see section 2.2.2
Other Nodes.
If two NFS servers are deployed to work in active/standby
mode, when creating an NFS server, you must create the
shared, backup, and SBD volumes. When creating the
other NFS server, you do not need to create these disks.
You only need to attach these disks to the other NFS
server. For details, see section Formatting the Disks
Attached to an NFS Server.
VPC and Security Group Use the VPC and security group that are created in section
Applying for a Subnet and Configuring a Security Group.
NIC When only one NFS server is deployed, two NICs are
required.
The first NIC is the primary NIC, which is used on the
server/client plane. NFS Server uses the primary NIC
plane to synchronize time with the NTP server.
The second NIC is an internal communication plane
NIC. This NIC is added when you created an ECS or
HANA ECS. However, this NIC will not be used.
The third NIC is used on the cloud management or
backup plane.
When two NFS servers are deployed to work in
active/standby mode, a total of three NICs are required.
The first NIC is the primary NIC, which is used on the
server/client plane.
The second NIC is the first extended NIC, which is
used on the internal communication plane for
exchanging heartbeat signals between the active and
standby NFS servers.
The third NIC is the second extended NIC, which is
used on the cloud management or backup plane.
Create NICs based on the preceding requirements and the
deployment plan described in section 2.2.3 Network
Planning.
Elastic IP Address (EIP) Do not bind an EIP to the ECS.
Key Pair This parameter is available only when the Login Mode is
set to Key pair.
An SSH key certificate is used for authenticating users
who attempt to log in to an ECS. To create a key pair,
click View Key Pair and then click Create Key Pair on
the Key Pair page.
Ensure that the HANA ECSs/ECSs where SAP HANA,
SAP HANA Studio, NFS servers, and NAT servers are to
be deployed use the same key. Otherwise, SAP HANA
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
25
Parameter Description
installation will fail.
NOTE
If you choose an existing SSH key certificate from the drop-down
list, make sure that you have saved the certificate locally.
Otherwise, you may fail to log in to the HANA ECS or ECS.
If you want to create an SSH key pair, do as follows:
Click View Key Pair. On the Key Pair page that is displayed,
click Create Key Pair, specify the key pair name, and click OK.
In the Information dialog box that is displayed, click OK. Then,
you can query and save the private key as prompted.
User Data Injection This parameter is displayed only after you click Show
behind Advanced Settings.
Select Text and enter the following content in the text box
to assign permissions to user root so that the user can log
in to a HANA ECS using a private key through SSH:
#cloud-config
disable_root: false
runcmd:
- sed -i 's/^PermitRootLogin.*$/PermitRootLogin
without-password/' /etc/ssh/sshd_config
- sed -i '/^KexAlgorithms.*$/d' /etc/ssh/sshd_config
- service sshd restart
ECS Group This parameter is displayed only after you click Show
behind Advanced Settings.
Determine the policy of a HANA ECS group based on the
scenario:
Single NFS: You do not need to specify ECS Group.
Active/standby NFS servers: Two NFS servers must
belong to the same ECS Group and this ECS Group
cannot be the same as the ECS Group in which HANA
ECSs locate.
Quantity 1
Step 4 Click Buy Now. On the page that is displayed, click Submit Application.
Step 5 The system returns to the Elastic Cloud Server page. Check the status of the created task in
Task Status on the right of the page.
After the HANA ECS is created, you can view the ECS from the ECS list on the right of the
page.
Step 6 (Optional) Repeat Step 1 to Step 5 to create the other NFS server.
This operation is performed only when two NFS servers are deployed to work in
active/standby mode. When creating the other NFS server, you do not need to create a shared,
backup, or SBD volume for it.
Step 7 Change the password of user root for logging in to the NFS server.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
26
Securely keep the password of user root.
1. Log in to the NFS Server using the key file.
2. Run the following command to change the password of user root:
passwd
Enter the new password as prompted and confirm it.
----End
2.4.5 Configuring SSH Switching Permissions
Scenarios
To allow SSH switchovers between SAP HANA servers, NFS servers, and NAT servers, you
must configure the servers to be trusty.
Procedure
Step 1 Upload the key file to the NAT server.
1. On the local computer, generate the key file for logging in to the NAT server.
When creating the NAT server, you specify the certificate key file (.pem file) for the
NAT server.
The .pem file is used to generate the .ppk file. For details, see section Logging In to a
Linux ECS Using an SSH Key.
2. On the local computer, install the WinSCP software.
3. Upload the certificate private key file (.pem file) to the NFS server.
Use WinSCP to upload the certificate private key file (.pem file) to the /usr directory on
the NAT server using an elastic IP address. Ensure that user root and the key file (.ppk
file) are used for authentication.
4. Use PuTTY to log in to the NAT server. Ensure that user root and the key file (.ppk file)
are used for authentication.
5. Copy the certificate private key file (.pem file) to the /root/.ssh directory and rename the
file id_rsa.
For example, if the original file name is private.pem, run the following command to
rename it:
cp /usr/private.pem /root/.ssh/id_rsa
cd /root/.ssh/
chmod 600 id_rsa
6. Run the following command to generate a public key file:
cat authorized_keys >>id_rsa.pub
Step 2 Use the server/client plane IP address to allocate the locally stored public key file to an SAP
HANA node on which SAP HANA is to be installed.
Ensure that the SAP HANA node cannot be the planned standby node and the subsequent
operations to manually install SAP HANA must be performed on this node.
The command is in the following format:
scp /root/.ssh/id_rsa.pub Peer IP address:/root/.ssh/
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
27
For example, if the peer IP address is 10.0.3.102, run the following commands:
scp /root/.ssh/id_rsa.pub 10.0.3.102:/root/.ssh/
Step 3 Use the server/client plane IP address to allocate the locally stored private key file and
authorized_keys file to all HANA ECSs excepting the SAP HANA Studio. Specifically,
allocate the files to the SAP HANA nodes and NFS servers. The NFS servers are available
only in cluster scenarios.
The command is in the following format:
scp /root/.ssh/id_rsa Peer IP address:/root/.ssh/id_rsa
scp /root/.ssh/authorized_keys Peer IP address:/root/.ssh/
For example, if the peer IP address is 10.0.3.102, run the following commands:
scp /root/.ssh/id_rsa 10.0.3.102:/root/.ssh/id_rsa
scp /root/.ssh/authorized_keys 10.0.3.102:/root/.ssh/
Step 4 Verify the switching.
Use SSH to switch from the NAT server to all servers and HANA servers excepting the SAP
HANA Studio for verification. These servers and HANA servers include SAP HANA nodes
and NFS servers. The NFS servers are available only in cluster scenarios.
For example, the server/client plane IP address of the NFS server is 10.0.3.101. To switch
from the NAT server to the NFS server, run the following command:
ssh 10.0.3.101
After the switching, you must switch back to the NAT server. Then, verify the switching from the NAT
server to other nodes.
----End
2.4.6 Configuring an IP Address for a BMS NIC
Scenarios
After applying for a BMS for an SAP HANA cluster, you must manually configure an IP
address for the BMS NIC on the internal communication plane to meet SAP HANA
deployment requirements.
Procedure
Step 1 Use a key or password to log in to the BMS as user root.
Step 2 On the BMS CLI, view the NIC information.
ip link
Information similar to the following is displayed:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group
default
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
28
UP mode DEFAULT group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
UP mode DEFAULT group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state DOWN mode DEFAULT
group default qlen 1000
link/ether 38:4c:4f:89:55:8d brd ff:ff:ff:ff:ff:ff
5: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state DOWN mode DEFAULT
group default qlen 1000
link/ether 38:4c:4f:89:55:8e brd ff:ff:ff:ff:ff:ff
6: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 8888 qdisc noqueue state UP mode
DEFAULT group default
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
7: bond0.3133@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8888 qdisc noqueue state
UP mode DEFAULT group default
link/ether fa:16:3e:57:87:6e brd ff:ff:ff:ff:ff:ff
Step 3 Configure the udev rules.
Run the following command to create the 80-persistent-net.rules file:
cp /etc/udev/rules.d/70-persistent-net.rules /etc/udev/rules.d/80-persistent-net.rules
Write the NIC MAC address and name that are queried in Step 2 and that are not displayed in
80-persistent-net.rules to the file. In this way, after the BMS is restarted, the NIC name and
sequence will not change.
Ensure that NIC MAC address and name are lowercase letters.
vim /etc/udev/rules.d/80-persistent-net.rules
The file after the modification is as follows:
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="38:4c:4f:29:0b:e0",
NAME="eth0"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="38:4c:4f:29:0b:e1",
NAME="eth1"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="38:4c:4f:89:55:8d",
NAME="eth2"
SUBSYSTEM=="net", ACTION=="add", DRIVERS=="?*", ATTR{address}=="38:4c:4f:89:55:8e",
NAME="eth3"
After the modification, save the change and exit.
Step 4 Check the NIC IP address.
ifconfig
Information similar to the following is displayed, where bond0 and bond0.313 show the NIC
IP addresses automatically allocated by the system when you apply for the BMS:
bond0 Link encap:Ethernet HWaddr FA:16:3E:3D:1C:E0
inet addr:10.0.1.2 Bcast:10.0.1.255 Mask:255.255.255.0
inet6 addr: fe80::f816:3eff:fe3d:1ce0/64 Scope:Link
UP BROADCAST RUNNING MASTER MULTICAST MTU:8888 Metric:1
RX packets:852 errors:0 dropped:160 overruns:0 frame:0
TX packets:1121 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
29
RX bytes:125429 (122.4 Kb) TX bytes:107221 (104.7 Kb)
bond0.313 Link encap:Ethernet HWaddr FA:16:3E:57:87:6E
inet addr:10.0.3.2 Bcast:10.0.3.255 Mask:255.255.255.0
inet6 addr: fe80::f816:3eff:fe57:876e/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:8888 Metric:1
RX packets:169 errors:0 dropped:0 overruns:0 frame:0
TX packets:13 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:8684 (8.4 Kb) TX bytes:1696 (1.6 Kb)
eth0 Link encap:Ethernet HWaddr FA:16:3E:3D:1C:E0
UP BROADCAST RUNNING SLAVE MULTICAST MTU:8888 Metric:1
RX packets:428 errors:0 dropped:10 overruns:0 frame:0
TX packets:547 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:64670 (63.1 Kb) TX bytes:50132 (48.9 Kb)
eth1 Link encap:Ethernet HWaddr FA:16:3E:3D:1C:E0
UP BROADCAST RUNNING SLAVE MULTICAST MTU:8888 Metric:1
RX packets:424 errors:0 dropped:7 overruns:0 frame:0
TX packets:574 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:60759 (59.3 Kb) TX bytes:57089 (55.7 Kb)
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:65536 Metric:1
RX packets:8 errors:0 dropped:0 overruns:0 frame:0
TX packets:8 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:520 (520.0 b) TX bytes:520 (520.0 b)
Step 5 View the names of bonded NICs.
The in-service bonded NICs cannot be used on the internal communication plane. Therefore,
you must obtain them by name.
cd /etc/sysconfig/network
vi ifcfg-bond0
Information similar to the following is displayed, where bond0 is composed of NICs eth0 and
eth1:
BONDING_MASTER=yes
TYPE=Bond
STARTMODE=auto
BONDING_MODULE_OPTS="mode=4 xmit_hash_policy=layer3+4 miimon=100"
NM_CONTROLLED=no
BOOTPROTO=dhcp
DEVICE=bond0
USERCONTRL=no
LLADDR=fa:16:3e:3d:1c:e0
BONDING_SLAVE1=eth1
BONDING_SLAVE0=eth0
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
30
After the query, exit.
Step 6 View the statuses of all NICs.
ip link
Information similar to the following is displayed:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group
default
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
UP mode DEFAULT group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
UP mode DEFAULT group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state DOWN mode DEFAULT
group default qlen 1000
link/ether 38:4c:4f:89:55:8d brd ff:ff:ff:ff:ff:ff
5: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state DOWN mode DEFAULT
group default qlen 1000
link/ether 38:4c:4f:89:55:8e brd ff:ff:ff:ff:ff:ff
6: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 8888 qdisc noqueue state UP mode
DEFAULT group default
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
7: bond0.3133@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8888 qdisc noqueue state
UP mode DEFAULT group default
link/ether fa:16:3e:57:87:6e brd ff:ff:ff:ff:ff:ff
Step 7 Change NIC status qdisc mq state DOWN to qdisc mq state UP. In the following
commands, NICs eth2 and eth3 are used as examples.
ip link set eth2 up
ip link set eth3 up
Step 8 View the statuses of the NICs again.
ip link
Information similar to the following is displayed:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group
default
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
UP mode DEFAULT group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
UP mode DEFAULT group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
4: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group
default qlen 1000
link/ether 38:4c:4f:89:55:8d brd ff:ff:ff:ff:ff:ff
5: eth3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group
default qlen 1000
link/ether 38:4c:4f:89:55:8e brd ff:ff:ff:ff:ff:ff
6: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 8888 qdisc noqueue state UP mode
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
31
DEFAULT group default
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
7: bond0.3133@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8888 qdisc noqueue state
UP mode DEFAULT group default
link/ether fa:16:3e:57:87:6e brd ff:ff:ff:ff:ff:ff
Step 9 Check the statuses of the NICs in Step 8 and obtain the names of the NICs in qdisc mq state
UP state.
Only the NICs that are in qdisc mq state UP state and have not been used can be bonded. In
this example, such NICs are eth2 and eth3.
The LLADR values of NICs eth2 and eth3 are 38:4c:4f:89:55:8d and 38:4c:4f:89:55:8e,
respectively.
Step 10 Create the configuration files of NICs eth2 and eth3.
You can copy an existing NIC configuration file and modify it to improve the creation
efficiency.
cp ifcfg-eth0 ifcfg-eth2
cp ifcfg-eth1 ifcfg-eth3
Step 11 Modify the configuration files of NICs eth2 and eth3.
vi ifcfg-eth2
vi ifcfg-eth3
Modified configuration file of NIC eth2 is as follows.
In this configuration file, set MTU to 8888, BOOTPROTO to STATIC, and configure
DEVICE and LLADDR as required.
STARTMODE=auto
MTU=8888
NM_CONTROLLED=no
BOOTPROTO=STATIC
DEVICE=eth2
USERCONTRL=no
LLADDR=38:4c:4f:89:55:8d
TYPE=Ethernet
Modified configuration file of NIC eth3 is as follows:
STARTMODE=auto
MTU=8888
NM_CONTROLLED=no
BOOTPROTO=STATIC
DEVICE=eth3
USERCONTRL=no
LLADDR=38:4c:4f:89:55:8e
TYPE=Ethernet
After the modification, save the change and exit.
Step 12 Bond NICs eth2 and eth3 to a NIC, for example, bond1.
Run the following commands to create the ifcfg-bond1 file and modify the configuration file:
cp ifcfg-bond0 ifcfg-bond1
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
32
vi ifcfg-bond1
Modified configuration file of NIC bond1 is as follows.
In this configuration file, MTU is set to 8888, BONDING_MODULE_OPTS is set to
mode=1 miimon=100, BOOTPROTO is set to STATIC, DEVICE, BONDING_SLAVE1,
BONDING_SLAVE0, IPADDR, NETMASK, and NETWORK are configured as required,
and LLADDR is set to the LLADDR value of the BONDING_SLAVE1 NIC.
BONDING_MASTER=yes
TYPE=Bond
MTU=8888
STARTMODE=auto
BONDING_MODULE_OPTS="mode=1 miimon=100"
NM_CONTROLLED=no
BOOTPROTO=STATIC
DEVICE=bond1
USERCONTRL=no
LLADDR=38:4c:4f:89:55:8d
BONDING_SLAVE1=eth2
BONDING_SLAVE0=eth3
IPADDR=10.0.2.2
NETMASK=255.255.255.0
NETWORK=10.0.2.0
After the modification, save the change and exit.
Step 13 Make the configuration file take effect.
1. Run the following commands to create a temporary directory and copy the NIC
configuration file to this directory:
mkdir /opt/tmp/
mkdir /opt/tmp/xml
cp /etc/sysconfig/network/ifcfg* /opt/tmp/
cp /etc/sysconfig/network/config /opt/tmp/
cp /etc/sysconfig/network/dhcp /opt/tmp/
2. Run the following commands to stop NICs to form bond1:
ip link set eth2 down
ip link set eth3 down
3. Run the following command to convert the NIC configuration file to a configuration file
that can be recognized by the OS:
/usr/sbin/wicked --log-target=stderr --log-level=debug3 --debug all convert --output
/opt/tmp/xml /opt/tmp/
4. Restart the NICs to form bond1.
ip link set eth2 up
/usr/sbin/wicked --log-target=stderr --log-level=debug3 --debug all ifup --ifconfig
/opt/tmp/xml/eth2.xml eth2
ip link set eth3 up
/usr/sbin/wicked --log-target=stderr --log-level=debug3 --debug all ifup --ifconfig
/opt/tmp/xml/eth3.xml eth3
/usr/sbin/wicked --log-target=stderr --log-level=debug3 --debug all ifup --ifconfig /opt/tmp/xml/bond1.xml bond1
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
33
Step 14 Run the following command to view IP addresses:
ip addr show
IP addresses have been assigned to the NICs. Information similar to the following is
displayed:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
valid_lft forever preferred_lft forever
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
UP group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond0 state
UP group default qlen 1000
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
4: eth2: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond1 state
UP group default qlen 1000
link/ether 38:4c:4f:89:55:8d brd ff:ff:ff:ff:ff:ff
5: eth3: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 8888 qdisc mq master bond1 state
UP group default qlen 1000
link/ether 38:4c:4f:89:55:8d brd ff:ff:ff:ff:ff:ff
6: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 8888 qdisc noqueue state UP
group default
link/ether fa:16:3e:3d:1c:e0 brd ff:ff:ff:ff:ff:ff
inet 10.0.1.2/24 brd 10.0.1.255 scope global bond0
valid_lft forever preferred_lft forever
inet6 fe80::f816:3eff:fe3d:1ce0/64 scope link
valid_lft forever preferred_lft forever
7: bond0.3133@bond0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 8888 qdisc noqueue state
UP group default
link/ether fa:16:3e:57:87:6e brd ff:ff:ff:ff:ff:ff
inet 10.0.3.2/24 brd 10.0.2.255 scope global bond0.3133
valid_lft forever preferred_lft forever
inet6 fe80::f816:3eff:fe57:876e/64 scope link
valid_lft forever preferred_lft forever
8: bond1: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 8888 qdisc noqueue state UP
group default
link/ether 38:4c:4f:89:55:8d brd ff:ff:ff:ff:ff:ff
inet 10.0.2.2/24 brd 10.0.2.255 scope global bond1
valid_lft forever preferred_lft forever
inet6 fe80::3a4c:4fff:fe29:b36/64 scope link
valid_lft forever preferred_lft forever
Step 15 Run the following commands to delete the temporary directory:
cd /opt
rm -rf tmp/
Step 16 Perform the preceding operations to configure other BMSs.
----End
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
34
2.4.7 Configuring the Mapping Between SAP HANA Host Name and IP Addresses
Scenarios
During the SAP HANA installation, installation programs use host names for communication.
Therefore, you must configure the mapping between the host names and IP addresses.
Procedure
Step 1 Use PuTTY to log in to the NAT server with an elastic IP address bound. Ensure that user
root and the key file (.ppk file) are used for authentication. Then, use SSH to switch to the
HANA server on which the SAP HANA software is to be installed.
In cluster scenarios, you will switch to the first HANA server. Then, you will perform operations on this
HANA server when installing SAP HANA.
Step 2 Run the following command to open the hosts file:
vi /etc/hosts
Step 3 Press i to enter editing mode and write the names and IP addresses of all SAP HANA nodes
into the hosts file.
In cluster scenarios, IP-Address is the IP address of the internal SAP HANA
communication plane.
Both Full-Qualified-Hostname and Short-Hostname are the server name, for example,
hana001.
The format is "IP-Address Full-Qualified-Hostname Short-Hostname".
In an SAP HANA system, you must write the mapping between all SAP HANA node IP
addresses and node names into the hosts file.
Take the IP addresses 10.0.2.2 to 10.0.2.6 of the internal HANA communication plane for five
SAP HANA nodes as an example.
The edited content is as follows:
...
10.0.2.2 hana001 hana001
10.0.2.3 hana002 hana002
10.0.2.4 hana003 hana003
10.0.2.5 hana004 hana004
10.0.2.6 hana005 hana005
Step 4 After you complete editing, press Esc, enter :x, and press Enter to exit the hosts file.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
35
Step 5 Run the following command to transfer the configured /etc/hosts file to other SAP HANA
nodes:
The command is in the following format:
scp /etc/hosts Peer IP address:/etc/hosts
Step 6 Verify the SSH switching between SAP HANA ECSs.
Use SSH to switch from one SAP HANA node to all SAP HANA nodes (including the current
node) to ensure that the switching is correct.
For example, if the name of the peer SAP HANA node is hana002, run the following
command:
ssh hana002
----End
2.4.8 Modifying OS Configurations
Scenarios
To ensure the proper installation of the SAP HANA system, disable the OS firewalls of all
nodes before the installation.
Procedure
Step 1 Log in to the NAT server as user root using the key file. Then, use SSH to switch to SAP
HANA nodes.
Step 2 Run the following commands on the SAP HANA node to disable automatic firewall enabling
and disable the firewall:
If the OS is SUSE Linux Enterprise Server 11 SP4, run the following command:
chkconfig SuSEfirewall2_setup off
chkconfig SuSEfirewall2_init off
SuSEfirewall2 stop
If the OS is SUSE Linux Enterprise Server 12, run the following command:
SuSEfirewall2 off
SuSEfirewall2 stop
systemctl disable SuSEfirewall2_init.service
systemctl disable SuSEfirewall2.service
systemctl stop SuSEfirewall2_init.service
systemctl stop SuSEfirewall2.service
Step 3 Repeat the preceding step to disable the firewalls of all nodes in the SAP HANA system.
----End
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
36
2.5 Installing SAP HANA (Cluster Deployment and Single NFS Server Deployed)
2.5.1 Formatting the Disks Attached to an NFS Server
Scenarios
The disks attached to an NFS server must be formatted to comply with the SAP HANA
installation requirements.
Procedure
Step 1 Format the disks of the NFS servers.
1. Use PuTTY to log in to the NAT server with an elastic IP address bound. Ensure that
user root and the key file (.ppk file) are used for authentication. Then, use SSH to switch
to an NFS server.
2. Run the following command to query the disk to be formatted:
fdisk -l
3. Determine the disks of the shared, backup, and SBD volumes according to disk
capacities. The SBD volumes are used only when two NFS servers are deployed to work
in active/standby mode.
4. Run the following command to download the configuration script:
Download the readme.txt file at https://obs-sap.obs.myhwclouds.com/readme.txt on a local computer
to obtain the location where the software and configuration file are stored.
wget https://obs-sap.obs.myhwclouds.com/hana/cfgandscript/scale_out_script.zip -P
~
5. Run the following commands to switch to the folder and decompress the software
package:
cd ~
unzip scale_out_script.zip
6. Run the following command to modify the nfsserver.cfg file:
vi ~/scale_out_script/nfsserver.cfg
7. Press i to enter editing mode. Write the disk information and the SAP HANA ECS IP
addresses that are allowed to access the NFS servers into the nfsserver.cfg file.
The parameters are as follows:
− share: specifies the disk of the shared volume.
− backup: specifies the disk of the backup volume.
− sbd: specifies the disk of the SBD volume. This parameter is required only when
two NFS servers are deployed to work in active/standby mode.
− permithosts: specifies the SAP HANA ECS that is accessible to the NFS servers.
The IP address of this SAP HANA ECS is the cloud management or backup plane
IP address, for example, 10.0.1.x. When multiple IP addresses are available,
separate them with a comma (,).
− nfsServerIps: specifies the IP addresses (for example, 10.0.1.x) of the cloud management or backup planes for the NFS servers. When two NFS servers are
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
37
deployed to work in active/standby mode, set this parameter value to the server IP
addresses and separate them with a comma (,).
The content of the configuration file is as follows when only one NFS server is
deployed:
share=/dev/sda
backup=/dev/sdb
sbd=
permithosts=10.0.1.2,10.0.1.3,10.0.1.4,10.0.1.5,10.0.1.6
nfsServerIps=10.0.1.101
The content of the configuration file is as follows when two NFS servers are deployed to
work in active/standby mode:
share=/dev/sda
backup=/dev/sdb
sbd=/dev/sdc
permithosts=10.0.1.2,10.0.1.3,10.0.1.4,10.0.1.5,10.0.1.6
nfsServerIps=10.0.1.101,10.0.1.102
8. After you complete editing, press Esc, enter :x, and press Enter to exit the nfsserver.cfg
file.
9. Run the following commands to assign execute permissions, convert the file format, and
format the disk:
cd ~/scale_out_script
chmod 777 nfsserver.sh nfsserver.cfg
dos2unix nfsserver.sh nfsserver.cfg
sh nfsserver.sh
----End
2.5.2 Formatting an SAP HANA Disk
Scenarios
In SAP HANA cluster scenarios, all data volumes and log volumes form a data volume
resource pool and a log volume resource pool, respectively, for all SAP HANA nodes. In this
way, every data or log volume is attached to all SAP HANA nodes.
After the attachment, the data disks, including the data and log volumes and the shared and
backup volumes on the NFS servers, can be used only after they are formatted and attached to
required directories.
Procedure
Step 1 Attach all data and log volumes to all SAP HANA nodes.
1. On the management console, choose Computing > Bare Meta Server. On the left side
of the page, choose Bare Metal Server. Then, the system displays all BMSs on the right
side of the page.
2. Locate the first BMS by name and click its name.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
38
3. On the page providing details about the ECS, click the EVS tab and find the data disk to
be bound to other SAP HANA nodes (log volume or data volume). Then, click the data
disk.
4. In the expanded data disk pane, click the link following the data disk ID.
5. On the page that is displayed, click Mount Point and then Attach to switch to the
Attach Disk page.
6. On the Attach Disk page, select all the target SAP HANA nodes and attach the disk to
the SAP HANA nodes.
Ensure that the disk is attached to all SAP HANA nodes.
7. Perform the preceding operations to attach the second disk (data volume or log volume)
of the SAP HANA node to all other SAP HANA nodes.
8. Perform the preceding operations to attach the log and data volumes of each SAP HANA
node to all other SAP HANA nodes.
The last SAP HANA node is a standby node. When creating this BMS, you did not create or attach log
and data volumes to it. Therefore, you do not need to attach the disks of this BMS to other SAP HANA
nodes.
Step 2 Format the disks attached to an SAP HANA node.
1. Log in to the NAT server with an elastic IP address bound. Ensure that user root and the
key file are used for authentication. Then, use SSH to switch to the first SAP HANA
node and run the following command to check the disks that are not formatted:
fdisk -l
2. Determine the paths for the data and log volumes according to the disk capacities.
3. View the IP address (for example, 10.0.1.x) of the cloud management or backup plane
for the NFS server.
For details, see section A.4 Querying the NIC IP Address of an ECS.
4. View the internal communication plane IP addresses (for example, 10.0.2.x) of all SAP
HANA nodes.
For details, see section A.3 Querying the NIC IP Address of a BMS.
5. Download the configuration script.
wget https://obs-sap.obs.myhwclouds.com/hana/cfgandscript/scale_out_script.zip -P
~
6. Run the following commands to switch to the folder and decompress the software
package:
cd ~
unzip scale_out_script.zip
7. Run the following command to modify the hanaCluster.cfg file:
vi ~/scale_out_script/hanaCluster.cfg
8. Press i to enter editing mode. Write the path, cloud management or backup plane IP
address of the NFS server, and internal communication plane IP address of all SAP
HANA nodes into the hanaCluster.cfg file.
Parameters are as follows:
− partition_1_data__volname: specifies the path for the data volume attached to the
first SAP HANA node. Apply this rule to other data volumes. A standby node has
no data volumes attached and does not require recording.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
39
− partition_1_log__volname: specifies the path for the log volume attached to the
first SAP HANA node. Apply this rule to other log volumes. A standby node has no
log volumes attached and does not require recording.
− usrsaps: You do not need to configure this parameter.
− shareFSType: Set this parameter to NFS.
− sfsShared: You do not need to configure this parameter.
− sfsBackup: You do not need to configure this parameter.
− nfsserverIp: specifies the IP address (for example, 10.0.1.x) of the cloud
management or backup plane for the NFS server. When two NFS servers are
deployed to work in active/standby mode, the value of this parameter is the floating
IP address of the NFS servers.
− sid: specifies the SAP HANA system ID. After the configuration, ensure that the
system ID is the same as this value when installing SAP HANA.
− hosts: specifies the internal communication plane IP addresses (for example,
10.0.2.x) of all SAP HANA nodes. When multiple IP addresses are available,
separate them with a comma (,).
Take five SAP HANA nodes, four worker nodes and one standby node, as an example.
Ensure that every volume path is unique.
partition_1_data__volname=/dev/sda
partition_1_log__volname=/dev/sdb
partition_2_data__volname=/dev/sdc
partition_2_log__volname=/dev/sdd
partition_3_data__volname=/dev/sde
partition_3_log__volname=/dev/sdf
partition_4_data__volname=/dev/sdg
partition_4_log__volname=/dev/sdh
usrsaps=
nfsserverIp=10.0.1.103
shareFSType=NFS
sfsShared=
sfsBackup=
sid=S01
hosts=10.0.2.2,10.0.2.3,10.0.2.4,10.0.2.5,10.0.2.6
9. After you complete editing, press Esc, enter :x, and press Enter to exit the
hanaCluster.cfg file.
10. Run the following commands to switch to the directory where the hanaCluster.cfg file
is stored and convert the file format:
cd ~/scale_out_script
dos2unix installHanaCluster.sh hanaCluster.cfg
11. Assign operation permissions to the hanaCluster.cfg file and run the
installHanaCluster.sh command to format disks.
chmod 777 installHanaCluster.sh hanaCluster.cfg
sh installHanaCluster.sh
----End
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
40
2.5.3 Installing the SAP HANA Software
Scenarios
The SAP HANA software is installed using an installation script. This section uses the SAP
HANA SPS12 installation package as an example. You can download the package from the
official website.
You only need to install SAP HANA on one SAP HANA node. Ensure that the SAP HANA
node cannot be the planned standby node.
Prerequisites Required resources are available. For details, see section 2.3 Preparing for Resources.
You have created servers, formatted disks attached to them, and completed required
configurations.
You have synchronized the time of the ECS.
The firewall on the ECS where SAP HANA is to be installed has been disabled.
Procedure
Step 1 Generate the configfile.cfg file required for the installation.
1. Use PuTTY to log in to the NAT Server using an EIP. Ensure that user root and the key
file (.ppk file) are used for authentication. Use the SSH protocol to switch to the SAP
HANA node on which the SAP HANA software is to be installed. Ensure that the SAP
HANA node is the one to which the public key file id_rsa.pub is uploaded when SSH
switching permissions are configured. Assume that the host name is hana001.
2. On the CLI, run the following commands to download the SAP HANA installation
package:
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part1.exe
-P /hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part2.rar
-P /hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part3.rar
-P /hana/shared
3. Run the following command to decompress the installation package and navigate to the
directory where the installation file locates:
unrar x 51052383_part1.exe
4. Enter the directory where the installation file is stored.
For example, if the installation file is stored in
/51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64, run the following
command:
cd /51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64
5. Run the following command to assign permissions to the directory:
chmod 777 -R /hana
6. Run the following command to obtain the configfile.cfg file:
./hdblcm --action=install --dump_configfile_template=/hana/shared/configfile.cfg
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
41
After this command is executed, the configfile.cfg file is stored in the /hana/shared
directory.
7. Run the following command and press i to enter the editing mode and edit the
configfile.cfg file:
vi /hana/shared/configfile.cfg
8. Modify parameter settings.
The parameters are as follows:
− remote_execution: specifies a remote execution mode, the default value of which is
ssh.
− sapmnt: specifies the installation path, which is /hana/shared.
− hostname: specifies the ECS name, such as hana001.
− root_password: specifies the password of user root for logging in to the HANA
ECS.
− storage_cfg: specifies the path in which the configfile.cfg file is stored. Set the
value of this parameter to /hana/shared.
− sid: specifies the SAP HANA system ID, which must be the same as the sid value
configured in hanaCluster.cfg when the disk is formatted.
− number: specifies the SAP HANA instance ID. The instance ID is used in 2.2.3
Network Planning, which must be the same as the planned one.
− sapadm_password: specifies the password of the HostAgent software package
installed on the HANA ECS. Configure this parameter based on service
requirements.
An example is provided as follows:
[General]
# Location of SAP HANA Installation Medium
component_medium=
# Comma separated list of component directories
component_dirs=
# Directory root to search for components
component_root=
# Components ( Valid values: all | client | es | ets | lcapps | server | smartda
| streaming | rdsync | studio | afl | pos | sal | sca | sop | trd | udf )
components=
# Remote Execution ( Default: ssh; Valid values: ssh | saphostagent )
remote_execution=ssh
[Server]
# Enable the installation or upgrade of the SAP Host Agent ( Default: y )
install_hostagent=y
# Non-standard Shared File System
checkmnt=
# Installation Path ( Default: /hana/shared )
sapmnt=/hana/shared
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
42
# Local Host Name ( Default: hana001 )
hostname=hana001
# Install SSH Key ( Default: y )
install_ssh_key=y
# Root User Name ( Default: root )
root_user=root
# Root User Password
root_password=Admin@123
# Directory containing a storage configuration
storage_cfg=/hana/shared
# Internal Network Address
internal_network=
# SAP HANA System ID
sid=S01
# Instance Number
number=00
# Database Mode ( Default: single_container; Valid values: single_container |
multiple_containers )
db_mode=single_container
# Database Isolation ( Default: low; Valid values: low | high )
db_isolation=low
# System Usage ( Default: custom; Valid values: production | test | development |
custom )
system_usage=custom
# Location of Data Volumes ( Default: /hana/data/${sid} )
datapath=/hana/data/${sid}
# Location of Log Volumes ( Default: /hana/log/${sid} )
logpath=/hana/log/${sid}
# Restrict maximum memory allocation?
restrict_max_mem=
# Maximum Memory Allocation in MB
max_mem=
# Certificate Host Names
certificates_hostmap=
# SAP Host Agent User (sapadm) Password
sapadm_password=Password@123
...
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
43
9. After you complete editing, press Esc, enter :x, and press Enter to exit the configfile.cfg
file.
Step 2 Save the installation script in /hana/shared.
1. Run the following commands to copy the fusionStorageClient.py and
getdiskbyidentifier.sh files to /hana/shared:
cp ~/scale_out_script/fusionStorageClient.py /hana/shared
cp ~/scale_out_script/getdiskbyidentifier.sh /hana/shared
2. Run the following commands to convert the formats of fusionStorageClient.py and
getdiskbyidentifier.sh:
dos2unix /hana/shared/fusionStorageClient.py
dos2unix /hana/shared/getdiskbyidentifier.sh
3. Run the following command to identify the client.py file in the SAP HANA installation
directory:
find /hana/shared -name client.py
4. Run the following command to copy the client.py file into the directory where the
fusionStorageClient.py file is stored:
cp Source file Destination file
An example is provided as follows:
cp
/hana/shared/51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64/server/hdb
_ha/client.py /hana/shared/client.py
Step 3 Install the SAP HANA software.
1. Run the following command to switch to the SAP_HANA_DATABASE directory on
HANA ECS hana001:
cd /hana/shared/51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64
2. Run the following command to assign permissions to the directory:
chmod 777 -R /hana
3. Start the installation.
The parameters are as follows:
− Install the software on HANA ECS hana001 and ensure that no description of this
ECS is contained in the CLI.
− Every node must have the worker or standby role assigned.
Take five nodes as an example:
./hdblcm --action=install
--addhosts=hana002:role=worker,hana003:role=worker,hana004:role=worker,hana0
05:role=standby --configfile=/hana/shared/configfile.cfg
--ignore=check_signature_file
The following information is displayed:
...
Select additional components for installation:
...
4. Use default settings and press Enter.
The following information is displayed:
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
44
...
Enter comma-separated list of the selected indices [1]:
5. Use default settings and press Enter.
The following information is displayed:
...
Restrict maximum memory allocation? [n]:
...
6. Use default settings and press Enter.
The following information is displayed:
...
Enter Certificate Host Name For Host 'hana001' [hana001]:
7. Use default settings and press Enter.
The system displays a message, asking you to enter the Certificate Host Name of other
nodes. In such a case, use default settings until the system displays the following
information:
...
Enter System Administrator (s01adm) Password:
8. Set and confirm the password as prompted.
The following information is displayed:
...
Enter System Administrator User ID [1000]:
9. Use default settings and press Enter.
The following information is displayed:
...
Enter ID of User Group (sapsys) [79]:
10. Use default settings and press Enter.
The following information is displayed:
...
Enter Database User (SYSTEM) Password:
11. Set and confirm the password as prompted. The system displays the brief installation
information and a message asking you whether to continue the installation.
...
Do you want to continue? (y/n):
12. After verifying the installation, enter y and press Enter to start the SAP HANA software
installation.
13. After the installation is complete, the system displays the following information:
...
SAP HANA system installed.
Step 4 Check the software installation.
1. Run the following command to enter the SAP HANA system administrator mode:
su - $SIDadm
An example is provided as follows:
su - s01adm
2. Run an SQL query statement.
The format is as follows:
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
45
hdbsql -i Instance ID -u SYSTEM -p Password of the database administrator -C
"select * from M_LANDSCAPE_HOST_CONFIGURATION"
An example is provided as follows:
hdbsql -i 00 -u SYSTEM -p Password1234 -C "select * from
M_LANDSCAPE_HOST_CONFIGURATION"
If the system displays the queried data of each node, the SAP HANA software has been
installed.
----End
2.5.4 Installing the SAP HANA Studio on a Windows ECS
Scenarios
The SAP HANA Studio manages SAP HANA. After SAP HANA nodes are deployed, you
need to install the SAP HANA Studio and use it to manage the SAP HANA nodes.
Prerequisites Required resources are available. For details, see section 2.3 Preparing for Resources.
You have created ECSs, formatted disks attached to them, and installed the SAP HANA.
The firewall on the target ECS has been disabled.
Remote login to the target ECS has been enabled.
Procedure
Step 1 Use the Remote Desktop Protocol (RDP) and EIP to log in to the SAP HANA Studio ECS.
Use the username Administrator and the password obtained in section Obtaining the
Password for Logging In to a Windows ECS to log in to the SAP HANA Studio ECS.
Step 2 Download the SAP HANA Studio installation package at either of the following URLs:
https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part1.exe
https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part2.rar
https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part3.rar
Step 3 Decompress the installation package and navigate to the directory where SAP HANA Studio
is stored.
Step 4 On the Windows page, switch to the directory where the SAP HANA Studio installation
package is stored and double-click hdbsetup.exe to open the installation wizard page.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
46
Figure 2-7 SAP HANA Studio installation wizard page
Step 5 Select the installation path and click Next.
Step 6 On the Select Features page, select the features to be installed and click Next.
You are advised to select all features.
Step 7 Confirm all information on the Review & Confirm page and click Install.
Step 8 An installation page is displayed. Continue the installation. When the installation is complete,
the system displays the message "You have successfully installed the SAP HANA Studio."
Step 9 Click Finish.
----End
2.5.5 Installing the SAP HANA Studio on a Linux ECS
Scenarios
The SAP HANA Studio manages SAP HANA. After SAP HANA nodes are deployed, you
need to install the SAP HANA Studio and use it to manage the SAP HANA nodes.
This section describes how to install the SAP HANA Studio on a Linux ECS.
Prerequisites Required resources are available. For details, see section 2.3 Preparing for Resources.
You have created ECSs, formatted disks attached to them, and installed the SAP HANA.
The firewall on the target ECS has been disabled.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
47
Procedure
Step 1 Log in to the SAP HANA Studio ECS with an elastic IP address bound as user root using the
key file.
Step 2 On the CLI, run the following commands to download the SAP HANA installation package:
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part1.exe -P
/hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part2.rar -P
/hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part3.rar -P
/hana/shared
Step 3 Run the following command to decompress the installation package:
unrar x 51052383_part1.exe
Step 4 Navigate to the directory where the SAP HANA Studio is stored.
For example, SAP HANA Studio is stored in
/51052383/DATA_UNITS/HDB_STUDIO_LINUX_X86_64. Run the following command:
cd /51052383/DATA_UNITS/HDB_STUDIO_LINUX_X86_64
Step 5 Assign operation permissions to the directory where the installation packages are stored.
For example, if the directory is HDB_STUDIO_LINUX_X86_64, run the following
command:
chmod 777 -R HDB_STUDIO_LINUX_X86_64
Step 6 Switch to the directory and perform the installation. The installation page is displayed, as
shown in Figure 2-8.
./hdbsetup
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
48
Figure 2-8 SAP HANA Studio installation wizard page
Step 7 Select the installation path and click Next.
Step 8 On the Select Features page, select the features to be installed and click Next.
You are advised to select all features.
Step 9 Confirm all information on the Review & Confirm page and click Install.
Step 10 An installation page is displayed. Continue the installation. When the installation is complete,
the system displays the message "You have successfully installed the SAP HANA Studio."
Step 11 Click Finish.
Step 12 Go to Step 7 to select the installation path, edit the hdbstudio.ini file, and add parameters to
configure the GTK version.
vi hdbstudio.ini
Add the following parameters:
--launcher.GTK_version
2
An example is provided as follows:
-startup
plugins/org.eclipse.equinox.launcher_1.3.201.v20161025-1711.jar
--launcher.library
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
49
plugins/org.eclipse.equinox.launcher.gtk.linux.x86_64_1.1.401.v20161122-1740
--launcher.GTK_version
2
--launcher.XXMaxPermSize
512m
Step 13 (Optional) If the version is not configured in Step 12, perform the following operations before
starting hdbstudio on the Linux OS:
export SWT_GTK3=0
./hdbstudio
----End
2.5.6 Connecting SAP HANA Nodes to the SAP HANA Studio
Scenarios
SAP HANA nodes can be managed only after they are connected to the SAP HANA Studio.
This section uses a Windows ECS where the SAP HANA Studio is deployed as an example.
Procedure
Step 1 Start the SAP HANA Studio.
On the ECS where the SAP HANA Studio is deployed, choose Start > SAP HANA > SAP
HANA Studio. Then, the system displays the SAP HANA Studio management page and the
Workspace Launcher dialog box.
Figure 2-9 Workspace Launcher
Step 2 Specify the Workspace directory, select Use this as the default and do not ask me again,
and click OK.
Step 3 The Security Storage dialog box is displayed, as shown in Figure 2-10. Click No.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
50
Figure 2-10 Security Storage dialog box
Step 4 On the Overview page, click Open Administration Console to enter the SAP HANA
Administration Console page.
Step 5 Right-click the blank area under System, as shown in Figure 2-11.
Figure 2-11 SAP HANA Administration Console page
Step 6 Choose Add System. The Specify System page is displayed, as shown in Figure 2-12. Configure parameters.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
51
Key parameters are as follows:
Host Name: Enter the service or client plane IP address of the SAP HANA ECS. In the
cluster scenarios, you only need to enter the IP address of the first SAP HANA node, on
which installation has been performed.
Instance Number: Enter the number of the instance on the SAP HANA node.
Mode: Select a mode based on actual requirements. If HANA 2.0 is used, select
Multiple containers.
Figure 2-12 Specify System page
Step 7 Click Next to go to the System page, as shown in Figure 2-13. Choose Authentication by
database user and enter the username and password.
The username and password are those configured during SAP HANA software installation.
The username is consistently set to SYSTEM.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
52
Figure 2-13 System page
Step 8 Click Next and then Finish. Then, the SAP HANA Studio automatically connects to the
database.
If the connection fails, check whether the database instance on the SAP HANA node is
running.
Step 9 Under System of the SAP HANA Administration Console page, double-click the node to be
checked.
Step 10 Click the Landscape tab on the right of the page and check the status of each process on the
SAP HANA node.
Green indicates that the process is running properly.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
53
Figure 2-14 Landscape page
----End
2.5.7 Configuring the Backup Path
Scenarios
A path is required for backing up SAP HANA data.
This section uses a Windows ECS where the SAP HANA Studio is deployed as an example
when SAP HANA 1.0 is used.
Procedure
Step 1 On the HANA Studio ECS, choose Start > SAP HANA > SAP HANA Studio to start the
SAP HANA Studio software.
Step 2 In the System area on the left, right-click the database node and choose Backup and
Recovery > Open Backup Console, as shown in Figure 2-15.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
54
Figure 2-15 Open Backup Console menu
Step 3 Click the Configuration tab on the right and configure the backup path, as shown in Figure
2-16.
The backup path is the directory to which the backup volume is attached, that is, /hana/backup.
Click the save icon in the upper right corner to save the configuration.
Figure 2-16 Configuring the backup path
Step 4 In the System area on the left, right-click the database node and choose Backup and
Recovery > Back Up System.... The Specify Backup Settings page is displayed.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
55
Figure 2-17 Specify Backup Settings page
Step 5 Use default settings and click Next. The Review Backup Settings page is displayed. Confirm
configurations and click Finish. The system starts the backup task.
According to SAP requirements, data must be backed up immediately after the SAP HANA system is
installed.
Step 6 Click Close after the backup task is complete.
----End
2.5.8 Configuring SAP HANA Storage Parameters
Scenarios
Configure SAP HANA storage parameters based on SAP's requirements.
Only SAP HANA 1.0 needs to be configured because the default configurations of SAP HANA 2.0 meet the specified requirements.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
56
For more information, see as follows:
SAP Note 2186744 - FAQ: SAP HANA Parameters
SAP Note 2267798 - Configuration of the SAP HANA Database during Installation
Using hdbparam
SAP_HANA_Administration_Guide
SAP Note 2156526 - Parameter constraint validation on section indicies does not work
correctly with hdbparam
SAP Note 2399079 - Elimination of hdbparam in HANA 2
Procedure
Step 1 Log in to an SAP HANA node.
Step 2 Run the following command to switch to the SAP HANA administrator:
su - s00adm
Step 3 Configure SAP HANA storage parameters.
hdbparam --paramset fileio.async_read_submit=on
hdbparam --paramset fileio.async_write_submit_active=on
hdbparam --paramset fileio.async_write_submit_blocks=all
Step 4 (Optional) Configure storage parameters on other SAP HANA nodes according to the steps
provided in preceding paragraphs.
If multiple SAP HANA nodes exist, perform the same configuration for these parameters on
other SAP HANA nodes.
----End
2.6 Installing SAP HANA (Cluster Deployment and Two NFS Servers Deployed)
2.6.1 Configuring NFS Servers to Be Trusty to Each Other
Scenarios
Two NFS servers use HAE for active/standby deployment. HAE requires that the two NFS
servers be trusty to each other.
The two NFS servers have been configured to be trusty in section Configuring SSH Switching
Permissions. In this section, you only need to configure the mapping between the host names
of the HANA ECSs where the NFS servers are deployed and the cloud management or backup
plane IP addresses.
Procedure
Step 1 Modify the /etc/hosts file.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
57
Use PuTTY to log in to the NAT server with an elastic IP address bound. Ensure that user root
and the key file (.ppk file) are used for authentication. Then, use SSH to switch to an NFS
server.
Step 2 Modify the /etc/hosts file on the NFS server. Specifically, write the mapping between the
cloud management or backup plane IP addresses (for example, 10.0.1.x) of the two NFS
servers and the names of the HANA ECSs where the two NFS servers are deployed into this
file.
For details, see section Configuring the Mapping Between SAP HANA Host Name and IP
Addresses.
Step 3 Upload the local hosts file to the other NFS server.
The command is in the following format:
scp /etc/hosts Peer IP address:/etc/hosts
For example, if the peer IP address of the SAP HANA ECS is 10.0.3.102, run the following
command:
scp /etc/hosts 10.0.1.102:/etc/hosts
Step 4 Verify the trusty relationship.
For example, if the local NFS server is nfs01, and the name of the other NFS server is nfs02,
run the following commands:
ssh nfs02
ssh nfs01
When an SSH connection is set up between two ECSs (where the two NFS servers are deployed) for the
first time, the system displays the message "Are you sure you want to continue connecting (yes/no)?" on
the CLI. In such a case, you only need to select yes and press Enter.
----End
2.6.2 Formatting the Disks Attached to an NFS Server
Scenarios
The disks attached to an NFS server must be formatted to comply with the SAP HANA
installation requirements.
Procedure
Step 1 Attach the shared, backup, and SBD volumes to an NFS server.
This operation is performed only when two NFS servers are deployed to work in
active/standby mode. The reason is as follows: When one NFS server is created, the shared,
backup, and SBD volumes are attached to it. These disks must be attached to the other NFS
server.
1. On the management console, choose Computing > Elastic Cloud Server. On the left
side of the page, choose Elastic Cloud Server. Then, the system displays all ECSs on
the right side of the page.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
58
2. According to the HANA ECS name, locate the HANA ECS where the NFS server with
disks attached is deployed and click the HANA ECS name.
3. On the page providing details about the ECS, click the EVS tab and locate the first data
disk and click the target data disk.
4. On the page providing details about the data disk, view and record its mount point. Then,
click the data disk ID.
5. On the page that is displayed, click Mount Point and then Attach to switch to Attach
Disk page.
6. On the Attach Disk page, select the ECS on which the other NFS server is deployed,
ensure that the mount point of the ECS is the same as that of the ECS involved in Step
1.4, and attach the disk to the ECS.
7. Repeat Step 1.2 to Step 1.6 to attach all disks to the NFS server.
Step 2 Format the disks of the NFS servers.
1. Use PuTTY to log in to the NAT server with an elastic IP address bound. Ensure that
user root and the key file (.ppk file) are used for authentication. Then, use SSH to switch
to an NFS server.
2. Run the following command to query the disk to be formatted:
fdisk -l
3. Determine the disks of the shared, backup, and SBD volumes according to disk
capacities. The SBD volumes are used only when two NFS servers are deployed to work
in active/standby mode.
4. Run the following command to download the configuration script:
Download the readme.txt file at https://obs-sap.obs.myhwclouds.com/readme.txt on a local computer
to obtain the location where the software and configuration file are stored.
wget https://obs-sap.obs.myhwclouds.com/hana/cfgandscript/scale_out_script.zip -P
~
5. Run the following commands to switch to the folder and decompress the software
package:
cd ~
unzip scale_out_script.zip
6. Run the following command to modify the nfsserver.cfg file:
vi ~/scale_out_script/nfsserver.cfg
7. Press i to enter editing mode. Write the disk information and the SAP HANA ECS IP
addresses that are allowed to access the NFS servers into the nfsserver.cfg file.
The parameters are as follows:
− share: specifies the disk of the shared volume.
− backup: specifies the disk of the backup volume.
− sbd: specifies the disk of the SBD volume. This parameter is required only when
two NFS servers are deployed to work in active/standby mode.
− permithosts: specifies the SAP HANA ECS that is accessible to the NFS servers.
The IP address of this SAP HANA ECS is the cloud management or backup plane
IP address, for example, 10.0.1.x. When multiple IP addresses are available,
separate them with a comma (,).
− nfsServerIps: specifies the IP addresses (for example, 10.0.1.x) of the cloud management or backup planes for the NFS servers. When two NFS servers are
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
59
deployed to work in active/standby mode, set this parameter value to the server IP
addresses and separate them with a comma (,).
The content of the configuration file is as follows when only one NFS server is
deployed:
share=/dev/sda
backup=/dev/sdb
sbd=
permithosts=10.0.1.2,10.0.1.3,10.0.1.4,10.0.1.5,10.0.1.6
nfsServerIps=10.0.1.101
The content of the configuration file is as follows when two NFS servers are deployed to
work in active/standby mode:
share=/dev/sda
backup=/dev/sdb
sbd=/dev/sdc
permithosts=10.0.1.2,10.0.1.3,10.0.1.4,10.0.1.5,10.0.1.6
nfsServerIps=10.0.1.101,10.0.1.102
8. After you complete editing, press Esc, enter :x, and press Enter to exit the nfsserver.cfg
file.
9. Run the following commands to assign execute permissions, convert the file format, and
format the disk:
cd ~/scale_out_script
chmod 777 nfsserver.sh nfsserver.cfg
dos2unix nfsserver.sh nfsserver.cfg
sh nfsserver.sh
Step 3 Configure the nfsserver.sh configuration script not to automatically start when the NFS
server starts.
This operation is performed only when two NFS servers are deployed to work in
active/standby mode. Ensure that both the NFS servers require configuration.
When the NFS servers run the SLES 11 SP4 OS, run the following commands:
chkconfig nfsserver off
service nfsserver stop
When the NFS servers run the SLES 12 SP1 OS, run the following commands:
systemctl disable nfsserver.service
systemctl stop nfsserver.service
----End
2.6.3 Configuring HA on NFS Servers (SLES 12 SP1)
Scenarios
When NFS servers are deployed to work in active/standby mode, you must enable data
synchronization on the NFS servers so that they can properly support SAP HANA.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
60
Prerequisites
You have configured NFS servers to be trusty to each other and formatted to the disks
attached to NFS servers. For details, see sections Configuring NFS Servers to Be Trusty to
Each Other and Formatting the Disks Attached to an NFS Server.
Procedure
Step 1 Initialize the cluster.
1. Use PuTTY to log in to the NAT server with an elastic IP address bound. Ensure that
user root and the key file (.ppk file) are used for authentication. Then, use SSH to switch
to the NFS server that will work as the active node. The name of the NFS server is
assumed to be nfs01.
2. Run the following command to enter the cluster configuration page:
ha-cluster-init
3. When the system displays the following information, enter y and press Enter:
WARNING: NTP is not configured to start at system boot.
WARNING: No watchdog device found. If SBD is used, the cluster will be unable
to start without a watchdog.
Do you want to continue anyway? [y/N]
4. When the system displays the following information, enter n and press Enter:
/root/.ssh/id_rsa already exists - overwrite? [y/N]
5. Enter the internal NFS server communication network segment and press Enter.
Take network segment 10.0.2.0 as an example:
Network address to bind to (e.g.: 192.168.1.0) []
6. Retain the default settings and press Enter.
Multicast address (e.g.: 239.x.x.x) [239.97.171.218]
7. Retain the default settings and press Enter.
Multicast port [5405]
8. When the system displays the following information, enter n and press Enter:
Do you wish to use SBD? [y/N]
9. When the system displays the following information, enter n and press Enter:
To ensure system security, change the hacluster password after configuring HA.
...
Log in with username 'hacluster', password 'linux'
WARNING: You should change the hacluster password to something more secure!
Enabling pacemaker.service
Waiting for cluster........done
Loading initial configuration
...
Do you wish to configure an administration IP? [y/N]
10. During the cluster initialization, the system displays the following information:
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
61
Done (log saved to /var/log/ha-cluster-bootstrap.log)
Step 2 Add the standby NFS server to the cluster.
1. On active NFS server nfs01, synchronize the corosync.conf configuration file to the
standby NFS server.
To do so, run the following command:
scp /etc/corosync/corosync.conf Name of the standby NFS server:/etc/corosync/
An example is provided as follows:
scp /etc/corosync/corosync.conf nfs02:/etc/corosync/
2. Log in to the standby NFS server from the active one using SSH.
ssh nfs02
3. Run the following command to add the standby NFS server to the cluster:
ha-cluster-join
4. When the system displays the following information, enter y and press Enter:
WARNING: NTP is not configured to start at system boot.
WARNING: No watchdog device found. If SBD is used, the cluster will be unable
to start without a watchdog.
Do you want to continue anyway? [y/N]
5. When the system displays the following information, enter the internal communication
plane IP address of the active NFS server, for example, 10.0.2.101, and press Enter:
IP address or hostname of existing node (e.g.: 192.168.1.1) []
6. When the system displays the following information, enter n and press Enter:
/root/.ssh/id_rsa already exists - overwrite? [y/N]
7. During the cluster initialization after the standby NFS server is added to the cluster, the
system displays the following information:
......
WARNING: You should change the hacluster password to something more secure!
Enabling pacemaker.service
Waiting for cluster........done
Done (log saved to /var/log/ha-cluster-bootstrap.log)
Step 3 Configure basic cluster parameters.
1. Use VNC and user root password to log in to the active NFS server as user root and
switch to the CLI mode.
2. Run the yast2 command to enter the YaST2 Control Center page.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
62
Figure 2-18 YaST2 Control Center page
3. Choose High Availability > Cluster. The system displays a message, asking you to
install the hawk software package, as shown in Figure 2-19. Select Cancel and press
Enter.
Figure 2-19 Installing the hawk software package
4. The system displays the message "If you continue without installing required packages,
YaST may not work properly." Select Continue and press Enter. The configuration
information page of the cluster is displayed.
5. On the left side of the page, choose Communication Channels. Then, configure
parameters on the right side of the page. Figure 2-20 shows an example.
The parameters are as follows:
− Transport: Set this parameter to Unicast.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
63
− In the Channel pane, select the network segment for the NFS server heartbeat link,
which is the network segment of the internal communication plane, for example,
10.0.2.0, for Bind Network Address; retain the default setting of Multicast Port.
− Auto Generate Node ID: Select this parameter.
− Redundant Channel: Select this parameter. Select the network segment for the
cloud management or backup plane, for example, 10.0.1.0, for Bind Network
Address. Ensure that the Multicast Port value is the same as that specified in the
Channel pane.
− Member Address: Use Add to add the internal communication plane IP addresses
as well as the cloud management or backup plane IP addresses of the active and
standby NFS servers to Member Address.
Figure 2-20 Communication Channels page
6. On the left side of the page, choose Security. Then, configure parameters on the right
side of the page.
Enable Security Auth: Select this parameter.
Select Generate Auth Key File and press Enter. In the dialog box showing the file
generation message, select OK and press Enter to generate the authentication key file.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
64
Figure 2-21 Security page
7. On the left side of the page, choose Configure Csync2. Then, configure parameters on
the right side of the page.
Select a Sync Host node and Generate Pre-Shared-Keys and press Enter to generate
the file. In addition, ensure that Turn csync2 OFF in the lower right corner of the page
is displayed, indicating that Csync2 has been enabled.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
65
Figure 2-22 Configure Csync2 page
8. On the left side of the page, choose Service. Then, configure parameters on the right side
of the page.
The parameters are as follows:
− Booting: On is selected by default, which indicates that the cluster is automatically
started when the server starts.
− Switch on and off: Ensure that the cluster has been started.
− Firewall Settings: Ensure that Firewall is disabled is selected.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
66
Figure 2-23 Service page
9. Select Finish and press Enter to complete the configuration.
10. Copy the key_hagroup and csync2.cfg files on the active NFS server to the standby
NFS server.
Log in to the active NFS server as user root and run the following command:
scp /etc/csync2/key_hagroup /etc/csync2/csync2.cfg Name of the standby NFS
server:/etc/csync2
An example is provided as follows:
scp /etc/csync2/key_hagroup /etc/csync2/csync2.cfg nfs02:/etc/csync2
11. Run the following commands on the active NFS server:
systemctl enable csync2.socket
systemctl enable xinetd
systemctl restart xinetd
12. Log in to the standby NFS server as user root and run the following commands:
systemctl enable csync2.socket
systemctl enable xinetd
systemctl restart xinetd
13. Run the following command on the active NFS server to synchronize the configuration
file:
csync2 -xv
14. Run the following command on the active NFS server to check the cluster status:
systemctl status pacemaker
Ensure that the cluster status is active (running).
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
67
If the terminal display shows that Stonith has not been configured, ignore the message because Stonith
will be configured in follow-up operations.
15. Run the following commands on the active NFS server to restart the cluster:
systemctl stop pacemaker
systemctl start pacemaker
16. Run the following command on the standby NFS server to check the cluster status:
systemctl status pacemaker
Ensure that the cluster status is active (running).
If the terminal display shows that Stonith has not been configured, ignore the message because Stonith
will be configured in follow-up operations.
17. Run the following commands on the standby NFS server to restart the cluster:
systemctl stop pacemaker
systemctl start pacemaker
18. Run the following command on the active NFS server to check the cluster heartbeat
status:
corosync-cfgtool -s
The system displays the heartbeats of the active and standby links. Ensure that the
configurations are correct. An example is provided as follows:
Printing ring status.
Local node ID 167772739
RING ID 0
id= 10.0.2.101
status= ring 0 active with no faults
RING ID 1
id= 10.0.1.101
status= ring 1 active with no faults
19. Check the cluster status on the active NFS server.
crm_mon -r1
The terminal display is as follows, where two NFS servers are in Online state, indicating
that the cluster is functional:
Last updated: Wed Oct 19 17:40:31 2016
Last change: Tue Oct 18 15:32:00 2016 by root via cibadmin on hana01
Stack: classic openais (with plugin)
Current DC: nfs01 - partition with quorum
Version: 1.1.12-f47ea56
2 Nodes configured, 2 expected votes
0 Resources configured
Online: [ nfs01 nfs02 ]
Step 4 Configure the softdog.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
68
1. Run the following command on the active NFS server to open the file and press i to enter
editing mode:
vi /etc/init.d/boot.local
2. Add the following information to the last line of the file:
modprobe softdog >/dev/null 2>&1
3. Press Esc, enter :x!, and press Enter to exit the configuration.
4. On the CLI, run the following command:
modprobe softdog
5. Repeat Step 4.1 to Step 4.4 to configure the softdog on the standby NFS server.
Step 5 Configure SBD on the Fence devices.
1. Run the following command on the active NFS server to create an SBD partition:
sbd -d /dev/disk/by-id/Partition ID create
To query an SDB disk ID, run the ll /dev/disk/by-id |grep Partition name|grep scsi-3 command. For
example, if the partition name is /dev/sdb1, run the ll /dev/disk/by-id |grep sdb|grep scsi-3 command
to query the ID corresponding to sdb1.
An example is provided as follows:
sbd -d /dev/disk/by-id/scsi-36888603000000db7fa179ea56210049-part1 create
2. Run the following command to allocate the partition to the active NFS server:
sbd -d /dev/disk/by-id/Partition ID allocate Active node host name
An example is provided as follows:
sbd -d /dev/disk/by-id/scsi-36888603000000db7fa179ea56210049-part1 allocate
nfs01
3. Run the following command to allocate the partition to the standby NFS server:
sbd -d /dev/disk/by-id/Partition ID allocate Standby node host name
An example is provided as follows:
sbd -d /dev/disk/by-id/scsi-36888603000000db7fa179ea56210049-part1 allocate
nfs02
4. Run the following command to query the allocated partition:
sbd -d /dev/disk/by-id/Partition ID list
An example is provided as follows:
sbd -d /dev/disk/by-id/scsi-36888603000000db7fa179ea56210049-part1 list
The terminal display is as follows:
0 nfs01 clear
1 nfs02 clear
5. Run the following command to dump the partition:
sbd -d /dev/disk/by-id/Partition ID dump
An example is provided as follows:
sbd -d /dev/disk/by-id/scsi-36888603000000db7fa179ea56210049-part1 dump
6. Run the vi /etc/sysconfig/sbd command to open the file and add data to the file.
An example is provided as follows:
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
69
Enter the SBD partition in SBD_DEVICE. The partition in the following terminal display is for
reference only.
SBD_DEVICE="/dev/disk/by-id/scsi-36888603000000db7fa179ea56210049-part1"
SBD_OPTS="-W"
7. On active NFS server, synchronize the configuration file to the standby NFS server.
scp /etc/sysconfig/sbd Standby node host name:/etc/sysconfig
An example is provided as follows:
scp /etc/sysconfig/sbd nfs02:/etc/sysconfig
8. Run the following commands on the active NFS server to restart the cluster:
systemctl stop pacemaker
systemctl start pacemaker
9. Run the following commands on the standby NFS server to restart the cluster:
systemctl stop pacemaker
systemctl start pacemaker
10. Check cluster configurations.
crm_mon -r1
The terminal display is as follows, where two NFS servers are in Online state, indicating
that the cluster is functional:
Last updated: Wed Oct 19 17:40:31 2016
Last change: Tue Oct 18 15:32:00 2016 by root via cibadmin on hana01
Stack: classic openais (with plugin)
Current DC: nfs01 - partition with quorum
Version: 1.1.12-f47ea56
2 Nodes configured, 2 expected votes
0 Resources configured
Online: [ nfs01 nfs02 ]
Step 6 Add Fence SBD resources.
1. Run the following command on the active NFS server to switch to the CRM
configuration mode:
crm
Information similar to the following is displayed:
crm(live)#
2. Run the following command to switch to the configuration mode:
configure
3. Run the following commands to add SBD resources:
primitive r_sbd stonith:external/sbd params pcmk_delay_max=40
commit
Step 7 Assign a virtual service IP address.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
70
1. On the management console, choose Computing > Elastic Cloud Server. On the left
side of the page, choose Elastic Cloud Server to switch to the Elastic Cloud Server
page.
2. Locate the HANA ECS on which an NFS server is deployed and click the HANA ECS
name. Then, the page for the HANA ECS details is displayed.
3. Click the NIC tab and then Manage Private IP in the row of the cloud management or
backup plane NIC. Then, the system displays the Virtual IP Address page.
4. Click Assign Virtual IP Address to assign a floating IP address. Locate the row that
contains the target IP address, click Bind to Server to bind the IP address to the target
ECS, and repeat this operation to bind the IP address to other ECSs.
5. Run the following commands on the active NFS server in CRM configuration mode:
primitive nfs_vip ocf:heartbeat:IPaddr2 params ip=10.0.1.103 op monitor
interval="10" timeout="20" on-fail=restart
commit
In the preceding commands, params ip is the virtual service IP address of the NFS
server. This value is for reference only.
Step 8 Create cluster shared volume file systems.
Run the following commands on the active NFS server to create the cluster shared disk file
system for the shared and backup volumes, respectively:
primitive nfs_shared ocf:heartbeat:Filesystem params
device="/dev/disk/by-id/scsi-368886030000000fffa17e21071682979-part1"
directory="/shared" fstype=xfs op monitor interval="10" timeout="40"
on-fail=restart
primitive nfs_bak ocf:heartbeat:Filesystem params
device="/dev/disk/by-id/scsi-36888603000000120fa17e21071682979-part1"
directory="/backup" fstype=ext3 op monitor interval="10" timeout="40"
on-fail=restart
commit
In the preceding commands, params device and directory specify the IDs and paths of the
shared and backup disks, respectively, and the values provided are for reference only. For
instructions about how to query a disk ID, see Step 5.1.
Step 9 On the active NFS server in CRM configuration mode, create the nfsserver resource.
If the floating IP address (specified by nfs_ip) of the NFS server is 10.0.1.103, run the
following commands:
primitive nfsserver nfsserver params nfs_ip=10.0.1.103 nfs_shared_infodir="/shared"
nfs_shared_infodir="/backup" operations $id=nfsserver-operations op monitor
interval="10" timeout="20" on-fail=restart
commit
Step 10 Run the following commands on the active NFS server in CRM configuration mode to create
resource group nfs_group for the resource created in Step 7.5 to Step 9:
group nfs_group nfs_vip nfs_shared nfs_bak nfsserver
commit
Step 11 Run the following commands on the active NFS server in CRM configuration mode to create
Clone Ping:
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
71
primitive r_ping ocf:pacemaker:ping params multiplier="100" dampen="5"
host_list="10.0.1.1" op monitor interval="15" timeout="60" start="60" op start
interval="0" timeout="60"
clone PING r_ping
commit
The parameters are as follows:
r_ping: specifies the Clone Ping resource name.
Multiplier: an efficiency enhancer that uses the benchmark of 100. When packet loss
occurs during the ping operation between two nodes, the multiplier decreases a certain
value from 100 to determine the service NIC communication status of the nodes.
dampen: indicates that the ping operation is performed once every 5s.
10.0.1.1: indicates the gateway IP address of the cloud management or backup plane
network segment for the NFS servers.
monitor: monitors resources; interval: indicates that if pinging the upper-layer switch
fails within 15s, the system will migrate resources at most 60s (timeout) later.
Step 12 Run the following commands on the active NFS server in CRM configuration mode to create
restraint nfs_conn on resource group nfs_group:
location nfs_conn nfs_group rule -inf: not_defined pingd or pingd lte 0
commit
Step 13 Configure other information.
1. Configure the Stonith parameter:
property startup-fencing=true
property stonith-enabled=true
property stonith-action=off
commit
2. Configure the failure timeout duration:
rsc_defaults failure-timeout=600
commit
3. Press Ctrl+C on the active NFS server to exit the CRM configuration mode.
----End
2.6.4 Formatting an SAP HANA Disk
Scenarios
In SAP HANA cluster scenarios, all data volumes and log volumes form a data volume
resource pool and a log volume resource pool, respectively, for all SAP HANA nodes. In this
way, every data or log volume is attached to all SAP HANA nodes.
After the attachment, the data disks, including the data and log volumes and the shared and
backup volumes on the NFS servers, can be used only after they are formatted and attached to
required directories.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
72
Procedure
Step 1 Attach all data and log volumes to all SAP HANA nodes.
1. On the management console, choose Computing > Bare Meta Server. On the left side
of the page, choose Bare Metal Server. Then, the system displays all BMSs on the right
side of the page.
2. Locate the first BMS by name and click its name.
3. On the page providing details about the ECS, click the EVS tab and find the data disk to
be bound to other SAP HANA nodes (log volume or data volume). Then, click the data
disk.
4. In the expanded data disk pane, click the link following the data disk ID.
5. On the page that is displayed, click Mount Point and then Attach to switch to the
Attach Disk page.
6. On the Attach Disk page, select all the target SAP HANA nodes and attach the disk to
the SAP HANA nodes.
Ensure that the disk is attached to all SAP HANA nodes.
7. Perform the preceding operations to attach the second disk (data volume or log volume)
of the SAP HANA node to all other SAP HANA nodes.
8. Perform the preceding operations to attach the log and data volumes of each SAP HANA
node to all other SAP HANA nodes.
The last SAP HANA node is a standby node. When creating this BMS, you did not create or attach log
and data volumes to it. Therefore, you do not need to attach the disks of this BMS to other SAP HANA
nodes.
Step 2 Format the disks attached to an SAP HANA node.
1. Log in to the NAT server with an elastic IP address bound. Ensure that user root and the
key file are used for authentication. Then, use SSH to switch to the first SAP HANA
node and run the following command to check the disks that are not formatted:
fdisk -l
2. Determine the paths for the data and log volumes according to the disk capacities.
3. View the IP address (for example, 10.0.1.x) of the cloud management or backup plane
for the NFS server.
For details, see section A.4 Querying the NIC IP Address of an ECS.
4. View the internal communication plane IP addresses (for example, 10.0.2.x) of all SAP
HANA nodes.
For details, see section A.3 Querying the NIC IP Address of a BMS.
5. Download the configuration script.
wget https://obs-sap.obs.myhwclouds.com/hana/cfgandscript/scale_out_script.zip -P
~
6. Run the following commands to switch to the folder and decompress the software
package:
cd ~
unzip scale_out_script.zip
7. Run the following command to modify the hanaCluster.cfg file:
vi ~/scale_out_script/hanaCluster.cfg
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
73
8. Press i to enter editing mode. Write the path, cloud management or backup plane IP
address of the NFS server, and internal communication plane IP address of all SAP
HANA nodes into the hanaCluster.cfg file.
Parameters are as follows:
− partition_1_data__volname: specifies the path for the data volume attached to the
first SAP HANA node. Apply this rule to other data volumes. A standby node has
no data volumes attached and does not require recording.
− partition_1_log__volname: specifies the path for the log volume attached to the
first SAP HANA node. Apply this rule to other log volumes. A standby node has no
log volumes attached and does not require recording.
− usrsaps: You do not need to configure this parameter.
− shareFSType: Set this parameter to NFS.
− sfsShared: You do not need to configure this parameter.
− sfsBackup: You do not need to configure this parameter.
− nfsserverIp: specifies the IP address (for example, 10.0.1.x) of the cloud
management or backup plane for the NFS server. When two NFS servers are
deployed to work in active/standby mode, the value of this parameter is the floating
IP address of the NFS servers.
− sid: specifies the SAP HANA system ID. After the configuration, ensure that the
system ID is the same as this value when installing SAP HANA.
− hosts: specifies the internal communication plane IP addresses (for example,
10.0.2.x) of all SAP HANA nodes. When multiple IP addresses are available,
separate them with a comma (,).
Take five SAP HANA nodes, four worker nodes and one standby node, as an example.
Ensure that every volume path is unique.
partition_1_data__volname=/dev/sda
partition_1_log__volname=/dev/sdb
partition_2_data__volname=/dev/sdc
partition_2_log__volname=/dev/sdd
partition_3_data__volname=/dev/sde
partition_3_log__volname=/dev/sdf
partition_4_data__volname=/dev/sdg
partition_4_log__volname=/dev/sdh
usrsaps=
nfsserverIp=10.0.1.103
shareFSType=NFS
sfsShared=
sfsBackup=
sid=S01
hosts=10.0.2.2,10.0.2.3,10.0.2.4,10.0.2.5,10.0.2.6
9. After you complete editing, press Esc, enter :x, and press Enter to exit the
hanaCluster.cfg file.
10. Run the following commands to switch to the directory where the hanaCluster.cfg file
is stored and convert the file format:
cd ~/scale_out_script
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
74
dos2unix installHanaCluster.sh hanaCluster.cfg
11. Assign operation permissions to the hanaCluster.cfg file and run the
installHanaCluster.sh command to format disks.
chmod 777 installHanaCluster.sh hanaCluster.cfg
sh installHanaCluster.sh
----End
2.6.5 Installing the SAP HANA Software
Scenarios
The SAP HANA software is installed using an installation script. This section uses the SAP
HANA SPS12 installation package as an example. You can download the package from the
official website.
You only need to install SAP HANA on one SAP HANA node. Ensure that the SAP HANA
node cannot be the planned standby node.
Prerequisites Required resources are available. For details, see section 2.3 Preparing for Resources.
You have created servers, formatted disks attached to them, and completed required
configurations.
You have synchronized the time of the ECS.
The firewall on the ECS where SAP HANA is to be installed has been disabled.
Procedure
Step 1 Generate the configfile.cfg file required for the installation.
1. Use PuTTY to log in to the NAT Server using an EIP. Ensure that user root and the key
file (.ppk file) are used for authentication. Use the SSH protocol to switch to the SAP
HANA node on which the SAP HANA software is to be installed. Ensure that the SAP
HANA node is the one to which the public key file id_rsa.pub is uploaded when SSH
switching permissions are configured. Assume that the host name is hana001.
2. On the CLI, run the following commands to download the SAP HANA installation
package:
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part1.exe
-P /hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part2.rar
-P /hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part3.rar
-P /hana/shared
3. Run the following command to decompress the installation package and navigate to the
directory where the installation file locates:
unrar x 51052383_part1.exe
4. Enter the directory where the installation file is stored.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
75
For example, if the installation file is stored in
/51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64, run the following
command:
cd /51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64
5. Run the following command to assign permissions to the directory:
chmod 777 -R /hana
6. Run the following command to obtain the configfile.cfg file:
./hdblcm --action=install --dump_configfile_template=/hana/shared/configfile.cfg
After this command is executed, the configfile.cfg file is stored in the /hana/shared
directory.
7. Run the following command and press i to enter the editing mode and edit the
configfile.cfg file:
vi /hana/shared/configfile.cfg
8. Modify parameter settings.
The parameters are as follows:
− remote_execution: specifies a remote execution mode, the default value of which is
ssh.
− sapmnt: specifies the installation path, which is /hana/shared.
− hostname: specifies the ECS name, such as hana001.
− root_password: specifies the password of user root for logging in to the HANA
ECS.
− storage_cfg: specifies the path in which the configfile.cfg file is stored. Set the
value of this parameter to /hana/shared.
− sid: specifies the SAP HANA system ID, which must be the same as the sid value
configured in hanaCluster.cfg when the disk is formatted.
− number: specifies the SAP HANA instance ID. The instance ID is used in 2.2.3
Network Planning, which must be the same as the planned one.
− sapadm_password: specifies the password of the HostAgent software package
installed on the HANA ECS. Configure this parameter based on service
requirements.
An example is provided as follows:
[General]
# Location of SAP HANA Installation Medium
component_medium=
# Comma separated list of component directories
component_dirs=
# Directory root to search for components
component_root=
# Components ( Valid values: all | client | es | ets | lcapps | server | smartda
| streaming | rdsync | studio | afl | pos | sal | sca | sop | trd | udf )
components=
# Remote Execution ( Default: ssh; Valid values: ssh | saphostagent )
remote_execution=ssh
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
76
[Server]
# Enable the installation or upgrade of the SAP Host Agent ( Default: y )
install_hostagent=y
# Non-standard Shared File System
checkmnt=
# Installation Path ( Default: /hana/shared )
sapmnt=/hana/shared
# Local Host Name ( Default: hana001 )
hostname=hana001
# Install SSH Key ( Default: y )
install_ssh_key=y
# Root User Name ( Default: root )
root_user=root
# Root User Password
root_password=Admin@123
# Directory containing a storage configuration
storage_cfg=/hana/shared
# Internal Network Address
internal_network=
# SAP HANA System ID
sid=S01
# Instance Number
number=00
# Database Mode ( Default: single_container; Valid values: single_container |
multiple_containers )
db_mode=single_container
# Database Isolation ( Default: low; Valid values: low | high )
db_isolation=low
# System Usage ( Default: custom; Valid values: production | test | development |
custom )
system_usage=custom
# Location of Data Volumes ( Default: /hana/data/${sid} )
datapath=/hana/data/${sid}
# Location of Log Volumes ( Default: /hana/log/${sid} )
logpath=/hana/log/${sid}
# Restrict maximum memory allocation?
restrict_max_mem=
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
77
# Maximum Memory Allocation in MB
max_mem=
# Certificate Host Names
certificates_hostmap=
# SAP Host Agent User (sapadm) Password
sapadm_password=Password@123
...
9. After you complete editing, press Esc, enter :x, and press Enter to exit the configfile.cfg
file.
Step 2 Save the installation script in /hana/shared.
1. Run the following commands to copy the fusionStorageClient.py and
getdiskbyidentifier.sh files to /hana/shared:
cp ~/scale_out_script/fusionStorageClient.py /hana/shared
cp ~/scale_out_script/getdiskbyidentifier.sh /hana/shared
2. Run the following commands to convert the formats of fusionStorageClient.py and
getdiskbyidentifier.sh:
dos2unix /hana/shared/fusionStorageClient.py
dos2unix /hana/shared/getdiskbyidentifier.sh
3. Run the following command to identify the client.py file in the SAP HANA installation
directory:
find /hana/shared -name client.py
4. Run the following command to copy the client.py file into the directory where the
fusionStorageClient.py file is stored:
cp Source file Destination file
An example is provided as follows:
cp
/hana/shared/51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64/server/hdb
_ha/client.py /hana/shared/client.py
Step 3 Install the SAP HANA software.
1. Run the following command to switch to the SAP_HANA_DATABASE directory on
HANA ECS hana001:
cd /hana/shared/51052383/DATA_UNITS/HDB_SERVER_LINUX_X86_64
2. Run the following command to assign permissions to the directory:
chmod 777 -R /hana
3. Start the installation.
The parameters are as follows:
− Install the software on HANA ECS hana001 and ensure that no description of this
ECS is contained in the CLI.
− Every node must have the worker or standby role assigned.
Take five nodes as an example:
./hdblcm --action=install
--addhosts=hana002:role=worker,hana003:role=worker,hana004:role=worker,hana0
05:role=standby --configfile=/hana/shared/configfile.cfg
--ignore=check_signature_file
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
78
The following information is displayed:
...
Select additional components for installation:
...
4. Use default settings and press Enter.
The following information is displayed:
...
Enter comma-separated list of the selected indices [1]:
5. Use default settings and press Enter.
The following information is displayed:
...
Restrict maximum memory allocation? [n]:
...
6. Use default settings and press Enter.
The following information is displayed:
...
Enter Certificate Host Name For Host 'hana001' [hana001]:
7. Use default settings and press Enter.
The system displays a message, asking you to enter the Certificate Host Name of other
nodes. In such a case, use default settings until the system displays the following
information:
...
Enter System Administrator (s01adm) Password:
8. Set and confirm the password as prompted.
The following information is displayed:
...
Enter System Administrator User ID [1000]:
9. Use default settings and press Enter.
The following information is displayed:
...
Enter ID of User Group (sapsys) [79]:
10. Use default settings and press Enter.
The following information is displayed:
...
Enter Database User (SYSTEM) Password:
11. Set and confirm the password as prompted. The system displays the brief installation
information and a message asking you whether to continue the installation.
...
Do you want to continue? (y/n):
12. After verifying the installation, enter y and press Enter to start the SAP HANA software
installation.
13. After the installation is complete, the system displays the following information:
...
SAP HANA system installed.
Step 4 Check the software installation.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
79
1. Run the following command to enter the SAP HANA system administrator mode:
su - $SIDadm
An example is provided as follows:
su - s01adm
2. Run an SQL query statement.
The format is as follows:
hdbsql -i Instance ID -u SYSTEM -p Password of the database administrator -C
"select * from M_LANDSCAPE_HOST_CONFIGURATION"
An example is provided as follows:
hdbsql -i 00 -u SYSTEM -p Password1234 -C "select * from
M_LANDSCAPE_HOST_CONFIGURATION"
If the system displays the queried data of each node, the SAP HANA software has been
installed.
----End
2.6.6 Installing the SAP HANA Studio on a Windows ECS
Scenarios
The SAP HANA Studio manages SAP HANA. After SAP HANA nodes are deployed, you
need to install the SAP HANA Studio and use it to manage the SAP HANA nodes.
Prerequisites Required resources are available. For details, see section 2.3 Preparing for Resources.
You have created ECSs, formatted disks attached to them, and installed the SAP HANA.
The firewall on the target ECS has been disabled.
Remote login to the target ECS has been enabled.
Procedure
Step 1 Use the Remote Desktop Protocol (RDP) and EIP to log in to the SAP HANA Studio ECS.
Use the username Administrator and the password obtained in section Obtaining the
Password for Logging In to a Windows ECS to log in to the SAP HANA Studio ECS.
Step 2 Download the SAP HANA Studio installation package at either of the following URLs:
https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part1.exe
https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part2.rar
https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part3.rar
Step 3 Decompress the installation package and navigate to the directory where SAP HANA Studio
is stored.
Step 4 On the Windows page, switch to the directory where the SAP HANA Studio installation
package is stored and double-click hdbsetup.exe to open the installation wizard page.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
80
Figure 2-24 SAP HANA Studio installation wizard page
Step 5 Select the installation path and click Next.
Step 6 On the Select Features page, select the features to be installed and click Next.
You are advised to select all features.
Step 7 Confirm all information on the Review & Confirm page and click Install.
Step 8 An installation page is displayed. Continue the installation. When the installation is complete,
the system displays the message "You have successfully installed the SAP HANA Studio."
Step 9 Click Finish.
----End
2.6.7 Installing the SAP HANA Studio on a Linux ECS
Scenarios
The SAP HANA Studio manages SAP HANA. After SAP HANA nodes are deployed, you
need to install the SAP HANA Studio and use it to manage the SAP HANA nodes.
This section describes how to install the SAP HANA Studio on a Linux ECS.
Prerequisites Required resources are available. For details, see section 2.3 Preparing for Resources.
You have created ECSs, formatted disks attached to them, and installed the SAP HANA.
The firewall on the target ECS has been disabled.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
81
Procedure
Step 1 Log in to the SAP HANA Studio ECS with an elastic IP address bound as user root using the
key file.
Step 2 On the CLI, run the following commands to download the SAP HANA installation package:
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part1.exe -P
/hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part2.rar -P
/hana/shared
wget https://obs-sap.obs.myhwclouds.com/hana/software/sps12/51052383_part3.rar -P
/hana/shared
Step 3 Run the following command to decompress the installation package:
unrar x 51052383_part1.exe
Step 4 Navigate to the directory where the SAP HANA Studio is stored.
For example, SAP HANA Studio is stored in
/51052383/DATA_UNITS/HDB_STUDIO_LINUX_X86_64. Run the following command:
cd /51052383/DATA_UNITS/HDB_STUDIO_LINUX_X86_64
Step 5 Assign operation permissions to the directory where the installation packages are stored.
For example, if the directory is HDB_STUDIO_LINUX_X86_64, run the following
command:
chmod 777 -R HDB_STUDIO_LINUX_X86_64
Step 6 Switch to the directory and perform the installation. The installation page is displayed, as
shown in Figure 2-25.
./hdbsetup
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
82
Figure 2-25 SAP HANA Studio installation wizard page
Step 7 Select the installation path and click Next.
Step 8 On the Select Features page, select the features to be installed and click Next.
You are advised to select all features.
Step 9 Confirm all information on the Review & Confirm page and click Install.
Step 10 An installation page is displayed. Continue the installation. When the installation is complete,
the system displays the message "You have successfully installed the SAP HANA Studio."
Step 11 Click Finish.
Step 12 Go to Step 7 to select the installation path, edit the hdbstudio.ini file, and add parameters to
configure the GTK version.
vi hdbstudio.ini
Add the following parameters:
--launcher.GTK_version
2
An example is provided as follows:
-startup
plugins/org.eclipse.equinox.launcher_1.3.201.v20161025-1711.jar
--launcher.library
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
83
plugins/org.eclipse.equinox.launcher.gtk.linux.x86_64_1.1.401.v20161122-1740
--launcher.GTK_version
2
--launcher.XXMaxPermSize
512m
Step 13 (Optional) If the version is not configured in Step 12, perform the following operations before
starting hdbstudio on the Linux OS:
export SWT_GTK3=0
./hdbstudio
----End
2.6.8 Connecting SAP HANA Nodes to the SAP HANA Studio
Scenarios
SAP HANA nodes can be managed only after they are connected to the SAP HANA Studio.
This section uses a Windows ECS where the SAP HANA Studio is deployed as an example.
Procedure
Step 1 Start the SAP HANA Studio.
On the ECS where the SAP HANA Studio is deployed, choose Start > SAP HANA > SAP
HANA Studio. Then, the system displays the SAP HANA Studio management page and the
Workspace Launcher dialog box.
Figure 2-26 Workspace Launcher
Step 2 Specify the Workspace directory, select Use this as the default and do not ask me again,
and click OK.
Step 3 The Security Storage dialog box is displayed, as shown in Figure 2-27. Click No.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
84
Figure 2-27 Security Storage dialog box
Step 4 On the Overview page, click Open Administration Console to enter the SAP HANA
Administration Console page.
Step 5 Right-click the blank area under System, as shown in Figure 2-28.
Figure 2-28 SAP HANA Administration Console page
Step 6 Choose Add System. The Specify System page is displayed, as shown in Figure 2-29. Configure parameters.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
85
Key parameters are as follows:
Host Name: Enter the service or client plane IP address of the SAP HANA ECS. In the
cluster scenarios, you only need to enter the IP address of the first SAP HANA node, on
which installation has been performed.
Instance Number: Enter the number of the instance on the SAP HANA node.
Mode: Select a mode based on actual requirements. If HANA 2.0 is used, select
Multiple containers.
Figure 2-29 Specify System page
Step 7 Click Next to go to the System page, as shown in Figure 2-30. Choose Authentication by
database user and enter the username and password.
The username and password are those configured during SAP HANA software installation.
The username is consistently set to SYSTEM.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
86
Figure 2-30 System page
Step 8 Click Next and then Finish. Then, the SAP HANA Studio automatically connects to the
database.
If the connection fails, check whether the database instance on the SAP HANA node is
running.
Step 9 Under System of the SAP HANA Administration Console page, double-click the node to be
checked.
Step 10 Click the Landscape tab on the right of the page and check the status of each process on the
SAP HANA node.
Green indicates that the process is running properly.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
87
Figure 2-31 Landscape page
----End
2.6.9 Configuring the Backup Path
Scenarios
A path is required for backing up SAP HANA data.
This section uses a Windows ECS where the SAP HANA Studio is deployed as an example
when SAP HANA 1.0 is used.
Procedure
Step 1 On the HANA Studio ECS, choose Start > SAP HANA > SAP HANA Studio to start the
SAP HANA Studio software.
Step 2 In the System area on the left, right-click the database node and choose Backup and
Recovery > Open Backup Console, as shown in Figure 2-32.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
88
Figure 2-32 Open Backup Console menu
Step 3 Click the Configuration tab on the right and configure the backup path, as shown in Figure
2-33.
The backup path is the directory to which the backup volume is attached, that is, /hana/backup.
Click the save icon in the upper right corner to save the configuration.
Figure 2-33 Configuring the backup path
Step 4 In the System area on the left, right-click the database node and choose Backup and
Recovery > Back Up System.... The Specify Backup Settings page is displayed.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
89
Figure 2-34 Specify Backup Settings page
Step 5 Use default settings and click Next. The Review Backup Settings page is displayed. Confirm
configurations and click Finish. The system starts the backup task.
According to SAP requirements, data must be backed up immediately after the SAP HANA system is
installed.
Step 6 Click Close after the backup task is complete.
----End
2.6.10 Configuring SAP HANA Storage Parameters
Scenarios
Configure SAP HANA storage parameters based on SAP's requirements.
Only SAP HANA 1.0 needs to be configured because the default configurations of SAP HANA 2.0 meet the specified requirements.
SAP HANA
User Guide (BMS Cluster Deployment) 2 Deployment
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
90
For more information, see as follows:
SAP Note 2186744 - FAQ: SAP HANA Parameters
SAP Note 2267798 - Configuration of the SAP HANA Database during Installation
Using hdbparam
SAP_HANA_Administration_Guide
SAP Note 2156526 - Parameter constraint validation on section indicies does not work
correctly with hdbparam
SAP Note 2399079 - Elimination of hdbparam in HANA 2
Procedure
Step 1 Log in to an SAP HANA node.
Step 2 Run the following command to switch to the SAP HANA administrator:
su - s00adm
Step 3 Configure SAP HANA storage parameters.
hdbparam --paramset fileio.async_read_submit=on
hdbparam --paramset fileio.async_write_submit_active=on
hdbparam --paramset fileio.async_write_submit_blocks=all
Step 4 (Optional) Configure storage parameters on other SAP HANA nodes according to the steps
provided in preceding paragraphs.
If multiple SAP HANA nodes exist, perform the same configuration for these parameters on
other SAP HANA nodes.
----End
SAP HANA
User Guide (BMS Cluster Deployment) A Appendix
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
91
A Appendix
A.1 Logging In to a Linux ECS Using an SSH Key
Prerequisites You have obtained the key file of the ECS.
An EIP has been bound to the ECS.
You have configured the inbound rules of the security group.
Logging In to the Linux ECS from a Windows Computer
This section describes how to log in to the Linux ECS from a Windows computer.
The following operations use PuTTY as an example to log in to the ECS. Before the login,
you must convert the private key format.
1. Visit the following website and download PuTTY and PuTTYgen:
http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html
PuTTYgen is a private key generator, which is used to create an SSH key pair that consists of a public
key and a private key for PuTTY.
2. Run PuTTYgen.
3. In the Actions area, click Load and import the private key file stored during HANA ECS
creation.
The file to be imported must be in the format of "All files (*.*)".
4. Click Save private key.
5. Save the converted private key to the local computer. For example, kp-123.ppk.
6. Run PuTTY.
7. Choose Connection > data and enter :root in Auto-login username.
8. Choose Connection > SSH > Auth. In the last configuration item Private key file for
authentication, click Browse and select the private key converted in step 5.
9. Click Session and enter the EIP of the ECS under Host Name (or IP address).
10. Click Open.
The ECS is logged in.
SAP HANA
User Guide (BMS Cluster Deployment) A Appendix
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
92
Logging In to the Linux ECS from a Linux Computer
This section describes how to log in to the Linux ECS from a Linux computer. The following
operations use private key file kp-123.pem as an example to log in to the ECS.
1. On the Linux CLI, run the following command to change the permission:
chmod 600 /path/kp-123
In the preceding command, path specifies where the private key file is saved.
2. Run the following command to log in to the ECS:
ssh -i /path/kp-123 root@EIP
In the preceding command, path specifies where the private key file is saved.
EIP is the EIP bound to the ECS.
A.2 Obtaining the Password for Logging In to a Windows ECS
Scenarios
Password authentication mode is required to log in to a Windows ECS. Therefore, you must
use the key file used when you created the ECS to obtain the administrator password
generated when the ECS was initially installed. The administrator user is Administrator or
the user configured using Cloudbase-Init. This password is randomly generated, offering high
security.
You can obtain the initial password for logging in to a Windows ECS using either the
management console or API. For details, see this section.
Obtaining the Password Using the Management Console
1. Obtain the private key file (.pem file) used when you created the ECS.
2. Log in to the management console.
3. Under Computing, click Elastic Cloud Server.
4. In the ECS list, select the ECS whose password you want to delete.
5. In the Operation column, click More and choose Get Password.
6. Use either of the following methods to obtain the password through the key file:
− Click Select File and upload the key file from a local directory.
− Copy the key file content to the text field.
7. Click Get Password to obtain a random password.
Obtaining the Password Using APIs
1. Obtain the private key file (.pem file) used when you created the ECS.
2. Set up the API calling environment.
3. For details, see section "API Calling" in the Elastic Cloud Server API Reference.
4. Obtain the ciphertext password.
SAP HANA
User Guide (BMS Cluster Deployment) A Appendix
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
93
Call the password obtaining APIs to obtain the ciphertext password of the public key
encrypted using RSA. The API URI is in the format "GET
/v2/{tenant_id}/servers/{server_id}/os-server-password".
For instructions about how to call an API, see section "Retrieving the Password for Logging In to a
Windows ECS (Native OpenStack API)" in Elastic Cloud Server API Reference.
5. Decrypt the ciphertext password.
Use the private key file used when you created the ECS to decrypt the ciphertext
password obtained in step 4.
a. Run the following command to convert the ciphertext password format to ".key
-nocrypt" using OpenSSL:
openssl pkcs8 -topk8 -inform PEM -outform DER -in rsa_pem.key -out
pkcs8_der.key -nocrypt
b. Invoke the Java class library
org.bouncycastle.jce.provider.BouncyCastleProvider and use the private key file
to edit the code decryption ciphertext.
A.3 Querying the NIC IP Address of a BMS
Scenarios
The IP address of a BMS is required.
Procedure
Step 1 On the management console, choose Computing > Bare Meta Server. On the left side of the
page, choose Bare Metal Server. Then, the system displays all BMSs on the right side of the
page.
Step 2 Click the name of the BMS to be queried. Then, the page for the BMS details is displayed.
Step 3 Click the NIC tab. On the page providing detailed information, In the expanded area, check
the IP address information.
----End
A.4 Querying the NIC IP Address of an ECS
Scenarios
The IP address of an ECS is required.
Procedure
Step 1 On the management console, choose Computing > Elastic Cloud Server. On the left side of
the page, choose Elastic Cloud Server. Then, the system displays all ECSs on the right side
of the page.
Step 2 Click the name of the HANA ECS to be queried. Then, the page for the HANA ECS details is
displayed.
SAP HANA
User Guide (BMS Cluster Deployment) A Appendix
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
94
Step 3 Click the NIC tab. On the page providing detailed information, click target NIC. In the
expanded area, check the IP address information.
EIP: specifies the EIP bound to the ECS.
Private IP Address: specifies the private IP address of the ECS NIC.
----End
SAP HANA
User Guide (BMS Cluster Deployment) B Change History
Issue 03 (2018-02-12) Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
95
B Change History
Change History Release On
This issue is the third official release, which incorporates the following
changes:
Added SUSE Linux Enterprise Server 12 SP2 for SAP.
2018-02-12
This issue is the second official release, which incorporates the following
changes:
Supported SAP HANA 2.0.
2018-01-24
This issue is the first official release. 2018-01-10
top related