User Guide (Cluster Deployment)€¦ · If high availability (HA) is required, construct such architecture using two single nodes through system replication. This system architecture

SAP HANA

User Guide (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 (Cluster Deployment) Contents



ii

Contents

1 Introduction.................................................................................................................................... 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 Scenarios .................................................................................................................................................. 4

2.1.1 Cluster Scenario (Single NFS Server Deployed) ....................................................................................................... 5

2.1.2 Cluster Scenario (Two NFS Servers Deployed to Work in Active/Standby Mode) ................................................... 7

2.1.3 Cluster Deployment (SFS) ......................................................................................................................................... 8

2.2 Data Planning.............................................................................................................................................................. 10

2.2.1 SAP HANA Nodes ................................................................................................................................................... 10

2.2.2 Other Nodes ............................................................................................................................................................. 11

2.2.3 Network Planning .................................................................................................................................................... 12

2.3 Preparing for Resources .............................................................................................................................................. 19

2.3.1 Software and Tools .................................................................................................................................................. 19

2.3.2 License ..................................................................................................................................................................... 20

2.3.3 Creating a VPC ........................................................................................................................................................ 21

2.3.4 Applying for a Subnet and Configuring a Security Group ....................................................................................... 21

2.4 Creating ECSs ............................................................................................................................................................. 22

2.4.1 Creating an SAP HANA Node ................................................................................................................................. 22

2.4.2 Creating an SAP HANA Studio ECS ....................................................................................................................... 26

2.4.3 Creating a NAT Server ............................................................................................................................................. 28

2.4.4 Creating an NFS Server ........................................................................................................................................... 30

2.4.5 Creating SFS File System ........................................................................................................................................ 32

2.4.6 Configuring SSH Switching Permissions ................................................................................................................ 33

2.4.7 Configuring the Mapping Between SAP HANA Host Name and IP Addresses ...................................................... 35

2.4.8 Modifying OS Configurations ................................................................................................................................. 36

2.5 Installing SAP HANA (Cluster Deployment and Single NFS Server Deployed) ....................................................... 37

2.5.1 Formatting the Disks Attached to an NFS Server .................................................................................................... 37

2.5.2 Formatting an SAP HANA Disk .............................................................................................................................. 38

2.5.3 Installing the SAP HANA Software ......................................................................................................................... 41

2.5.4 Installing the SAP HANA Studio on a Windows ECS ............................................................................................. 46

SAP HANA

User Guide (Cluster Deployment) Contents



iii

2.5.5 Installing the SAP HANA Studio on a Linux ECS .................................................................................................. 48

2.5.6 Connecting SAP HANA Nodes to the SAP HANA Studio ...................................................................................... 50

2.5.7 Configuring the Backup Path ................................................................................................................................... 54

2.5.8 Configuring SAP HANA Storage Parameters .......................................................................................................... 56

2.6 Installing SAP HANA (Cluster Deployment and Two NFS Servers Deployed to Work in Active/Standby Mode) ... 57

2.6.1 Configuring NFS Servers to Be Trusty to Each Other ............................................................................................. 57

2.6.2 Formatting the Disks Attached to an NFS Server .................................................................................................... 58

2.6.3 Configuring HA on NFS Servers (SLES 12 SP1) .................................................................................................... 60







2.6.10 Configuring SAP HANA Storage Parameters ........................................................................................................ 90

2.7 Installing SAP HANA (Cluster Deployment Where SFS Is Used) ............................................................................. 91







2.7.7 Configuring SAP HANA Storage Parameters .......................................................................................................... 94

3 Management and Monitoring ................................................................................................... 95

4 Backing Up and Restoring Data ............................................................................................... 96

4.1 Obtaining the Backup Size ......................................................................................................................................... 96

4.2 Configuring the Backup Path ...................................................................................................................................... 97

4.3 Creating a Backup Task .............................................................................................................................................. 98

4.4 Canceling a Backup Task .......................................................................................................................................... 100

4.5 Checking Backup File Integrity ................................................................................................................................ 100

4.6 Restoring SAP HANA Data ...................................................................................................................................... 102

5 FAQs ............................................................................................................................................ 105

5.1 What Should I Do If I Cannot Switch to an ECS or HANA ECS Using SSH? ........................................................ 105

6 Appendix .................................................................................................................................... 107

6.1 Obtaining the Password for Logging In to a Windows ECS ..................................................................................... 107

6.2 Logging In to a Linux ECS Using an SSH Key ........................................................................................................ 108

6.3 Querying the NIC IP Address of an ECS .................................................................................................................. 109

A Change History ......................................................................................................................... 110

SAP HANA

User Guide (Cluster Deployment) 1 Introduction



1

1 Introduction

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




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




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 (Cluster Deployment) 2 Deployment



4

2 Deployment

2.1 Deployment Scenarios

SAP HANA can be deployed in the following scenarios:

Single-node deployment: applies in OLTP scenarios. You can choose HA configuration

as required. SAP HANA receives and processes data quickly, provides processing results

in a short period of time, and rapidly responds to user operations.

Cluster deployment: applies in OLAP scenarios. As a data warehouse, SAP HANA offers

support for decision-makers and senior managers. It can quickly and flexibly process

complex queries on a large amount of data based on the analysts' requests. It can also

provide decision makers with the query results intuitively and clearly. Using SAP HANA,

decision makers can obtain accurate information about the enterprise operating status,

learn object demands, and make correct decisions.

Table 2-1 lists recommended deployment modes based on systems.

Production (PRD): indicates the production system where HANA is formally used.

Quality Assure (QAS): indicates the quality assurance system where HANA functions,

performance, and reliability are fully verified.

Development (DEV): indicates the development system where the development

engineers configure and verify the compatibility between application software and SAP

HANA and continuously optimize the application software.

Training (TRN): indicates the training and demonstration system where you provide a

training or demonstration after deploying HANA.

Test (TST): indicates the test system where the development engineers test application

software and HANA to verify the functions of application software after application

software development is complete.

In Suite on HANA (SoH) scenario, SAP HANA works with SAP business suites, such as Enterprise

Resource Planning (ERP) or Customer Requirement Management (CRM). In this scenario, SAP HANA

provides OLTP functions. The SAP HANA process latency is the key concern.

In Business Warehouse on HANA (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.

SAP HANA




5

Table 2-1 Systems and deployment schemes

System SoH BWoH (Single-Node) BWoH (Cluster)

PRD Single-node scenario

where HA is required

Single-node scenario

where HA is required

Cluster scenario

where two NFS

servers are deployed

QAS Single-node,

regardless of whether

HA is required or not

Single-node, regardless

of whether HA is

required or not

Single node (no

HA), cluster (two

NFS Servers), or

cluster (single NFS

Server)

DEV Single-node scenario

where HA is not

required


where HA is not

required


where HA is not

required

TRN Single-node scenario

where HA is not

required


where HA is not

required


where HA is not

required

TST Single-node scenario

where HA is not

required


where HA is not

required


where HA is not

required

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:

SAP HANA




6

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 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.

SAP HANA




7

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.

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:

SAP HANA




8

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

2.1.3 Cluster Deployment (SFS)

Figure 2-5 shows the cluster scenario where a file system is mounted.

SAP HANA




9

Figure 2-5 Cluster deployment (SFS) scenario

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:

SFS (Scalable File Service): uses a file system to provide users with shared file storage

service. The shared and backup volumes are provided by SFS for sharing them with SAP

HANA ECSs.

Figure 2-6 shows the deployment flowchart in the cluster scenario where SFS is required.

Figure 2-6 Cluster deployment (SFS) scenario

SAP HANA




10

2.2 Data Planning

Before installing the SAP HANA, you must plan the following data:

Data of SAP HANA nodes: includes the OSs, specifications, and disk capacities of the

SAP HANA nodes used in SoH and BWoH scenarios.

Data of other nodes: includes the OSs, specifications, and disk capacities of the SAP

HANA Studio, NAT server, and NFS servers.

Network data: includes subnets and security group rules.

SAP HANA installation data: planned based SAP HANA requirements.

2.2.1 SAP HANA Nodes

SAP poses specific requirements on HANA ECSs 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 2-2 lists the requirements on HANA ECSs in BWoH scenario.

Table 2-2 Large-memory E2 ECS specifications

ECS Type vCPUs Memory (GB) Flavor

Large-memory 12 256 e2.3xlarge

18 445 e2.4xlarge

36 890 e2.9xlarge

Requirements on OSs and Disks

SAP HANA ECS volumes include log volumes, data volumes, shared volumes, and backup

volumes.

The following volume sizes are recommended. For more configuration information, see the official

SAP documents.

A shared disk can be attached to multiple ECSs, while a non-shared disk can only be attached to one

ECS.

Table 2-3 Requirements on HANA ECS OS in cluster deployment scenarios

Scenario Specifications

OS SUSE Linux Enterprise Server 11 SP4 for SAP

SAP HANA




11

Scenario Specifications

SUSE Linux Enterprise Server 12 SP1 for SAP

SUSE Linux Enterprise Server (SLES) 12 SP2 for SAP

Table 2-4 Requirements on E1 and E2 HANA ECS disks in cluster deployment scenarios

Disk Type Sharing Mode

Size

OS

volume Ultra-high I/O (Latency

optimized)

Non-share

d disk

N/A

Log


optimized)

Shared

disk

When the memory size is

less than or equal to 512

GB, the log volume

capacity is half of the

memory size and rounded

up for decimal places.

When the memory size is

greater than 512 GB, the

log volume capacity is 512

GB.

Data


optimized)

Shared

disk

Same as the memory size

/usr/sap


optimized)

Non-share

d disk

50GB

2.2.2 Other Nodes

Other nodes include the NAT server, SAP HANA Studio, and NFS server nodes. Table 2-5

lists the requirements on these nodes.

Table 2-5 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 12 SP1

SAP HANA




12

Node Specifications

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 (SLES)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, 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: ultra-high I/O, SCSI, shared disk. The

recommended size is at least three times that of the memory

size.

− SBD volume: 10 GB, high I/O, SCSI, shared disk This

volume is required only when NFS servers are deployed to

work in cluster or active/standby mode.

Table 2-6 lists the disk specifications when the shared volume and backup volume are

provided by the SFS.

Table 2-6 SFS disk requirements

Disk Size

Shared volume Number of worker nodes/4 x Memory size. The value of number of

worker nodes/4 is rounded up for decimal places.

Backup volume The recommended size is at least three times that of the memory size.

2.2.3 Network Planning

Network Plane Planning in the Cluster Scenario Where a Single NFS Server Is Deployed

Figure 2-7 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




13

Figure 2-7 Network plane planning in the cluster scenario where a single NFS server is deployed

Table 2-7 lists the entire network planning.

Table 2-7 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.



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.



HANA nodes as an

example)

NFS server: 10.0.1.101

SAP HANA




14

Network Plane Planning in the Cluster Scenario Where Two NFS Servers Are Deployed to Work in Active/Standby Mode

Figure 2-8 shows the network plane planning in the cluster scenario where two NFS servers

are deployed to work in active/standby mode.


Figure 2-8 Network plane planning in the cluster scenario where two NFS servers are deployed to

work in active/standby mode

Figure 2-8 lists the entire network planning.

Table 2-8 Network planning in the cluster scenario where two NFS servers are deployed to work

in active/standby mode


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 nodes as an

example)

SAP HANA




15


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


EIP Allows you to access the SAP

HANA Studio, NAT Server,

NFS Server, and SAP HANA

node.


IP address of the internal

communication plane

Allows SAP HANA nodes,

each of which is equipped with

an SR-IOV NIC, to exchange

data.



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.



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

Network Plane Planning (SFS)

Figure 2-9 shows the network plane planning in the cluster scenario where SFS is used.


SAP HANA




16

Figure 2-9 Network plane planning in the cluster scenario where SFS is used

Table 2-9 lists the entire network planning.

Table 2-9 Network information planning (cluster scenario where SFS is used)


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 SFS and

ERP) or SAP HANA Studio

client software.



HANA nodes as an

example)


EIP Allows you to access the

SAP HANA Studio and

NAT server.


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.



HANA nodes as an

example)

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.

SAP HANA




17

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 2-10 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.

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


Allows all peers to access SAP

https://help.sap.com/viewer/575a9f0e56f34c6e8138439eefc32b16/2.0/en-US/616a3c0b1cc748238de9c0341b15c63c.html

https://help.sap.com/viewer/575a9f0e56f34c6e8138439eefc32b16/2.0/en-US/616a3c0b1cc748238de9c0341b15c63c.html

SAP HANA




18


HANA.

Table 2-11 Security group rules (SAP HANA Studio)


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.

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





with each other.

Outbound



Allows all peers to access the

SAP HANA Studio.

Table 2-12 Security group rules (NAT server)


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



SAP HANA




19


system It enables ECSs in the same


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.

Table 2-13 Security group rules (NFS server)


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





with each other.

Outbound



Allows all peers to access the

NFS server.

2.3 Preparing for Resources

2.3.1 Software and Tools

Table 2-14 lists the software and tools to be obtained.

SAP HANA




20

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 2-14 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

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.

SAP HANA installation

package. You can visit

the SAP official website

or access the OBS

bucket to download the

installation package.

For detailed paths where the files

can be downloaded, see the sections

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_script.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.

https://support.sap.com/home.html

SAP HANA




21

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

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.

SAP HANA




22

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.

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 ECSs

2.4.1 Creating an SAP HANA Node

Scenarios

SAP HANA software runs on SAP HANA ECSs. Depending on deployment scenarios, you

need to create one or multiple HANA ECSs for deploying the SAP HANA software.

For details about the number of HANA ECSs to be created and their planned data, see

sections 2.1 Deployment Scenarios and 2.2 Data Planning. If multiple HANA ECSs are

created in cluster scenarios, you need to create multiple HANA ECSs which will function as

the worker nodes, and create another one HANA ECS which will function as the standby

node.

SAP HANA




23

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 HANA ECS to switch to Create HANA ECS page.

Step 3 Configure the parameters listed in Table 2-15.

Table 2-15 HANA ECS parameters

Parameter Description

ECS Name Specifies the ECS name. The name contains a maximum

of 13 characters.

When you create ECSs in batches, the number in the ECS

Name is generated automatically in ascending order based

on the Quality value that you filled in. For example, if you

fill hana in ECS Name, the first ECS is hana-0001, and

the second ECS is hana-0002.

AZ Specifies the AZ in which HANA ECSs locate. The AZ

must support SAP HANA. Choose the AZ as required.

All ECSs in the SAP HANA system must be within the

same AZ.

ECS Type Retain the default setting Large-memory for ECS Type.

vCPU and Memory Set these parameters as required.

For details, see section 2.2.1 SAP HANA Nodes.

Image Type and Image Choose Public Image and an SAP HANA ECS image.

Disk In cluster deployment mode, one system disk and two data

disks are required.

NOTE

If a standby node is created in cluster deployment mode, only a

system disk is required. In such a case, you must delete the two

shared volumes that are selected by default.

If additional disks are required, you can click Add Data

Disk to add more disks.

For details about disk requirements, 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 according to the

deployment plan provided in section 2.2.3 Network

Planning.

The first NIC is the primary NIC, which is the

server/client plane NIC.

The second NIC is the internal communication plane

NIC, which is required only in cluster scenarios.

SAP HANA




24


The third NIC is the cloud management or backup

plane NIC, which is required only in cluster scenarios.

EIP Do not bind an EIP.

Key Pair An SSH key certificate is used for authenticating users

who attempt to log in to a HANA 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.

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 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


Specifies a HANA ECS group. When you create ECSs, the

system will allocate the HANA ECSs in the same server

group to different physical servers to ensure the running

reliability of these HANA ECSs.

Determine the policy of a HANA ECS group based on the

scenario:

Single-node scenarios where HA is not required: You

do not need to specify ECS Group.

Single-node scenario where HA is required: The two

HANA ECSs must belong to the same ECS Group.

Cluster scenario: All HANA ECSs must belong to the same ECS Group.

SAP HANA




25


NOTE

If no ECS group is available, perform the following operations to

create one:

Click View ECS Group. On the page that is displayed, click

Create ECS Group, specify the ECS group name, and click OK.

Agency Name Specifies the agency name. When ECSs access the public

cloud platform, they obtain a temporary access credential

from the agency.

Data Provider is an SAP indicator collector on the public

cloud platform to collect key performance data of ECS and

CES in the SAP NetWeaver system and then surface the

data to SAP applications. Specify the

DataproviderAccess agency to ECSs where SAP HANA

and SAP NetWeaver are all deployed.

Before using the agency, log in to the public cloud

management console as the tenant administrator and create

the DataproviderAccess agency.

NOTE

Log in to the public cloud management console, choose Mgmt &

Deployment > Identity and Access Management. On the left

side of the page, choose Agency on the left. Then, click Create

Agency on the right side of the page.


Agency Name: DataProviderAccess

AgencyType: Cloud service

Cloud Service: ECS

Validity Period: Retain the default setting.

Permissions: Locate the Region and Project in which the

target HANA ECS locates and configure the Base > Tenant

Guest permission for the HANA ECS.

Quantity Set this parameter as required.

NOTE

For the HANA ECSs created in batches in cluster scenarios, the

HANA ECS that functions as the standby node does not require

the data and log volumes. This configuration is different from that

of the HANA ECSs functioning as the worker nodes. Due to this

requirement, you are recommended to create the HANA ECSs that

function as the worker nodes and then create the HANA ECS that

functions as the standby node.

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 Create other HANA ECSs as required.

Step 7 Change the password of user root for logging in to all HANA ECSs.

SAP HANA




26

Securely keep the password of user root. In addition, ensure that all HANA ECSs use the

same root password.

1. Use the key to log in to the SAP HANA ECSs.

2. Run the following command to change the password of user root:

passwd

Enter the new password as prompted and confirm it.

----End

2.4.2 Creating an SAP HANA Studio ECS

Scenarios

An SAP HANA system requires an ECS for deploying the SAP HANA Studio software.

Procedure



Step 2 On the right side of the page, click Buy ECS to switch to the Create ECS page.


Table 2-16 ECS parameters



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.

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

SAP HANA




27


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.





NOTE




If you want to create an SSH key pair, do as follows:





File Injection This parameter is displayed only after you click Show


Do not configure them in this scenario.



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:





Quantity 1


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.

SAP HANA




28

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.





of 13 characters.

AZ Specifies the AZ where the ECS is located. Select an AZ

as required.




vCPU and Memory s1.medium (1 vCPU and 4 GB memory capacity) or higher


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.

SAP HANA




29


This parameter is configured as required only when

Elastic IP Address is set to Buy Now.


set to Key pair.




the Key Pair page.





NOTE








File Injection This parameter is displayed only after you click Show


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





#cloud-config

disable_root: false

runcmd:





Quantity

Quantity

1


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.

SAP HANA




30

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 2 On the right side of the page, click Buy ECS to switch to Create ECS page.


Table 2-18 Parameters of the ECS where the NFS server is deployed



of 13 characters.

AZ Specify the AZ in which an ECS locates. Select an AZ as

required.




vCPU and Memory s1.xlarge (4 vCPUs and 16 GB memory capacity)


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 2.6.2 Formatting the Disks

Attached to an NFS Server.

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 When only one NFS server is deployed, two NICs are

required.

The first NIC is the primary NIC, which is used on the

SAP HANA




31


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.


set to Key pair.




the Key Pair page.





NOTE




If you want to create an SSH key pair, do as follows:










#cloud-config

disable_root: false

SAP HANA




32


runcmd:





ECS Group This parameter is displayed only after you click Show


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 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.

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 Creating SFS File System

Scenarios

In the SAP HANA system, if the shared volume and backup volume are provided by SFS,

create a file system to provide a shared path for SAP HANA ECSs.

SAP HANA




33

Procedure

Step 1 Log in to the management console and choose Storage > Scalable File System to switch to

the SFS page.

Step 2 Click Create File System. The Create File System dialog box is displayed.




Name Specifies the file system name.

For example, sfs-shared and sfs-backup.

AZ Specifies the AZ in which the file system is located. Select

an AZ as required.

Type Specifies the file system type. Set this parameter to NFS.

VPC Select the VPC where SAP HANA ECSs reside.

Total Capacity Specifies the maximum capacity of a single file system.

For details, see section 2.2.2 Other Nodes.

Quantity 1

Step 4 Click Create Now, read and agree to the service agreement, and click Submit.

Step 5 The application returns back to the Scalable File Service page. In the file system list, locate

the new file system by its name. In the Shared Path column, obtain the shared path.

Step 6 Log in to the SAP HANA ECS and check whether the IP address of the DNS server is

configured in the /etc/resolv.conf file. If not, write the IP address of the DNS server into the

/etc/resolv.conf file.

----End

2.4.6 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.

SAP HANA




34

The .pem file is used to generate the .ppk file. For details, see section 6.2 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/

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.


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.

SAP HANA




35

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

2.4.7 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:

...

SAP HANA




36

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.

Step 5 Run the following command to transfer the configured /etc/hosts file to other SAP HANA

nodes:


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

SAP HANA




37

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

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:



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.


− share: specifies the disk of the shared volume.

− backup: specifies the disk of the backup volume.

SAP HANA




38

− 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

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.

SAP HANA




39

Prerequisites

The firewall on the ECS where SAP HANA is to be installed has been disabled.

Procedure

Step 1 Attach all data and log volumes to all SAP HANA nodes.

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.

2. Locate the first HANA ECS by ECS name and click its name.

3. On the page providing details about the ECS, click the EVS tab and find the first data

disk (log volume or data volume). Then, click the target 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 Attach

Disk page.

6. Select all target HANA ECSs and attach the disk to them.

Ensure that the disk is attached to all SAP HANA nodes.

7. Perform the preceding steps to attach the second disk (data or log volume) of the HANA

ECS to all other SAP HANA nodes.

8. Perform the preceding steps to attach the log and data volumes of all HANA ECSs to

other SAP HANA nodes.

The last HANA ECS is a standby node. When creating this HANA ECS, you did not create or attach log

and data volumes to it. Therefore, you do not need to attach the disks of this HANA ECS to other SAP

HANA ECSs.

Step 2 Upload the local hosts file to other SAP HANA ECSs.



to the first HANA ECS.

2. The /etc/hosts file has been configured in section 2.4.7 Configuring the Mapping

Between SAP HANA Host Name and IP Addresses. Therefore, you only need to transfer

this file to other SAP HANA ECSs.



For example, if the peer IP address of the HANA ECS is 10.0.3.3, run the following

command:

scp /etc/hosts 10.0.2.2:/etc/hosts


Use SSH to switch from every HANA ECS to any other SAP HANA ECS and verify that the

switching is correct.

For example, if the name of the peer SAP HANA ECS is hana002, run the following

command:

ssh hana002

Step 4 Format the disks attached to an SAP HANA node.

SAP HANA




40

1. Run the following command on the first SAP HANA ECS to query the disk to be

formatted:

fdisk -l

2. Determine the disks of the /usr/sap volumes, data volumes, and log volumes according to

the disk capacity.

3. Query the IP address (for example, 10.0.1.x) of the cloud management or backup plane

for the NFS server.

For details, see section 6.3 Querying the NIC IP Address of an ECS.

4. Query the internal communication plane IP addresses (for example, 10.0.2.x) of all

HANA ECSs.




~


package:

cd ~


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 disk information, cloud management or backup

plane IP address of the NFS server, and internal communication plane IP address of all

SAP HANA ECSs into the configuration file.

Parameters are as follows:

− partition_1_data__volname: specifies the disk of the data volume attached to the

first SAP HANA ECS. 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 disk of the data volume attached to the


log volumes attached and does not require recording.

− usrsaps: specifies the disks of /usr/sap volumes of all SAP HANA ECSs. The ECS

sequence is the same as that in the hosts 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 ECSs. When multiple IP addresses are available,


Take five SAP HANA ECSs, four worker nodes and one standby node, as an example.

Ensure that every volume path is unique.

SAP HANA




41

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=/dev/xvdb,/dev/xvdc,/dev/xvdd,/dev/xvde,/dev/xvdf

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:


dos2unix installHanaCluster.sh hanaCluster.cfg

11. Assign operation permissions to the hanaCluster.cfg file and run

installHanaCluster.sh to format disks.

chmod 777 installHanaCluster.sh hanaCluster.cfg

sh installHanaCluster.sh

----End

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.


SAP HANA




42

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


-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

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.


− 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.

SAP HANA




43

− number: specifies the SAP HANA instance ID. The instance ID is used in Security

Group Rules, 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

# 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

SAP HANA




44

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

...

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

SAP HANA




45


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


chmod 777 -R /hana

3. Start the installation.


− 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.


...

Enter comma-separated list of the selected indices [1]:



...

Restrict maximum memory allocation? [n]:

...



...

Enter Certificate Host Name For Host 'hana001' [hana001]:


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.


SAP HANA




46

...

Enter System Administrator User ID [1000]:



...

Enter ID of User Group (sapsys) [79]:



...

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


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"


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.


SAP HANA




47

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 6.1 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


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.

Figure 2-10 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.

SAP HANA




48

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



This section describes how to install the SAP HANA Studio on a Linux ECS.




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


/hana/shared

Step 3 Run the following command to decompress the installation package:


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 2-11.

SAP HANA




49

./hdbsetup









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


SAP HANA




50

-startup

plugins/org.eclipse.equinox.launcher_1.3.201.v20161025-1711.jar

--launcher.library

plugins/org.eclipse.equinox.launcher.gtk.linux.x86_64_1.1.401.v20161122-1740


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 2-12 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 2-13. Click No.

SAP HANA




51

Figure 2-13 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 2-14.

Figure 2-14 SAP HANA Administration Console page

Step 6 Choose Add System. The Specify System page is displayed, as shown in Figure 2-15. Configure parameters.

SAP HANA




52

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 2-15 Specify System page

Step 7 Click Next to go to the System page, as shown in Figure 2-16. 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




53

Figure 2-16 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




54

Figure 2-17 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 2-18.

SAP HANA




55

Figure 2-18 Open Backup Console menu

Step 3 Click the Configuration tab on the right and configure the backup path, as shown in Figure

2-19.

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 2-19 Configuring the backup path


Recovery > Back Up System.... The Specify Backup Settings page is displayed.

SAP HANA




56

Figure 2-20 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




57

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


----End

2.6 Installing SAP HANA (Cluster Deployment and Two NFS Servers Deployed to Work in Active/Standby Mode)

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 2.4.6 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.

https://help.sap.com/viewer/6b94445c94ae495c83a19646e7c3fd56/2.0.01/en-US

SAP HANA




58

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 2.4.7 Configuring the Mapping Between SAP HANA Host Name and

IP Addresses.

Step 3 Upload the local hosts file to the other NFS server.



For example, if the peer IP address of the SAP HANA ECS is 10.0.3.102, run the following

command:


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.


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.




SAP HANA




59

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.


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.



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.





~


package:

cd ~


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.


− 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,


− 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




60

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:


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.


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




61

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 2.6.1 Configuring NFS Servers to Be Trusty

to Each Other and 2.6.2 Formatting the Disks Attached to an NFS Server.

Procedure

Step 1 Initialize the cluster.



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]


Do you wish to use SBD? [y/N]


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




62

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/


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) []


/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




63

Figure 2-21 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 2-22. Select Cancel and press

Enter.

Figure 2-22 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 2-23 shows an example.


− Transport: Set this parameter to Unicast.

SAP HANA




64

− 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 2-23 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




65

Figure 2-24 Security page

7. On the left side of the page, choose Configure Csync2. Then, configure parameters on


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




66

Figure 2-25 Configure Csync2 page

8. On the left side of the page, choose Service. Then, configure parameters on the right side

of the page.


− 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




67

Figure 2-26 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


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




68

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:



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




69

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.


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


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


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


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


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.


SAP HANA




70

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


scp /etc/sysconfig/sbd nfs02:/etc/sysconfig

8. Run the following commands on the active NFS server to restart the cluster:



9. Run the following commands on the standby NFS server to restart the cluster:



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




71


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


Clone Ping:

SAP HANA




72

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


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.


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


Scenarios




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




73

Prerequisites


Procedure










Disk page.









HANA ECSs.

Step 2 Upload the local hosts file to other SAP HANA ECSs.



to the first HANA ECS.

2. The /etc/hosts file has been configured in section 2.4.7 Configuring the Mapping

Between SAP HANA Host Name and IP Addresses. Therefore, you only need to transfer

this file to other SAP HANA ECSs.



For example, if the peer IP address of the HANA ECS is 10.0.3.3, run the following

command:



Use SSH to switch from every HANA ECS to any other SAP HANA ECS and verify that the

switching is correct.

For example, if the name of the peer SAP HANA ECS is hana002, run the following

command:

ssh hana002


SAP HANA




74

1. Run the following command on the first SAP HANA ECS to query the disk to be

formatted:

fdisk -l

2. Determine the disks of the /usr/sap volumes, data volumes, and log volumes according to

the disk capacity.

3. Query the IP address (for example, 10.0.1.x) of the cloud management or backup plane

for the NFS server.



HANA ECSs.




~


package:

cd ~




8. Press i to enter editing mode. Write the disk information, cloud management or backup

plane IP address of the NFS server, and internal communication plane IP address of all

SAP HANA ECSs into the configuration file.










− 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.







Ensure that every volume path is unique.

SAP HANA




75










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











----End


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.


You have created servers, formatted disks attached to them, and completed required

configurations.

You have synchronized the time of the ECS.


SAP HANA




76

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


-P /hana/shared


-P /hana/shared

3. Run the following command to decompress the installation package and navigate to the

directory where the installation file locates:


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


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.


− 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.

SAP HANA




77

− number: specifies the SAP HANA instance ID. The instance ID is used in Security

Group Rules, 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.


[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

# 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

SAP HANA




78

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

...

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

SAP HANA




79


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


chmod 777 -R /hana

3. Start the installation.


− 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


...

Select additional components for installation:

...



...

Enter comma-separated list of the selected indices [1]:



...

Restrict maximum memory allocation? [n]:

...



...

Enter Certificate Host Name For Host 'hana001' [hana001]:


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.


SAP HANA




80

...

Enter System Administrator User ID [1000]:



...

Enter ID of User Group (sapsys) [79]:



...

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


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"


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


Scenarios




SAP HANA




81



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 6.1 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



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




82




----End


Scenarios



This section describes how to install the SAP HANA Studio on a Linux ECS.




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


/hana/shared


/hana/shared

Step 3 Run the following command to decompress the installation package:


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 2-28.

SAP HANA




83

./hdbsetup









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:


2


SAP HANA




84

-startup

plugins/org.eclipse.equinox.launcher_1.3.201.v20161025-1711.jar

--launcher.library

plugins/org.eclipse.equinox.launcher.gtk.linux.x86_64_1.1.401.v20161122-1740


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


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 2-29 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 2-30. Click No.

SAP HANA




85

Figure 2-30 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 2-31.

Figure 2-31 SAP HANA Administration Console page

Step 6 Choose Add System. The Specify System page is displayed, as shown in Figure 2-32. Configure parameters.

SAP HANA




86

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 2-32 Specify System page

Step 7 Click Next to go to the System page, as shown in Figure 2-33. 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




87

Figure 2-33 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




88

Figure 2-34 Landscape page

----End


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.


Recovery > Open Backup Console, as shown in Figure 2-35.

SAP HANA




89

Figure 2-35 Open Backup Console menu

Step 3 Click the Configuration tab on the right and configure the backup path, as shown in Figure

2-36.

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 2-36 Configuring the backup path


Recovery > Back Up System.... The Specify Backup Settings page is displayed.

SAP HANA




90


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


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




91

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


----End

2.7 Installing SAP HANA (Cluster Deployment Where SFS Is Used)


Scenarios




After the attachment, the data disks, including the data and log volumes as well as the shared

and backup volumes provided by SFS, can be used only after they are formatted and attached

to required directories.

Procedure



SAP HANA




92









Disk page.









HANA ECSs.


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 disks of the /usr/sap volumes, data volumes, log volumes, shared volumes,

and backup volumes according to the disk capacity.


HANA ECSs.






~


package:

cd ~




7. Press i to enter editing mode. Write the disk information and internal communication

plane IP address of all SAP HANA ECSs into the configuration file.


SAP HANA




93









− nfsserverIp: You do not need to configure this parameter.

− shareFSType: Set this parameter to SFS.

− sfsShared: specifies the SFS shared path of the shared volume.

− sfsBackup: specifies the SFS shared path of the backup volume.







Ensure that each disk of volumes is unique.










nfsserverIp=

shareFSType=SFS

sfsShared=[sfs-domain]:/share-599ab856

sfsBackup=[sfs-domain]:/share-02c088b6

sid=S01

hosts=10.0.2.2,10.0.2.3,10.0.2.4,10.0.2.5,10.0.2.6











SAP HANA




94

----End


For details, see 2.6.5 Installing the SAP HANA Software.


For details, see 2.6.6 Installing the SAP HANA Studio on a Windows ECS.


For details, see 2.6.7 Installing the SAP HANA Studio on a Linux ECS.


For details, see 2.6.8 Connecting SAP HANA Nodes to the SAP HANA Studio.


For details, see 2.6.9 Configuring the Backup Path.


For details, see 2.6.10 Configuring SAP HANA Storage Parameters.

SAP HANA

User Guide (Cluster Deployment) 3 Management and Monitoring



95

3 Management and Monitoring

Managing the Public Cloud Resources Used by SAP HANA

You can use Cloud Eye on the public cloud platform to monitor and manage public cloud

resources, such as Elastic Cloud Server (ECS) and Elastic Volume Service (EVS) resources,

used by SAP HANA.

For example, Cloud Eye monitors the CPU usage and disk read/write rate of the ECS where

SAP HANA is deployed and generates alarms when the metrics exceed alarm thresholds.

For more information, see Cloud Eye introduction and Cloud Eye User Guide.

Managing SAP HANA

SAP provides customers with various tools and instructions to manage SAP HANA.

To manage SAP HANA, SAP supports:

− Landscape-, system-, and database-level monitoring and management.

− Security-related monitoring and configuration.

− High reliability and scalability management

For details, see SAP HANA Administration Guide released by SAP.

To back up and restore data, SAP supports:

− Complete backup.

− Incremental backup.

− Backup using a third-party tool.

− Backup lifecycle management.

For details, see SAP HANA Database Backup and Recovery released by SAP.


https://help.sap.com/viewer/6b94445c94ae495c83a19646e7c3fd56/2.0.01/en-US/c4663eabbb571014923a90c70cec566c.html

SAP HANA

User Guide (Cluster Deployment) 4 Backing Up and Restoring Data



96

4 Backing Up and Restoring Data

To ensure data reliability, back up SAP HANA data.

This section describes how to use SAP HANA Studio to back up SAP HANA data into

/hana/backup and use the data to restore SAP HANA so that you can learn typical backup

and restoration operations.

For details about requirements and operation guide, see related documents released by SAP at

their official website.

4.1 Obtaining the Backup Size

Before installing SAP HANA, make sure that the available space in /hana/backup on SAP

HANA nodes is at least three times the SAP HANA memory capacity.

Obtain the backup size before backing up data. The size is displayed on the wizard page for

creating a backup task, as shown in Figure 4-1. Ensure that the available space in

/hana/backup meets the backup size requirement.

SAP HANA




97

Figure 4-1 Obtaining the backup size

4.2 Configuring the Backup Path

Before backing up data, configure the default backup path.

For details, see section "Configuring the Backup Path" in 2.5 Installing SAP HANA (Cluster

Deployment and Single NFS Server Deployed) and 2.6 Installing SAP HANA (Cluster

Deployment and Two NFS Servers Deployed to Work in Active/Standby Mode).

SAP HANA




98

4.3 Creating a Backup Task

Scenarios

Create a data backup so that it can be used to restore SAP HANA if an error occurs, thereby

ensuring SAP HANA reliability.

This section uses an SAP HANA system with SAP HANA Studio installed on a Windows

ECS as an example to describe how to completely back up SAP HANA data.

For details about operation requirements and notes, see SAP HANA Database Backup and

Recovery released by SAP.

Procedure

Step 1 On the SAP HANA Studio web page, right-click the SAP HANA system to be backed up and

choose Backup and Recovery > Back Up System from the shortcut menu.

Figure 4-2 Backup entry

Step 2 On the Specify Backup Settings page, configure backup parameters.


Backup Type: specifies the type of the backup. In this section, select Complete Data

Backup, indicating that the backup contains all data required for restoring SAP HANA.

Destination Type: specifies the target backup type. In this section, select File, indicating

that the backup data is stored in a file.

Backup Destination: specifies the target backup path. The default backup path,

/hana/backup, is the path you have specified.



SAP HANA




99

For file backup, ensure that the target backup path has sufficient space to store the

backup file. You can change the target backup path as needed.

Backup Prefix: specifies the backup file prefix. Time or date is recommended, for

example, COMPLETE_DATA_BACKUP_20170622.


Step 3 Click Next.

Brief information about the backup configuration is displayed.

Step 4 Confirm the information and click Finish.

The system starts the backup.

Step 5 After the backup is complete, click Close.

Step 6 Log in to the SAP HANA node as user root, switch to /hana/backup, and run the following

command:

ls -l | grep COM

In the preceding command, COM indicates that the files with prefix COM are to be displayed.

SAP HANA




100

Displayed backup files are as follows:

-rw-r----- 1 s01adm sapsys 163840 Jun 23 16:22

COMPLETE_DATA_BACKUP_20170622_databackup_0_1











----End

4.4 Canceling a Backup Task

Scenarios

Cancel a backup task during the backup process if the backup task is not required.



Procedure

Step 1 On the SAP HANA Studio web page, right-click the SAP HANA system that is being backed

up and choose Backup and Recovery > Open Backup Console from the shortcut menu.

The backup task that is in process is displayed on the Overview page.

Step 2 Click Cancel Backup to cancel the backup task.

----End

4.5 Checking Backup File Integrity

Scenarios

Before restoring data, check backup file integrity, preventing errors during the restoration

process.







SAP HANA




101

Procedure

Step 1 On the SAP HANA Studio web page, right-click the target SAP HANA system and choose

Backup and Recovery > Open Backup Console from the shortcut menu.

Step 2 On the page that is displayed, click Backup Catalog, select the backup file to be checked, and

obtain the path in which the file is stored.

Figure 4-4 Obtaining the path in which the backup file is stored

Step 3 Log in to the ECS where the backup file is stored as user root and switch to the administrator

account for logging in to the SAP HANA system.

For example, if the SAP HANA SID is s01, the administrator username for logging in to the

SAP HANA system is s01adm.

su - s01adm

Step 4 Run the following command to switch to the directory in which the backup file is stored:

cd /hana/backup

Step 5 Run the following command to view the name of the backup file:

ls -l | grep COM

In the preceding command, COM indicates that the files with prefix COM are to be displayed.

Displayed backup files are as follows:







SAP HANA




102







Step 6 Run the following command to check the integrity of the backup file:

hdbbackupcheck COMPLETE_DATA_BACKUP_20170622_databackup_0_1

The following information is displayed if the check result is normal:

Backup '/hana/backup/COMPLETE_DATA_BACKUP_20170622_databackup_0_1'

successfully checked.

Step 7 Repeat Step 6 to check the integrity of other backup files.

----End

4.6 Restoring SAP HANA Data

Scenarios

SAP HANA must be restored if the data volume or log volume of the SAP HANA system is

unavailable, or other exceptions occur in the SAP HANA system.

This section uses an SAP HANA system with SAP HANA Studio installed on a Windows

ECS as an example to describe how to use the backup data to restore SAP HANA.



Procedure

Step 1 On the SAP HANA Studio web page, right-click the SAP HANA system to be stopped and

choose Configuration and Monitoring > Stop System from the shortcut menu.

Step 2 On the page that is displayed, set Shutdown Type to Soft and click OK.

Step 3 Enter the administrator account for logging in to the SAP HANA system as prompted.

For example, if the SAP HANA SID is s01, the administrator username for logging in to the

SAP HANA system is s01adm.

Step 4 Right-click the SAP HANA system to be restored and choose Backup and Recovery >

Recover System from the shortcut menu.



SAP HANA




103

Figure 4-5 Restoration entry

Step 5 On the Specify Recovery Type page, select Recovery the database to a specific data

backup or storage snapshot and click Next.

Step 6 On the Specify Backup Location page, retain default settings and click Next.

SAP HANA




104

Figure 4-6 Specifying the path in which the backup file is to be stored

Step 7 On the Select a Backup page, select a backup record for recovery and click Next.

Step 8 On the Other Settings page, retain the default settings and click Next.

Step 9 On the page showing the brief information, click Finish to start the restoration.

Step 10 After the restoration is complete, click Close.

On the SAP HANA Studio web page, you can find that SAP HANA has started and is running

properly.

----End

SAP HANA

User Guide (Cluster Deployment) 5 FAQs



105

5 FAQs

5.1 What Should I Do If I Cannot Switch to an ECS or HANA ECS Using SSH?

Symptom

When I switched from a Linux ECS/HANA ECS to another Linux ECS/HANA ECS using

SSH, the system displayed a message indicating the switching failed.

The message is as follows:

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

@ WARNING: REMOTE HOST IDENTIFICATION HAS CHANGED! @

@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@

IT IS POSSIBLE THAT SOMEONE IS DOING SOMETHING NASTY!

Someone could be eavesdropping on you right now (man-in-the-middle attack)!

It is also possible that a host key has just been changed.

The fingerprint for the RSA key sent by the remote host is

2c:d0:17:8a:82:4c:23:d6:14:be:d0:1d:88:8b:8b:03 [MD5].

Please contact your system administrator.

Add correct host key in /root/.ssh/known_hosts to get rid of this message.

Offending ECDSA key in /root/.ssh/known_hosts:1

You can use following command to remove all keys for this IP:

ssh-keygen -R fanhana-0002 -f /root/.ssh/known_hosts

RSA host key for fanhana-0002 has changed and you have requested strict checking.

Host key verification failed.

Possible Causes OpenSSH has been reinstalled on the target ECS/HANA ECS.

The node IP address or name has been changed.

Other issues have occurred.

SAP HANA

User Guide (Cluster Deployment) 5 FAQs



106

Handling Method

Delete the known_hosts file on the local end to resolve this issue. To do so, perform the

following operations:

1. On the local ECS/HANA ECS, switch to the CLI as user root.

2. Run the following command to delete the known_hosts file:

rm /root/.ssh/known_hosts

After the file deletion, when you attempt to switch to the target ECS/HANA ECS using SSH, the system

displays the fingerprint as well as the message "Are you sure you want to continue connecting

(yes/no)?". In such a case, enter yes and continue the switching.

SAP HANA

User Guide (Cluster Deployment) 6 Appendix



107

6 Appendix

6.1 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




108

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.

6.2 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.

SAP HANA




109

9. Click Session and enter the EIP of the ECS under Host Name (or IP address).

10. Click Open.

The ECS is logged in.

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.

6.3 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.

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 (Cluster Deployment) A Change History



110

A Change History

Description Released On

This issue is the third official release, which incorporates the following

changes:

Supported SAP HANA 2.0.

Added SUSE Linux Enterprise Server 12 SP2 for SAP.

2018-02-12

This issue is the second official release, which incorporates the

following changes: which incorporates the following changes:

Updated the network planning information.

Updated paths for obtaining installation packages.

Deleted the time synchronization scheme.

Added the /usr/sap volume.

Changed the installation path from /usr/sap to /hana/shared.

Added the SFS scheme.

Added modifying OS configurations.

2018-01-10

This issue is the first official release. 2017-11-10

Documents

User Guide (Cluster Deployment)€¦ · If high availability (HA) is required, construct such architecture using two single nodes through system replication. This system architecture