DEV Community

Cong Li
Cong Li

Posted on

GBase数据库 | Quick Setup Guide for GBase 8s SSC Shared Storage Cluster

#database

Hello, readers! GBase 8s is one of the few mature databases with advanced SSC (Shared Storage Cluster) functionality. Are you looking for a simple way to validate the features of a shared storage cluster? This article will provide a guide to setting up a cluster and quickly verifying functionality. Let’s dive into deploying an SSC shared storage cluster in a single-machine environment.

Preparation

1. Prepare the Database Installation Package:
Download the package from the GBase official website according to your architecture requirements:

https://www.gbase.cn/download/gbase-8s-1?category=INSTALL_PACKAGE

2. Prepare Test Machines (VM or Physical Machines):

Steps

1. Install the Database

Use a non-instantiated installation (select the option not to create an instance after running ids_install) or silent installation.

Note: For silent installation, ensure the gbasedbt user is created beforehand with the following command:

# useradd -m -U gbasedbt
Enter fullscreen mode Exit fullscreen mode

For silent installation, extract the database installation package with tar -xvf and then execute:

# ./ids_install -i silent -DLICENSE_ACCEPTED=TRUE -DUSER_INSTALL_DIR={install_path}
# Replace {install_path} with the desired installation path, such as /data/ssc
Enter fullscreen mode Exit fullscreen mode

2. Create Environment Variable Files

Prepare two environment variable files.

Assume the instance names are Server1 for database instance 1 and Server2 for instance 2. Here’s the configuration:

For Instance 1:

# touch /data/ssc/profile.server1
# vi /data/ssc/profile.server1

export GBASEDBTDIR=/data/ssc
export ONCONFIG=onconfig.server1
export GBASEDBTSQLHOSTS=/data/ssc/etc/sqlhosts.server1
export GBASEDBTSERVER=server1
export PATH=/data/ssc/bin:/data/ssc/sbin:/data/ssc/extend/krakatoa/jre/bin:$PATH
Enter fullscreen mode Exit fullscreen mode

For Instance 2:

# touch /data/ssc/profile.server2
# vi /data/ssc/profile.server2

export GBASEDBTDIR=/data/ssc
export ONCONFIG=onconfig.server2
export GBASEDBTSQLHOSTS=/data/ssc/etc/sqlhosts.server2
export GBASEDBTSERVER=server2
export PATH=/data/ssc/bin:/data/ssc/sbin:/data/ssc/extend/krakatoa/jre/bin:$PATH
Enter fullscreen mode Exit fullscreen mode

3. Prepare Data Storage Directories and rootdbs File for Instances 1 and 2 

# mkdir -p /data/ssc/storage
# chown gbasedbt:gbasedbt /data/ssc/storage
# chmod 755 /data/ssc/storage
# cd /data/ssc/storage
# touch rootdbs
# chown gbasedbt:gbasedbt rootdbs
# chmod 660 rootdbs
Enter fullscreen mode Exit fullscreen mode

4. Modify sqlhosts 

# touch /data/ssc/etc/sqlhosts.server1
# vi /data/ssc/etc/sqlhosts.server1

g_cluster_1     group    -       -       i=1
server1     onsoctcp        localhost    50000    g=g_cluster_1
server2     onsoctcp        localhost    50001    g=g_cluster_1
Enter fullscreen mode Exit fullscreen mode

The sqlhosts file for server2 should match server1:

# cp /data/ssc/etc/sqlhosts.server1 /data/ssc/etc/sqlhosts.server2
Enter fullscreen mode Exit fullscreen mode

5. Configure onconfig

Update onconfig files with SSC cluster settings.

# cd /data/ssc/etc
# cp onconfig.std onconfig.server1
# cp onconfig.std onconfig.server2
Enter fullscreen mode Exit fullscreen mode

For Primary Node (server1):

ROOTPATH rootdbs
LTAPEDEV /dev/null
DBSERVERNAME server1
SERVERNUM 130
SDS_TEMPDBS sdstmpdbs1, /data/ssc/tmp1/sdstmpdbs1,2,0,16000
SDS_PAGING /data/ssc/tmp1/sdstmp1,/data/ssc/tmp1/sdstmp12
SDS_ENABLE
Enter fullscreen mode Exit fullscreen mode

For Secondary Node (server2):

ROOTPATH rootdbs
LTAPEDEV /dev/null
DBSERVERNAME server2
SERVERNUM 131
SDS_TEMPDBS sdstmpdbs1, /data/ssc/tmp2/sdstmpdbs1,2,0,16000
SDS_PAGING /data/ssc/tmp2/sdstmp1,/data/ssc/tmp2/sdstmp12
SDS_ENABLE 1
Enter fullscreen mode Exit fullscreen mode

Note: Ensure /data/ssc/tmp1 and /data/ssc/tmp2 directories are created beforehand.

6. Set Up SDS Cluster

On the primary node, run:

# source /data/ssc/profile.server1
# cd /data/ssc/storage
# oninit -ivwy
# onmode -d set SDS primary server1
Enter fullscreen mode Exit fullscreen mode

On the secondary node, run:

# source /data/ssc/profile.server2
# cd /data/ssc/storage
# oninit -vwy
Enter fullscreen mode Exit fullscreen mode

7. Verify Cluster Setup

To verify, run the following on the primary node. The successful output should resemble the below:

# onstat -g sds

Your evaluation license will expire on 2025-10-21 00:00:00
On-Line -- Up 00:00:59 -- 156976 Kbytes

Local server type: Primary
Number of SDS servers:1

SDS server information

SDS srv      SDS srv      Connection        Last LPG sent        Supports
name         status       status            (log id,page)        Proxy Writes
server2      Active       Connected               6,2872         N
Enter fullscreen mode Exit fullscreen mode

Following these steps, you have successfully set up an SSC shared storage cluster. We hope this guide has helped you better understand and utilize shared storage cluster technology. If you have any questions or need further assistance about GBase database (GBase数据库), feel free to reach out. Thank you for reading, and we look forward to seeing the results of your implementation!

Top comments (0)

Read next

cezz profile image

Why should you upgrade your PostgreSQL today?

Cezzaine Zaher -

sachingaggar profile image

React Native with WatermelonDB: A Lightweight and Reactive Database for Scalable Apps

Sachin Gaggar -

esproc_spl profile image

Java, perform COUNT on each group of a large csv file #eg33

Judy -

mrcaption49 profile image

Common Table Expression (CTE) in PLSQL | Part 1

Pranav Bakare -