Create a standby instance|V4.3.0|OceanBase Deployer|docs|OceanBase Documentations

This topic describes how to use obd to create a standby instance of the database named seekdb.

Concepts

The primary/standby architecture is a high-availability solution provided by seekdb. It synchronizes data from the primary instance to the standby instance, which is a read-only instance. When the primary instance is unavailable due to planned or unplanned reasons, the standby instance can take over the service to minimize service downtime and reduce potential data loss.

The primary instance (PRIMARY) is a business instance created by the user and supports complete database capabilities, including queries, DML, and DDL operations. The standby instance (STANDBY) only provides disaster recovery and read-only capabilities.

Prerequisites

The primary instance of seekdb must be of V1.2.0 or later and managed by obd.

The default primary instance of seekdb is managed by obd. If the primary instance of seekdb is not managed by obd, you can use obd to take over the corresponding instance. For more information, see the Take over the instance section in Use obd to take over an instance.
The primary instance must be running. You can execute the obd seekdb display command to check the status of the primary instance of seekdb.
For a simple experience, the primary instance must have at least 1C1GB memory and 3GB of log disk space. For long-term use, we recommend that the primary instance have at least 1C4GB memory.

Procedure

This section describes how to create a standby instance in three scenarios.

Scenario 1: One primary instance and one standby instance

In this scenario, two instances of the seekdb type exist: seekdb_A is the primary instance, and seekdb_B is the standby instance of seekdb_A.
Scenario 2: One primary instance and two standby instances

In this scenario, three instances of the seekdb type exist: seekdb_A is the primary instance, and seekdb_B and seekdb_C are the standby instances of seekdb_A.
Scenario 3: Cascading primary and standby instances

In this scenario, three instances of the seekdb type exist: seekdb_A is the primary instance, seekdb_B is the standby instance of seekdb_A, and seekdb_C is the standby instance of seekdb_B.

Scenario 1: One primary instance and one standby instance

Scenario 2: One primary instance and two standby instances

Scenario 3: Cascading primary and standby instances

Create a standby instance

Execute the following command to create a standby instance. This command starts an interactive deployment process. In the interactive deployment process, configure the information of the seekdb standby instance and select the deployment name of the primary instance. For more information about how to execute this command, see the Interactive deployment of seekdb section in Deploy seekdb.
```
obd seekdb install --standby
```

View the primary-standby relationship of the seekdb instance

obd seekdb display seekdb_B -g

The -g parameter displays the primary-standby instance topology. The output is as follows:

Get local repositories and plugins ok
Open ssh connection ok
Connect to seekdb 10.10.10.1:2881 ok
Wait for seekdb init ok
+--------------------------------+
|             seekdb             |
+---------------+---------+------+
| ip            | version | port |
+---------------+---------+------+
| 10.10.10.1    | 1.2.0.0 | 2881 |
+---------------+---------+------+
obclient -h10.10.10.1 -P2881 -uroot -p'********' -Doceanbase -A

Cluster Topology:

seekdb_A
└── seekdb_B

Verify the synchronization

MySQL [(oceanbase)]> select * from __all_virtual_server_stat\G

The output is as follows:

*************************** 1. row ***************************
                 svr_ip: 127.0.0.1
               svr_port: 2882
               sql_port: 2881
               rpc_port: 2882
           cpu_capacity: 8
       cpu_capacity_max: 8
           cpu_assigned: 2
       cpu_assigned_max: 2
           mem_capacity: 1073741824
           mem_assigned: 1073741824
     data_disk_capacity: 150403547136
       data_disk_in_use: 31457280
data_disk_health_status: NORMAL
data_disk_abnormal_time: 0
      log_disk_capacity: 3221225472
      log_disk_assigned: 3221225472
        log_disk_in_use: 134217728
   rpc_cert_expire_time: 2089754466000000
        rpc_tls_enabled: 1
           memory_limit: 1073741824
    data_disk_allocated: 21474836480
     data_disk_assigned: NULL
     start_service_time: 1774423270740291
            create_time: 1774423267357315
                   role: STANDBY
      switchover_status: NORMAL
     log_restore_source: 10.10.10.1:2882
               sync_scn: 1774427027950470000
           readable_scn: 1774427027950470000

Create a standby instance seekdb_B for the primary instance seekdb_A

After you execute this command, an interactive deployment process starts. In the interactive deployment process, configure the information of the seekdb_B standby instance and select the deployment name of the primary instance. For more information about how to execute this command, see the Interactive deployment of seekdb section in Deploy seekdb.
```
obd seekdb install --standby
```
Repeat the previous operation to create a standby instance seekdb_C for the primary instance seekdb_A.

View the primary-standby relationship of the seekdb instance

Execute the following command to view the primary-standby relationship of the seekdb instance. Here, the seekdb_B instance is used as an example.

obd seekdb display seekdb_B -g

The -g parameter displays the primary-standby instance topology. The output is as follows:

Get local repositories and plugins ok
Open ssh connection ok
Connect to seekdb 10.10.10.1:2881 ok
Wait for seekdb init ok
+--------------------------------+
|             seekdb             |
+---------------+---------+------+
| ip            | version | port |
+---------------+---------+------+
| 10.10.10.1    | 1.2.0.0 | 2881 |
+---------------+---------+------+
obclient -h10.10.10.1 -P2881 -uroot -p'********' -Doceanbase -A

Cluster Topology:

seekdb_A
├── seekdb_B
└── seekdb_C

Verify the synchronization

MySQL [(oceanbase)]> select * from __all_virtual_server_stat\G

The output is as follows: For more information, see View the synchronization status of an instance in the seekdb documentation.

*************************** 1. row ***************************
                 svr_ip: 127.0.0.1
               svr_port: 2882
               sql_port: 2881
               rpc_port: 2882
           cpu_capacity: 8
       cpu_capacity_max: 8
           cpu_assigned: 2
       cpu_assigned_max: 2
           mem_capacity: 1073741824
           mem_assigned: 1073741824
     data_disk_capacity: 150403547136
       data_disk_in_use: 31457280
data_disk_health_status: NORMAL
data_disk_abnormal_time: 0
      log_disk_capacity: 3221225472
      log_disk_assigned: 3221225472
        log_disk_in_use: 134217728
   rpc_cert_expire_time: 2089754466000000
        rpc_tls_enabled: 1
           memory_limit: 1073741824
    data_disk_allocated: 21474836480
     data_disk_assigned: NULL
     start_service_time: 1774423270740291
            create_time: 1774423267357315
                   role: STANDBY
      switchover_status: NORMAL
     log_restore_source: 10.10.10.1:2882
               sync_scn: 1774427027950470000
           readable_scn: 1774427027950470000

Create a standby instance seekdb_B for the primary instance seekdb_A

After you execute this command, an interactive deployment process starts. In the interactive deployment process, configure the information of the seekdb_B standby instance and select the deployment name of the primary instance. For more information about how to execute this command, see the Interactive deployment of seekdb section in Deploy seekdb.
```
obd seekdb install --standby
```
Repeat the previous operation to create a standby instance seekdb_C for the instance seekdb_B.

View the primary-standby relationship of the seekdb instance

Execute the following command to view the primary-standby relationship of the seekdb instance. Here, the seekdb_B instance is used as an example.

obd seekdb display seekdb_B -g

The -g parameter displays the primary-standby instance topology. The output is as follows:

Get local repositories and plugins ok
Open ssh connection ok
Connect to seekdb 10.10.10.1:2881 ok
Wait for seekdb init ok
+--------------------------------+
|             seekdb             |
+---------------+---------+------+
| ip            | version | port |
+---------------+---------+------+
| 10.10.10.1    | 1.2.0.0 | 2881 |
+---------------+---------+------+
obclient -h10.10.10.1 -P2881 -uroot -p'********' -Doceanbase -A

Cluster Topology:

seekdb_A
└── seekdb_B
    └── seekdb_C

Verify the synchronization

MySQL [(oceanbase)]> select * from __all_virtual_server_stat\G

Output:

*************************** 1. row ***************************
                 svr_ip: 127.0.0.1
               svr_port: 2882
               sql_port: 2881
               rpc_port: 2882
           cpu_capacity: 8
       cpu_capacity_max: 8
           cpu_assigned: 2
       cpu_assigned_max: 2
           mem_capacity: 1073741824
           mem_assigned: 1073741824
     data_disk_capacity: 150403547136
       data_disk_in_use: 31457280
data_disk_health_status: NORMAL
data_disk_abnormal_time: 0
      log_disk_capacity: 3221225472
      log_disk_assigned: 3221225472
        log_disk_in_use: 134217728
   rpc_cert_expire_time: 2089754466000000
        rpc_tls_enabled: 1
           memory_limit: 1073741824
    data_disk_allocated: 21474836480
     data_disk_assigned: NULL
     start_service_time: 1774423270740291
            create_time: 1774423267357315
                   role: STANDBY
      switchover_status: NORMAL
     log_restore_source: 10.10.10.1:2882
               sync_scn: 1774427027950470000
           readable_scn: 1774427027950470000

References

For more information about seekdb, see seekdb documentation.
For more information about how to perform a primary-standby switch or decoupling by using the obd command, see Role switching and decoupling.