RECOVER STANDBY

Purpose

After physical recovery is completed or a physical standby tenant is created, you can use the ALTER SYSTEM RECOVER STANDBY statement to specify the restore endpoint for the standby tenant's logs. This allows the standby tenant to either replay logs in segments or continuously replay logs.

Privilege requirements

This statement must be executed by the root user of the sys tenant (root@sys) or the administrator user of each tenant. Specifically:

• In MySQL mode, the default administrator user is root.
• In Oracle mode, the default administrator user is SYS.

Syntax

ALTER SYSTEM RECOVER STANDBY
    [TENANT [=] tenant_name] UNTIL { TIME='timestamp' | SCN=scn_no | UNLIMITED };

Parameters

Examples

• For the sys tenant:

  • In a backup and restore scenario, the sys tenant specifies the standby tenant restore_oracle_tenant to replay logs up to the specified timestamp '2023-06-01 00:00:00'.

    obclient [oceanbase]> ALTER SYSTEM RECOVER STANDBY TENANT restore_oracle_tenant UNTIL TIME='2023-06-01 00:00:00';
    

  • In a physical standby database scenario, the sys tenant specifies the standby tenant restore_oracle_tenant to enter continuous synchronization mode.

    obclient [oceanbase]> ALTER SYSTEM RECOVER STANDBY TENANT restore_oracle_tenant UNTIL UNLIMITED;
    

• For a user tenant:

  • In a backup and restore scenario, the standby tenant restore_oracle_tenant replays logs up to the specified timestamp '2023-06-01 00:00:00' for itself.

    obclient [oceanbase]> ALTER SYSTEM RECOVER STANDBY UNTIL TIME='2023-06-01 00:00:00';
    

  • In a physical standby database scenario, the standby tenant restore_oracle_tenant puts itself into continuous synchronization mode.

    obclient [oceanbase]> ALTER SYSTEM RECOVER STANDBY UNTIL UNLIMITED;
    

References

• Segmented log replay for standby tenants

• Set the log restore endpoint