Stop / start redo apply or redo transport with the broker; configure delayed apply
OracleData GuardCommand
# Redo apply on the standby:
EDIT DATABASE '<standby_db_unique_name>' SET STATE='APPLY-OFF';
EDIT DATABASE '<standby_db_unique_name>' SET STATE='APPLY-ON';
# Redo transport on the primary (to all standbys):
EDIT DATABASE '<primary_db_unique_name>' SET STATE='TRANSPORT-OFF';
EDIT DATABASE '<primary_db_unique_name>' SET STATE='TRANSPORT-ON';
# Delayed apply (minutes; protection against logical errors):
EDIT DATABASE '<standby_db_unique_name>' SET PROPERTY DelayMins=240;
SHOW DATABASE '<standby_db_unique_name>' DelayMins;What it does
Persistently stops/starts redo apply (MRP) on the standby or redo shipping on the primary through the broker; DelayMins holds redo for the given time before applying it.
When to use it
For maintenance or backups on the standby, planned network outages, controlling transport during large bulk loads on the primary, or building a delayed standby against accidental deletes.
Caution
While the broker is enabled, do not use ALTER DATABASE RECOVER MANAGED STANDBY DATABASE CANCEL in SQL; the broker restores the state. During APPLY-OFF the apply lag and the failover time (RTO) grow; during TRANSPORT-OFF data generated on the primary is unprotected (RPO). If FSFO is enabled these operations can prevent a failover.
Related
- Show the Data Guard broker configuration and database status (DGMGRL)
- Check primary/standby role and switchover readiness with SQL
- Measure transport/apply lag, apply rate and Data Guard processes on the standby
- Validate the standby with the broker before switchover/failover (VALIDATE DATABASE)
- Perform a planned switchover (DGMGRL, role change without data loss)
- Perform a failover after losing the primary (DGMGRL)