#Restart Instance

If an instance encounters operational faults, excessive connection counts, or other unknown issues that cannot be identified during its operation, problems can be resolved by restarting the instance.

#Prerequisites

The instance must be in a Running state.

Note:

During the restart process, Pods will be restarted one by one, which may cause a temporary interruption of services.

If the instance is in a Processing state, it indicates that the instance might be starting normally; please be patient.

If the instance fails to start or is abnormal, meaning the instance is in a Processing or Error state, it will automatically restart after the health check times out; please be patient.

#Procedure

kubectl patch mysql -n <namespace> <name> --type='merge' --patch '
spec:
    pxc:
      pxc:
          annotations:
              percona.com/Restart-time: '$(date -u +'%Y-%m-%dT%H:%M:%SZ')'
      proxysql:
          annotations:
              percona.com/Restart-time: '$(date -u +'%Y-%m-%dT%H:%M:%SZ')'    
'