Service Migration and Switchback After the Production Center Recovers from a Fault

In scenarios where application resources and persistent volumes are replicated across clusters, if data or applications in the production center cannot be used due to disasters or faults, you can perform fault recovery to quickly start services in the DR center. You can only perform this operation on the eReplication system in the DR center.

After the fault in the production center is rectified, if you want to switch services back from the DR center to the production center, you need to synchronize the data generated in the DR center during the fault period to the production center.

Prerequisites

Protection policies and recovery plans have been configured.
Ensure that the Kubernetes cluster at the DR site has the same image as that at the production site.

Context

If data or applications become unavailable due to disasters or faults in the production center, services will be quickly switched to and started in the DR center. After the production center recovers, the services will be switched back to it by performing planned migrations.

Procedure

Perform a fault recovery.
1. On the menu bar, choose Utilization > Data Restore.
2. Select the recovery plan for which you want to perform fault recovery. In the Operation area, choose More > Fault Recovery.
3. Perform fault recovery based on protected object types.
  - When the protected object type is Application Resource:
    Optional: Select Post-hook.
    
    Click OK.
  - If the protected object type is Persistent Volume, click OK.
In the DR center, check the application startup.
After the fault is rectified, check whether the application and data are normal. If the application or data is abnormal, contact Huawei technical support.
Perform reprotection to protect services switched to the DR center.
After the fault recovery is complete, the application system is running at the DR site and the protected group becomes invalid. You must perform reprotection to recover the replication status and synchronize the data in the DR site to the production site. Then, the original DR center becomes the new production center.
1. On the menu bar, choose Utilization > Data Restore.
2. Select the recovery plan with data to be reprotected, click More in the Operation area and then click Reprotection.
In the scenario of cross-cluster replication for application resources, service loads related to application resources in the original production cluster are deleted during reprotection.
In the confirmation dialog box that is displayed, carefully read the content and click OK.
- After the reprotection is complete, if you plan to run services in the DR center for a long time, you do not need to perform 5 to 6.
- Log in to the service system at the original DR site and check the service startup conditions.
Perform planned migration on eReplication.
Perform planned migration to switch services back from the DR center to the production center. For details, see 1 and 2 in Planned Migration and Switchback of Services in the Production Center.
Perform reprotection again on eReplication.
Copy data from the production center to the DR center to recover the protected group status.