Scenarios
When the production servers and disks in the primary AZ become faulty due to force majeure and a failover has been performed, enable the DR servers and disks to provide services. After the fault in the faulty AZ has been restored, users can make API calls to reprotect a replication consistency group, that is, to enable the restored resources in the faulty AZ to function as DR servers and DR disks.
EVS replication APIs have been deprecated. If you need to use the replication function, see Storage Disaster Recovery Service User Guide and Storage Disaster Recovery Service API Reference.
Prerequisites
- A failover for the replication consistency group has been performed.
- The faulty AZ has been restored.
- The replication consistency group status is available.
- The restored production server has been stopped.
Procedure
- Perform the reprotection for the replication consistency group to synchronize the disk data of the DR servers to the servers in the faulty AZ.
For details, see Reprotecting a Replication Consistency Group in the Elastic Volume Service API Reference.
- Query the details about the replication consistency group.
For details, see Querying Details About a Replication Consistency Group in the Elastic Volume Service API Reference.
After the request has been delivered, the replication status of the replication consistency group changes to copying. Once the replication status changes to active, the data synchronization is complete.
The time required for the data synchronization within a replication consistency group varies depending on the number of EVS replication pairs in the group, EVS disk capacity, and amount of differential data generated after a primary AZ fault occurred. Normally, the data synchronization of a reprotection takes a relative long period of time. Please wait patiently.
- (Optional) If the services need to be migrated back to the restored primary AZ, you need to perform a planned migration. For details, see Planned Migration (Deprecated).