Disaster Recovery

For disaster recovery, you must have at least one additional array that is configured as a replication partner. Replication partners can serve data to the original initiators while the original array is inaccessible. By including multiple arrays in your network, you can quickly restore access to data even in case of a catastrophic failure.

In the unusual event of a complete failure of the array, set the replication partner online and point your initiators to the volumes on the replication partner. Your volume data is available to applications during the recovery of the data to the failed array.

There are two methods to move operations from one volume collection to its replication partner; the handover method and the promotion and demotion method. If the original array is accessible, handover is always preferred.

NOTE: You cannot use the promotion and demotion method with synchronous replication.

In case of a complete failure, within the Site Recovery Manager (SRM) the "failback" procedure is to delete the failed VM from vCenter, then replicate the backed-up LUNs to a new VM.

IMPORTANT: Follow the procedures in this section to perform disaster recovery using the array only. For information on disaster recovery in HPE Cloud Volumes, refer to the HPE CV portal documentation.