Running a failover procedure

The Storwize® Family SRA completely automates the failover procedure.

Note:

For IBM SVC Stretched Cluster, the synchronization between the two copies of a stretched volume is managed by SVC stretched cluster, so the active-active replication within a stretched volume is always running, unless one of the copies is offline.
For IBM HyperSwap, the synchronization between the master volume and the aux volume of a hyperswap volume is managed by the HyperSwap function, so the active-active replication within a hyperswap volume is always running, unless one of the sites is offline.

For a planned failover, the Metro Mirror/Global Mirror direction is switched. Global Mirror with Cycling Mode is stopped with the target volume writable.

For an unplanned failover, the Metro Mirror/Global Mirror/Global Mirror with Cycling Mode is stopped with the target volume writable.

The Metro Mirror/Global Mirror/Global Mirror with Cycling Mode target volume is prompted to the vSphere Server as a production volume. During an SRA preconfigured failover procedure, the preconfigured FlashCopy® is resynchronized.

SRA Failover procedure prerequisites:

Ensure that all the steps that are outlined in Configuring preconfigured environments are finished.
Ensure that all the steps that are outlined in Configuring non-preconfigured environments are finished.