Failback Test

  To perform the failback test

1. Reconnect the ESCON cable and reboot the original R1 hosts. You must manually resynchronize the device, which can be done only if both sides are write-disabled. This requires taking the service group offline.
2. If you are running this test in a replicated data cluster, run the following command from any host:
   # hagrp -offline grpname -any

   If you are running the test in a global cluster, run the command from hostc or hostd.

3. After the service group goes offline, run the following command:
   # symrdf -g device_group restore

   The variable device_group represents the name of the RDF device group at the R2 side. The restore command determines which tracks to merge between the R1 and R2 arrays and initiates the resynchronization. The operation of this command write disables both sides; use this command only when a brief downtime is acceptable.

4. Bring the service group online at the R1 side:
   # hagrp -online grpname -sys hosta

The devices synchronize, and the environment state will be the same as when the test began.