During a failback, if prepare_failback was executed successfully on the secondary companion but the primary companion fails to boot, perform the following to rollback and then reissue the prepare_failback command:
Check the primary companion’s error log the HP MC/ServiceGuard package log, or the system log to find the reason the server failed to boot, and correct the problems.
If the package for the primary companion is running on the primary node, halt the package.
Login to the secondary companion and issue:
dbcc ha_admin ("", "rollback_failback")
dbcc ha_admin ("", "rollback_failover")
Verify secondary companion is in normal companion mode
As root, start up the package for the primary companion to run on secondary node.
/usr/sbin/cmrunpkg -n secondary_node primary_companion_package_name
Your secondary companion is now in failover mode. Once you verify that everything is ready for the primary companion to failback to normal companion mode, you can issue sp_companion...prepare_failback.
