Currently,
if the secondary companion is brought down when its in failback status
(that is, @cmpstate = 14)
and Failback is set to Automatic, the resource group cannot be brought
back online by FailSafe.
To recover from this situation, temporarily set IRIX FailSafe to Manual Failback, bring the resource group online, then switch back to Automatic Failback, then you can continue with the Automatic Failback from whichever step you last left off. To recover:
Set IRIX FailSafe to manual failback on the secondary node. As root, enter the following at the command line:
cd /var/cluster/ha/resource_types/SYBASE_DB
As root, edit the SWITCHES file so that AUTO_FAILBACK is set to 0.
FailSafe does not need to be restarted for the new setting
to take effect.
Bring the resource group online.This is the same as described in “Bring the primary resource group online”, but use sec_rg instead of pri_rg for the Resource Group name.
Set IRIX FailSafe to automatic failback on the secondary node. As root, move to the following directory:
cd /var/cluster/ha/resource_types/SYBASE_DB
As edit the SWITCHES file to set AUTO_FAILBACK to 1.
