Recovering from a failed prepare_failback

During fail back, if prepare_failback executes successfully on the secondary companion but the primary companion also fails, roll back and then reissue the prepare_failback command:

1. Check the cluster system error logs, callback error logs, high availability agent fault monitor error logs, and Adaptive Server error logs to find the reason the failback failed, and correct any problems.

2. Clear any error states in the resource group. To determine the states of resource group, enter:

   scha_resourcegroup_get -O RG_STATE -G resource_group_name
   

   For example:

   scha_resourcegroup_get -O RG_STATE -G rg_MONEY1
   

   To determine the states of resource group, enter:

   scha_resource_get
   -O RESOURCE_STATE_NODE
   -R resource_name node_name
   

   For example, to find the state of the resource ase_MONEY1 on node2:

   scha_resource_get
   -O RESOURCE_STATE_NODE -R ase_MONEY1 node2
   

   Issue the following command to clear the STOP_FAILED state:

   scswitch -c -h node_name -j resource_name -f STOP_FAILED
   

3. Log in to the secondary companion and issue:

   dbcc ha_admin("", "rollback_failback")