Re: [Veritas-bu] backing up MS cluster

Configure a virtual server on your cluster for each database instance. Assign a dedicated IP address for this virtual server. Make those drives a dependency of that cluster resource and back them up that way. We've been doing this for a long time and it works fairly well. You'll have one policy that backs up the static data on each physical node ( e.g. the C: drive and the system_state) and then 1 policy per database instance that backs up the appropriate drives associated with that instance. As a node fails and the virtual server fails over to another node, the drives and the backup IP will go with it. NetBackup won't notice the difference unless it's in the middle of a backup in which case the job dies, restarts/retries, and continues right on going.

.../Ed

On 12/13/07, X_S <netbackup-forum AT backupcentral DOT com> wrote:

We have 4 MS 2003 servers with MS multi-node cluster configurations for SQL 2005. What is the best way to configure the policies? We have tried using both the All Local Drives directive and specifying each individual drive letters (pain when resources fail over). But what concerns us how NB reacts when resources are moved or failed over. we have had instances where resources were moved but NB didn't pick the drive up on the new owner-node.

--
Ed Wilts, Mounds View, MN, USA
mailto:ewilts AT ewilts DOT org

_______________________________________________
Veritas-bu maillist  -  Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu