Optimizing Exchange Cluster failover backups and restores (4047895)

Date - 01/2008
Affected Product & Version - NetVault: Backup 7.4.x and above
Affected Module & Version - Exchange Server APM 3.1.x
OS Version - Windows 2003
Application Information - Exchange 2003/2007


Description:

This article shows how to optimize your Exchange server backup strategy in a failover cluster using the Job retries
option in NetVault: Backup.


Symptoms:

Many customers have inquired about the NetVault's ability to keep backing up a cluster node even if this acitve node fails in the middle of a backup.
According to their expectations, NetVault would have to be 'intelligent' enough to proceed backing up the same data/file from the second
failover node once it is up, without failing the job.

This is not supported and not currently possible with NetVault.

Once the communication with the target client is lost, the data channel previously established to stream data, will inevitably be closed and the job fail.
Furthermore another variable that makes it impossible is the failover recovery time to the other node that is never a definite value.

Bearing in mind that we communicate with the active node in a cluster via a Virtual Client that has its own virtual IP address,
it is however possible to implement a simple workaround that will allow the same backup job to occur regardless of which the node is active:
Retrying the job.