I need some advice from those using Backup Exec to backup highly available virtual machines running on a Hyper-V cluster using Cluster Shared Volumes.
The admin guide is limited on it's Hyper-V information, it explains how to backup VM's running on Hyper-V hosts but there is not much information on Hyper-V clusters.
Is there a more in depth article somewhere? If not, I am just starting down this path and so far have these questions.
- According to the admin guide "When virtual machines are configured for high availability, they are moved to a new node in the backup selection list. Clustered virtual machines appear under the name of the cluster, in the Highly Available Hyper-V Machines node." I assume that if I deploy the RAWS agent to all hyper-v hosts that are part of a cluster, the cluster name it's self can then be expanded as per the above article for VM selection? i.e Backup Exec will instantly recognize it as a cluster? Obviously I can't do a push install of the RAWS agent to the cluster name it's self and expect magic to happen although that would be easier.
- GRT technology is used for restoring individual files from within a full VM backup. Do these GRT restores behave the same way that Exchange GRT restores do in that if the backup has gone straight to tape the restoration of any file will require the full GRT tape backup to be restored to an image file on disk before the individual file/s themselves can be restored?
- If a HA VM is restored in full back to a cluster (say the VM was accidentally deleted), does the VM get restored onto the Hyper-V host it was originally running on or does it decide at the time which host is best suited based on current load?
More information from Symantec on Backup Exec behavior with Hyper-V and Cluster Shared Volumes would be welcome as I believe there would be a lot of people making use of LiveMigration and HA VM's, Cluster Shared Volumes is the only way to achieve this.
