07-23-2013 06:26 AM
When a Backup Exec 2012 job on our Windows 2008 R2 Hyper-V cluster fails (in our case usually due to "activity on a volume is preventing all volumes from being snapped" or "unable to get minimum quiet time for physical volumes"), Backup Exec leaves the CSV volume in a redirected access state. The volume will say "Online (Backup In Progess)" and it never gets released from this redirected state. The backup software is required to send proper call (according to MSDN it is the ClusterClearBackupStateForSharedVolume) to the cluster to release the volume when it is done, and it appears it is not doing this when the job fails. The only way for me to get the volume out of redirected access is to simulate a failure on the resource, forcing it to move the volume to another node and bring it back online. If any virtual machines are running on this volume, they will crash when simulating the failure unless I shut them down prior to simulating the failure.
This is becoming a hassle to have this happen repeadily during the course of a week and Symantec needs to address this issue.
07-23-2013 06:43 AM
See if this technote helps
http://www.symantec.com/docs/TECH205414
07-23-2013 06:59 AM
I had already seen that, we aren't using SCVMM 2012, and the 2nd "solution" is to disable backing them up. WTH kind of answer is that?
07-23-2013 08:15 AM
I suggest you to open a support case with Symantec for faster solution.