Hi,
Currently we have a HyperV Cluster with 8 nodes all accessing a Clustered Shared Volume. This is HyperV 2008 R2.
I understand the limitations of the Clustered Shared Volume so on that basis I have created a job for each node in the HyperV cluster.
The problem is as follows.
Hyper V Host A has 6 guest Machines. Server 1,2,3,4,5,6 etc
The initial job is created and the backup runs successfully
If I then live migrate guest Server 6 to Host B, I have to manually change the job setup for Host A. If I don''t then I will get a 156 error as only one host can run VSS on the Clustered Shared Volume. So either one guest will fail or the 5 guests on the first node will fail. I know I can add a timeout value for the Clustered Shared Volume access but that will not really solve this problem because of the number of Guests we run on our 8 nodes.
Is it possible that Hyper V backups can query their nodes prior to each backup so that it only backups up the guests which are only running on that host. That way a reasonable backup windows can be defined for each host without the problem of multiple hosts trying to create a snapshot on the Clustered Shared Volume and failing.