VMware Cloud Community
Ed000
Contributor
Contributor
‎10-17-2014 03:41 AM

SQL Cluster Group - Active node affected when restarting passive node

I have a SQL cluster group on windows 2012 and esxi 5.5 and find that the active node is affected when I restart my passive node. It Initially caused an active node restart due to disks becoming unavailable for writing to on the active node and I believe it is covered by this article - VMware KB: The RDM LUN resets while powering on the virtual machine  [note - reference updated]

The article doesn't make reference to a clustered environment though and as it doesn't seem correct behavior for a cluster, I thought I'd check to validate that this is indeed expected. Is it and if so, what can I do to counter this?

Thanks

Tags (3)
0 Kudos
1 Reply
JPM300
Commander
Commander
‎10-17-2014 05:09 AM

Hey Ed000,

We had this issue with one of our UCS clusters awhile back.  The hosts did take an excessive amount of time to boot that had RDM volumes connected to it prior to putting the fix in, but it did not cause the cluster to be affected the way yours is.  Are all your Clustered disks setup properly and are RDM's?  Do the RDM's have access to both hosts?  Are you using any MSCS File shares?  If your using MSCS File Shares they HAVE to be created through the failover cluster snap in, otherwise the other host will not see them when a failure occurs. 

0 Kudos