Server 2012 R2 failover cluster

Hi Guys,

I created a 2 node cluster using server 2012 R2 ..I have VMM 2012 R2 running on both nodes with SQL 2012 SP1

VMM is configured for High availability and SQL always-on feature is also enabled.

Now when I stop cluster service on one of the node, the roles and quorum disk failover to the other node but the quorum disk does not come back online and as a result the cluster fails until I bring back the cluster service on the current owner (owner of the quorum disk)

I'm at a loss for what to do, any help will be greatly appreciated.

Richard

December 18th, 2013 2:37pm

Richard,

How is the quorum disk shared to the nodes? (i.e. iSCSI, FC)

Also have you configured MPIO?

Free Windows Admin Tool Kit Click here and download it now
December 18th, 2013 3:29pm

the disk is shared to both nodes vis FC (I believe you mean fibre channel by this?)

I have also configured MPIO from control panel on both servers

December 18th, 2013 3:35pm

When I move the quorum disk from one node to another, it moves but fails (it refuses to come back online) until i move it back.
Free Windows Admin Tool Kit Click here and download it now
December 18th, 2013 3:53pm

Is MPIO correctly configured on both nodes?  What's the output of mpclaim -s -d and then the output of mpclaim -s -d # where # is the number of the quorum disk?  Same on both nodes?

Validation report show any warning on the quorum disk?  I will often run the validation report just against the quorum disk because that does not impact the running of the clustered application.

I don't know if this is a problem or not, but everything I read about making SCVMM highly available talks about the SQL environment being on a different set of systems.  I have not seen anything that says it's fine to install HA SQL and HA SCVMM on the same hosts.  You might want to ask about that over in the SCVMM forum.  There is even a forum specific to VMM clustering - http://social.technet.microsoft.com/Forums/en-US/home?forum=virtualmachinemgrclustering

December 18th, 2013 11:29pm

Richard,

Just to follow up with what Tim said, more for clarification. Your VMM and SQL servers are not on the Hyper-V hosts are they?

I am just trying to understand your environment a little better.

Free Windows Admin Tool Kit Click here and download it now
December 19th, 2013 12:05am

Hi Guys,

when i run the said commands on both  nodes I get the same result ..see below

Validation report shows no warning on the quorum disk

VMM and SQL are not on hyper-v hosts. I have 2 physical machines, each running VMM 2012 R2 (HA) and SQL 2012 SP1 (Always On). Do you believe hosting both on the same server could be the issue? I will look that up too.

The cluster is fine when Node2 owns the quorum disk.

Roles on Node1 successfully fail over to Node2 when I stop cluster service on Node1

When I stop Cluster service on Node2 roles on Node2 move to Node1 (including the quorum disk) but the Disk fails to come online and as a result quorum is lost and the cluster fails until I resume cluster service on Node2.

It looks to me like Node2 owns the quorum Disk explicitly ..

December 19th, 2013 11:30am

Why is this marked as an answer to my post???
Free Windows Admin Tool Kit Click here and download it now
January 13th, 2014 6:14am

This topic is archived. No further replies will be accepted.

Other recent topics Other recent topics