I have a 3 node cluster all running Windows Server 2008 R2. Roughly once a month I see my vInstance become degraded and attempt to fail-over. Everything is good as long as it fail-over to SQL01 or SQL02. However if it attempts to fail-over to SQL03, it does not come online

Quick resolution is to move it manually to SQL01 or SQL02. What could be causing it to fail every time on SQL03.

A couple points:

I did not build the environment.

I am not a DBA.

I only have general knowledge of SQL clustering.

I always get two EVENT ID's: 1069

Cluster resource 'SQL Server (VSQL04)' in clustered service or application 'SQL Server (VSQL04)' failed.

and then

EVENT ID 1205

The Cluster service failed to bring clustered service or application 'SQL Server (VSQL04)' completely online or offline. One or more resources may be in a failed state. This may impact the availability of the clustered service or application.

Where should I begin to look for issues?

can you check the event viewer log or the windows cluster log .. it should tell you the reason why it was not able to bring online on that??

Here is the cluster event prior to offline state. I will have to go check the cluster log.

The Cluster service failed to bring clustered service or application 'SQL Server (VSQL04)' completely online or offline. One or more resources may be in a failed state. This may impact the availability of the clustered service or application.

Here is the cluster event prior to offline state. I will have to go check the cluster log.

The Cluster service failed to bring clustered service or application 'SQL Server (VSQL04)' completely online or offline. One or more resources may be in a failed state. This may impact the availability of the clustered service or application.

i do not think this helps.. it just says..a resource in offline state.. you need to dig more and see which one it is and why it did not come banck on ..it should be mentioned in the log and\or event viewer.

Hello,

When this happens you'll want to run (it may still be in the log) the following:

import-module failoverclusters
get-clusterlog -Destination "c:\mylogfolder\"

This will grab the cluster logs from all of the nodes and put them in the destination path. Post back with the log when it happens again. It may still be in the cluster log, so it may be wise to run it now and check.

-Sean