May 20, 2011 at 8:31 am
We have 2 two-node clusters running Windows Server 2008 R1 and SQL Server 2008 R1, and we are in the process of upgrading to SQL Server 2008 R2. The first (Dev/Test) cluster has nine instances, and the second (Prod) cluster has six instances.
Prior to upgrading, I googled the web and found a couple of fairly extensive Microsoft articles which outlined what appeared to be the proper procedure for upgrading to R2, and we have successfully completed three of the first nine instances. We had a bunch of Registry permission issues for one of the upgrades, but we operate in an extremely locked down environment, and, unfortunately, Operating System permissions tend to “bite us” frequently because of it. In all other ways, the upgrades of those three instances seemed to go well.
A couple of days ago I ran across a white paper, “Applying Updates to a Clustered Instance of SQL Server 2008 or SQL Server 2008 R2,” on the Quest site written by Allan Hirt, the author of “Pro SQL Server 2008 Failover Clustering,” which was an exceptionally good book for us when we created the R1 clusters. In his white paper, Allan writes, “Before you do any installations of the SQL Server update, you need to reconfigure an instance’s network name resource to remove the node(s) that will be patched.”
I have not seen similar advice from any other source. Is Allan’s recommendation valid? Does anyone have experience with this for R2? I hesitate to go “dinking around” with a cluster that works if I don’t need to do so. I hate to invite Murphy’s Law into anything…
May 20, 2011 at 9:51 am
I would tend to listen to Allan as he's had a lot of experience with clusters.
However, I would use VMWare/VirtualBox/HyperV to set up a test cluster and experiment there.
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply