Snapshot Windows Failover Cluster Servers

This post was originally published on this site

What is the best way to snapshot the nodes of a Windows Failover Cluster that has SQL Server Availability Groups installed.  I want to make sure I can revert if needed

I am hoping someone can provide some insight into this subject


I have been working with SQL Server for several years, but have not had the pleasure of working much with the Windows Failover Cluster Service and Availability Groups.  In my new role I am responsible for both WFCS and AG and have a question


I need to patch SQL Server and usually take VMWare snapshots in case there is a problem.  It the past I have been using a stand alone SQL installation and no WFCS and I now have WFCS being used by SQL 2012, 2014 and 2016 Availability Groups


My question:

1: Will I be able to revert to snapshot when using a WFCS and AG just as easy as it was using non WFCS and Stand alone instances – my concern is with the cluster and AG synchronizing if I had to revert to the snapshot


2: What is the best process for taking a snapshot, more specifically

Should I snapshot both server before patching and revert both servers if there are any failures.


should I use a process recommended by one of the server opt team member.  They said use the following steps

A: Shut down SQL Server Service on the Secondary Node and snapshot it

B: Install your patch, reboot and test as needed

C: Fail the AG over to that Secondary node

D: Shut down SQL Server Service on the first node, snapshot and then patch and test that node.

E: Fail everything back.


Your feedback is appreciated.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.