Skip to main content

We do a DR test every year and most of our servers are Virtual which makes it SUPER easy.  However we have 2 servers that are physical.  Last year I had an issue once I got Commvault finally installed on the new physical server and named the server the same name and IP and everything as what is on our production system I couldn’t get it to connect with commvault to restore to.  Its possible there was issues at the facility, but I am wondering best practices for how to make this work smooth.  Should we be adding commvault to a blank server and then connecting it back in with the same name or add the server as a generic name and push software to it from commvault and then rename it to what our system normally sees.  I hope this makes a little sense and is not just a rambling question, if so forgive me for that one.  Thanks for all the help.

That’s a good question.  In a real situation, that server would not be online, so you’d be bringing up the server with the correct name, IP, etc., then install the agent.  At that point, you’d do the full restore and be on your way.

For a test, you’d want to replicate the environment as much as possible without impacting the actual production environment (i.e. a lab/isolated environment).

I’ll defer to some of our longstanding admins and experts on how they run their tests.


Just to make clear this is all in a test environment.  Just bringing up week old VM’s and data to prove we can do it in a true emergency.  We start from scratch setting up Commvault on 2 servers with our DR from tape.


In a DR test we either:

Prepare a server, install the file system agent and perform full recovery.

Or run a 1-touch recovery, not so 1 touch, but does the job well for both Linux and Windows.


Let me try to explain, since this is a weird thing we have to do with this server and this is so I am not spending a ton of time at our next DR test.  The server is just a glorified file server.  It just has all of our data on it that is used by our homebrew systems.  Things like Excel files, PDFs stuff like that.  Last year when I tried to make Commvault understand to restore our data to that it wasn’t liking it named the same.   Its possible that the location we were was having issues with their own systems, so I just put out a general question of what people normally do for this type of restore.  Don’t want anyone to lose sleep over trying to answer this question :-)  Thanks everyone for all your input already.  I feel like I need to look more into the 1-touch.