If we had a catastrophic failure on our system which hosts our Atlassian services, and it had to be restored on a new (VM) with a different host name, what all would need to change?

[Bill Kepner]

My boss always challenges someone when they claim they had a system backed up. Can you restore it? So I'm off to prove I can. The scenerio is the OS is dead. Don't have time to reinstall from scratch. But one hitch, IT takes a backup of the entire system, but when it would be restored, it would need to be done with a different host name. My inital guess is it ain't so simple and there would be a couple configuration items I would need to change before everything would work.

Can't find anything related with google. Has anybody had to do this?

[Dennis Kromhout van der Meer · Accepted solution]

If you know what hostname it's going to be, I would do a simple search & replace through all backups for the old hostname to make sure it all remains the same. If you have all the binaries and data folders the rest should be fairly easy to deploy, just be sure to hook it up to the new database. Let me know if you have any product specific hurdles you're running into.

[Bill Kepner]

The information in the knowledge base regarding "Client Host is Invalid" is a good reference.

http://confluence.atlassian.com/display/CROWDKB/Client+Host+is+Invalid

I also had to change crowd.properties