Configuration Backup Restore to different server testing

[tjdoc]

Hi,
I have setup Configuration Backup on our VBR server to store its .bco files on another Windows VM, which is part of the same domain / local network etc.
What I'd like to do is test simulating the restore of the Configuration database to the other VM, which has a vanilla Veeam install (same version etc).
The reason for this is to test how we would recover all settings to another server, should the primary VBR server corrupt or fail.
I dont actually want to break my primary, just check the recovery process should it ever be needed.
I believe this approach is best practice for ensuring its possible to recover the VBR server. (?)

So a couple of questions on this please:
1) When I go through the Restore Configuration Backup wizard on the new VM, do I select "Restore" or "Migrate"?
I dont really want to stop using the original server so I don't think Migrate is correct, also I think this will break my Immutable Linux repo as well, but just a bit unsure.

2) If I choose the restore option and then end up with two VBR servers running with the same configuration, will the original server break in some way? As I'm only testing should I just shut the original server down?

I'm sure this must have been done lots of times before...just looking to understand a bit more about the right way to go about it...

Many Thanks
Tim

[david.domask]

Hi TimD,

Just to confirm quick, you plan to test actual operation of the DR (secondary) Veeam Backup and Replication Server, or you just want to confirm that the configuration restore works?

Your caution about having two VBR servers utilizing the same resources is valid as sharing components is not supported and can introduce issues, so if you plan to test operations on the DR server, I would actually recommend using the Migrate option. Take a fresh configuration backup, power down the original server, migrate to the DR server, perform your testing, then reverse it and do a fresh configuration backup from the DR server, migrate back to the production.

However, if you just want to test that the configuration restore works and don't plan to test operations frequently, then I would just do the Configuration Restore to a VBR server with the network disconnected -- confirm that everything looks appropriate, and revert.

Our User Guide page here explains the use cases for restore vs migrate a bit more, but main point is don't have both servers active at the same time with access to the same shared resources, as that can (and likely will) introduce problems as noted in the first link above.

[tjdoc]

Hi David, thank you for replying on this.
I'm only looking to validate that the config restore works - I don't want to actually start using the DR server at all, it's enough just to define / document a procedure that could be followed should it ever be needed. Actually implementing it fully to test end to end might be something we do a bit further down the line.
I will follow your suggestion to disconnect the DR server from the network and check the config restores as expected.

Thanks for the link to further info on restore v migrate - will try to understand a bit more about each method so we know a bit more in the future.

Cheers
Tim