Usage of the backup copy repository in the case of a disaster

[christian.naenny]

Hello everyone!

I would like to get some thoughts on this idea: Use the backup copy repository located in the second (disaster recovery) datacenter as a repository in case of a disaster.

Our setup:
We're operating two different datacenters, one primary datacenter for production and a second datacenter for dev, test and disaster recovery. We have 2 productive Veeam Servers. One for each datacenter. Both are connected to a local, hardened Linux repository (physical server with all the LUNs via SAN). The hardened repository of the second datacenter hosts all the backups for the second datacenter but also hosts the backup copy repository for the first datacenter (on different LUNs and file systems, of course).

Now the scenario:
The first datacenter is no longer operational. But fortunately, we have all the backups in the backup copy repository in the second datacenter, right?

My question:
Is it feasible in the disaster scenario to just connect the backup copy repository to the Veeam server in the second datacenter, run an inventory of the backup copy repository and after the inventory finishes, start restoring VMs backed up in the primary datacenter in the second datacenter if necessary?

Prerequisites:

• The productive Veeam infrastructure in both datacenters is always kept on the same version of Veeam.

• The sharing of the backup copy repository is NOT permanent! It would only happen in the case of a disaster.

Any thoughts on this setup?

Thanks!
Christian

[Mildur]

Hi Christian

Yes, that's completely fine and supported (Veeam Blog Article). Many customers have a standby backup server prepared for such scenarios.
If you store your configuration backup on the backup copy repository, you can also restore the entire configuration from it instead of manually connecting a repository and performing rescans.

Best,
Fabian

[christian.naenny]

Hi Fabian

My first thought was: I'd rather not use the productive Veeam server of the second datacenter to restore the configuration of the Veeam server of the primary datacenter to!
But of course you just gave me an idea: I might want to utilize our test Veeam server as a standby backup server. As we're testing the disaster scenario twice a year, I might want to simply take a snapshot of the test Veeam server before the disaster exercise and revert back to the snapshot after the exercise.

Thanks for the idea and the input!

Cheers
Christian

[Mildur]

Hi Christian

Yes, overwriting a production/test server may not always be a good option. But in case of a disaster situation it may be an option.

I might want to simply take a snapshot of the test Veeam server before the disaster exercise and revert back to the snapshot after the exercise.

For testing purposes its better to "just connect/rescan your copy repository" instead of restoring the configuration and then reverting a snapshot. Configuration restore will restore the entire configuration, such as backup jobs. It's possible that it will interfere with the primary servers operations.

Best,
Fabian