Disaster Recovery Scenario with Object Storage

[shebang]

Hi community

We are architecting a B&R solution and our scenario is that we have all our jobs running to a SOBR with the performance tier running in our primary site and and capacity (private S3 object storage) tier in our secondary site. The idea is to also have a small amount of block storage in our DR site so that we can backup the B&R configuration to this. In the event of a primary site outage, we can restore the B&R config to a new VM in the DR site (as per https://helpcenter.veeam.com/docs/backu ... ml?ver=110).

My question is that when we do restore the config database, in order to restore VM backups, can we simply import the capacity tier as a standalone object storage repository and start to restore VMs (as per https://helpcenter.veeam.com/docs/backu ... ml?ver=110) ?

If so, I assume that if we want to continue protecting VMs in our DR site we will need to create a new SOBR with performance tier and backup to this (until V12 is out where we could potentially backup directly to object store).

Thanks in advance!

[Mildur]

Hi Shebang

My question is that when we do restore the config database, in order to restore VM backups, can we simply import the capacity tier as a standalone object storage repository and start to restore VMs (as per https://helpcenter.veeam.com/docs/backu ... ml?ver=110) ?

If you restore the configuration backup, you don't have to import the object storage.
With the restore of the configuration database the new veeam server will have all the information which backups are stored in the capacity tier.

If so, I assume that if we want to continue protecting VMs in our DR site we will need to create a new SOBR with performance tier and backup to this (until V12 is out where we could potentially backup directly to object store).

The benefit of configuration backup restore is to have a running Veeam Server again.
A new SOBR is normally not needed because the configuration backup will restore the SOBR configuration. There is one exception, if you loose your old performance tier extends, you have to rebuild the performance tier.
But that's simply adding a new extend to the existing SOBR and rescan the SOBR to download the archive indexes. When the download is finished, you can start doing backups again (with an active full) to the original SOBR, and Veeam will only offload changed blocks.

[shebang]

Brilliant, thanks for the info and confirmation. When we say add a new extent to download the archive indexes, is it downloading them from the capacity tier back onto the performance tier? As you say, in a DR situation we will have lost the primary site performance extents so I just want to confirm what we are rebuilding from.

[Mildur]

Your welcome.

When we say add a new extent to download the archive indexes, is it downloading them from the capacity tier back onto the performance tier?

Yes, the SOBR rescan will download the indexes from the capacity tier and create the index and stub files on the performance tier extend. The archive indexes do contain the information about the offloaded blocks.
VBR will then compare the blocks from the new active fulls with the archive indexes to see what data already exists in the capacity tier. Only new Data will be offloaded. Even in Case of this active full.

If you build a new SOBR with a new Bucket or use new backup jobs, everything has to be offloaded again.
So make sure to use the old SOBR and old backup configuration You can put your VBR Server in the DR site. In case of a disaster at the primary location, you don't have to restore the VBR configuration first.