Some advice is need on our Veeam Backup Azure Storage Options

[gabbas]

We are reviewing our Veeam Backup for Microsoft Azure design to optimise Azure storage costs while maintaining best practices. Currently, our Azure VM backups are stored in GRS storage accounts, with daily backups (31-day retention), monthly backups, yearly backups in a separate GRS repository, and hourly snapshots. We also have a 125TB Veeam Data Vault subscription and use Backup Copy Jobs to maintain an immutable offsite copy of all backups. Given that we already have a secondary immutable copy in Veeam Data Vault, we are questioning whether GRS is still providing sufficient additional value to justify the extra cost compared to LRS.

We are considering creating new LRS-based repositories and directing future backups there while leaving existing backups in the current GRS repositories until they age out naturally. Is this the recommended approach, and would restores continue to work from the existing GRS repositories during the transition? Also, if we retain GRS, can someone explain the practical recovery process during a regional outage (e.g., UK South failure) and how Veeam Backup for Microsoft Azure would access and restore data from the replicated copy in UK West? We would appreciate any guidance on whether this proposed redesign aligns with Veeam best practices and cost-optimisation recommendations. Thanks in advance.

[nielsengelen]

Hi,

There is normally also an option to switch back to LRS from GRS. You can of course create new repositories but you'll have double the storage used for a while.
If you create new repositories but leave the GRS repositories available in VB for Azure, the old backups should be visible and you could perform a restore from them.

If there would be an outage (and maybe your VB for Azure would be out as well), you can setup a new/temporary VB for Azure and import the replicated repositories and since all metadata will be there - you will be able to perform restores.

Does this answer all your questions?

[gabbas]

Yes and thank you @nielsenglen

[gabbas]

Thanks, if we leave our current setup as it is with GRS repos and in case our primary region becomes unavailable, what do we need to do to get this backup data to get up & running in our secondar region? Do we need to spin up a VBR server and add the Azure Veeam Appliance, have our configuration backup ready to import and how can we add the backup repos (storage account in our secondary region) inside Veeam VBR? Are there any steps that we need to complete before we add those backups / repos and start the restore of our VMs in the secondary region? We would need this info for any region outage type scenario. Below are the steps that we think we need to follow in case of our primary region outage and like to confirm if this is a valid approach:

If UK South Suffers a Regional Outage

In theory, the process would be:

UK South becomes unavailable.
Microsoft makes the GRS replica available (depending on the outage scenario and Microsoft DR processes).
Deploy a Veeam Backup & Replication server in UK West (This can be deployed in advance and leave it in de-allocated state)
Deploy a new Veeam Backup for Azure appliance in UK West using the VBR server (This can also be deployed in advance).
Restore or reconfigure the Veeam Azure appliance configuration.
Add/import the repository using the storage account containing the replicated backup data from UK West.
Veeam imports the backups and makes restore points available.
Restore VMs into UK West.