Error Failed to find the backup file.

[mroldan]

Currently, we have a production environment with a VMware cluster, with the Veeam Backup & Replication (VBR) server on one VM and the VeeamONE server on another. In Veeam, we perform backups to a Scale-Out Backup Repository (SOBR), where the performance tier has a local repository, and the capacity tier has an Azure blob.

Our project is to implement a disaster recovery (DR) site in Azure without discontinuing the use of the local VBR. To achieve this, we have installed two servers in Azure: one for VBR and another for Veeam Recovery Orchestrator (VRO), connected to our on-premises environment via a VPN to collect information from VMware. Additionally, the Azure VBR is connected via the Veeam agent.

Our plan is that, in case of an incident requiring DR execution, we will attach the blob to the Azure VBR server, perform a rescan, and then execute the DR plan on the Azure VRO server. We conducted a test to restore the VM from the Azure blob using VBR, and it was successful. However, when executing the DR plan, it fails with the following error:

[Backup existence] Failed to find the backup file.

Case Number: 07328555

[Alec King]

Hello Manuel, at first glance I think you need to use the VBR server which is co-installed on the VRO server in this scenario. The co-installed or 'embedded' VBR on the VRO server has certain special capabilities including awareness of repository rescans. Using a remote VBR won'r work.
Meantime I will look at your support case, and get some details. Thanks!

[mroldan]

hi alec

My question is, if I can take the blob from the on-premises VBR and attach it to the VBR in Azure to perform the restoration to Azure, why can't the orchestrator perform it? Would it be necessary to connect the Azure VBR to the vCenter, considering that in the event of an incident, the entire on-premises environment would be offline

[Alec King]

Hi, the only reason is that VRO currently does not support the "imported backups" scenario for remote VBRs - it's only supported for the embedded VBR.
We're planning that any VBR connected to VRO can have this capability. Not yet sure which version that will be released in though.
So:
- You don't need to connect to the source vCenter or source VBR - just use the embedded VBR.
- Connect the Azure blob to the embedded VBR on the VRO server, and rescan (you probably also need to wait around 5- 10 minutes for all the data to sync internally from VBR - ONE - VRO)
...and then your Recovery Plan should work. You will also get notification if you run a Readiness Check first that the embedded VBR will be used for recovery.

The VBR server co-installed on VRO is fully-featured, I'd advise using it for this situation as it will save you money - you don't need to deploy another VBR in Azure, just a VRO server.

[mroldan]

Thanks, Alec. I will be testing what you suggested this afternoon, and I will let you know how it went.

[mroldan]

alec good afternoon do the actions that you indicate me add the blob repository to the vbr integrated in VRO, after adding it perform a rescan but it ends without finding any backup file, but if I import it by hand if they are detected. has something to do that the backup job in its origin are encrypted by the console vbr of origin.

[mroldan]

I did the test of generating a new backup job in the source vbr without encrypting but the error persists. I copy screenshot of the error.

[Alec King]

Hi Manuel,

If the backups were encrypted on a different VBR server then you will not be able to restore them. VRO has no password or other method to decrypt them. We are looking at adding such capability in future.

I am not clear on your second test though, did you create a new backup job on the VRO server's embedded VBR? That should of course work as normal - but you would have to create a new plan that uses that backup job in VRO. Note that, even if you create a job with the same name, VRO will not automatically 'attach' those backups to your existing plan - it will be seen by VBR as a separate backup.

Edit: I see you've already got a support case, however if you could attach the logs from your recent tests that would help us anlayse the problem. Thanks!

[mroldan]

perform these tests with unencrypted backups and the error persists.

The test I performed was to put the blob in maintenance in the source VBR and then add it to the integrated VBR in the VRO server. am I missing a step?

I will be uploading the logs to the support ticket.