Veeam v12 SMB Backup to Hardened SOBR

[bcbkupadm]

Hello,
Case: 06303507
Posting this issue here in hopes we can get this resolved.
Environment:
We are running Veeam 12. We have our Veeam server (Windows 2022), a proxy server (Server 2022) and a Linux server (redhat9).
The linux server is connected via SAS to a Dell ME storage. We added several hardened repositories and added them to a SOBR.
On Veeam we added several SMB shares hosted on a Dell Isilon.

The issue:
When we create a file backup job, and the target is the hardened SOBR the job mostly fails, except some times. And when it does complete, the immutability date for the whole backup chain is reset. When it does not work, we get the error message below:

Error: Agent: Failed to process method {NasMaster.LoadJobBackupMeta}: Failed to connect to the endpoint [127.0.0.1:3000]. Connection refused

This does NOT happen if the target is a hardened repository.


Any help would be appreciated.

[HannesK]

Hello,
I'm a bit confused about "This does NOT happen if the target is a hardened repository." What does that mean? Is it possible that you added the same paths multiple times to different repositories? For example:
/mnt/repo/backups/
/mnt/repo/backups/production/

The recommended setup is one extent per server. That would mean the Dell ME storage should present one volume to the Linux server and that one volume can be part of a SOBR. No way to directly access the Hardened Repository. One can only access via SOBR.

If the maximum volume size is too small to present all disk storage as one volume, then the storage can present multiple volumes to the Linux machine. Then one can have multiple Hardened Repositories on one server and put all extents together in one SOBR. Same here: No way to directly access the Hardened Repository. One can only access via SOBR.

Best regards,
Hannes

[bcbkupadm]

We divided all the storage into smaller volumes for manageability. Is there a document I can reference that recommends a single volume?

Our current setup
Single Linux server with 6 volumes.
/hardenedvol1
/hardenedvol2
/hardenedvol3
/hardenedvol4
/hardenedvol5
/hardenedvol6
All 128TB in size.

One SOBR with vol1 to 4. The SMB job fails when the target is this SOBR
If I create an SMB backup job and the target is vol5 or vol6 for example, the job runs fine.

[HannesK]

Hello,
okay, the setup is good so far. I'm not sure whether there is an official document on the "one extent per machine" recommendation. I mean, with 6 extents / repositories it's not too bad, so it should "just work". I just asked because I saw a case recently with nested repositories.

I'm afraid we have to wait what support finds out (at least I have no more guesses)

Best regards,
Hannes