Veeam and XO backup CBT. How to avoid conflict?

zorro

I'm replicating VMs with XO Incremental Replication (NBD on, "Purge snapshot data when using CBT" on). The VMs also have a Veeam backup using the Veeam Plug-in for Xen, which relies on CBT. Disks are qcow2.

The XO full seed caused no problems in Veeam. After the first XO delta run on a VM, the next Veeam job finished with Success but showed this warning:

Failed to use CBT: [Task 542e5367-9d6b-9567-bcab-c2448d26223a (Async.VDI.list_changed_blocks) failed: . SR_BACKEND_FAILURE_460. . Failed to calculate changed blocks for given VDIs. [opterr=Source and target VDI are unrelated]

VMs that only had the XO seed showed no warning. Veeam falls back to reading the whole disk, and I don't want that every night. XO's own deltas still work and don't fall back to a full.

Questions:

How can I avoid this conflict between XO replication and Veeam on the same VMs?
If I turn off "Purge snapshot data when using CBT", will XO stop touching the CBT state Veeam depends on?

Setup: XCP-ng XCP-ng 8.3.0, XO 6.9.1 - XOA build: 20251219, Veeam 13.1.2.57.

Thanks

acebmxer

@zorro

Hello if you have not seen this post - https://xcp-ng.org/forum/topic/12407/veeam-13.1-rocky9-linux-appliance-potential-data-loss-with-cbt-and-workers-with-expired-tokens Myself and another user reported issues with Veeam and CBT.

My latest update from Veeam was no eta on the next update release to fix the issues. That has been some time and there might be a eta. I have not followed back up as of yet.

On thing that seems to have helped was using Active Full backup opposed to Synthetic full backups.