Failed backup jobs since updating

peo

Yesterday, I installed the patches on my three hosts and rebooted them, directly followed by updating XO on my Debian machine to the latest (2cd5b at that time).
Since then, both the evening-replication (21:00) of my Deb12-XO host failed, as well as the morning-replication (09:00). This one has never failed before and is (was) being replicated to another host on the same network.

The mail-report says:
Debian 12 XO self-install

Pool id: 4cc74549-71c3-31d5-f204-7106e90acd1e
UUID: 30829107-2a1b-6b20-a08a-f2c1e612b2ee
Start time: Friday, March 13th 2026, 9:00:01 am
End time: Friday, March 13th 2026, 9:00:28 am
Duration: a few seconds
Error: _removeUnusedSnapshots don't handle vdi related to multiple VMs Deb12-XO - Deb12-XO - (20260312T200004Z) and [XO Backup Deb12-XO] Deb12-XO

As my XO-installation is outdated, I will update again and let it retry the replication without any other changes.

Also, this is probably nothing else than an annoying incorrect error message from microsoft:

What I did: followed the link in "About" to the current commit, went back to main, and clicked the link to all commits to see what I've missed (4 commits after I update XO)

peo

I thought I solved it.. but now the problem is back again:

Yesterday, 14 March, I did the following:

deleted the snapshots on the running Deb12-XO machine, so it was clean
deleted the clones of the machine
ran the backup job maually - with success (transferred about 30G to the other host)
(still on the "old" XO version)

Checked the destination host - no clone (backup) of the VM found.

manually created two (full, not quick) clones of the VM so it won't be lost of anything goes sideways
updated XO to the latest (2aff8)
let it do the relication according to the schedule
21:00 replication success

Next morning (15 Mar) 09:00 replication also ran without any issues - at least according to the log.

no additional copy of the VM on the destination host (retention is set to 2, so I should have one 21:00 and room for the 09:00 too)

Evening replication, 21:00, failed. Got that same error message as before:

VM Backup report

Global status : failure

Job ID: 883e2ee8-00c8-43f8-9ecd-9f9aa7aa01d1
Run ID: 1773604800005
Mode: delta
Start time: Sunday, March 15th 2026, 9:00:00 pm
End time: Sunday, March 15th 2026, 9:00:13 pm
Duration: a few seconds
Successes: 0 / 1
Transfer size: 126 MiB

1 Failure
Deb12-XO
Debian 12 XO self-install

Pool id: 4cc74549-71c3-31d5-f204-7106e90acd1e
UUID: 30829107-2a1b-6b20-a08a-f2c1e612b2ee
Start time: Sunday, March 15th 2026, 9:00:02 pm
End time: Sunday, March 15th 2026, 9:00:13 pm
Duration: a few seconds
Error: _removeUnusedSnapshots don't handle vdi related to multiple VMs Deb12-XO - Deb12-XO - (20260315T080004Z) and [XO Backup Deb12-XO] Deb12-XO

=
I notice the name of the snapshot seems odd:
[XO Backup Deb12-XO] Deb12-XO - Deb12-XO

Maybe just a new naming convention, but I found the old one better (for my Admin VM):
Admin Ubuntu 24 - Admin Ubuntu 24 - (20260201T125508Z)
Admin Ubuntu 24 - Admin Ubuntu 24 - (20260208T125508Z)
(I assume the double "Admin Ubuntu 24" is because the backup job name is the same as the machine name)

ph7

@peo said:

Error: _removeUnusedSnapshots don't handle vdi related to multiple VMs

I had the same error when updated to 449E7
It ran 2 CR, then threw the error.
There was other problems so I rolled back to 5bdd7
https://xcp-ng.org/forum/topic/11969/timestamp-lost-in-continuous-replication