After upgrading pve-qemu-kvm 11.1.3-1 to pve-qemu-kvm 11.1.3-2:
VMs stopped or starting after this upgrade fail backup consistently.
VMs that where not shut down after the upgrade continue to behave normally but once shut down and restarted the same VM fails backup.
Running vzdump with --performance 'max-workers=1' gives inconsistent results so I don't recommend it as a work-around.
Downgrading to pve-qemu-kvm 11.1.3-1 fixes the problem completely:
I pinned version 11.1.3-1 until this nasty problem gets fixed (took me entire night to find the culprit) !
Environment:
3 node Proxmox cluster on 3 x DELL Poweredge 640 - 192GB Ram - Dual Intel Xeon Gold 6154 @ 3.00GHz (all nodes identical hardware)
ProxMox VE 9.2.28
PBS 4.2.12 on dedicated DELL Poweredge 730
VMs stopped or starting after this upgrade fail backup consistently.
VMs that where not shut down after the upgrade continue to behave normally but once shut down and restarted the same VM fails backup.
Code:
vzdump 217 --storage BCKP-STS --mode stop
217: 2026-11-20 11:33:15 INFO: Starting Backup of VM 217 (qemu)
217: 2026-11-20 11:33:15 INFO: status = stopped
217: 2026-11-20 11:33:15 INFO: backup mode: stop
217: 2026-11-20 11:33:15 INFO: ionice priority: 7
217: 2026-11-20 11:33:15 INFO: VM Name: RustServ
217: 2026-11-20 11:33:15 INFO: include disk 'scsi0' 'VMs-STS:vm-217-disk-0' 15G
217: 2026-11-20 11:33:15 INFO: creating Proxmox Backup Server archive 'vm/217/2026-11-20T10:33:15Z'
217: 2026-11-20 11:33:15 INFO: starting nvm to execute backup task
217: 2026-11-20 11:33:16 INFO: starting backup via QMP command
217: 2026-11-20 11:33:17 INFO: started backup task '2ab84c48-6f07-464b-bfdf-ac94312a7349'
217: 2026-11-20 11:33:22 ERROR: VM 217 qmp command 'query-pbs-bitmap-info' failed - got timeout
217: 2026-11-20 11:33:22 INFO: aborting backup job
217: 2026-11-20 11:34:22 INFO: stopping kvm after backup task
217: 2026-11-20 11:34:24 ERROR: Backup of VM 217 failed - VM 217 qmp command 'query-pbs-bitmap-info' failed - got timeout
Running vzdump with --performance 'max-workers=1' gives inconsistent results so I don't recommend it as a work-around.
Downgrading to pve-qemu-kvm 11.1.3-1 fixes the problem completely:
root@pve0:~# apt install pve-qemu-kvm=11.1.3-1The following packages were automatically installed and are no longer required: proxmox-kernel-7.0.14-24-pve-signed proxmox-kernel-7.0.14-26-pve-signedUse 'apt autoremove' to remove them.DOWNGRADING: pve-qemu-kvmSummary: Upgrading: 0, Installing: 0, Downgrading: 1, Removing: 0, Not Upgrading: 20 Download size: 0 B / 32.7 MB Freed space: 1,024 BContinue? [Y/n] Ydpkg: warning: downgrading pve-qemu-kvm (11.1.3-2) to (11.1.3-1)(Reading database ... 98231 files and directories currently installed.)Preparing to unpack .../pve-qemu-kvm_11.1.3-1_amd64.deb ...Unpacking pve-qemu-kvm (11.1.3-1) over (11.1.3-2) ...Setting up pve-qemu-kvm (11.1.3-1) ...Processing triggers for man-db (2.13.1-1) ...root@pve0:~#root@pve0:~# vzdump 217 --storage BCKP-STS_Lab --mode stopINFO: starting new backup job: vzdump 217 --mode stop --storage BCKP-STS_LabINFO: Starting Backup of VM 217 (qemu)INFO: Backup started at 2026-11-20 10:20:35INFO: status = stoppedINFO: backup mode: stopINFO: ionice priority: 7INFO: VM Name: CopyOf-WrongIP-012-RustServINFO: include disk 'scsi0' 'VMs-STS:vm-217-disk-0' 15GINFO: creating Proxmox Backup Server archive 'vm/217/2026-11-20T09:20:35Z'INFO: starting kvm to execute backup taskINFO: starting backup via QMP commandINFO: started backup task '6cd4ecd0-02ba-486e-a5b7-1537cf6731a3'INFO: scsi0: dirty-bitmap status: created newINFO: 4% (660.0 MiB of 15.0 GiB) in 3s, read: 220.0 MiB/s, write: 4.0 MiB/sINFO: 6% (1.0 GiB of 15.0 GiB) in 6s, read: 122.7 MiB/s, write: 2.7 MiB/sINFO: 8% (1.3 GiB of 15.0 GiB) in 9s, read: 86.7 MiB/s, write: 0 B/sINFO: 11% (1.8 GiB of 15.0 GiB) in 13s, read: 131.0 MiB/s, write: 9.0 MiB/s...INFO: 93% (14.1 GiB of 15.0 GiB) in 2m 2s, read: 86.7 MiB/s, write: 20.0 MiB/sINFO: 95% (14.4 GiB of 15.0 GiB) in 2m 6s, read: 73.0 MiB/s, write: 18.0 MiB/sINFO: 97% (14.7 GiB of 15.0 GiB) in 2m 9s, read: 97.3 MiB/s, write: 17.3 MiB/sINFO: 100% (15.0 GiB of 15.0 GiB) in 2m 12s, read: 118.7 MiB/s, write: 25.3 MiB/sINFO: backup is sparse: 632.00 MiB (4%) total zero dataINFO: backup was done incrementally, reused 13.96 GiB (93%)INFO: transferred 15.00 GiB in 132 seconds (116.4 MiB/s)INFO: stopping kvm after backup task trying to acquire lock... OKINFO: Finished Backup of VM 217 (00:02:16)INFO: Backup finished at 2026-11-20 10:22:51INFO: Backup job finished successfullyINFO: notification matched the following matchers: default-matcher.The following targets will be notified: TimeNetINFO: notified via target `TimeNet`I pinned version 11.1.3-1 until this nasty problem gets fixed (took me entire night to find the culprit) !
Environment:
3 node Proxmox cluster on 3 x DELL Poweredge 640 - 192GB Ram - Dual Intel Xeon Gold 6154 @ 3.00GHz (all nodes identical hardware)
ProxMox VE 9.2.28
PBS 4.2.12 on dedicated DELL Poweredge 730