VM Fails to backup and job gets stuck

[brokebit]

Hey folks,

I have a strange problem. I have one VM in my cluster that just won't backup. No issues backing up other VMs. This is on a three node 9.1.6 cluster that is current on all updates. The VM is a linux machine with a 256GB root disk. QCOW2 format. VMs are stored on an NFS mount from a Unifi UNAS. Backups go to another NFS mount on another UNAS machine.

The backup status gets to this point and then just hangs indefinitely and there is no traffic to/from the UNAS boxes:

INFO: starting new backup job: vzdump 119 --storage local --compress zstd --notification-mode notification-system --remove 0 --mode snapshot --notes-template '{{guestname}}' --node neutron
INFO: Starting Backup of VM 119 (qemu)
INFO: Backup started at 2026-04-04 13:20:56
INFO: status = running
INFO: VM Name: newkali
INFO: include disk 'scsi0' 'unas-pro8-proxmox:119/vm-119-disk-0.qcow2' 256G
INFO: backup mode: snapshot
INFO: ionice priority: 7
INFO: creating vzdump archive '/var/lib/vz/dump/vzdump-qemu-119-2026_04_04-13_20_56.vma.zst'
INFO: started backup task 'f0aec091-1e38-4f2e-b57f-48c0f01fb156'
INFO: resuming VM again
INFO: 0% (0.0 B of 256.0 GiB) in 3s, read: 0 B/s, write: 0 B/s

Any ideas?

 

[brokebit]

Here is the VM Config:

root@neutron:/var/log/pve/tasks# qm config 119
boot: order=scsi0;net0
cores: 12
cpu: x86-64-v2-AES
lock: backup
memory: 30518
meta: creation-qemu=10.1.2,ctime=1774718981
name: newbox
net0: virtio=BC:24:11:0F:ED:23,bridge=vmbr0,firewall=1,queues=8
numa: 1
ostype: l26
scsi0: unas-pro8-proxmox:119/vm-119-disk-0.qcow2,aio=native,iothread=1,size=256G
scsihw: virtio-scsi-single
smbios1: uuid=2e0d66e2-6e54-45af-9184-5a0336358541
sockets: 1
vmgenid: 4dde0ca6-5f68-4753-b4da-040bbb512bca

 

[powersupport]

Hi,


From the config, the VM is currently showing lock: backup, which suggests a previous backup task may be stuck. That alone can prevent new backups from progressing.


Also, since the disk is QCOW2 on NFS and you’re using snapshot mode, it can sometimes hang at 0% if there’s an issue with snapshot creation or NFS responsiveness.


You could try clearing the backup lock and retrying. If the issue persists, testing with stop mode would help confirm whether it’s snapshot/NFS related.

 

[brokebit]

Hi, thanks for the response. The lock was because I had a stuck backup going when I dumped the config. I aborted the backup, cleared any locks, and tried again, but no luck. Also, I stopped the VM, tried the backup, and got the same result.

I have a bunch of other VMs with the exact same config and disk size, and they are all working fine. It just seems like this VM, and I cannot find any difference between this one and the ones that do work.

Some other bits of data:


root@neutron:~# qemu-img check /mnt/pve/unas-pro8-proxmox/images/119/vm-119-disk-0.qcow2
No errors were found on the image.
4194304/4194304 = 100.00% allocated, 0.00% fragmented, 0.00% compressed clusters
Image end offset: 274920112128

root@neutron:~# qm listsnapshot 119
user config - ignore invalid acl role 'MultiP'
`-> current You are here!

Curious, you mentioned qcow on NFS. Is that not a recommended configuration?