ES | EN

Proxmox CT Locked (snapshot-delete): Quick Fix for Backup Errors

By Jose FYS •
Proxmox
#proxmox#lxc#snapshot#troubleshooting#error

The “CT is locked (snapshot-delete)” error in Proxmox locks LXC containers and stops automatic backups, showing a padlock icon in the GUI. This issue arises when a snapshot deletion operation gets stuck, leaving the container unusable until it is manually unlocked.

Causes of the CT Locked Snapshot-Delete Error

Proxmox marks containers as “locked” during snapshots to prevent concurrent changes. If the deletion fails (due to a full thin pool, interruptions, or bugs), the lock persists, and backups like vzdump fail with “CT is locked (snapshot-delete)”. This affects nightly jobs, leaving containers like Grafana or n8n without backups.

  • Logs show: ERROR: Backup of VM 303 failed - CT is locked (snapshot-delete) as shown in the following images:

  • Common in LVM-thin with oversubscription warnings: “Sum of all thin volume sizes exceeds size of thin pool”

Steps to Fix CT Locked in Proxmox

Access the affected CT’s node via SSH and execute these commands step by step.

1. Verify the Lock

With the following command, we will see if the container is truly still locked:

pct config 303 | grep lock

If we see this response, we know exactly that the container is locked.

2. Unlock the Container

If you are sure there is no active snapshot task running (check the “Tasks” tab of the node in the GUI), you can remove the lock.

Remove the lock from the configuration without affecting data:

pct unlock 303

3. Delete Stuck Snapshots (If Any)

If you see stuck snapshots that haven’t been deleted, you must delete them manually.

  • In GUI: CT > Snapshots > Delete.
  • LVM CLI: lvremove pve/data/snap_vm-303-disk-0-NAME
  • ZFS CLI: zfs destroy poolname/vm-303-disk-0@snapname

Restart the backup job and verify.

Troubleshooting and Technical Considerations

Underlying volume still in use Sometimes, the unlock works but you can’t start the container. This happens because the ZFS dataset or Logical Volume (LVM) remains mounted as a failed snapshot. You will need to manually clean the orphaned snapshot with zfs destroy or lvremove.

Danger of data corruption Never run pct unlock if a legitimate vzdump or ZFS replication task is running in the background (even if it seems slow). Only do it if the task explicitly failed in the logs.

Another problem solved so we can continue with our server. If this post has helped you, share it with other administrators and engineers who can benefit. And follow me for more real experiences from the homelab trenches.