How to Fix 'VM is locked (backup/snapshot)' Error in Proxmox VE

The error TASK ERROR: can't lock file '/var/lock/qemu-server/lock-XXX.conf' - got timeout or the persistent notification VM is locked (backup) / VM is locked (snapshot) in the Proxmox Virtual Environment (PVE) web console occurs when an automated or manual maintenance task terminates abnormally before releasing the internal safety lock.

Proxmox automatically places safety locks on QEMU virtual machines and LXC containers during critical operations such as live backups via Proxmox Backup Server (PBS), snapshot creations, or cross-node migrations to prevent concurrent disk writes and filesystem corruption. If the target backup storage encounters a network timeout or the job is forcefully killed, this administrative lock remains active indefinitely, preventing administrators from starting, stopping, or configuring the instance from the web interface.

Step-by-Step Solution

1. 1

   Step 1: Check for and terminate orphaned backup processes

   Before forcibly releasing the lock, make sure no background worker process is actively streaming data to or from the virtual disk. Access the Proxmox node shell over SSH or via the Web GUI console and list all running jobs associated with your VM ID:

   BASH Copy

   # Replace 105 with your specific VMID
   ps aux | grep qm | grep 105
   ps aux | grep vzdump
   

   If an inactive or hung vzdump process is still lingering in the process table with no I/O progression, terminate it:

   BASH Copy

   kill -9 <PROCESS_PID>
   

2. 2

   Step 2: Release virtual machine locks using qm unlock

   For KVM/QEMU virtual machines, invoke the native Proxmox qm unlock CLI utility:

   BASH Copy

   qm unlock 105
   

   Once executed, refresh the Proxmox Web GUI. The yellow padlock icon will immediately vanish, and power actions (Start, Shutdown, Reboot) will be restored.

3. 3

   Step 3: Unlock LXC containers using pct unlock

   If the locked instance is an LXC container rather than a full virtual machine, use the pct management command:

   BASH Copy

   pct unlock 200
   

   Confirm the container has returned to an operational state:

   BASH Copy

   pct status 200
   

4. 4

   Step 4: Manually clear persistent locks from configuration files

   In rare scenarios where the unlock command returns no errors but the GUI still indicates that the instance is locked, manually inspect the configuration file:

   BASH Copy

   # For QEMU Virtual Machines:
   nano /etc/pve/qemu-server/105.conf
   
   # For LXC Containers:
   nano /etc/pve/lxc/200.conf
   

   Locate the line beginning with lock: (such as lock: backup or lock: snapshot), delete that line completely, save with Ctrl + O, and exit with Ctrl + X. The Proxmox cluster filesystem (pmxcfs) automatically syncs the edit across the cluster in real time.

Frequently Asked Questions

Is it safe to run qm unlock while a virtual machine is running?

Yes. The qm unlock command strictly removes Proxmox's metadata lock flag. However, ensure no background backup is actively transferring disk blocks to avoid producing a partially corrupt backup file.

Why does Proxmox lack an unlock button in the web GUI?

This design choice is intentional. Forcing administrators to access the shell ensures that hung processes are manually reviewed before releasing the safeguard, preventing accidental dual-write scenarios on virtual disks.