Vm - destroy unrefrenced disks owned by guest issue

[SamoPM]

I have an issue when deleting a VM on a Proxmox cluster.

I am getting an error TASK ERROR: rbd error: rbd: listing images failed: (2) No such file or directory if I check the destroy unrefrenced diskd owned by guest.
The VM seems to be removed, but then the job fails. The VM is also visible in the GUI.

If I don't check the second box the VM seems to be removed fine.
I checked that there is no config for the VM left on the system.


Env:
PM 9.0.6
Ceph 19.2.3

 

[dakralex]

Hi!

Are there any errors shown for Ceph or when listing images manually for that rbd storage?

 

[SamoPM]

Hi!

No, no errors for Ceph...

 

[cfgmgr]

Ironically I'm having this same issue after we did a graceful shutdown of the entire cluster and a startup the other week. I'm trying to clean up some infra that is no longer being used.

Everything is online and running post maint, ceph seems healthy.

PVE 8.4.5
Ceph 19.2.3

 

[cfgmgr]

Ironically I'm having this same issue after we did a graceful shutdown of the entire cluster and a startup the other week. I'm trying to clean up some infra that is no longer being used.

Everything is online and running post maint, ceph seems healthy.

PVE 8.4.5
Ceph 19.2.3

Actually fixed this. There was a previous clone that was either stoped mid-clone or it finished and it was deleted. It was locked via the UI, and sort of in a strange state.

I unlocked it on the CLI as I knew it was not being used and could be cleaned up. Once the VM was gone from the UI, it still had a referenced disk that didn't get cleaned up. It had several disk (based on what was cloned) but there was 1 straggler.

# rbd ls ceph-vm | grep 140
vm-140-disk-0
# rbd rm ceph-vm/vm-140-disk-0
Removing image: 100% complete...done.

Once that ran and it was cleaned up... the rest of the VM cleanup I was doing worked without issues