VM migration failed and VM is locked and cannot be unlocked because destination node doesn't have qm

[EzzatQ]

Hi all,
I have a 2 node proxmox setup:
My first node is a Mac Mini 2012 running version: pve-manager/8.0.4/d258a813cfa6b390 (running kernel: 6.2.16-3-pve)

I recently for an old HP ProLiant DL380 G7 server and wanted to create a proxmox cluster with it. For some reason I was not able to boot the server with pve 8 installer, i was only able to with pve 6.4. After successfully installing pve 6.4, I was able to create a cluster and join the 2 nodes.

I know I was supposed to update to pve 7 then pve 8, but I forgot and migrated one of my vms to the HP server (very stupid). The migration copied the VM's LV to the HP server, and I saw the VM mvoe to the HP server in the UI. However, the migration failed in the last stage, when it tried to execute "qm unlock 103" with error: "qm: command not found"

Now anything I try to do with this VM, start, migrate, backup, ... doesn't work, because the VM is locked, and I can't unlock it because qm does not exist on the HP server for some reason.

Weirdly enough, I'm also not able to execute "pveversion", it errors with: "pveversion: command not found". I figure that qm and pveversion don't exist on the destination node because it is running an older version of pve.

Now I'm trying to recover the VM but all tries failed and I am not sure what to do. I unfortunately don't have any backups of this VMs because my drive that stores the backups failed .

I have almost lost hope, but this VM is very important and I have to recover it! Does anyone know how I can recover the VM?

 

[bbgeek17]

i dont know how you got to the current state. Your second node sounds compromised.

You can check first if the source volumes still exist on original server. Assuming it was LVM you can list them with "lvs"

If the volumes exist, you can try to move the VM configuration file located in /etc/pve/qemu-server to the first node. Make sure you save the config file, then scp it over. Afterwards, you can try to unlock and/or start it on original node.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[EzzatQ]

Thanks for the quick reply.
The LV does exist (I can see it in "lvs"), but I am not sure how to transfer that back to the old node. Any idea?
The VM config file is also available under /etc/pve/qemu-server/103.conf. I didn't completely understand what exactly is locked? is it the conf file that's locked? the LV? or the combination of the two?

In other words, do I need to transfer both the conf file and the LV to the old node to attempt "qm unlock"?