VMs do not start after migration to another node

Jun 1, 2023
35
14
13
We have a 3 node cluster and no central storage. The disks of our virtual machines, all zfs volumes, are replicated using pvesr with local jobs to all three nodes of our cluster. We are still using Proxmox 7.x, the update to 8.x will hopefully happen next week.

If i migrate a vm from one node to another with the command

ha-manager crm-command migrate vm:121 node2

I can see with the

ha-manager status

command that the vm is migrated to the new node and that the vm is also started on the new machine.

On the new node I can see that the vm is running using the

qm list | grep 121

command.

However, the vm is not reachable via the network and using

qm terminal 121

opens a terminal, but no system is runing and no output is shown.

Stopping the migrated vm via

qm stop 121

on the new node and starting the vm with

qm start 121

is working and the vm gets started normaly and is also reachable via network.

I have no idea what is going wrong. It seems that the migration is working and the vm gets started on the new node, but for what reason ever the vm stucks during the start on the new system. I can find nothing in the logs, no error, all looks good to me. Because I am blind and can therefore not use the Proxmox GUI I have to gou through the logs on the command line and because these logs are not really easy to read it might be that I am missing something.

Is anyone having an idea what could be going one and what could be the problem?