vm 7.4 to 9.1 migrate

[seko]

I have two different clusters.
One is 7.4-3,
the other is 9.1.4.

Both clusters are connected to the same NFS storage with the same names and paths.
I want to move the virtual machines from the 7.4 cluster to the 9.1 cluster.

If I shut down the virtual machine on the 7.4 cluster,
copy the file /etc/pve/nodes/OLD01/qemu-server/123.conf
to /etc/pve/nodes/NEW01/qemu-server/123.conf on the 9.1 cluster,
and then start it, will I encounter any problems?

Is this the correct method?

Or should I use Proxmox backup as suggested by AI?

 

[aaron]

Both clusters are connected to the same NFS storage with the same names and paths.

That is generally not a great idea as each Proxmox VE cluster will assume it is the sole user of a storage. If you would remove a VM and check the "Destroy unreferenced disks owned by guest" checkbox for example, and have a VM with the same VMID in the other cluster, those virtual disks would get removed.


Regarding your actual question: if everything the VM config references (storage name, used bridge (vmbrX)), then this should work. Not recommended because it is very manual, but nonetheless technically possible.

The safer approach would be backup → restore or a remote live migration. I think 7.4 was the first version with that feature.

Maybe try it first with a small test VM

 

[spirit]

copy the file /etc/pve/nodes/OLD01/qemu-server/123.conf
to /etc/pve/nodes/NEW01/qemu-server/123.conf on the 9.1 cluster,

move the file (delete the vm config on source node) to be sure to not restart it again on old01 by mistake, when it's running on NEW01, or you'll break your vm filesystem

 

[seko]

Thanks for your suggestion.
The 9.1 cluster is currently empty, and what we actually want to do is move everything from the 7.4 cluster to the 9.1 cluster, then delete the 7.4 one and fully switch to 9.1.


I already know that the VMs will work after the transfer.
What I want to know is whether anyone has done this before, their experiences, or any points I should pay attention to.