Unable to backup VM after hardware failure

[proxcp]

Hello,

I recently had a motherboard fail so I built a new system, the previous build was using an intel chip the new one is using an amd ryzen.

I simply moved over the drives from the old machine to the new one it boots up fine, I changed the iommu setting to amd instead of intel

the problems started when I couldn't get a network connection I've verified that I'm setting it up correctly in /etc/network/interfaces based on the interface id, after applying the settings and restarting the network service the machine would freeze so I would need to reboot, it would come up and freeze on the login page, if I unplugged the ethernet cable I could login and change the interfaces file back to the original and things would work as normal granted without any network access.

In my frustration I put in a new drive into the server made a clone of the boot drive and did a clean install and networking works as expected and I was able to join it to my 2 server cluster after deleting the old one as well as the old ssh keys etc.

I still have the original cloned boot drive which has 1 VM that I would like to save if possible

I unplugged the ethernet cable, and put in the old cloned boot drive, I can see my VM but when I try to do a vzdump I get a permission denied error and I've searched everywhere and I can't find a clear answer, some say that this is related to the quorum issue, when I do

pvecm expected 1

the server ramps up and the monitor shuts off, I can't do anything until I hard reset the server

is there any way to save my VM? or did I screw this up badly enough that I just have to re-build the machine?

thanks for your time and any help is appreciated.

 

[fiona]

Hi,
that's strange behavior with pvecm. Is there anything in the syslog?

You can also copy over the virtual drives manually. If they are just files, simply use cp, otherwise the command depends on the storage type. The VM configuration can be found in /etc/pve/nodes/<node>/qemu-server/<ID>.conf.

On the cluster, place the configuration in /etc/pve/nodes/<node>/qemu-server/<new ID>.conf with an ID that doesn't exist yet in the cluster and place the virtual drives on one of your storages and rename it in case you had to pick a new ID. Then modify the configuration to reflect the new location of the drive (I'm assuming that the storage is not called the same).