Move VM and CT to another host

[TomNick]

Hello everybody, I am running 3 instances in a cluster. Every VM or CT on instance 1 and 2 are running their root disk on a shared drive on instance 3. Now instance 3 is messed up and I would like to move all the shared root disks to the local disk on instance 1 and 2. So I did copy all the files to the "dump" dir of the other instances and tried:

qmrestore /var/lib/vz/dump/vm-108-disk-0.raw 108

I got

cluster not ready - no quorum?

So tried:

qm unlock 108

brought me:

Configuration file 'nodes/pve1/qemu-server/108.conf' does not exist

So tried to delete the VM and got:

rying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
trying to acquire cfs lock 'storage-pve-shared' ...
TASK ERROR: cfs-lock 'storage-pve-shared' error: no quorum!

I am stuck right now, anybody any idea how to handle my problem? Thanks in advance.

 

[Dunuin]

You should check why there is no quorum. Without it PVE will be read-only. 2 of your 3 nodes are still up and running?

 

[TomNick]

You should check why there is no quorum. Without it PVE will be read-only. 2 of your 3 nodes are still up and running?

Yes they are up and running but the VMS and CTS are "dead" since the root data are located on node 3

 

[Dunuin]

You have to fix the cluster first. Without quorum those two remaining PVE nodes will be read-only.

 

[TomNick]

You have to fix the cluster first. Without quorum those two remaining PVE nodes will be read-only.

Meaning to complete delete node 3 and then adding it again to the cluster of 1 and 2?

 

[Dunuin]

Meaning to complete delete node 3 and then adding it again to the cluster of 1 and 2?

Check your cluster status. With 2 of 3 votes you should got quorum so probably something isn't working.

 

[TomNick]

Check your cluster status. With 2 of 3 votes you should got quorum so probably something isn't working.

Name:             T
Config Version:   5
Transport:        knet
Secure auth:      on

Quorum information
------------------
Date:             Thu Nov 30 14:12:51 2023
Quorum provider:  corosync_votequorum
Nodes:            2
Node ID:          0x00000001
Ring ID:          1.341
Quorate:          No

Votequorum information
----------------------
Expected votes:   5
Highest expected: 5
Total votes:      2
Quorum:           3 Activity blocked
Flags:           

Membership information
----------------------
    Nodeid      Votes Name
0x00000001          1 192.168.35.205 (local)
0x00000002          1 192.168.35.206

I got this one which I do not understand.....

 

[Dunuin]

Then you probably edited your number of votes to more than 1 vote per node.
5 votes are expected and you only got 2 votes from the remaining nodes and 3 of 5 votes are required for quorum.

In case you were never running a 5 node clustee you maybe gave the failed node 3 votes so it could run stand alone?