Removing a node, re-imaging it and re-adding to cluster

[dougknows]

I can't seem to find this process documented anywhere. I tried to delete and re-add node2:

• # pvecm delnode node2
• bring node2 down
• put a fresh install of the latest Proxmox on node2 (all nodes are running this version)
• start node2
• from node2: # pvecm add node1

I got all kinds of key errors, and any attempt I made at fixing it without re-installing every node in the cluster failed.

Is there a step I'm missing here? No amount of reboots helped this.

Thanks,
Doug

 

[udo]

I can't seem to find this process documented anywhere. I tried to delete and re-add node2:

• # pvecm delnode node2
• bring node2 down
• put a fresh install of the latest Proxmox on node2 (all nodes are running this version)
• start node2
• from node2: # pvecm add node1

I got all kinds of key errors, and any attempt I made at fixing it without re-installing every node in the cluster failed.

Is there a step I'm missing here? No amount of reboots helped this.

Thanks,
Doug

Hi,
use -f for force...

Udo

 

[dougknows]

Hi,
use -f for force...

Udo

When I tried that, I got:

root@swlvsh1:/etc/init.d# pvecm add node1 --force
root@node1's password: 
I/O warning : failed to load external entity "/etc/pve/cluster.conf"
ccs_tool: Error: unable to parse requested configuration file


command 'ccs_tool lsnode -c /etc/pve/cluster.conf' failed: exit code 1
unable to add node: command failed (ssh node1 -o BatchMode=yes pvecm addnode swlvsh1 --force 1)

(I've replaced the full hostname, which resolves correctly, with node1 for this example.) /etc/pve/cluster.conf was a valid XML file, and it looked correct.

But you think the steps I followed are correct and should have worked, something else just got messed up along the way?

 

[dietmar]

You wrote earlier that you stopped the pve-cluster service?