Hi, I screwed up when creating a cluster with an existing node. Now I can't figure out how to fix it.
Earlier this week I set up my first Proxmox node, it's been running fine with 5 VMs ever since.
I just got the second node today, and was therefore going to set up a cluster. So I followed this wiki page to make a cluster with a dedicated cluster network: https://pve.proxmox.com/wiki/Separate_Cluster_Network
Immediately after running:
I realized I was an idiot and screwed up the "-ring0_addr one-corosync". The wikipage even mentions this as the first entry under "Troubleshooting". But it doesn't mention how you actually correct this, or revert the clustering to do it over again.
Now I'm struggling to figure out how to fix this. As I have existing virtual machines on this main node that I really wish to keep.
I cannot run any pvecm commands as the current cluster votes is 0. As such, commands like "pvecm expected 1" just gives me "Cannot initialize CMAP service" and does nothing. The node also doesn't seem to be capable of deleting itself from a cluster.
How do I reverse the actions of a "pvecm create" command?
Or alternatively fix this issue, if it's possible without screwing up the cluster.
Any help would be much appreciated.
EDIT: Forgot to mention that I'm aware I can stop "pve-cluster", but I still can't figure out how to correct the issue after that. I tried changing "one-corosync" to my node's name or "<nodename>-corosync" in /etc/corosync/corosync.conf. But once I restart pve-cluster, it gets changed back to "one-corosync".
Earlier this week I set up my first Proxmox node, it's been running fine with 5 VMs ever since.
I just got the second node today, and was therefore going to set up a cluster. So I followed this wiki page to make a cluster with a dedicated cluster network: https://pve.proxmox.com/wiki/Separate_Cluster_Network
Immediately after running:
Code:
pvecm create <clustername> -bindnet0_addr <my node's IP> -ring0_addr one-corosyncNow I'm struggling to figure out how to fix this. As I have existing virtual machines on this main node that I really wish to keep.
I cannot run any pvecm commands as the current cluster votes is 0. As such, commands like "pvecm expected 1" just gives me "Cannot initialize CMAP service" and does nothing. The node also doesn't seem to be capable of deleting itself from a cluster.
How do I reverse the actions of a "pvecm create" command?
Or alternatively fix this issue, if it's possible without screwing up the cluster.
Any help would be much appreciated.
EDIT: Forgot to mention that I'm aware I can stop "pve-cluster", but I still can't figure out how to correct the issue after that. I tried changing "one-corosync" to my node's name or "<nodename>-corosync" in /etc/corosync/corosync.conf. But once I restart pve-cluster, it gets changed back to "one-corosync".
Last edited: