Hello everyone,
I’d like to know if someone could help me by answering a few questions.
First, the two nodes are in version 8.4.11. in attach files have a list of descriptions and commands.
After having done this once before, I went through the process again of changing the IP of my second node in a 2-node cluster. Yes, I am aware that having only two nodes is not recommended and that there is a quorum loss when one of them goes down. However, due to company needs, I initially set up the cluster this way.
The issue is that after changing the IP address of my second node for the second time, I couldn’t get the quorum files to synchronize. At least it doesn’t seem to have fully replicated, because on the second node the file at /etc/pve/corosync.conf is still outdated, even though I copied the correct file (as it exists on pve1) to /etc/corosync/corosync.conf.
What I’ve already set up:
systemctl stop corosync
systemctl stop pve-cluster
pmxcfs -l
That allowed me to interact with node1 again (I couldn’t even access its GUI before). After that, I was also able to reboot node2. Since it was already very late, I didn’t make further changes and just left everything running.
My question is:
Based on what I’ve done, what should I do next? Eventually, I plan to rebuild the servers and bring them up without clustering, but for now I need to fix this situation at least enough to back up the VMs — Veeam is currently returning strange errors in this state.
I’d like to know if someone could help me by answering a few questions.
First, the two nodes are in version 8.4.11. in attach files have a list of descriptions and commands.
After having done this once before, I went through the process again of changing the IP of my second node in a 2-node cluster. Yes, I am aware that having only two nodes is not recommended and that there is a quorum loss when one of them goes down. However, due to company needs, I initially set up the cluster this way.
The issue is that after changing the IP address of my second node for the second time, I couldn’t get the quorum files to synchronize. At least it doesn’t seem to have fully replicated, because on the second node the file at /etc/pve/corosync.conf is still outdated, even though I copied the correct file (as it exists on pve1) to /etc/corosync/corosync.conf.
What I’ve already set up:
- Both nodes can resolve each other’s names correctly
- Both can communicate properly, despite being on different networks
- /etc/hosts is configured correctly on both sides with the right IPs
- On pve1, both /etc/pve/corosync.conf and /etc/corosync/corosync.conf are correct
- On pve2, only /etc/corosync/corosync.conf is correct, since I haven’t directly edited /etc/pve/corosync.conf, so it hasn’t replicated yet
systemctl stop corosync
systemctl stop pve-cluster
pmxcfs -l
That allowed me to interact with node1 again (I couldn’t even access its GUI before). After that, I was also able to reboot node2. Since it was already very late, I didn’t make further changes and just left everything running.
My question is:
Based on what I’ve done, what should I do next? Eventually, I plan to rebuild the servers and bring them up without clustering, but for now I need to fix this situation at least enough to back up the VMs — Veeam is currently returning strange errors in this state.
