Mistake clustering

[Icery]

I am re-configuring my homelab from VMWare to Proxmox, and currently have one node (Node A) that has 10 disks passed through to a Truenas Scale vm for a ZFS pool, along with 5 vms migrated from another node (Node B), that has a fresh install of Proxmox 8. Given I am relatively new to Proxmox, I wanted to try out clustering to make moving the 5 vms easier, but mistakenly tried to make a cluster on Node A while MFA was active so Node B gets an error when it tries to join. I've found how to remove a node from a cluster, but the process freezes since there are vms still on the node.

So here are my questions:

1. Is there a way to remove Node A from the cluster it made and delete the cluster without losing data?
2. If not, what would be the best way to move things around to resolve?

 

[bbgeek17]

There are a few documents and threads covering this information:
https://forum.proxmox.com/threads/remove-or-reset-cluster-configuration.114260/
https://forum.proxmox.com/threads/proxmox-ve-6-removing-cluster-configuration.56259/
https://pve.proxmox.com/wiki/Cluste...g.-,Remove a Cluster Node,-Read the procedure

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[Icery]

There are a few documents and threads covering this information:
https://forum.proxmox.com/threads/remove-or-reset-cluster-configuration.114260/
https://forum.proxmox.com/threads/proxmox-ve-6-removing-cluster-configuration.56259/
https://pve.proxmox.com/wiki/Cluster_Manager#_remove_a_cluster_node:~:text=Cluster Join dialog.-,Remove a Cluster Node,-Read the procedure

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

Thanks for the quick reply. These are what I found initially when searching; Is it a requirement to move all the vms off the node? Most of the examples are of new installs or they have nothing on them.
I'm able to vzdump and manually move most of them over to Node B, but the vm with the multiple drives would be a problem.

 

[Maximiliano]

Hello,

Yes, you cannot join a cluster with VMs on it. You will have to backup them and move them to the cluster at a later point. See [1] for more information.

[1] https://pve.proxmox.com/pve-docs/chapter-pvecm.html#pvecm_join_node_to_cluster

 

[bbgeek17]

Thanks for the quick reply. These are what I found initially when searching; Is it a requirement to move all the vms off the node? Most of the examples are of new installs or they have nothing on them.

I am unclear on what the issue is. You asked:

1. Is there a way to remove Node A from the cluster it made and delete the cluster without losing data?

The first two links explain how to remove cluster configuration from the node, which is what you asked about. Neither of them mentions anything about new node or removing VMs.

In the end, the VM configuration is a text file located in /etc/pve. You can save it (along with things like storage.cfg), move it away and have a "clean" node. Later you can move the file back when you established the cluster. Obviously this can get very involved quickly, depending on your configuration. So the disclaimer is - dont do it if you are not comfortable and backup/restore your data.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox