Proxmox VE: Remove a Dead Node and Go Back to Standalone

20 January 2026

1. Diagnose

pvecm nodes                                     # cluster members
pvecm status                                    # quorum
ls -l /etc/pve/nodes                            # node folders
ls -l /etc/pve/nodes/DEAD_NODE/qemu-server      # VMs defined on the dead node

2. Remove the dead node

Only if it is never coming back with its current configuration.

pvecm expected 1              # regain quorum with a single surviving node (2-node cluster)
pvecm delnode DEAD_NODE

If its VMs had disks on shared storage, move their configs to a live node before cleaning up:

mv /etc/pve/nodes/DEAD_NODE/qemu-server/*.conf /etc/pve/nodes/LIVE_NODE/qemu-server/
rm -rf /etc/pve/nodes/DEAD_NODE

3. Go back to standalone (optional)

To dissolve the cluster on the surviving node:

systemctl stop pve-cluster corosync
pmxcfs -l                     # local mode, lets you edit without quorum
rm /etc/pve/corosync.conf
rm -rf /etc/corosync/*
killall pmxcfs
systemctl start pve-cluster

A reboot at the end doesn't hurt. If you use Ceph, handle its monitors and OSDs separately.

Docs: Proxmox cluster manager.