A few steps to remember when tearing down a cluster.

• Stop the k0s service on the nodes (this stops all containers that k0s started too)
• Delete cluster state
• Delete partition tables on disks usedd by Ceph

Something like this on each node:

# Stop services
rc-service k0scontroller stop
rc-service containerd stop

# Remove old cluster config/state.
rm -rf /var/lib/k0s /etc/k0s

# Delete partition tables on Ceph disks
sgdisk --zap-all /dev/XXX