Hi,
the boot SSD of a node in our Proxmox cluster is failing and I'm looking to replace it. From what I've read before, the safest bet would be backing up to Proxmox Backup Server, ejecting the node from the cluster, reinstalling Proxmox and rejoining it to the cluster, then restoring from backup.
The boot drive uses LVM with the VMs on this node being stored on the local-lvm lv. I know this isn't ideal, and this will change with the rebuild of this server.
So, in short, my questions are as follows:
Thanks, and a nice weekend everyone!
the boot SSD of a node in our Proxmox cluster is failing and I'm looking to replace it. From what I've read before, the safest bet would be backing up to Proxmox Backup Server, ejecting the node from the cluster, reinstalling Proxmox and rejoining it to the cluster, then restoring from backup.
The boot drive uses LVM with the VMs on this node being stored on the local-lvm lv. I know this isn't ideal, and this will change with the rebuild of this server.
So, in short, my questions are as follows:
- When I backup to PBS and eject the node from the cluster, the backups aren't touched, right? Backups from nodes that no longer exist aren't purged automatically, I assume? Do I need to set the backup as protected before?
- The VMs in question are few, but they do have some physical disks passed through via /dev/disk/by-id/... - apart from maybe having to change the paths after a backup restore, is this likely to give me any problems?
- I'm planning to keep the dying SSD if any of this goes wrong. Should I run the pvecm delnode while the node is still online (as in will it delete any configurations on that particular node when I do it) or should I just shut down, delnode, replace the drive and install from scratch?
Thanks, and a nice weekend everyone!