Proxmox 6 - Chep - Cluster HA

[Fabio]

Good evening, I installed proxmox 6 with clustered chep.
I have 3 nodes, each with 2 HD SATA. Diso one contains the operating system proxmox, disk 2 used for OSD by chep.
In fact, the cluster seems to work if I disconnect the network card of node 1 where I created the virtual machine, it is automatically migrated. The virtual machcian's HD is on the chep pool.
I would like to understand a few things: if disk 1 of node 1 (the disk in which I have the proxmox operating system) fails, and I change it, do I have to reinstall the whole node?
Is there a system to backup the configuration of node 1 so that I only change the failed disk?
If disk 2 (one of the OSDs) fails, how do I replace it? Is there a procedure to follow?

 

[spirit]

for disk os:

you can backup /var/lib/pve-cluster/config.db && /etc/corosync/corosync.conf.
then, when you reinstall the node (with same hostname/ip), restore theses files, reboot, and it should be ok.



for the osd, if the disk crash, you need to remove/destroy osd in proxmox ceph gui && recreate a new osd with the new disk.

 

[Fabio]

Good morning,

I did the backsup and the indicated files and reinstalled the node with the same name and ip addresses. Restored the indicated files but the node does not restart.
After reinstalling the node I also installed ceph but without configuring it. Did I omit something?
Could you be more precise on the procedure to follow for the recovery of a proxmox operating system that is part of a cluster with ceph?