Disaster recovery - proper way to re-install a failed node?

[e100]

I read that when removing a node from the cluster you must turn that node off and never turn it on again.
So I was wondering what is the proper procedure is for recovering a failed node.

Lets say my storage gets messed up and only option is to reinstall and restore VM data from backups.

The name of the node is still the same, but now it exists stand-alone and needs re-joined to the cluster.
How do I re-join the cluster after re-installing?

Can I use the same node name or do I need to use a new name?

Are there specific files we should be backing up on the node so that if this happens we can more easily re-join the cluster?

 

[dietmar]

Can I use the same node name or do I need to use a new name?

Yes, that should work.

Are there specific files we should be backing up on the node so that if this happens we can more easily re-join the cluster?

Well, maybe the network config (depends how complex that is). But most infos are now on the distributed FS in /etc/pve, so you won't loose them.