OK. After reading wiki, this forum, and other sources, I think I'm almost ready.
I'm doing my homework to avoid issues. My home lab environment as 3x PVE nodes in a cluster. All nodes are fully updated to v8.4.12 and all my VMs/LXCs are backed up. I'm unclear as to best way to backup my host configs. In the event of an upgrade issue, I want a record of my configs. I've tried to keep the hosts as stock as possible but there are some things there that I'd like to protect if a fresh install was needed (e.g. UPS monitoring, PCI passthrough config, driver blacklisting, etc). 2x of the nodes are ZFS booting so my understanding is that Clonezilla isn't an option
I've yet to run "pve8to9 -full" to see what comes back... my plan was as follows:
1) bring down all VMs/LXCs
2) change Debian and Proxmox repositories to Trixie
3) issue "apt update"
4) issue "apt dist-upgrade"
5) pray
My biggest question is, because this is a cluster, whether I can do this node by node, bringing #1 backup under v9 BEFORE proceeding to the next node (and then the next)? Or, do I need to update all 3x nodes at the same time? I'm trying to understand whether, in the event of an issue, I can pause the upgrading to fix while the other nodes are up...
I'm doing my homework to avoid issues. My home lab environment as 3x PVE nodes in a cluster. All nodes are fully updated to v8.4.12 and all my VMs/LXCs are backed up. I'm unclear as to best way to backup my host configs. In the event of an upgrade issue, I want a record of my configs. I've tried to keep the hosts as stock as possible but there are some things there that I'd like to protect if a fresh install was needed (e.g. UPS monitoring, PCI passthrough config, driver blacklisting, etc). 2x of the nodes are ZFS booting so my understanding is that Clonezilla isn't an option
I've yet to run "pve8to9 -full" to see what comes back... my plan was as follows:
1) bring down all VMs/LXCs
2) change Debian and Proxmox repositories to Trixie
3) issue "apt update"
4) issue "apt dist-upgrade"
5) pray
My biggest question is, because this is a cluster, whether I can do this node by node, bringing #1 backup under v9 BEFORE proceeding to the next node (and then the next)? Or, do I need to update all 3x nodes at the same time? I'm trying to understand whether, in the event of an issue, I can pause the upgrading to fix while the other nodes are up...
