I recently upgraded my proxmox cluster that is a bunch of NUC and NUC class (2 NUC, 1 Beelink) devices. They had been rock solid on 7.x and after using Debian 12 now for a few months as my primary Dev Workstation, I said let's move forward. However, the Beelink device is incredibly unstable. It may be something in it's NIC or otherwise as I see some errors in the syslog referencing interface down events. I have swapped cabling and interfaces to no solution, the node keeps losing connectivity and then getting fenced.
Is there some other t-shooting that can go into it? My thoughts of next steps would be:
1) I see there is a kernel update out today, going to try to get that in first
2) Attempt to reformat the node to 8.0.2 from a boot ISO
3) Attempt to downgrade the cluster to 7.x
How I would look to accomplish the downgrade, I don't know if this is possible or not. I would look to take one node out at a time and reformat with a 7.4 ISO image, going node to node and then rejoining to the cluster after the 7.4 image is on the host. Would the cluster support such a move?
Is there some other t-shooting that can go into it? My thoughts of next steps would be:
1) I see there is a kernel update out today, going to try to get that in first
2) Attempt to reformat the node to 8.0.2 from a boot ISO
3) Attempt to downgrade the cluster to 7.x
How I would look to accomplish the downgrade, I don't know if this is possible or not. I would look to take one node out at a time and reformat with a 7.4 ISO image, going node to node and then rejoining to the cluster after the 7.4 image is on the host. Would the cluster support such a move?
