Random kernel panics under proxmox

Hi everybody,
I’m running HA under Proxmox, and I’m experiencing random kernel panics every now and then (may be weeks, more frequently days).

See attached screenshot for details.

I have no clue where to investigate, I’ve already tried to disable selectively all my addons, but no luck.
I’ve also noticed that these panics occur after hours/days of > 90% Ram usage (allocated 4GB). Immediately after startup memory usage is at ~40%.

What can I do to investigate more, and who can be the culprit for such a behaviour?

Thanks in advance

1 Like

Have a look here

Thanks @tteck, the issue reported in that thread is slightly different than mine, that OP speaks about “restarts”, whereas I’m experiencing panics that can be resolved only by logging into proxmox and hard resetting the HA VM.

However I’ll give it a try, thanks again.

Just thinking that the PVE edge kernel would be a good test.

Just thinking that the PVE edge kernel would be a good test.

Sure, I’ve just installed it. Will report about how it goes, fingers crossed :slight_smile:

1 Like

I suspect that I am having a similar issue as I see the “lost connection” or “reconnecting” messages on the dashboard with everything unresponsive. I added the uptime sensor and it doesn’t jive with what I thought was a crash/reboot. Does this sound similar to what you were experiencing?

Obviously, given the thread topic, I too am running HA in Proxmox.

Yes. This is what I’m experiencing. When I see “lost connection” or “reconnecting”, I open proxmox, go to HA VM console and see the screenshot I posted above.

Of course, at that point, the console is unresponsive as well, and logs unrecoverable. The only thing I can do is reset the VM from proxmox.

Have you come to any conclusion yet?

Hi @tteck , so far so good. HA has been up for 10 days now, will continue monitoring it.
I’m still experiencing a very high memory usage (94% at the time being), but no kernel panics so far.