Random kernel panic.

[hjorden]

Fairly new, but know a bit.
I got a home server in the garage running different things , Home Assistant, a family tree and different things. Been going fine, could Google the stuff I didn't know. But now I'm in deep water.

I've upgraded from 18 I think, to 22. Everything seemed to work fine, but I get some kernel panic errors.
They come regularly, maybe once a day or every other day.

See error: https://ibb.co/1d3jt6m

It doesn't really tell me much. I don't really have anything to go on. I don't know what to do, to get closer to s solution.

I've cleaned up some stuff, run a RAM and HDD check without any problems.

Restarting the machine daily is driving me nuts. Any pointers and any help is greatly appreciated.

[MAFoElffen]

Is it server edition?

Please Run the UbuntuForums 'system-info' script in my signature line, and post the URL of where it uploads to.

That will tell us what your hardare is, and something about how it is installed, including the kernel boot line.

It appears to be power, pstate, and apci related. Need more info to suggest a solution.

[hjorden]

Yes it is.

I've run the script:
https://paste.ubuntu.com/p/Yxwpz3KTkk/
It said my system is missing some utilities, but it didn't state which ones. So I hope this is fine.

Got a new error earlier, maybe it's different: https://ibb.co/9WxSKWd

The help is much appreciated

[MAFoElffen]

File a Bug Report with LaunchPad.

Why am I saying this so early (in the 4th Post)?

Because look at this, in line #41 of the report:

 *-processor UNCLAIMED

I have "never" seen that before. Nor can I find that kind of error on a CPU reported anywhere on the internet. Usually devices going unclaimed are GPU's, network devices,etc. Not processors.

Let me ask a friend with an AMD CPU if he can run something and what his says...

[Doug S]

The kernel taint code codes are also a worry. Out of tree and unsigned module.

[MAFoElffen]

Been diving into some things with this "kernel stack corrupted" error...

Before blaming it on the kernel in relation to your hardware, we need to rule a few things out first.

That error often comes up when the image is corrupted due to filesystem errors or mount errors (which also points back to filesystem errors).

Is this a ext4 filesystem? Does it get this if booted from on an older kernel?

What I would try first is, if ext4, then boot from a rescue mode boot option under Grub2 > Advanced Options and do an fsck on the filesystems. If that checks out okay, then do 'network', which will remount the filesystem as r/w, then drop to a root prompt using 'root' > <Enter>, then do a reinstall of the current latest kernel... This will reinstall the latest installed kernel of releases still in support

VERSION=$(apt list linux-image-{5,6}* --installed 2>/dev/null | tail -n 1 | awk -F "-" '// {print $3 "-" $4}')
sudo apt install --reinstall -y linux-image-$VERSION-generic linux-headers-$VERSION-generic Linux-modules*-$VERSION-generic

Then test again.

If it still gets errors, then report as a Bug Report.

[hjorden]

Before the update, the kernel was 4-something, does that sound right?
I didn't get this error. Once a month or maybe every other, I lost contact to the server and had to restart. Never investigated it.

When doing fsck, I got the following:
https://ibb.co/TbtBCFg

When trying the update I get:
https://paste.ubuntu.com/p/45tPWW2xF8/

[1fallen]

Please also show this:

 apt list nvidia* --installed

Your failed upgrade looks to be installing all drivers ie:

The following packages have unmet dependencies:
 linux-modules-nvidia-450-server-5.15.0-106-generic : Depends: nvidia-kernel-common-450-server (>= 450.248.02) but it is not installable
 linux-modules-nvidia-470-5.15.0-106-generic : Depends: nvidia-kernel-common-470 (>= 470.239.06) but it is not installable
 linux-modules-nvidia-470-server-5.15.0-106-generic : Depends: nvidia-kernel-common-470-server (>= 470.239.06) but it is not installable
 linux-modules-nvidia-535-5.15.0-106-generic : Depends: nvidia-kernel-common-535 (>= 535.171.04) but it is not installable
 linux-modules-nvidia-535-open-5.15.0-106-generic : Depends: nvidia-kernel-common-535 (<= 535.171.04-1) but it is not installable
                                                    Depends: nvidia-kernel-common-535 (>= 535.171.04) but it is not installable
 linux-modules-nvidia-535-server-5.15.0-106-generic : Depends: nvidia-kernel-common-535-server (>= 535.161.08) but it is not installable
 linux-modules-nvidia-535-server-open-5.15.0-106-generic : Depends: nvidia-kernel-common-535-server (<= 535.161.08-1) but it is not installable
                                                           Depends: nvidia-kernel-common-535-server (>= 535.161.08) but it is not installable
 linux-modules-nvidia-550-5.15.0-106-generic : Depends: nvidia-kernel-common-550 (>= 550.67) but it is not installable
 linux-modules-nvidia-550-open-5.15.0-106-generic : Depends: nvidia-kernel-common-550 (<= 550.67-1) but it is not installable
                                                    Depends: nvidia-kernel-common-550 (>= 550.67) but it is not installable
 linux-modules-nvidia-550-server-5.15.0-106-generic : Depends: nvidia-kernel-common-550-server (>= 550.54.15) but it is not installable
 linux-modules-nvidia-550-server-open-5.15.0-106-generic : Depends: nvidia-kernel-common-550-server (<= 550.54.15-1) but it is not installable
                                                           Depends: nvidia-kernel-common-550-server (>= 550.54.15) but it is not installable
E: Unable to correct problems, you have held broken packages.

That could and will cause corruption.

[hjorden]

Doesn't look like I have any.
I'm just getting

Listing... Done

But if thats the cause, how come the system can run fine in sometimes over a day?

[1fallen]

I don't see how it would run period. I've not seen any Nvidia card run all drivers.
EDIT: And your system-info return MAFoElffen had you run don't even show you have an Nvidia card at all.
What recent changes have you made
Please now show this:

sudo apt autoremove --purge nvidia*