After a month out on fieldwork, I’m back and decided to upgrade before working and… I’m becoming rather scared of upgrading.
Kernel panic.
“double fault, error code 0x0”
“CPU: 2 PID: 2719 Comm: fc-cache32 Tainted: G”
I’m trying to upload an image from the phone.
I cannot boot into a previous snap either, as the screen flickers constantly and won’t let me log in.
Anyone has any ideas? I might lose some of my recent work, so I’m worried.
Mwwh, kernel panic triggered by zypper removing fetchmsttfonts-11.4-22.1.noarch it seems.
On Double fault - Wikipedia I read:
As double faults can only happen due to kernel bugs, …
That does not look good.
Can you boot in rescue mode?
Alright, I managed to complete a botched dup (I don’t know why it was botched, but I fear a power issue) and while I could update rather well, and rebuild the rpm database, I still cannot log into the DE. Related or not, it seems apparmor is having problems with loading profiles:
(1518278266.439:2):apparmor=" STATUS" operation=“profile_load” profile=“un confined” name=“ping”
(there are about 5 names)
I was wondering if there’s any way to fix that? Thanks in advance!
Not sure I can follow you; to rebuild the rpm database you need to be logged in, so what is it, “DE”, where you can not log in to?
I cannot run gdm nor load Gnome at all. When booting I get several apparmor error messages (the ones I previously mentioned). I can log into text mode, but not graphic mode. The graphic server doesn’t seem to load, the screen keeps flickering and trying to type the root password to even reboot becomes next to impossible. I’m sorry if my typing is weird, I’m posting from the phone.
Thanks in advance for any help you can provide!
The default TW takes snapshots of the root partition before any zypper activity (and at time intervals). From the Grub scren you should be able to roll back to the system as it was just before the botched “zypper dup”.
Yes, indeed. But that doesn’t work.
Alright, I don’t seem to find errors in the log files, except for apparmor related errors, but I don’t seem to be able to regenerate the questioned profiles successfully. I still cannot log into the graphic mode. I did run systemctl status apparmor.service and it seems to be running fine, which is odd.
I wonder what it could be that causes this ordeal.
Initially I got several python related errors, but after re dupping they seemed to be corrected. I’m at a bit of a loss.
Sorry for all the trouble.
What video card are you using?
Did you reinstall the drivers?
The flickering at login can be caused when you do a zypper dup
without first uninstalling the proprietary NVIDIA drivers (installed “the hard way”).
I am not sure, but I think that your current state is:
You have logged into a text-mode console (e.g. tty1) and everything appears normal. You have network access. All graphics modes result in an unusable screen. The latter means that you could not use Grub to roll-back to a previous snapshot.
As an aside you could temporarily remove the grub graphics packages and you might have a useful text (curses) menu.
# zypper rm gfxboot gfxboot-branding-openSUSE
What I am guessing has happened is that your graphics card’s memory has become corrupted. Try powering off and removing the mains power lead completely for a few minutes. If it is a laptop remove or disconnect the battery. Most machines will keep standby power to the motherboard, graphics and network devices etc. while there is a physically connected power source. A few minutes without power will normally reset the device scratchpad ram and enable a clean boot.
Hi! Thank you for all your answers. I’m not using nvidia, but amdgpu. I’ve à AMD A8-6410 2.0GHZ QUAD Core, Mullins, and the rade on driver never seemed to work well with it.
It seems the last error I get is amdgpu related, and complains about kfd not being supported. I cannot boot into graphic mode with 4.15 nor 4.14.12 nor 4.14.11 which are the snaps I have.
I reinstalled amdgpu and ran mkinitrd, but to no avail.
Sounds rather bad…
Hi! Thank you for all your answers. I’m not using nvidia, but amdgpu. I’ve à AMD A8-6410 2.0GHZ QUAD Core, Mullins, and the rade on driver never seemed to work well with it.
It seems the last error I get is amdgpu related, and complains about kfd not being supported. I cannot boot into graphic mode with 4.15 nor 4.14.12 nor 4.14.11 which are the snaps I have.
I reinstalled amdgpu and ran mkinitrd, but to no avail.
Sounds rather bad… I will try powering it off completely for a while. Thank you for the idea!
If you cannot read the Grub screen, then the problem is occurring before the kernel and any GPU drivers are loaded.
Thank you for all your kind answers. Yes I can read the grub screen, it seems I just cannot get into graphic mode. Amdgpu keeps showing the same error about kfd not being supported right when it seems the gdm login screen is going to load, then the screen flickering starts.
Other than that, the unconfined apparmor profiles are noted, but I doubt this would prevent the graphic mode from starting.
I’m at a loss.
How can I do a reinstall without losing my files? Backing everything up (just in case) would take forever.
Truly at a loss here.
Boot to a terminal log in as root run zypper dup to get things back in line
If there was a power outage you could have a partial update. Redoing should bring it back to normal.
Have you tried booting to previous kernel or tried booting with nomodeset
Thank you. Like I mentioned before, I did all of these things but sadly I cannot load the graphic interface anymore.
I recently reinstalled, using the same default filesystem,
but chose a new user name – my old home folder is still accessible.
I have no idea if that’s going to work for you.
Thank you. I’ll consider that.
I noticed on Xorg log that I get a glx error, it might be the cause, as having most everything suspicious reinstalled, this is the only error I still get.
I’m posting from the phone so perhaps I cannot quote perfectly:
AIGLX : reverting to software rendering
AIGLX error: dlopen of /usr/lib64/dri/swrast_dri.so failed (/usr/lib64/libllvm.so.5: file too short)
GLX: could not load software renderer
GLX: no usable GL providers found for screen 0
I tried force reinstalling libllvm5, but it doesn’t seem to fix the problem.
Thanks in advance for all the help you can provide.
You could try making a backup copy of /etc/X11/xorg.conf
and then deleting the original.
I haven’t tried it but I see:
X -configure
https://www.x.org/releases/X11R7.7/doc/man/man1/Xorg.1.xhtml