4

Today I got lot of unusual looking updates to Ubuntu22 (with NVIDIA GT 630), similar as describe in other fresh question. After 4th round (typically 1-2 x to install and remove unused) of message to install updates/remove old, I cancelled it and rebooted. However, then boot failed somehow, and I was left with back screen after normal Ubuntu had listed some items loaded/started. Further, the starting logo was in low res during the load. I only managed to get to safe without driver for display or sound.

General error in distributed updates?

Should I wait for new updated and assume the situation is corrected, or what is the process to correct this?

Some details:

$ lsb_release -a  
No LSB modules are available.  
Distributor ID: Ubuntu  
Description:    Ubuntu 22.04.2 LTS  
Release:    22.04  
Codename:   jammy

$ hostnamectl
Static hostname: ... Icon name: computer-desktop
Chassis: desktop
Machine ID: ...
Boot ID: ...
Operating System: Ubuntu 22.04.2 LTS
Kernel: Linux 6.1.0-1015-oem
Architecture: x86-64
Hardware Vendor: System manufacturer
Hardware Model: System Product Name
$
$ sudo ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:04.0/0000:01:00.0 == modalias : pci:v000010DEd00001284sv....
vendor : NVIDIA Corporation
model : GK208 [GeForce GT 630 Rev. 2]
driver : nvidia-driver-470 - distro non-free recommended
driver : nvidia-driver-450-server - distro non-free
driver : nvidia-driver-390 - distro non-free
driver : nvidia-driver-418-server - distro non-free
driver : nvidia-driver-470-server - distro non-free
driver : xserver-xorg-video-nouveau - distro free builtin

jmizv
  • 103
mtp
  • 49
  • What’s your kernel release number (uname -r) ? Does it end with -generic ? Most people (like me) use the generic kernel stack, but it seems that on 22.04 and 23.04 at least, update-manager suddenly wants to install other kernel stacks. Something seems to have gone wrong with Canonical’s update system… – NovHak Jul 01 '23 at 23:11
  • I have not selected to go for anything else then generic, but is seems update-manager has changed something: $ uname -r ->
    6.1.0-1015-oem
    – mtp Jul 02 '23 at 14:50
  • Indeed, it’s not a generic kernel. What’s the output of apt list --installed 'linux-*' ? You should probably delete all linux image, headers, modules, and Nvidia objects and signatures, that are not generic. – NovHak Jul 02 '23 at 20:59
  • I posted in a comment below what I removed in my case where lowlatency and oracle stacks had been installed, I suppose you should adapt to the unrequested kernel stacks (i.e. at least oem from what you wrote) that got installed on your system, and reboot after all. Of course, make sure there’s at least one generic kernel remaining beforehand, or you may run into surprises… but it’s likely still there. – NovHak Jul 02 '23 at 21:06

4 Answers4

3

Exactly the same situation here. The solution that worked for me: Explanation: there are no kernel modules installed. Please try installing the kernel headers

sudo apt install linux-headers-$(uname -r)

and then post the output of

dkms status

Reboot and good. Thanks to all who took the time to search and share.

2

I too ran into problems with the new 6.1.0-1015-oem kernel. I finally got it to work with the latest NVidia drivers (535.54.03) by following Tooster's advice in this post.
Basically, enter:

sudo apt install linux-headers-$(uname -r)

while running the latest drivers from NVidia and using the new linux kernel. Reboot and profit.

zx485
  • 2,426
1

My system initially : GeForce GT 730 with 'nvidia-driver-470 - distro non-free recommended'.

Since the last Ubuntu upgrade I had an issue with booting. Hanging on 'Plymouth bootscreen'. Purging the NVIDIA-drivers - as suggested elsewhere - solved that, but had some side effects.

As for a possible broken update : linux-firmware:all 20220329.git681281e4-0ubuntu3.14 was published minutes ago. So I installed that. Just prior to that I had installed the kernel headers as mentioned here earlier.

No issues now with booting. My suspicions are on a broken Linux-update.

PJW
  • 11
1

I saw the same error with updates you described and ended with a dysfunctional Nvidia driver, but with new kernel 6.1.0. Purging Nvidia, reinstalling, different versions: didn't work, then I applied this command:

sudo dkms autoinstall

And reboot. Maybe it helps for your problem too. After this, I installed the Nvidia driver again (which I previously uninstalled), rebooted again, working great. Ubuntu 22.04.2, Kernel 6.1.0-1015-oem, Nvidia driver 535.

w-sky
  • 1,286