Only Black Screen after login (after kernel/nvidia update)

General system Kernel, boot, graphics & hardware
1

Hi,

after updating my system and therefor the kernel (6.11.1.arch1-1) and nvidia drivers (nvidia-560.35.03-9), I only get a black screen after logging into KDE Plasma and Wayland. X11 works “fine”. I had no issues whatsoever before this update.
I already tried downgrading the kernel and nvidia packages without success. After downgrading, I even don’t get a login screen anymore. It’s just stuck on the console.
Can anybody recommend any next steps to try? I would really like to have wayland back. :slight_smile:

Here is a current inxi output.

Greetings
stibbons

1 Like
2

The same here

3

Me too, wayland has a black screen + my mouse, x11 seems to be “ok” I have a 4070.

4

Me three, Wayland not working but X11 does.

Black screen + cursor, terminal appears to be working in the background as Ctrl+Alt+T and the reboot works (although I can’t see it!), and the shutdown printout appears as expected.

I only have an AMD R3 3200G and 1050Ti so naturally I would much prefer Wayland over X11. Nvidia, EOS and kernel versions are identical to the @stibbons inxi output, however my GPU temperature sensor appears to be missing whereas it was fine before the update.

5

I too have this same issue. Black screen and a cursor. I used to get the KDE logo and then desktop. Now I get nothing, not even the logo. I didn’t know I could use Ctrl+Alt+T so I have had to use the long press on the power to shut down, short press didn’t work. Kwin(Wayland) here was also the jump to Kernel 6.11.1 and nvidia 560. This is the first real issue I have had with Linux since I started using 3 months ago.

6

Same issue on my end

7

Tried that and it’s the same for me. I can see the cursor and use CTRL+ALT+T and reboot that way. But there isn’t anything being rendered on screen except the black background and the mouse cursor.
If I switch to a TTY with CTRL+ALT+F3 and then back to CTRL+ALT+F1 I can see a fraction of my wallpaper for a short moment. Then all goes black again.

8

@Pepan @stibbons @Joe_Mama @ColonelKielbasa @Dre9872 @lll_Death_lll
Welcome to the community! :vulcan_salute: :enos_flag:

Have you all downgraded correctly?

sudo downgrade nvidia nvidia-utils linux linux-headers

Or did you do:

sudo downgrade nvidia linux

The first one is correct, and the second will cause issues.

You can finish with the below command, though it may be unnecessary:

sudo mkinitcpio -P && sudo grub-mkconfig -o /boot/grub/grub.cfg
1 Like
9

Please show the output of:

cat /var/log/pacman.log

Or use the below to create a file in your home directory.

cat /var/log/pacman.log > pacman_log.log
10

Hmm, I followed the Arch Linux wiki article and used:

cd /var/cache/pacman/pkg
sudo pacman -U linux-6.10.10.arch1-1-x86_64.pkg.tar.zst linux-headers-6.10.10.arch1-1-x86_64.pkg.tar.zst
sudo pacman -U nvidia-560.35.03-6-x86_64.pkg.tar.zst 
reboot

Should I give your way a try though?

Edit1: I didn’t downgrade nvidia-utils because it wasn’t upgraded in the process.

1 Like
11

This method works too. And yes, if it (nvidia-utils) wasn’t upgraded, then you wouldn’t need to downgrade it. So, there seems to be something else that is causing the issue.

1 Like
12

I just updated to the latest kernel and had the same issue of black screen on Nvidia. I reverted to an earlier snaphot with previous kernel.

13

By reverted to a snapshot, do you mean with timeshift or a similar tool? Or do you mean using downgrade?

14

I have btrfs with btrfs-assistant,snapper-support and btrfsmaintenance with grub-btrfs.

Edit: So yes I’m able to boot on a previous snapshot and then use btrfs-assistant to revert to a previous snapshot. Not necessarily the one i booted on either.

Edit2: Currently running on the following kernel and Nvidia driver

 [ricklinux@asus-tuff ~]$ uname -a
Linux asus-tuff 6.10.10-arch1-1 #1 SMP PREEMPT_DYNAMIC Thu, 12 Sep 2024 17:21:02 +0000 x86_64 GNU/Linux

 Device-2: NVIDIA GP104 [GeForce GTX 1060 6GB] vendor: ASUSTeK
    driver: nvidia v: 560.35.03 alternate: nouveau,nvidia_drm non-free: 550.xx+
    status: current (as of 2024-09; EOL~2026-12-xx) arch: Pascal code: GP10x
    process: TSMC 16nm built: 2016-2021 pcie: gen: 3 speed: 8 GT/s lanes: 4
    link-max: lanes: 16 ports: active: none off: HDMI-A-1,HDMI-A-2
    empty: DP-1,DP-2,DVI-D-1 bus-ID: 02:00.0 chip-ID: 10de:1b83 class-ID: 0300
1 Like
15

Yeah, so not the current standard EOS setup. If these members have the standard setup, they may need to follow the guide. Let’s wait for their responses.

1 Like
16

Oh shoot, I just reviewed the log and it seems that I never downgraded the nvidia driver version (forgot to paste it from my prepared text file…). Thanks for pointing me into the right direction! :slight_smile:
I just repeated the same procedure and this time I did the nvidia driver downgrade too. I now have a working wayland session at least.

But of course the root issue with the new kernel and nvidia driver still persists.

17

You’re welcome.

This is fine. It just means you’ll have to hold off on updating for a few days or maybe even a week. See the guide below on how to track updates to critical packages.

How to Track Bug Reports

18

It’s the standard install of EOS with btrfs on grub and then you add the other packages to be able to use snapshots with grub.

19

I meant for the newer ISOs. Don’t they use Dracut and Systemd-boot?

1 Like
20

Okay, I will do that then.
Is there a way to exempt the kernel and nvidia driver from update runs? So that I can install all other updates except these? Or should I completely forget about updates for a few days or a week?

1 Like