Just experienced an issues with update to any NVIDIA-Linux-x86_64-46XXXX driver version. Have been running NVIDIA-Linux-x86_64-455.45.01.run for a while and everything worked fine. The update procedure to NVIDIA-Linux-x86_64-46XXXX (tried 4 different version) works without any error. Manually starting X (via sddm) after the installation is finished also works fine and the system is running as expected. However, after reboot X won’t start and I get an error message “Failed to initialize the NVIDIA kernel module. … No devices detected” in the Xorg.0.log. Reinstalling the same driver makes it work for the same session but after a reboot X won’t start. Reverting back to 455.45.01 solved the problem and everything works fine again.
System:
Gefore 1070 (desktop, not mobile)
Debian Testing
SDDM Display Manger
Kernel 5.10.1 from kernel.org (vanilla)
As currently there’s a shortage in crystal balls, a nvidia-bug-report would be required to help. Best run (nvidia-bug-report.sh - as root) after the problem occurs.
However it’s generally recommended to use the distros packaged drivers not the .run file installer. https://wiki.debian.org/NvidiaGraphicsDrivers#Installation
Thanks for the fast response. NVIDIA customer care told me to post the issue here … they did not mention the crystal ball issue, otherwise I would not have posted it. Just wanted to help, I can live with the working version for now. Thanks for the links.
Oh, that was just me trying to be amusing :-/
All I wanted to say is, that more information is needed to help you. The nvidia-bug-report.sh script, which is installed alongside with the driver, collects a lot of information useful for debugging issues.
Hello!
I am trying to install Debian 12 on my system when it comes to drivers I have problems every step I go. After updating the kernel from 6.1.0-13 to 6.5.8 Zebbly I was able to install drivers, but after restart I still have black screen. if I run “startx” command the system freezes immediately. If I run “startxfce4” I get error message “Server is already active on display0” & “unable to connect to X server. Resource temporary unavailable”.
If there’s still shortage of crystal balls, logs are attached. ;) nvidia-bug-report.log (1.1 MB)
Looking at your log, it appears that your system has both an AMD GPU and an NVIDIA one. Your xorg.conf file is telling the X server to load only the NVIDIA driver, but your NVIDIA GPU doesn’t have any display devices attached.
I would recommend deleting /etc/X11/xorg.conf and letting the X server autoconfigure itself. That will use the AMD GPU as the primary one rendering the desktop and then you can use PRIME render offload to run specific applications on the NVIDIA GPU. If you want to use the NVIDIA GPU as the primary rendering device with display offloaded to the integrated GPU, please see chapter 34 of the README.