hey thanks for your valuable post can you please elaborate the driver installation method from the mid if error occur at the same time.
I checked /var/log/nvidia-installer.log
but this file does not exist at all, and /var/log/nvidia/
is an empty folder. The specific information is in myquension. Can you help me? Yes, thank you
You are installing in a WSL environment. There is no driver install in a WSL environment, which is why you can’t find various log files. You should follow the instructions here
But how can I close this server. I also wanna give it a try but don’t know what cli instruction to use
@codeguyross , I had the same issue during my installation, this is the way I fixed it:
-
A NVIDIA kernel module is this case
nvidia-drm
is already loaded, in my case I stopped the display manager and turn off the graphical interface executingsudo systemctl stop gdm
or pressingCtrl + Alt + F3
(another combination could beCtrl + Shift + F3
) this will take you to a text-based login screen. -
Log in with your username and password
-
execute
sudo rmmod nvidia-drm nvidia-modeset nvidia
, this will unload the NVIDIA modules -
check if any modules are still loaded:
lsmod | grep nvidia
-
run the CUDA installation again, in my case:
sudo sh cuda_12.6.1_560.35.03_linux.run
-
the installer is most likely located in the folder
Downloads
, so you can executesudo sh ./Downloads/cuda_12.6.1_560.35.03_linux.run
-
run
nvcc --version
andnvidia-smi
to see if you installed everything correctly -
turn on the graphical interface
Ctrl + Alt + F2
orCtrl + Shift + F2
or executingsudo reboot
I hope this can be helpful