Unable to install NVIDIA driver on various GCP Ubuntu VM s with Tesla K80 GPU

Question

I have followed the guide on GCP below with Ubuntu 18 and 20 ( have also tried Ubuntu Lite , Debian and Centos 7 ):

https://cloud.google.com/compute/docs/gpus/install-drivers-gpu

Unfortunately, after completing the lengthy install I get this:

me@gpu:~$ nvidia-smi
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running

Have tried installing via the script and via the direct downloads from Nvidia site for Cuda 10. Ready to pull my hair out if that helps! I don't understand how a company that builds a bazillion gpu's can't make the installation process robust?

Answer 1

The mistake you are making here is not doing the pre installation steps before running the cuda_10.1.243_418.87.00_linux.run script The users must do all the pre installation steps specially I had to do this for Ubuntu 18.

sudo nano /etc/modprobe.d/blacklist-nouveau.conf
blacklist nouveau
options nouveau modeset=0
sudo update-initramfs -u
reboot

This seems like bit of a hack so I don't why nvidia can't make the process more robust.