Nvidia driver 530.30.02 causes random freezing/crash on Ubuntu 20.04

My ubuntu randomly freezes/crashes after 20 mins to ~2hrs of using it. (If I don’t use GPU - no crashes)

Attached is Nvidia bug log and output from journalctl -b -1 -e

Ubuntu version: Linux rashmi 5.15.0-75-generic #82~20.04.1-Ubuntu SMP Wed Jun 7 19:37:37 UTC 2023 x86_64 x86_64 x86_64 GNU/Linux

Nvidia Driver 530.30.02, CUDA 12.1

I am using Python 3.11, Pytorch 2.1 for CUDA 12.1, same problem persists even if I use stable PyTorch. I have tried different Nvidia Driver 525.85. as well - but still no resolution.

I would really appreciate any help or if there are other relevant forums where I could post this.
Thank you very much
nvidia-bug-report.log.gz (441.9 KB)

1 Like