Hello NVIDIA community,
I have a Linux PC with Ubuntu 20.04.6 and a GeForce RTX 2080 Ti with the driver 520.61.05. In the past 4 months it has already happened 3 times that the GPU has stopped working. No GPU intensive programs have been run during the crash. Each time, the screen goes black and the GPU fans start spinning at maximum speed. After restarting the PC, the GPU is recognised again and the PC can be used again.
In /var/log/kern.log I can see that the following error message is output:
GPU 0000:03:00.0: GPU has fallen off the bus.
You can also find the nvidia-bug-report in the attachment.
What could be the reason that the GPU produces this error about every 4 weeks and cannot be used afterwards?
Many thanks for your help!
Best,
Marvin
nvidia-bug-report.log.gz (152.5 KB)