GPU at 0000:02:00.0 has fallen off the bus

I use nvenc transcoding my audio and video files, because I have a lot of documents, it takes a long time to transcoding, but in the use of a week after the following errors are always reported as follows:

Oct 14 17:15:17 localhost kernel: NVRM: GPU at 0000:02:00.0 has fallen off the bus.
Oct 14 17:15:17 localhost kernel: NVRM: GPU is on Board 0420516010519.
Oct 14 17:15:17 localhost kernel: NVRM: A GPU crash dump has been created. If possible, please run
Oct 14 17:15:17 localhost kernel: NVRM: nvidia-bug-report.sh as root to collect this data before
Oct 14 17:15:17 localhost kernel: NVRM: the NVIDIA kernel module is unloaded.

nvidia-smi
Unable to determine the device handle for GPU 0000:02:00.0: GPU is lost. Reboot the system to recover this GPU


I observe that the temperature of GPU is too high to cause self protection! But the fan speed is too slow, there is no way to improve the speed of the fan in the Linux system?

Looking forward to your reply, thank you.

You can enable manual fan control by setting

Option "<a target='_blank' rel='noopener noreferrer' href='"http://us.download.nvidia.com/XFree86/Linux-x86/370.28/README/xconfigoptions.html#Coolbits"'>Coolbits</a>" "4"

in xorg.conf and then control it using the nvidia-settings thermal monitor page.