Desktop
OS: Ubuntu 24.04.1 LTS
GPU: NVIDIA GeForce RTX 4060 Ti from Gigabyte eagle (not OC not manually OC)
Drivers: 550.107.02
NVML Version: 12.550.107.02
Happens at random times without any load at the GPU and the temp is at 38C when it happens.
Also when it happens reaches 4400 RMP and manually at the max percent the fan goes only to 3400 RMP and when the problem occurs I tried to manually set the fan to 30% and the fan not responding (keeping the 100%).
Any ideas?
Usually, the fan ramps up to maximum when some gpu error occurs. Anything like “Unknown Error” in nvidia-smi -q output when it happens?
I didn’t check for any logs because I didn’t know where to look. I will try to see the logs from the command you propose. To start with, I want to clarify if this is a driver/software problem or a hardware problem.
Also, my GPU has 3 fans and the 2 of them work together as one. The two fans have the name fan1 and only to them does the problem occur.
Might be both, the nvidia-smi output from when the issue occurs should make things clearer.