[Question] How to kill a procedure named no-such-process

XuehaiPan / nvitop

An interactive NVIDIA-GPU process viewer and beyond, the one-stop solution for GPU process management.

https://nvitop.readthedocs.io

Apache License 2.0

4.8k stars 150 forks source link

[Question] How to kill a procedure named no-such-process #134

Open HuXiLiFeng opened 3 months ago

HuXiLiFeng commented 3 months ago

Required prerequisites

[X] I have read the documentation https://nvitop.readthedocs.io.
[X] I have searched the Issue Tracker that this hasn't already been reported. (comment there if it has.)
[X] I have tried the latest version of nvitop in a new isolated virtual environment.

Questions

after exec nvitop 1 56752 C N/A 32.34GiB 100 N/A N/A N/A No Such Process However, when i exec nvidia-smi, i cant find this process 56752. I tried "kill -9 56752", but it didn't works. I also tried "fuser -v /dev/nvidia1 | xargs -t -n 1 kill -9", it didn't works.

Question: how to kill 56752 process?

XuehaiPan commented 1 week ago

@HuXiLiFeng If you are running in a docker container, the PID from the NVIDIA driver is the host PID rather than the one in the container. So, you got No Such Process. You can add --pid when using docker:

docker run ... --pid=host ...

If you are running in a physical machine, try this:

sudo lsof -t /dev/nvidia*