Killing Plex even when it's not using GPU

[CorneliousJD]

I'm running into an issue where this script (scheduled hourly) is killing my Plex server - the container itself keeps running but Plex crashes and restarts itself within the container.

I only noticed this when I added live TV/DVR to Plex - it keeps causing DVR recordings to end abruptly and sometimes interferes w/ media being played too.

This is all happening on unraid 7.0.0 and 7.0.1 -- is there no need for this script now?
Is it safe to run it just once at array startup instead?

[CorneliousJD]

Checking for process termination eligibility...
No Docker containers are using the GPUs. Terminating non-Docker processes using NVIDIA devices...
USER PID ACCESS COMMAND
/dev/nvidia-uvm: nobody 30733 F.... Plex Media Serv
/dev/nvidia0: nobody 30733 F...m Plex Media Serv
/dev/nvidiactl: nobody 30733 F.... Plex Media Serv

Are you still having this issue? I've got the same issue in that if I'm watching plex (looks limited to the (samsung) TV) when the script runs it results in the video stream crashing. I've changed the script to run at 3AM as a stop gap

[CorneliousJD]

Are you still having this issue? I've got the same issue in that if I'm watching plex (looks limited to the (samsung) TV) when the script runs it results in the video stream crashing. I've changed the script to run at 3AM as a stop gap

I outright stopped using the script. It doesn't seem necessary anymore, the GPU will idle immediately upon not being used so it doesn't seem to be saving power anymore at all - even booting the server from a cold boot, it'll go into idle low power mode right away, so there was no need for the script anymore at all from what I could see.

I've done the same, I'll assume it's a change in the drivers as I'm sure it's not been doing it forever. Thanks for replying too.