Is the UUID you have in the .yaml the same as the UUID you get when you use the nvidia-smi -L command?
Also, in Plex you have to configure the transcoding option to use the GPU. Simply enabling GPU access when mounting the application will not make Plex use it; it just makes the GPU available to the app.
EE no longer uses the ixapplications dataset. It creates a hidden dataset on your apps pool and then mounts that hidden dataset to /mnt/.ix-apps, so this will be the default path for all installs
that command gives me 0000:02:00.1. However, if I set this Plex won’t even start. If I set 0000:02:00.0 instead, Plex will start, but hw transcoding doesn’t work.
Can I make a suggestion? Maybe starting a new thread with your issue and posting the diagnostic information (like above, or in the stickied post) will help. Your issue is related to transcoding but is definitely a different issue, so you might have better luck with people focusing solely on that.
Thanks to everyone for their input. The workaround from @CtagaDev worked for me, I’m marking it as a solution to this topic.
+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| 0 N/A N/A 280687 C /usr/lib/jellyfin-ffmpeg/ffmpeg 316MiB |
+-----------------------------------------------------------------------------------------+
Sorry but I am a real newbie when it comes to this but I have just upgraded to 24.10 and I am having the same issues which adding my GPU to my Plex app. I have installed the Nvidia drivers and can see the GPU but I don’t know how to edit the command line to add in the UUID. Is someone able to step me through this process? Thanks