Server Version#: 1.30.2.6563 (Linxserver.io Docker, Unraid 6.9.2)
Player Version#: Multiple (Web, Windows Desktop, Android)
I have had hardware transcoding set up with my server’s i5 -11400 for a few years now with no issues. Today I am suddenly getting the following error message (~30 seconds after starting a transcode):
This does not occur all the time, but does occur every time I try to transcode 4K content and only some of the time with 1080p content. When I start trying to transcode the video playback stops, and ~30 seconds later I get the error.
From the server playback view nothing appears to be wrong. Hardware transcoding is shown and the time for the video playing continues to pass even though the screen is blank. There is about a 30 second period between when I start transcoding and the error message shows up.
In addition I am seeing much higher CPU usage during my attempts to transcode than I normally see when hardware transcoding is working properly.
I have confirmed this happens with both remote and local connections to my server, and across windows desktop client, web client and android (which makes sense as the transcoding issue appears to be happening on the server, not client side).
I already tried uninstalling and reinstalling the plex docker but no luck. Any help is greatly appreciated.
I’m trying to hunt down a problem everyone is seeing with Hardware Trancoding related playback.
Is it possible for you to easily switch to the 1.29.2.6364-6d72b0cf6 image and retest ?
Your screenshot above looks like normal playback but having your DEBUG logs which capture the playback failure would be great. It’ll let us see what is actually happening.
Confirmed still broken on the latest plex version 1.32.0.6918.
Each time I test I change the linuxserver.io docker repository to lscr.io/linuxserver/plex and the issue returns. Only setting to linuxserver/plex:1.29.2 fixes the issue.
You don’t have it working in Docker on Unraid.
I have PMS in Docker on Unraid and it works.
I’m using plexinc/pms-docker.
I have a problem with the LSIO - 1.32.0 image. I don’t know why; haven’t had time to dig into it. ( maybe that’s for them ? )
I’ve lost my train of thought here about where we were going.
Can you restart PMS and give me a fresh set of DEBUG logs.
I’m asking because I’ve gotten a lot smarter about the underpinnings.
I started with 1.30.0 and then did a ‘docker pull’ to upgrade it.
Linux 5.19.17-Unraid.
root@BabyNUC:~# cd /dev/dri
root@BabyNUC:/dev/dri# ls -la
total 0
drwxrwxrwx 3 root root 100 Apr 3 20:51 ./
drwxr-xr-x 17 root root 3400 Apr 3 20:55 ../
drwxrwxrwx 2 root root 80 Apr 3 20:51 by-path/
crwxrwxrwx 1 root video 226, 0 Apr 3 20:51 card0
crwxrwxrwx 1 root video 226, 128 Apr 3 20:51 renderD128
root@BabyNUC:/dev/dri# uname -a
Linux BabyNUC 5.19.17-Unraid #2 SMP PREEMPT_DYNAMIC Wed Nov 2 11:54:15 PDT 2022 x86_64 Intel(R) Core(TM) i3-7100U CPU @ 2.40GHz GenuineIntel GNU/Linux
root@BabyNUC:/dev/dri# docker container list
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
1060595ada26 d119d2302d10 "/init" 7 weeks ago Up 8 days (healthy) Plex-Media-Server
root@BabyNUC:/dev/dri#
I just wanted to confirm that I am seeing the transcode fail issue on 1.32 as well. I’ve tested various releases as they’ve come out in recent weeks, but 1.29.2.6364-6d72b0cf6-ls146 is what I keep having to go back to.
EDIT
Just wanted to say that I switched my repo to plexinc/pms-docker and am still seeing the transcoding fail issue on v1.32.0.6918.
I can get these logs for you over the next couple of days. My server is seeing heavy usage and I hate to kick users off so I’m trying to wait for some downtime.
To clarify, you would like the debug logs from running on v1.32.0.6918 covering the transcoding failure, correct?
Here are the logs. Looks like the latest LSIO release is v1.31.0.6635-8942ba8cc because that is what I get when I pull the latest docker image. Edit: I am seeing a different Plex version listed in settings (1.32.0.6918) than in logs, I am not sure why.
Let me know if these logs have what you need to troubleshoot or if there is anything else I can help with. Logs.zip (5.9 KB)
