“Logs” directory, on Linux platforms, is directly under “Plex Media Server”
Please be advised:
This is a HW transcoding thread –
You’ve stated Ryzen 3600 (which has no HW trancoding internally)
You’ve stated you have no external GPU.
I think you might want to be in your own thread ?
I have moved all Intel Arc GPU posts to their own thread .
Please keep this thread for generalized HW transcoding issues.
I’m having the same problem. Is this being fixed?
I’m running unraid 6.12.11 which uses linux kernel version 6.1.99. I’m using the official docker container (plexinc/pms-docker version Version 1.40.4.8679).
Hardware transcoding works without tonemapping enabled. If I enable it then it will start using the cpu.
I fixed my issue when I did the following:
I removed all nvidia gpu referenced in the docker settings:
The Nvidia gpu still showed up in Plex but now when I select the alder lake gpu it uses that gpu and tone mapping works.
Just seen 1.40.5.8854 is available for PlexPass Plex Media Server - #636 by gbooker02
With this fix
However it doesn’t seem to have solved it for me, anyone else tried it yet?
I’m running Unraid 6.12.11 and Intel 12600 CPU.
The workaround of disabling tone mapping still works.
I running my plex under docker (unraid). Version 1.40.5.8854
When I play a video file in my browser and force it to transcode with my nividia GPU it shortly starts a transcode process:
and then disappears:
When I play from my android via app it works just fine. When I play from my android browser I get the same issue.
The video stucks and it doesn’t going on anymore. It seems to me, that it plays the little temp file which he transcoded and the file ends it stops. But he doesn’t going on to transcode.
PMS will probe the hardware momentarily after you hit play to determine if it can handle all the tasks needed for playing this video.
If it can then it launches FFMPEG with the appropriate command line options to use the GPU.
PMS logs ZIP file of the playback attempt tell us why it decided what it did (logs don’t last long… 5 rollovers)
Thank you for your reply.
I played the file and directly after I downloaded the log file.
There are some error messages regarding readonly database 
Is this connected to such an issue? Because transcoding is working on android apps?
Plex Media Server.log (16.8 KB)
I already tried your repair script (thank you for that 
) - but this didn’t fix this db readonly error.
My script will keep the UID:GID which the database exists as.
It can’t fix this kind of error because it doesn’t know what the correct UID, GID, or perms should be.
Given this is in a container, the container has a UID and GID
(or PUID : PGID or PLEX_UID:PLEX_GID )
the error comes from either;
Look at where /config maps in the real filesystem and start tracking down the owning UID:GID to see what’s not set correctly.
You’ll probably find either a chown or two find commands will fix it.
allright, thank you for that hint.
I use unraid, so I did a reset of the permissions.
The new log file
Plex Media Server.log (15.1 KB)
There are entries like “ERROR - SQLITE3:0x80000001, 11, statement aborts at 7: [SELECT coalesce((SELECT max(blockid) FROM ‘main’.‘fts4_metadata_titles_icu_segments’) + 1, 1)] database disk image is malformed”
but when I run your script it says:
Checking the PMS databases
Check complete. PMS main database is OK.
Check complete. PMS blobs database is OK.
Is there any other things I can check?
Edit: I run your script now in “automatic”, will report back
After automatic repair the malformed messages are gone.
Plex Media Server.log (12.5 KB)
The transcoding issue still exists. I can’t really see any error messages in this log. Are there any other log files which I can check?
I changed the debug level to “all”, maybe there is something you can read out of that?
Plex Media Server.log (516.7 KB)
Maybe the web player isn’t reporting back to the server that he needs more data and the server thinks the web player stopped playback?
Edit:
There is an entry
Aug 15, 2024 07:15:06.995 [23266085317432] DEBUG - [Req#10d6/Transcode/gyi7gu5tsgds5k0bi7exxd5n] TPU: hardware transcoding: enabled, but no hardware decode accelerator found
but then
Aug 15, 2024 07:15:08.551 [23266381015864] DEBUG - [Req#f66/Transcode] Codecs: hardware transcoding: testing API nvdec for device 'pci:0000:01:00.0' (GA106 [RTX A2000])
around
Aug 15, 2024 07:15:08.726
it’s trying to transcode
When I play via android app:
nvidia-smi looks fine here
web player peaks at 100mb for a second and then disappears:
I already deleted the Codecs folder to let recreate things.
I just had a crash on a preview build for subtitles, going to update to the beta version as now I see that has come out. There was no dump even though it said there was.
Looking in logs the time of crash don’t really show anything, post launch says about unable to start driver. What can I do now in case of a crash next time, there was quite a few people streaming at the time perhaps the GPU (Intel) was overloaded?
Provide the Debug server logs so we can look at it and figure out what happened – is the best course of action. 
Sure, but I need to enable debug logging first right? I will update to the latest beta and enable debug logging.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
