Server Version#: Version 1.40.3.8555 (corrected)
Server Description: Ubuntu 22.04.4 LTS
Release: 22.04
Codename: jammy
HDR Tonemapping: Disabled
My Plex Server runs in a docker container on an Intel NUC11PAHi5.
It uses linuxserver/plex:latest as it’s base - I use this mostly because I read somewhere it is somehow ‘easier’ to configure.
I was just watching an episode that was encoded in x265 last night with no issues. 12 hours later, when I try to resume playing it on plex it says “Conversion failed. The transcoder exited due to an error”. This behavior happens on every device/platform I use be it windows app, web, phone or tv.
I have another show that I recently imported that refuses to play and returns an error code: 4294967283. The show’s episodes are encoded in h264. Considering that I was able to watch the x265 episode 12 hours ago, I assume these two issues are not related. As I understand it, this problem will be resolved in a future fix that is currently being worked on.
I used to analyze files hoping that they would work after the analysis, but this has not been working as of late.
Any advice? What’s going on? As far as I am aware, I have not made any adjustments in the last 24 hours. Let me know if logs are needed. I have captured some logs, but within 1 minute, there are about 3-4k lines (debug log mode on). Is there a recommended filter to use?
UPDATE: oh it’s playing again. don’t know what has changed. oh well.
UPDATE2: oh wow, not sure how but the x265 files are playing fine now too.
UPDATE3: and we’re back to square one - was watching clarson’s farm x264 from plex not 1 hour ago. trying to resume or play another episode and now i have transcoder issues again. in between these updates i made some changes to the secure server settings because I noticed tautulli wasn’t connecting for some reason so did some troubleshooting for that. but other than that, nothing else.
UPDATE4: and it’s playing again. honestly, don’t know what’s going on.
UPDATE5: spoke too soon. some episodes still don’t work.
UPDATE6: OK, so I’ve tried deleting codecs and rebooting. Didn’t notice any change in behavior. ran sudo apt update and upgrade on my server. Restarted the thing. As soon as Plex was up, I tried playing some videos and everything seems to work. A few minutes later, everything doesn’t work again. The server is currently scanning and updating metadata. not sure if this is relevant. Is it possible that the library needs to finish scanning everything for things to work? again, honestly don’t know what is different. will probably give up in a few weeks and do a clean install (fingers crossed that works) or it’s time to leave plex.
The files you are having issues with what audio format are they and do you have subtitles enabled?
Hi @morganj,
Audio Formats: AAC Stereo and EAC3 5.1
Subtitles: Usually OFF unless there are words or phrases that i’m not familiar with or i’m watching anime.
UPDATE: not sure if this helps but I decide to spin up my old synology 918+ plex server (Version 1.40.4.8598). no issues with the same files.
So what is your current server version where you are having the issue?
The Synology is on a pretty new server version.
Have you noticed any pattern with the files not playing? For example does it only happen when trying to resume them or does it also happen when trying to play from start?
The reason I asked about the audio format and subs is there is a known issue where EAC3 audio that requires transcoding while subtitles are enabled will cause playback to hang, but disabling the subtitles playback will start.
I realize I made a mistake with my previous reporting of the problem server version - It should be Version 1.40.3.8555. I reported the version number of the windows app used on my desktop instead.
Thanks for pointing out the synology server version being pretty new. I updated it after I noticed all the problem files playing properly. It hasn’t been up since I transitioned servers a few months ago, so it probably started on an older version.
Should I update the problem server? Would this resolve the issue? Pretty sure I’m on the ‘latest’ version of the docker image. I’ll have to figure that one out.
On the problem server, I’ve noticed that the problem files start showing signs of issues when imported - the files usually have issues detecting codec details i.e. audio, video and sub data says ‘none’. After some time or after I analyze a file, the codec details will be populated (i’m not sure of the accuracy). When I play a file, regardless of resume or from the beginning, the file typically doesn’t play. If I somehow manage to get a ‘problem’ video file (video A) playing, the next video (video B,C,D, etc) will usually play if it continues from the previous episode. Sometimes if I select video B (C,D,etc) from the menu, it won’t play. Going back to video A usually works, but sometimes does not.
The reason I asked about the audio format and subs is there is a known issue where EAC3 audio that requires transcoding while subtitles are enabled will cause playback to hang, but disabling the subtitles playback will start.
This means that as long as I don’t enable subtitles when audio is using the EAC3 codec, it should be fine, right? I’m pretty sure I am avoiding exactly that combination, but I will pay more attention.
1.40.3.8555 is the latest production version of the app and 1.40.4+ is currently the beta version which you have on your Synology it seems. It would be interesting to see if upgrading your docker version to the beta version if those files work or not.
The problem files are they direct playing or transcoding? And if transcoding are you using HW or SW transcoding?
It would be great if you could share both your client’s Android logs as well as your PMS logs after reproducing the issue. It would also be helpful if you could share a problem file.
And yes if you disable subs when using EAC3 audio it should be fine, but also the issue is only if it’s transcoding. Direct play with that combination is fine.
I’ll try updating the docker version after this.
It should be directplay and HW transcoding, if any transcoding is happening.
The following is the Plex Transcoder Settings.
- Transcoder temporary directory: /transcode
- Enable HDR tone mapping: OFF
- Use hardware acceleration when available: TICKED
- Use hardware-accelerate video encoding: TICKED
- Hardware transcoding device: TigerLake-LP GT2 [Iris Xe Graphics]
docker volumes include:
/dev/dri:/dev/dri
/tmp:/transcode
Android Logs here.
PMS logs here
I had a quick look at the logs and I have no idea how to shorten it, so I’ll just leave it as is.
The episode details are:
Clarkson’s Farm - S3 • E7 (AAC)
Filename: Clarksons.Farm.S03E07.720p.AMZN.WEBRip.x264-GalaxyTV (here)
The Acolyte - S1 • E5 (EAC3)
Filename: The Acolyte S01E05 Night 1080p DSNP WEB-DL DDP5 1 H 264-NTb[TGx]
I directplay as much as I can, unless it’s anime. I can’t say the same for my users - from what little research I’ve done, it looks like people don’t bother changing the defaults or users prefer to use as little data as possible,
*links expire in two weeks.
Thanks, I was able to download the logs but not the file it keeps failing around 70mb.
Do you want me to try upload it elsewhere or is the filename sufficient for now?
Figured out how to update my intel NUC docker plex version to 'Version 1.40.4.8626 '.
Playing the same episodes and some of the known problem files, indicates that this resolves the issue.
However, playing a movie file that was also problematic still did not work but it seems to play after a server restart. I’m concerned that the same behavior as before may still be an issue i.e. the file will start to show the same issue after some time has passed since the last server shutdown/restart.
Will report back later.
Ok, that’s promising that it at least seems better now. Please let me know if it does start up again. Looking at the logs you gave me it was failing to direct play on the client device and then tried transcoding on the server but seems like it was failing to transcode. There were some fixes for transcoding in the latest PMS build but it was related to tone mapping which you stated was disabled.
Ran into the problems again last night and tested against windows app, android phone & tablet and a Sony smart TV.
Yesterday’s movie file no longer plays as is the same for Clarkson’s Farm S3E7.
The Acolyte S1E5 plays fine.
You mention tone mapping, so I included that in my test cases, just FYI as you might see that in the logs. I disabled the feature again after one test failed.
logs and video
UPDATE: I noticed on chrome that when I play a EAC3 film, it gets stuck loading (perpetual loading wheel). I tested two films. One managed to play after what feels like a minute of loading (this video was resumed and had subtitles enabled). The other did not play (this video was played from the start and does not have subs enabled).
UPDATE2: I’m playing The Boys S4E3 that is a EAC3 video. It plays on Android (smartphone&tablet) but on chrome, it plays after the loading wheel is stuck for a few minutes.
filename: The Boys S04E03 Well Keep the Red Flag Flying Here 1080p AMZN WEB-DL DDP5 1 H 264-NTb[TGx].mkv
I was able to download the file this time. I will try to reproduce the issue. So you had the issue on all those devices?
Yes, the issue is present on all the devices I mentioned.
Thanks @morganj
Would you mind enabling Debug logging on your server then reproduce the issue and share your server logs again? The debug logs will give us a lot more information.
got you both the android and the server logs here.
UPDATE: Tried playing the shared problem video file today on the windows and android (phone) apps. Playback successful. I noticed the windows app prompted me for UAC so assume it’s updated to a new version (fyr: 1.96.0.177-dc5df788). Not sure if the android app was updated. Not sure if anything else is different.
So looking at your server logs your server seems to have issues with permissions to access files and your GPU. There may be an issue with your setup.
I know you have said playback sometimes works, during those working sessions are you direct playing the content?
Try forcing it to direct play and see if its successful even if playback buffers, then try forcing it to transcode by selecting a lower quality.
OK, I can take a closer look and reconfigure permissions.
As for the GPU, any advice on how to approach this?
All the previously reported problem files are currently behaving, so I have no way to continue testing at the moment.
I, however, got some new anime files in today and they show the typical ‘none-none’ I usually see with newly imported files.
I tried playing two episodes with the plex for windows app and got error code of numbers (didn’t note this set of numbers; not sure if the same or different).
I’ve included the PMS logs and a screenshot of the episode in plex (here).
Playing the anime episode a few seconds/minutes later (after I collected the logs), and on chrome, gives a new error code: s1001 (Network).
Let me know if android logs would be useful too.
Thanks.
I had a look at my permissions and made some corrections to the container’s PGID and PUID (1000,1000 - which are the IDs for the ubuntu user account running docker). The smb mount credentials were fine but I added a UID to the fstab (the mount credentials UID i.e. nas user uid), for safe measure.
Reset mount and the plex container, analyzed a few new problem files yesterday and tried to play the Boys episode 7 just now, which failed.
I also had a look at the /dev/dri permissions and based on this and reviewing the permissions from within the docker container, suggests the user ‘abc’ (the default user account managed by the docker image) has permission to /dev/dri via the group ‘video6ncc’ (this is for the directory ‘renderD128’, if that matters).
I am attaching logs here for your reference.
Could you please advise me on how best to diagnose/track this problem as well as verify that the previous issues have been resolved.
Please and thank you.
EDIT: For a former problematic EAC5.1 video with and without subtitles on, if I directplay it at original quality and then set the quality to 480p, the video still plays successfully.
So two weeks ago, I completely redid my network folder mounts and everything seems to be working smoothly.
The main difference is that this new mount is CIFS v3 as opposed to v2 with some CIFS options (like file and directory modes).
Hope this helps someone encountering similar problems.
1 Like
