Server Crash, Memory Full, transcoder error

Server Version#: 1.20.2.3370
Player Version#: 4.43.1

Can you help me ? I sometimes find myself with this kind of error, which fills my server RAM, which becomes unusable!

Error :

Sep 27, 2020 23:13:31.683 [0x7f37177fe700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-6jc3hze5s0mfjaerwiue9rf4-b5af3835-cf06-426d-964b-c3b5b39a07c1 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1

Installing in Docker - DSM 6.2.3

Look : Plex Media Server.5.log
I have to restart the docker for this to stop and the RAM to stop increasing.
On the plex dashboard, the RAM is not seen as used by Plex, but by the "system"Plex Media Server Logs_2020-09-28_16-45-34.zip (3,9 Mo)

King Regards,

Might want to see if your issue is similar to this thread: Memory leak since 1.15.x on 300+ Plex Server Instances The OP is also on Docker (could be different, but figured worth a look).

-Shark2k

Please forgive my basic question?

Where are /config and all the other Plex directories outside the container mapped?
Are they on NAS / network storage or is everything resident on the machine?

Thx for your reply @shark2k and @ChuckPa

all the folders are on disks of the machine hosting docker, no network disk

Server Version#: 1.20.4.3517

I still have the same problem.
It becomes very penalizing.
(File removed)

This only seems to happen with 1 user, reading through Plex Web (Chrome)

Are you 100% certain that file is correct and you’re not running out of space?

All 5 rollover logs are filled with the same thing:

Nov 02, 2020 11:44:54.929 [0x7fe35874c700] INFO - Plex Media Server v1.20.4.3517-ab5e1197c - Docker Docker Container (LinuxServer.io) x86_64 - build: linux-x86_64 debian - GMT 00:00
Nov 02, 2020 11:44:55.119 [0x7fe35874c700] INFO - Linux version: 4.4.59+, language: en-US
Nov 02, 2020 11:44:55.323 [0x7fe35874c700] INFO - Processor Intel(R) Core(TM) i5-8600K CPU @ 3.60GHz
Nov 02, 2020 11:44:55.323 [0x7fe35874c700] INFO - /usr/lib/plexmediaserver/Plex Media Server
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1
Nov 02, 2020 11:44:17.696 [0x7fe2e9ffb700] DEBUG - [TranscodeOutputStream] Changed instance plex-transcode-uyvbvg71ykoppku3m9r882wc-d93cd175-fad7-448b-b10a-856883f62544 which had to -1; was at chunk -1 with offset -1.000000 now at chunk -1

Might have just found a damaged file that FFMPEG/transcoder doesn’t know how to handle.

Thx for your reply

I’m not running out of space (free: ~ 800Go / used: ~ 80Go)

No, I’m not 100% sure the file is good, how can I be?

But this does not necessarily happen on the same le, but always with the same user.
I don’t have a problem with another user playing the same file.

Probably the easiest way is to copy the file in FFMPEG and change the container.

something like:

ffmpeg -i input_file.mp4  -c copy output_file.mkv

It will minimally force read the container info.
if there are major errors there, then the file is definitely in trouble.

The only other means is to play it

The file is obviously not involved. ffmpeg does not return an error and reading from my pc (via Firefox, or Chrome, Plex Web) works without problems.

Do you think there might be a problem with the chrome browser or the user’s pc causing me problem?

Yes, that’s the next logical piece of the puzzle.

Is there any way to find out what that person was doing - what was observed at time of failure?

For the trick – Can it be recreated ?
If so, would you care to make your log file retention longer so we might be able to see the start of the failure?

Not going to be super helpful, but just want to add/let you know I actually experienced a similar issue and it involved 1 user that was using Chrome and my memory would fill up and then Plex would basically crash. I took a slightly extreme route by using Tautulli to trigger a kill stream script when that user started playing something using Chrome (and I put a message saying there was an issue with the Chrome player and to download Plex for [their OS] and use that instead. That user hasn’t used Plex Web on Chrome since (as I can see when the script is triggered) and I haven’t had the issue. I realize it’s kind of extreme, but honestly for me, I would prefer my users to use Plex for [their OS] because it is going to handle more codecs vs Plex Web as well as it should be able to handle some transcoding on their side (I can’t confirm that just speculation on my end).

All that was to let you know that I have seen this issue before, but I “can’t” reproduce it any longer because I got the user to stop using Plex Web for Chrome. Also, I know that user has logged into their account for a couple family members so I don’t actually know who was using it during the crash.

-Shark2k

the user was playing a video lowering the quality to 2Mbps.
After the transcoder failure, the container restarts, and sometimes after several tries, the reading proceeds correctly.

This only seems to happen with the user’s computer, I was not able to reproduce it with my machine.

Also, it is not always possible to reproduce it, sometimes playing multiple episodes does not report bugs. (even when re-playing a video that previously caused a crash)

Yes

Thank you for your answer, indeed, the solution seems to be not to read via Plex Web and Chrome! But banning it via Tautulli is a bit “radical”, I would like not to come to that even if I will have to do it if there is no solution because regular crashes are very penalizing for other users.

If needed, could you help me set up the script?

thx.

If you would please stop Plex.
open a terminal session and navigate to where the Preferences.xml is stored

In the XML, add LogNumFiles="xx", where “xx” is the number to retain (suggest 20) before the closing />

It will look like this when done: SomePref="Value" LogNumFiles="20" />
Now start Plex.
it will retain 20 logs instead of the usual 5.

If it comes to it where you need to setup the script and you already have Tautulli installed, I will be glad to help you get the script up and running.

However, hopefully you will be able to get logs for Chuck that will help show the issue and allow it to be resolved.

-Shark2k

Now 5 days that I have set up in order to have more log, and the problem does not recur!

I remain vigilant, and will come back to you as soon as it occurs again.

@shark2k yes, i’m have Tautulli installed :wink:

1 Like

I don’t know if this is similar but I am having a memory issue as well. I am using Plex Version 1.20.4.3517 in a LXC container in Proxmox. I am having stead memory usage increase over time.

11/12/20 11:49 — 61.3% 2.45GB of 4GB of ram…
11/13/20 09:12 — 70.15% 2.81GB of 4GB ram…

Im not sure what logs you need. I used proxmox’s menu to see if memory is released, it might happen a little bit but it continues throughout the day. I do have a GPU pass-through, and I do use trans-coding while I DVR from my HDHomeRun. If I can provide any information to help diagnose the problem as well let me know.

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.