Thank you. I will have a look at these steps.
Regarding the J: Drive, it’s an external (USB) SSD. I would not be shocked if the issue is related to it in some way because my issues started around the time I moved the Plex Folder to the J drive and updated the Reg.
The F: Drive is an external (USB) HDD. I’ve done read/write speed tests on the drives, and they should be plenty fast to handle things. I will send a PM and get you the stuff you requested. Thanks so much!
I have this same issue - Plex running on Windows Server 22 randomly crashing. The icon remains in the task bar but when connecting to the localhost website it never loads. Exiting / reloading fixes the issue.
Used to work just fine ~2 months ago and have been battling with this ever since.
Running on a r730 Poweredge -
OS: ESXi → Windows Server 22
Updated to Version 1.28.0.5999 this morning. Not holding my breath. From reading here it seems logs are useless in diagnostic?
In the meantime I am going to create a task to kill / start the Plex executable daily…
For what it is worth, I have been running 1.28.0.5999 since 8/1 with no deadlocks yet. Logs are not useless to diagnose a deadlock, but additional steps are needed to provide the devs with all of the info that they need to see what is going on. Logs are just one piece of the puzzle; you would also need to provide the connections text file and the process dump for Plex Media Server. So if you do experience a deadlock on the latest version, here is the way to get all of the diagnostics that the Plex team needs to see what is going on:
Figured I would mention the process again since it is so far up in this thread.
I have uninstalled Plex and installed it again a few times, went back and tried with 3 different builds, optimized etc and its now running again without locking up but when it eventually does freeze up ill grab connections, dump, logs etc and add them here.
The file that was being transcoded at the time was
F:\Stuff\TV\Fargo\Fargo S01-S04 web 10bit ddp hevc-d3g\Fargo S01\Fargo.S01E10.WEBRip.10Bit.1080p.DDP5.1.H265-d3g.mkv
Are there any subtitles ?
Is the issue easily reproducible ?
Was it with Plex Web ?
The transcoder files were on the J Drive
As mentioned we need to establish if it was a lockout / hang or performance / speed issue on the J Drive that led to requests being repeated and building up
Could be to do with photos scanning. Not sure
Could you look for files in %TEMP% named as follows:
836efb6e-85f5-447a-a69b-58957b7a48af
5fb034e5-9c19-449f-8665-9644b61e4244
27ac58f5-cb8f-4ac2-9c39-09ad22584b3f
If you find them please zip the files and attach the zip file
and please when providing logs - do not attach individual log files - but a zip file. In fact you can download a logs zip file after restarting following a crash through the Plex Web interface for downloading server logs
Could you also enable memory logging in Plex Media Server. See the regedit outlined on this article for LogMemoryUse
Advanced, Hidden Server Settings | Plex Support
Hey there thanks for the response. I’ve been really impressed with your responsiveness in this thread. I was out of town over the weekend and didn’t get a chance to run the deadlock diagnostic…
Some context: before the 1.28.0.5999 update, this sort of crash / deadlock (server unreachable, but the app itself not closing, and remaining in the tray) was occurring about once a day. 48 hours was the longest it went between crashes/deadlocks during the 3 week period when it was frequently occurring. Since that update, this was the only time that it happened. I am starting to think that it might have been a separate one-off issue. If it happens again, I will go through the deadlock diagnostic steps outline above and follow up with you. To answer your questions:
Are there any subtitles ? Yes. I was actually making edits to SRT subtitles at the time of the crash because they weren’t aligned.
Is the issue easily reproducible ? No. Since the update, this is the only time it has occured. Before the update, it was happening at roughly the same time every/most nights (usually not scheduled task hours).
Was it with Plex Web ? Yes, because I was testing something, I was just on Web when this happened.
The transcoder files were on the J Drive
As mentioned we need to establish if it was a lockout / hang or performance / speed issue on the J Drive that led to requests being repeated and building up.
If it happens again, this will be my next step. I have also been rebuilding my thumbnail cache. When I moved my Plex folder to my J: drive, I deleted some thumbs on accident. Another theory that I have (based on what I’ve seen in logs) is that my crashing was somehow tied to missing thumbs. But I also don’t really know what I’m talking about, so maybe that’s nonsense.
I wonder if this locked the files and triggered the issue
Anyway - see if you can reproduce the problem and if it can then may be you can get it to fail with a smaller sample file and srt file
Yes it is different. The deadlocks / crashes due to the compiler bug that we got round was masking other issues - or at least happening much more frequently not giving other issues the chance to surface !
I’m so glad I found this thread. I just want to say that it’s happening to me A LOT lately. Ever since I migrated my server to a brand new M1 Mac Mini, Plex regularly becomes unresponsive. It’s not a “crash” as the Plex icon is still in the Menu Bar, but no client can connect to the server. I need to close and reopen the Plex server manually.
I have a hard time reproducing the issue at will. Sometimes it happens overnight while nothing is accessing the server, sometimes it happens while I’m in the middle of editing metadata using the web client.
To give more details, before the server migration I was running version 1.25.9.5721 on an Intel Macbook and never encountered this kind of issue. But on the M1 Mac Mini it’s happening a lot. The version I installed was 1.28.0.5999. I am now running the 1.28.1.6041 beta, and it’s still happening.
I just enabled debug logging, do you still need logs? What else would you need to help resolve the issue?
Ok so its started happening again after a few days, locking up every few minutes or so.
I have grabbed all of the logs, dump, connections etc and sent them to sa2000 through wetransfer.
The starting point for investigating this is to get diagnostics.
If the server process does not crash and remains running but unresponsiveness, then we can treat this as a dealock and get diagnostics for a deadlock - which would be:
Keep this value of the token until you get a deadlock/hang.
Then on the server itself, open a browser and type this url
http://127.0.0.1:32400/connections?X-Plex-Token=xxxxxxxxxxxx
where xxxxxxxxxxxx is the token string
Copy out the browser response text into a text file and save it to add to the provided diagnostics
3, Restart the server and download the server logs
Send me zip of the process dump and list of connections by private message and attach the logs zip. If the dump zip is too big for the forum posts, please upload to google drive / dropbox / etc and send me a link
If you are getting crashes as opposed to deadlocks, then just initially attach the debug server logs. You can also look in these paths for any plex related crash reports
~/Library/Logs/CrashReporter/
~/Library/Logs/DiagnosticReports/
Reinstalled yesterday, restarted etc multiple times, back to normal for about 18 hours and now locking up again after a few minutes each time, I have uninstalled and installed different builds over the past two hours and nothing will stick this time, constant locking up.
How can this not be fixed yet? Lots of logs have been submitted by people but this issue is still happening.
As someone else said earlier on, we pay for this software to work and right now it isnt, this needs to be sorted asap.
It may not be all caused by same issue. We have already fixed one cause in 1.28.0.5999 and the 1.28.1 beta
A fix is being worked on now for a bug that we found from diagnostics from @Malcolm_John
Your lockout looks different and looks like some performance issue where requests were taking a long time to complete and queue starts building up
Is the C Drive an HDD / SDD ? anything special in the setup?
Any disc related events in the Windows System Event log (Event Viewer) ?
Is the plex media server database large ? - the database files are here
C:\Users\Plex\AppData\Local\Plex Media Server\Plug-in Support\Databases\com.plexapp.plugins.library.db
C:\Users\Plex\AppData\Local\Plex Media Server\Plug-in Support\Databases\com.plexapp.plugins.library.db-shm
C:\Users\Plex\AppData\Local\Plex Media Server\Plug-in Support\Databases\com.plexapp.plugins.library.db-wal
I will send you a build with the fix for the problem we discovered this week and then wait for new occurrence of the problem and get me fresh logs and dump and connections list and also zipped copy of the database files
Thanks for this info, I did not get my token prior to another deadlock so unfortunately i have to wait again. New version, 1.28.0.5999 still crashes. Server recently updated, freshly rebooted 24 hours prior to this.
Today I have updated my Windows Server 22 instance of PMS to 1.28.1.6041-738907df3
For example when it crashes I cannot access the server web frontend. I need to also kill the task.
This could be a deadlock. In this state, request for the connections list should hopefully get processed - it would be
http://127.0.0.1:32400/connections?X-Plex-Token=xxxxxxx
I suggest waiting for availability of a Plex Media Server build with the fix for the deadlock bug discovered this week and then see if we have any remaining issues and get diagnostics then - connections list, process dump, debug server logs
Please Media Server 1.28.2 has just been released as a beta release
It has a fix for a deadlocks issue that was identified last week
Hi, i update to The last versión and The server continue downing and disspaear the tray icon, i must start again to work again. This happen about The same time every days . What could be ? This happened whit all versions, when i update dont happen every day but few days later appear The problem again.
Please make sure you have debug logging enabled on the server - see https://support.plex.tv/articles/201643703-reporting-issues-with-plex-media-server/
and after next crash please restart and downloads the server logs and attach the zip here
Hi, about 1:48 today 27/08/2022 , again down and dissapear the tray icon . I send The logs, Hope be usefull , i can take them again if you need, thank you
Latest versión, 1.28.2.6106 Windows 10 x64
Thanks for the logs
The logs show that we stopped adding to the log file at 01:38:55 and that you re-launched Plex Media Server at 01:55:59 and when it launched there were a number of running Plex Transcoder processes from the earlier period
There is no evidence of a crash detected by Plex Media Server crash reporting.
Could you check the windows event log through Event Viewer and select the Windows Application Event Log and see what Program Error events got logged between 01:38 am and 01:50 am and paste here the error data
Probably need to enable the settings in windows to get Windows to capture the crash dump - see the registry edit outlined here Crash Logs: Plex Media Server | Plex Support and also create a directory named “CrashDumps” below C:\Users\Xtrem3\AppData\Local\ and then we wait for next crash and as before capture the logs but also look for any dmp files within
C:\Users\Xtrem3\AppData\Local\CrashDumps and zip what you find for Plex Media Server and upload on dropbox or google drive etc and send me a link privately. Also check the Windows Application Event Log and see what errors get logged
Thanks
