Credit to @Raazer for narrowing the issue down, and everyone else involved for confirmations. 
1 Like
I was having this issue before going to 64bit version, I deleted my music library and it seemed to go away for a few weeks and then come back after rebuilding the library. I found a Youtube video that showed how to rebuild corrupt libraries and that worked for awhile the trouble went away till the build .6483. I deleted the music library again and the trouble went away. Ill try this new temporary fix of turning off the Music Analysis…
Thanks @AmazingRando24 - Just standard troubleshooting after using Plex for 10 years now.
2 Likes
thanks to all who have contributed to this issue.
turned off loudness analyze and the server is stable now. no crash in the last 24h
But still I have for plex guys a question not answered: how can I perform a firmware downgrade???
1 Like
As far as I know, there’s no straightforward way to downgrade. It’s been covered in this forum before, so try poking around here.
Can confirm that the server is still running strong with “Analyze audio tracks for loudness” disabled. Everything else is set default.
@terierul on Synology you can downgrade by uninstalling the app in Package Center and manually installing the version you want. It won’t remove any of the Plex data, I just tested it a few days ago and it worked perfectly for me.
You can find a list of archived download links for previous Plex versions here (just choose the file that matches your device/OS/CPU architecture).
Here are the full instructions from the Synology FAQs:
I still have the issue after setting Music Analyze to Never… The only way I am able to fix it is to delete Music library from Plex. Any suggestions?
This can be easily tested and reproduced by manually triggering the LoudnessAnalysis butler task with Python PlexAPI. If anyone wants a basic script to blow up their *nix based system, here you go. Drop that code into a file ending with .py, update your plex token in the code, make it executable (chmod +x), and run the python script. If you are on 1.31, you should have now successfully crashed your server!
Dump attached.
12-24-2022.dmp.zip (74.9 KB)
For future debugging, there isn’t any need to wait on butler tasks to run over time as scheduled. You can manually test trigger them one by one with PlexAPI allowing for much quicker diagnosis.
Same issue here with v 1.30.1.6497 on Synology 6 (native package, not docker). I’ve disabled loudness analysis for now since from my logs it appears that’s been causing the crash:
Dec 24, 2022 06:58:58.392 [0x7f048be1db38] ERROR - LoudnessJobManager: job unable to complete in allotted time, killing
Dec 24, 2022 06:58:58.392 [0x7f048be1db38] DEBUG - Killing job.
Dec 24, 2022 06:58:58.392 [0x7f048be1db38] DEBUG - Signalling job ID 22300 with 9
Dec 24, 2022 06:58:58.399 [0x7f04944fab38] DEBUG - Jobs: '/volume1/@appstore/Plex Media Server/Plex Media Scanner' exit code for process 22300 is -9 (signal: Killed)
Dec 24, 2022 06:58:58.399 [0x7f048be1db38] ERROR - LoudnessJobManager: job for item 202122 failed (-9)
Dec 24, 2022 06:58:58.400 [0x7f048be1db38] DEBUG - Activity: Ended activity 0d12f17f-7f20-4f5d-a381-4e6894ba59e5.
Dec 24, 2022 06:58:58.401 [0x7f048be1db38] DEBUG - Activity: ended sub-activity 7cbef811-1324-4199-b88f-a12ade4e2eca parent: daea3611-9cc4-4c0c-bf40-1c0864d77094 overall progress: 100.0% (1188/1188)
Dec 24, 2022 06:58:58.401 [0x7f048be1db38] DEBUG - Activity: updated activity daea3611-9cc4-4c0c-bf40-1c0864d77094 - completed 100.0% - Butler tasks
Dec 24, 2022 06:58:58.402 [0x7f048be1db38] DEBUG - Activity: registered new sub-activity faec9189-d25a-4400-be57-9360cbfedce3 - "LoudnessAnalysis" parent: daea3611-9cc4-4c0c-bf40-1c0864d77094 overall progress: 99.9% (1188/1189)
Dec 24, 2022 06:58:58.402 [0x7f048be1db38] DEBUG - Activity: updated activity daea3611-9cc4-4c0c-bf40-1c0864d77094 - completed 99.9% - Butler tasks
Dec 24, 2022 06:58:58.403 [0x7f048be1db38] DEBUG - Butler: Scheduling loudness analysis for album 206497.
Dec 24, 2022 06:58:58.653 [0x7f048be1db38] WARN - Crash: Crash reporting disabled
Hoping this gets resolved soon, I want to get a ‘clean’ system and backup again before I try to migrate to Synology DSM 7.
Depending on your OS, the Plex service should restart on its own after a crash (ie systemd). However, if it doesn’t come back up you can use basic python watchdog script to check that Plex is up and restart the service if it is down. Here is an example for *nix based servers. You can set it to run at your desired intervals with cron.
Same issue here. Posted on another thread and someone linked this one. Disabled the loudness check and Plex server has remained up and running overnight.
We have a problem with the database ending up with missing entries for some music albums leading to a crash when loudness analysis is triggered for the album = the issue arises when the track entries exist in the database but the album entry is not present. The crash itself will be fixed in Plex Media Server 1.30.2 which is currently in alpha
Anyone getting these crashes immediately after logging this line in the Plex Media Server log file
Butler: Scheduling loudness analysis for album can send me a zip of their PMS database to see when the albums with the missing db entries were created - to try and help find the issue started
DB zip can be downloaded through Plex Web Troubleshooting server setting interface
1 Like
Sorry, the file you are trying to upload is too big (maximum size is 30 MB).
Zip of DB is 10x too large to upload and send 
forum has limits - would need to be uploaded to google drive / dropbox or other similar services and link sent to me by private message
PM Sent
Thank you. I now need the debug server logs for one of the crashes so I can look at the database and see what data is involved in the crash
Attached.
Plex Media Server Logs_2022-12-29_08-20-50.zip (365.5 KB)
The database I have does not match what is in the logs
You have 2 servers. I suspect you sent me the database from the other server
The one that I have the logs for is the Over The Top one
The one that I have the database from is the other server
Can you send me the database for the Over The Top server and if the other server is also crashing - then debug logs after a crash from it plus any dmp files
