You seem to be having a similar issue to @Indian9990 with the database being locked up.
Also debug logging is not enabled on your system, so I cannot determine much more.
Feel free to DM me your database file so I can investigate.
For anyone posting here without server logs, Iām just going to ignore it.
Also that volume you think is running up against āfile watch limitsā (which is not a term Google thinks exists; sounds like something you made up) has a terabyte of free space on it. So more incompetence.
On Linux itās the inode limit (file watch is what we refer to this process internally).
You should find some info on the forums and general web searches on how to change this limit.
Not sure how staff are not seeing this as an issue. There are other forums posts with the same behavior. Thereās no way everybody changed/messed up their environment to break the agents. 1.41.5.9522-a96edc606-x86_64 is the last server version where the agents are matching content properly. Every newer version of plex media server Iāve tried breaks the agents ability to match content.
So far Iāve seen only two logs showing an issue (thatās very rare and being investigated).
The other logs have shown other underlying issues causing the database to get locked up and causing the agents not being able to update in the database.
The more logs I have to check the better to see if there are any other causes.
I have so far not had any reports from anyone internally with this issue.
What a rude arrogant response. And you clearly donāt know what you are talking about about nor do you know how to use Google to search.
2 Likes
Iām not staff, but I am running two PMS instances in docker on Synology and Ubuntu 22.04, both have been fine on 1.41.6.9685 and now 1.41.7.9799. I have been adding a lot of media with no issue. Iām not discrediting what youāre saying, but thereās probably a small variable that is effecting some that isnāt effecting everyone.
Iāll grab logs from both 1.41.5.9522-a96edc606-x86_64 and 1.41.7.9799-5bce000f7-x86_64 (after upgrading) and see if the employee can find a difference. I made no changes to my environment besides upgrading the server.
1 Like
Here the logs from the working server (1.41.5.9522) versus the non-working server 1.41.7.9799. Iāll stay on 1.41.7.9799 for the time being. Let me know if anyone requires more logs. The agents are currently not matching any content.
Uploading: Plex Media Server Logs_2025-05-21_10-21-43 (1.41.5.9522-a96edc606-x86_64).zipā¦
Plex Media Server Logs_2025-05-21_10-21-43 (1.41.5.9522-a96edc606-x86_64).zip (605.6 KB)
Plex Media Server Logs_2025-05-21_10-30-00 (1.41.7.9799-5bce000f7-x86_64).zip (778.8 KB)
There you are. Please donāt attack people directly on forums. No matter if you feel youāre correct or not.
2 Likes
My console wonāt stop with the logs after upgrading to 1.41.7.9799-5bce000f7-x86_64 versus being empty or a few logs here and there with 1.41.5.9522-a96edc606-x86_64.
Naw, heās doing his job. The rude ignorant one is @tubesockfan
2 Likes
Thanks thatās really useful having these logs.
Are you using a legacy agent for the search in your screenshot?
Iāve noticed an error in your log that comes up sometimes with Windows users where PMS just cannot communicate with the Python stack for some reason. Iāve seen this a few times over the years and have never figured out what causes it, and it usually just starts working randomly again.
The lines related to that issue in your log:
May 21, 2025 10:29:34.390 [18548] ERROR - [Req#3] Timeout or error reading status line from plug-in pipe [com.plexapp.agents.imdb], we're killing it.
May 21, 2025 10:29:34.390 [18548] DEBUG - [Req#3] Stopping [com.plexapp.agents.imdb].
May 21, 2025 10:29:34.390 [18548] DEBUG - [Req#3] [com.plexapp.agents.imdb] Killing.
May 21, 2025 10:29:34.390 [18548] ERROR - [Req#3] Error received reading configuration for com.plexapp.agents.imdb
This would break the legacy agents but the default Plex agents seem to be working fine though and they loaded up as I would expect:
May 21, 2025 10:26:44.220 [18544] DEBUG - [MediaProviderManager/MetadataAgentManager] [tv.plex.agents.movie] provider added to map
May 21, 2025 10:26:44.220 [18544] DEBUG - [MediaProviderManager/MetadataAgentManager] [tv.plex.agents.series] provider added to map
May 21, 2025 10:26:44.220 [18544] DEBUG - [MediaProviderManager/MetadataAgentManager] [tv.plex.agents.music] provider added to map
May 21, 2025 10:26:44.220 [18544] DEBUG - [MediaProviderManager/MetadataAgentManager] [org.musicbrainz.agents.music] provider added to map
So those agents should work fine.
In your previous log (still on 1.41.7) I did see that you also had a slow database during start up and it caused the same time out I saw in 2 of the other logs posted.
May 21, 2025 10:26:08.951 [11924] ERROR - [MediaProviderManager/MetadataAgentManager] Waited over 10 seconds for a busy database; giving up.
Iām going to try and make this process more resilient so that we retry more often after failures like this. It should help with cases like yours.
Just to update you all and know that somebody is looking into this, so far Iāve come across 3 or 4 different causes for the agents to not find any matches.
- Database issues. The database is slow and/or there is a separate issue during startup causing the database to lock up and not be able to update the metadata agents tables.
- The request to our metadata backend services is timing out (408 errors). I cannot tell whatās causing this. You may have been really unlucky and hit a service outage exactly when you were starting your server, or there is something causing an issue on your local internet connection.
- A very rare deadlock condition. Iāve seen 2 logs hinting at this happening. I have not been successful in reproducing this issue but it is being looked into and Iāll probably have a custom PMS build for those affected to try out next week.
- Issues with the legacy agents - as these are no longer being maintained Iām unfortunately not going to include this in this investigation. There was a known issue where it was completely broken in one beta build but that has been fixed. Other than that, please test with the default Plex agents only if reporting issues here.
For points 1 and 2, Iām going to make some changes to make the process of discovering the agents more resilient, this should hopefully allow things to work correctly even when there are very long database locks (you may just have to wait a bit longer).
As for point 3 Iāve said Iāll try to provide a test build to some folks next week with some changes that should help that.
Feel free to post more logs if youāre seeing issues using the default Plex agents.
3 Likes
No iām using the new agent. As I mentioned before, everything works as expected if I downgrade to 1.41.5.9522-a96edc606-x86_64 so iām not sure what the issue is. The server will still work fine but I just canāt match any content with the agents. I have database logs if youād like to take a look. Let me know and iāll upload them.
Thanks.
I just want a log from a clean PMS start up, wait a couple minutes, then try match/fix match something and then grab the log.
Your latest log had the fresh start up but not the match/fix match request.
Sure, give me a sec. Iāll reboot my server and grab fresh logs.
The logs have been attached.
Plex Media Server Logs_2025-05-21_12-55-01 (1.41.7.9799-5bce000f7-x86_64 After A Reboot).zip (2.1 MB)
I am happy to send ya the database file via DM. I tried to message you but looks like youāre not accepting DMs ATM.
