Plex Using 100+ GB of RAM during Schedule Tasks

No thanks! I rolled back to plex-media-server-1-21-3-4021 and have not had a single issue. I do appreciate you asking, because I also want to know if they fixed the crashing issues.

I uninstalled the newer plex media servers and installed plex-media-server-1-21-3-4021 and have not had a single issue since going rolling back to this version. I had tried rolling back to other versions, but still had issues. One issue with another suggested older version is my newly added TV shows weren’t being found. This version I suggested has had no issues with crashing or scanning for a week now.

https://forums.plex.tv/t/plex-media-server/30447/479
https://forums.plex.tv/t/plex-media-server/30447/480
Two updates in the past week and I’m still crashing during scheduled tasks.

I’ve been considering options. I’m reluctant to just do a fresh install because of all the manually metadata changes I’ve made over the years. I’ve been considering rolling back to a stable version for myself and may still just to have some sanity again.

I was curious though if anyone had any suggestions for metadata export tools (1st or 3rd party); the idea being that if I can find a way to store all my hard work locally, then at least if I can run a fresh install, it shouldn’t be quite the same headache as starting completely over. I’ve googled around and have found a bunch of different export tools, but I wasn’t sure which to use, hence my asking if anyone has any suggested ones that they’ve had experience with.

It’s a thought that I was curious about, so I just wanted to throw it out there and see what the possibilities were.

I saw that Plex Media Server 1.25.4.5426 was released on the beta channel and touted a bunch of fixes for databases and the like, so I excitedly decided to give it a try in hopes that it helped with the issues I’ve been experiencing here.

Alas, no. It helps nothing. I’m still crashing everyday during scheduled tasks from a burst of increased RAM usage. I’m still waiting for a fix to the issue. Unfortunately, my server host is killing my Plex instance before a crash report can be generated (out of fairness for the other users on the server). So all I can offer at this time is my logs.
Plex Media Server Logs_2022-01-24_14-13-56.zip (4.4 MB)

I’ve still been holding out for an update fix to this issue because I haven’t been able to find a workable solution to save and reimplement the metadata on my server. I have a significant amount of shows and movies that I’ve either had to correct a mismatch from Plex on or manually supply the metadata and posters for. I really don’t want to have to lose that and start over. I don’t want to forget which ones had these particular concerns and have them just become lost in the database that Plex would attempt to recreate from scratch. I also have concerns of the many different playlists created that are attached to my server losing everything. Again, so much of this has come down to all of the time and effort put into maintaining this information over the past several years. I’m being stubborn in giving those up simply because Plex implemented a database update that screwing things up for on my server host that they have yet to fix.

This has been incredibly frustrating as it feels like I’ve been left out to the wolves to fend for myself and the preservation for my server by the Plex engineers. I’ve been using Plex for nearly a decade at this point and solidified my relationship with the software when I purchased my Plex Pass Lifetime subscription back in 2017. While there have been many issues over the years varying in degrees of frustration, this has been the only time that I’ve experience something that I can’t just out right ignore and move on from. I’m becoming a bit desperate as I’m essentially begging for any help/assistance in correcting this issue so that I can continue forward with the server I first set up many years ago and not just have to throw it all away and start over. Please help.

We’re on the same boat, buddy. Curious to see if/when they will do something about it. Uninstalling my PMS and installing it again is completely out of question since I have tons of manually inputed metadata. :worried:

So I’ve been reading in other posts in the forum and folks have been saying just keep optimizing your database in the newest version in the beta channel and you should see the size of your database go down and the crashing should stop.

The problem for me has been that I’ve been optimizing in the background for the past couple days and I’ve seen no change in my database size and it’s still crashing during scheduled tasks. I’m at a loss.

@ChuckPa do you happen to have any thoughts or suggestions with this?

I decided to give the new update a try and I get crashing every single night again. Why can’t Plex fix this? This use to not be an issue years ago. I have had crashing issues for the past 2 years now.

I am about to reinstall windows and plex, but I also hate the idea of losing out on my fixed match movies. I would have no idea which ones would need correcting.

1 Like

Another update in the beta channel. Another attempt made to see if this fixes things. Short answer… nope. It crashed when installing/updating. Restarting Plex showed that the server was reflecting the new version. So again, I attempted optimizing database to reduce its size manually in hopes that it would finally end the awful experience I’ve been having lately with Plex. Nope. My database continues to show no change in size as I’ve endlessly optimize it over the past few days. It had still crashed during schedule tasks last night. We’ll see if it happens again tonight. At this point my belief is that the updates continue to fix nothing with the particular issue I’m facing.

Again, trying to fight against the feelings of being ignored and abandoned by Plex in the void that these forums have seemed to become. I’m greatly frustrated and disappointed.
Plex Media Server Logs_2022-01-27_00-34-14.zip (2.5 MB)

Update: it crashed again during scheduled tasks

I have no problems with plex ebuilds from here on Gentoo:

What does your ebuild look like?

This has been happening for years to me as well, I had it on auto pay for the yearly renew, and I should have just bought the one time right away, so I have paid hundred of dollars for this ā€œserviceā€ that just constantly crashes. I mean it goes in waves, but today it crashed 10 times, I cannot even keep it up for 5 minutes. I don’t understand how they expect this service to be paid for at this point. I know they will want logs and of course its my servers fault, but I look at all the people having issues and it’s all platforms. I am on my last nerve here, I just don’t see a solution. Do I fully start over with a fresh install and rescan all files, but then when I think of that, it’s like what’s the point, I am sure it will just start to crash again.

It’s too bad that they don’t support their loyal customers. Just keep putting out trash update and after update.

Sorry for being negative but after years, I don’t see any positivity.

1 Like

What platform ?

Edit: the size of the db’s is not the issue. sqlite is good for way larger DB’s…

See #14 here Implementation Limits For SQLite

Thanks to help from @Gloppie, I’ve learned that my server host whatbox uses OpenRC, and installs the RPM versions of Plex. I’ll be sure to include that information with future posts with logs.

I just saw this post Recurring database corruption on Ubuntu Server have you upgraded to the new plex metadata agents?

For the Dev’s looking into this, I’ve recently found in the logs when optimizing the db:

Jan 27, 2022 17:55:16.214 [0x7f1384331b38] WARN - [Database optimization/com.plexapp.plugins.library.db/DatabaseFixups] Held transaction for too long (/data/jenkins/server/3534426338/Library/DatabaseFixups.cpp:217): 0.310000 seconds
Jan 27, 2022 17:55:16.567 [0x7f1384331b38] WARN - [Database optimization/com.plexapp.plugins.library.db/DatabaseFixups] Held transaction for too long (/data/jenkins/server/3534426338/Library/DatabaseFixups.cpp:375): 0.310000 seconds
Jan 27, 2022 17:55:38.400 [0x7f1384331b38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:58): 22.530000 seconds  
Jan 27, 2022 17:55:42.660 [0x7f1384331b38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:71): 3.800000 seconds   
Jan 27, 2022 17:55:50.079 [0x7f1384331b38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:81): 7.550000 seconds   
Jan 27, 2022 17:55:54.189 [0x7f1384331b38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:94): 4.060000 seconds

And…

INFO - [Database optimization/com.plexapp.plugins.library.db] SQLITE3:0x80000001, 17, statement aborts at 57: [select * from metadata_items limit 1] database schema has changed

I change my individual library settings to the new Plex meta agents when they came out months ago. Do I need to do a full scan of my library though with the new agent?

That I don’t know. Maybe try it and report back?

This is what I have in my log with optimization

Jan 28, 2022 01:11:35.804 [0x7efc2d73eb38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:58): 61.330000 seconds
Jan 28, 2022 01:11:49.183 [0x7efc2d73eb38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:71): 12.030000 seconds
Jan 28, 2022 01:12:06.113 [0x7efc2d73eb38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:81): 16.660000 seconds
Jan 28, 2022 01:12:14.178 [0x7efc2d73eb38] WARN - [Database optimization/com.plexapp.plugins.library.db] Held transaction for too long (/data/jenkins/server/3534426338/Library/FullTextSearch.cpp:94): 7.840000 seconds

I’ll give a shot. It’s going to take a while though.

sigh so I refreshed all the libraries and nothing changed. It still crashed overnight during scheduled tasks.

update: I take it back; it didn’t finish refreshing the metadata when it crashed. So I’m refreshing bit by bit and keeping an eye on it. It still looks like this’ll take a while.

update 2 (1/30/22):
so I went through and refreshed the metadata on every library after double checking and make sure that all the current megadata agents were being used. Afterwards, I cleaned bundles, optimized, and emptied the trash. Sadly, it’s still crashing during scheduled tasks for the moment.

This SUCKS.

I went back to an older version and while it never crashed it wouldn’t recognize any new TV shows I added. It recognized any new movies I added though.

So then I changed to another version someone said had no issues and it found the new TV shows no problem, but it won’t allow enabling of Remote Access.

Come on man! And of course the latest version has constant crashing. I am so sick of this crap. I agree with another poster that I am not usually negative. This is so frustrating. How can they not make a stable version by now?