#Corrupted Plex DB

1 messages · Page 1 of 1 (latest)

native lichen
#

Hello folks. Unfortunately, the last week or so I’ve been having a constant corruption of my plex db.

I’ve tried several repairs and eventually decided to start over.

I had a ton of data on tracearr… of course the server was saying it couldn’t connect, and it didn’t give me options to edit it.

So I removed it… the warning said all session date will be retained. Unfortunately, all the watch data and everything else disappeared…

Is there a way for me to reinsert this data?

I’m happy to do it in SQL. Or I can restore from a backup from a couple days ago and do different steps

fervent oreBOT
# native lichen Hello folks. Unfortunately, the last week or so I’ve been having a constant cor...

Thanks for opening a support thread!
Please include the following so we can help:

Required Information

• Tracearr version
• Environment (Docker, LXC, Unraid, etc.)
• Docker compose file
• Logs (pastebin.com) or screenshots

Describe The Issue

• Expected behavior
• Actual behavior
• Steps to reproduce

Read The Docs

Docs · Logs · FAQ

native lichen
#

Tracearr Version: tracearr-supervised: latest
Environment: unRAID - Docker
Compose: n/a

finite sequoia
#

there is a known issue with the current latest plex update. they are working on a fix

#

can you be specific about what corruption you are seeing? is it data in tracearr, or in plex?

#

you removed the server from tracearr, and then re-added it?

#

did you perform a library sync?

#

lots of missing info here, please fill in the gaps

native lichen
#

Here's the corruption - it is NOT with the Tracearr db. Just plex

I removed the plex server from tracearr and then created a new server (because of the corruption) and added as new
library sync probably ran as a task already since I did this last night

#

The funny part is, when I was troubleshooting the SQL errors... I blew away just the com.plexapp.plugins.library.db. Then it came back with an integrity check of ok. So all good right? I readd the libraries, readd my users. About 6 hours later BOOM... there be corruption again.

And I know it isn't good because what happens is plex stops loading libraries. It doesn't full on crash, it just hangs indefinitely until I restart the container

#

My initial strategy was to go through the things that connect to plex... like Tracearr, like Pulsarr... and run tasks to see if plex freaks out and borks. Do simple things like add libraries, move things around, play around with labels.

I'm an IT guy... so I was trying various things to see if I can get to the bottom of the issue.

#

I did indeed notice the forum post, but the version mentioned wasn't the version I was on, so I ignored it.

#

Funny enough, I DID upgrade to the latest version and so far no corruption

finite sequoia
#

This really sucks, and im sorry it happened. but I'm not sure what we can do about it.

cedar summit
#

Did you downgrade?

native lichen
#

oh ya, I'm not looking for plex db help, that I'm working on. I'm looking to restore historical data from tracearr

cedar summit
#

You can not take data from tracearr and fix plex with it

#

if that is what you are asking

native lichen
#

no...

#

that's not what I'm asking

#

I just want the historical data within tracearr

cedar summit
#

its IN tracearr

native lichen
#

it isn't though, it's all gone 🙁

#

see... this is after I made a new database

#

the rest is all gone

cedar summit
#

What else did you do?

#

remove the server in tracearr?

native lichen
#

just removed the server and linked a new one

cedar summit
#

You deleted the data

native lichen
#

but the message says here that session data will be retained

#

that's why I did it

cedar summit
#

Retained in tables - not for your consumption

#

its a giant warning for a reason

native lichen
#

retained.. is a word I'd use to say that everything will stay, but new data from this server won't be added.

cedar summit
#

And lets read it together

#

Sessions

#

one tiny subset of data

#

Not the many many other tables needed to make it work

#

its not joinable to your new server

#

This is a "look before you leap" scenario

native lichen
#

right but that's not clear in the message.. it isn't clear that data will be removed... in fact, I think the message is misleading at best.

Anyways, I DO have a backup and can put tracearr back to before yesterday -- is there a way I can keep my historical data or do I just need to leave the old server in and add a new one as a secondary?

cedar summit
#

"All your shits gonna be gone" will be in the next pr

cedar summit
native lichen
#

Please, that would be helpful for the next dope 🙂

cedar summit
#

every id is going to be different in plex

#

for each time your DB corrupts

native lichen
#

so it'll be a mess is what you're saying

cedar summit
#

I dont think i said that

native lichen
#

SO FAR my db is good, so I think it is steady now

cedar summit
native lichen
cedar summit
#

if you didnt downgrade plex, you are cooked

native lichen
#

I'm on Version 4.159.0

#

released last night

#

that post was from before that, I believe

cedar summit
#

The post was at 7pm last night

native lichen
#

fair

cedar summit
#

Just use your head - if you are on the same version that borked your DB... fixing your db wont fix plex.

native lichen
#

no, I was on 4.143

cedar summit
#

and if plex hasnt come out with a patch that says "this fixes the issue that borks your db"

#

its probs not fixed

native lichen
#

now, I started a new db, on the already new version 4.159

#

I am having the same issue on Windows 11 and upgrading to v1.43.1.10576-06378. Rolling back to 1.43.0.10492 and it is able to launch.

#

ah, that's a different build number

cedar summit
#

what the hell is 4.159

#

v.1.43.1.10576-06378bdcd

#

that is a plex version number

native lichen
#

I was reading this, one sec

finite sequoia
#

thats the plex web client

native lichen
#

ya I just realized taht

#

oh goodie lol

finite sequoia
#

it mirrors plex, but its not plex

native lichen
#

but this isn't the version that corrupted my db

#

that was the prior version

finite sequoia
#

the string after the version is the git commit ref of the image

#

plex settings -> Settings -> General shows the server version

native lichen
cedar summit
#

@native lichen Please slow down, and stop leaning on us to do all the work for use

native lichen
#

ya same

cedar summit
#

that is the version.

#

This is so easy to find and confirm on your own.

native lichen
#

damn dude, I'm just chatting at this point

finite sequoia
#

i posted the plex forum link as my first reply too 🙂

native lichen
#

yes, and I read it

finite sequoia
#

in all fairness, i did not read it. but im pretty sure you shouldnt be doing surgery on your plex db

#

maybe im wrong and the plex staff said to do it

native lichen
#

I wasn't, I was doing integrity checks

#

because the plex server kept hanging

#

they have a kb on it and what to use

finite sequoia
#

im just spit balling and not helping, sorry 🙂

native lichen
#

I've had plex for 13 years, I've never had a corrupt db before lol so it isn't like I've seen this in the past.

#

so far, no issues.. and it looks like based on other forum responders that this is something that occured during the upgrade process. And that their servers wouldn't start at all

#

so while might still be related, is different symptoms than what I was experiencing.

#

and mine occured on the previous version, not on this latest version

#

Anyways, I guess I'll start fresh with Tracearr and hope for the best for a stable release (if that's actually the issue)