Server Version#: 4.138.0
Player Version#:all
Today 2 of my libraries are empty.
i checked my NAS, and all the files are there.
i checked where the library was pointing to in plex, and everything was good.
but for some reason TV Shows just says “something went wrong” on recommended tab.
on other devices it says the library is empty. but if i go to categories, all of the categories show up. some of them have pictures of the shows.
if i scan library, it scans though everything. i see it in the logs.
it just cant display any of it.
i had a user somehow continue playing an episode on accident, but when they backed out, it was gone.
this is the weirdest issue i have had because movies, and anime all work perfectly fine.
steps i took to try and fix it:
i updated plex.
i restarted plex.
i restarted the headless machine that runs plex
i checked the NAS and everything is healthy.
i have not rebooted the NAS yet.
this feels like a simple glitch, and i can restore from backup, but i dont want to do that until my back is against the wall.
anyone have any ideas?
so as soon as i did that, it imediatly did not work. it has the same “something went wrong” on the new database. its still scanning and building a collection. but i cant see the library. Plex Media Server Logs_2024-09-14_07-23-57.zip (5.3 MB)
Turn off IPv6 support. You have an IPv4 LAN . It’s complaining at the IPv6
Sep 13, 2024 22:45:20.389 [22566883785528] WARN - [CERT] TLS connection from [::ffff:192.168.29.247]:65093 came in with unrecognized plex.direct SNI name ‘192-168-29-171.4805ebe33f794685afb7cbd8c22f65b4.plex.direct’; using installed plex.direct cert
There are no other errors I can see anywhere in your logs.
i do not know what js blocker is, but the machine is headless, and the library is not working across all devises and users.
but all this is irrelevant now. after adding tv shows 2, the whole system became unhealthy.
i had a docker called APPdata backup that was running for 3 months, and backing up every month, but i guess what it was backing up was just some config files and not actually my data. so the backup did nothing, and now im going to have to rebuild the entire database from scratch again.
i dont know enough about this headless machine to do any tech support on it. so this will be the 6th “burn is all down and start over”
because i cant get into plex to make any changes. there is nothing more i can do.
oh, i had all the directions correct. but the backups worked way to fast. could not figure out why it only took a second to restore.
i had it pointing right at the folder. it just seems to have not backed up the actual files for some reason.
here is the backup information that should direct it what to back up. “hail mary” was the plex server location
here is the plex internal data. i can always go back to it if there is a way to fix it, but at this point i have just decided to rebuild the whole thing again from scratch.
I keep waiting for the and to walk by!
(yes, that rabbit hole is FAR too tempting lol)
Anyway.
peeling back the colorful names ,
Where is the container’s /config in your real filesystem?
That’s the directory you want to backup
You do NOT want any part of Plex’s internals , including transcoder temp or /tmp, to be on a remote server (file locking does not work and will corrupt)
I urge you to be a bit more constructive with CamelCaseNaming and avoid spaces wherever possible. I don’t know how reliable it is throughout the Unraid system. It should be all linux and not care
NOTE, I do not know how you properly list multiple sources in the backup.
(where you have “hail mary” and “Plex”. Is a comma needed ?? I never listed multiple directories per field.
For what I see here:
The changes you likely first need are container definition
Then, Knowing where Preferences.xml is, You can reset it.
Lastly, you can use my DBRepair tool to help recover database fron backups PMS keeps.
as i was just told one day to get the headless machine and hardware from someone who then ghosted me, i have had to figure all this out on my own… with no instruction.
so my understanding of what im doing is laughable at best. and anything that is the way it is, is held together with dried blood and finger nails.
i really appreciate the help though. i did not expect a response.
all the creative naming is from really late nights with no options, praying something will work. and when it does, i leave it alone.
this is my file structure.
this is the master folder where i operate everything. it is the plex database as far as i know.
i do not know where a config file is. this folder does not contain any media.
i assume if i had a exact copy of this stuff, and just dropped it over a corrupted version, it would just be fixed.
thats the “fixing” i have done in the past. sometimes it works. sometimes it does not.
my nas is linked in another old old old version of plex folder. setting up the “this is where the share is” was very tricky and i eventually got it working, so im never touching it again. if it aint broke.
this is probably the 6th time i have just started from scratch. i just go to the plex settings, and say the archive is in “INSERT NEW NAME” and it just makes a new folder and installs a new version of plex. then i populate it, and let it spent 2-3 weeks rebuilding, and i PRAY i dont have to do it again. cause i do not know what im doing.
the size of the folder is huge. just trying to calculate the size of the data takes an hour or 2. so backing up takes DAYS, because it runs out of resources as it copies, and it gets slower and slower and slower.
i was trying to check properties, but its at 470,000 files and 24 gb so far. i have no idea how long is has left.
the nas is close to 74 tb. but i dont think its a problem.
so… given all that, could you explain where in a file structure it would default put the config folder? but like… explain it like im 80.
or what this DBRepair tool is and where to find it…
its 97.6 gb and 1,586,772 files.
hence why it takes forever to backup and update.
but i had the backup pointing at this whole file and it backed up… like nothing…
but thats why i was filled with horror when it took less than a second to restore.
so i found out how to get into the docker with the comand line, i got dbrepair to the right place.
i ran it all. it kept failing. says error 137. says it cant export the blobs database.
Automatic Check,Repair,Index started.
Checking the PMS databases
Error: in prepare, file is not a database (26)
Check complete. PMS main database is damaged.
Killed
Check complete. PMS blobs database is damaged.
Exporting current databases using timestamp: 2024-09-15_14.48.29
Exporting Main DB
Exporting Blobs DB
Killed
Error 137 from Plex SQLite while exporting com.plexapp.plugins.library.blobs.db
Could not successfully export the blobs database to repair it. Please try restoring a backup.
Repair failed. Automatic mode cannot continue. Please repair with individual commands
so im guessing its truly cooked.
i will go back to rebuilding the database on a new folder. if only there was a way to import all of the intro detection, that would save me so much pain.
137 = variable in the DB (the scripts PMS installs) is being used before it’s declared. This is damage from a failed PMS installation & DB upgrade.
Recomnend you use the tool again,
Stay in the menu mode.
Look for the most recent backup you can restore from
It will check both blobs and main dbs , searching for a valid pair.
When it finds a valid pair (or pairs), it will present them to you to choose from as your restore point.
Select the one you want (they’re identified by date) and go from there.
and 
to walk by! 
