When you get into containers, the UID/GID matching becomes really important because you’re crossing namespaces.
Having media in home directories is usually problematic (no permission to see things) unless you set UID,PUID,PLEX_UID (and corresponding GID variables) to match your host’s UID & GID.
To take the minimum out of one instance to another –
Turn off all automatic library updating
Turn off automatically empty trash
– we don’t want database records lost while directory paths are updated –
Stop PMS
Copy Preferences.xml and the entire Databases directory
( sudo tar cf /somewhere/safe/plex.tar ./Preferences.xml ./Plug-in\ Support/Databases ) – Assumes you’re in the Plex Media Server directory.
Now create the new instance wherever it will be. No need to initialize it
Stop PMS
‘cd’ into the new “Plex Media Server” directory
Extract the tar file (it will overwrite the empty XML and Databases)
VERIFY owner & group matches (often needs chown -R command)
Start PMS
The new instance will come up, having assumed the identity of the previous one (don’t ever start the container again)
For each library section,
– EDIT the section
– ADD new directory paths – DO NOT delete OLD paths just yet
– SAVE
– AFTER the section finishes updating
– EDIT the section again
– DELETE the old directory paths
– SAVE
Advance to the next library section and repeat until done.
With all section paths updated,
– Scan files
– Empty Trash
– Clean Bundles
– Settings - Server - Library
– ENABLE those auto update settings previously disabled.
If I do it with this method, will it save the matches I had or will it re-match every metadata source? I had a lot of content that I had to manually fix so I wanted to try and avoid re-matching everything. I don’t mind regrabbing pictures and stuff
Instead of manually fixing matches, I recommend you make file naming adjustments to avoid the issue. I play around with Jellyfin and got tired of having to fix the missed/wrong matches on files repeatedly when redoing libraries.
If you find a movie not matched correctly, instead of manually fixing the match, look up the movie on TMDB and then add the correct {tmdb-xxxxx} tag to the end of the folder name for the directory the movie is in. Then rescan the library. Plex will rematch the folders you renamed and you wont have the same problem next time.
The Preferences file contains the server’s identification and a few basic settings.
The bulk of your settings, as well as ALL the matching information, is contained in the databases.
The only thing the databases don’t hold are the posters and descriptive texts.
When you do the restore, You hit “Refresh Metadata”.
This tells PMS to use the matching information it has and download posters and texts again (the “bundle” directories under “Media” and “Metadata” (Plex Media Server directory)
As stated above, the best solution it to have perfect naming so it doesn’t matter.
I use FileBot to ensure my naming 1) matches TMDB and can be found by Plex and 2) is renamed in Plex standard naming format. With this, I can completely “Fresh Match” my entire server in minutes.
And no worries, every Movie and TV Show I have on my Plex is named using the scheme Described here and I used filebot too (for every piece of content in my library) But for some reason, some movies still miss the mark!
Also - all the permissions issues seem to be fixed. I can add libraries successfully and the language box is there. Now I just have to figure out the transcoding issue and posters
Update: Weird, now getting this error. I’m cursed I tell you
Edit : I got everything working. My one question is though, how do I get it to generate the “Skip Intro” data. IS there a way to manually trigger it? It did not do it when I added new librarys. Or will it do it on its own? Thanks