from dmention7@lemm.ee to selfhosted@lemmy.world on 18 Feb 00:13
https://lemm.ee/post/55878453
I’ve been kind of piece-mealing my way towards cleaning up my media server, and could use a little advice on the next steps.
Currently I have a little under 10TB of torrented media that I have been downloading to / seeding from media library folders that Plex and Jellyfin monitor, using my desktop PC as the torrenting client. This requires a bit of manual maintenance–i.e., manually selecting the destination folder for the torrents in a way that Plex/Jellyfin can see.
I recently fired up qBittorrent on my media server (Unraid if that matters), and would like to try out some of the *arrs, but I’m not quite sure how to proceed without creating some kind of unholy mess.
I guess option A is just to import all of my current torrented content from desktop to media server client, and keep manually specifying the torrent destination. It’s not a huge deal, since I am typically only adding a few torrents per week, so it’s literal seconds or minutes of work to find the content I want.
Option B is to start “clean” and follow one of the many how-tos for starting up an *arr stack. But never having used the software, I don’t have a good sense for how it works, and whether there are any pitfalls to watch out for when trying to spin it up with an existing media library that includes both torrented and ripped content.
From a bit of reading, I think radarr for example will only care about new content. So I should be able to migrate all my existing torrents to the new client on my media server, including their existing locations amongst my media library, and then just let radarr locate and manage new content. Is that correct?
Any other advice or suggestions I should be considering?
threaded - newest
Look into running the arrs in containers. Makes it very easy to figure out which ones work for you without a ton of cleanup for the ones that don’t.
The arr tools I’ve played with all made it easy to import an existing media library for management.
Yeah, I’m using Unraid, so this would all be in dockers.
I think maybe I had a false understanding that radarr (for example) interposes itself between the actual media files and Plex/Jellyfin… which sounds like a PITA to undo, and was giving me pause.
I just switched from using Medusa and CouchPotato to Sonarr and Radarr. During the library import process, you can specify if the application should “monitor” the media which is what it means to download new content or try and replace with higher qualities. You can import entire libraries as “Unmonitored” so it will show it, but effectively ignore it unless you go back and change it. You can also just not import your library, and start “clean” if you wanted, and I believe it will just ignore the files for anything you don’t add.
Thanks, this is helpful!
If I do a “clean” install, can i later identify specific pieces of media within a library to monitor?
Yes you can. You select what it is at import.
I’d also suggest not setting up your torrent client (or not enabling it) u til you’ve finished your import, and setting what you are monitoring and what you are not.
This way you can just import, and enable/disable monitoring of items easily before you start any new downloads.
Got it, thanks!
I think it depends how your current files are organized. The arrs require a certain level of organization and naming to do a library import, they also want you to keep your downloads and your library copy as different folders but in a way that hopefully supports hardlinks so you don’t take up twice the space. ‘Trash guides’ is a great place to look at for getting started and they have an unraid guide as well.
The servarr (radarr, lidarr, readarr), sonarr, and trash guides discords are also really helpful if you don’t object to using discord.
If all your current files are still in the “download” folder, you could probably setup the arrs and qbit as recommended in the guides and then work your way through importing them to the arrs from the activity queue.
Yeah, that’s the rub… they are all currently in separate movies, shows, and music folders as Plex/Jellyfin want them to be.
But it’s sounding like the best bet is to leave the existing content alone for now and spin things up per the guides until I have a better handle on how it all works. Appreciate the input!
If they’re organized you can do library imports, after setting up the arrs for the most part. For example your movies folder would need to have a folder per movie named like ‘Movie Title (year)’ at a minimum and then contain the movie…just make sure what you’re seeding in qbit is not the same folder, but you can use hardlinks to keep seeding how qbit wants it while allowing the arrs to keep its copy nice and clean.
I’m pretty sure you can do a library import after you setup the arrs and use them for a while, you just need to utilize categories is qbit so the arrs only look at their specific category.
Read through their wiki getting started and faq and check out the trash guides.
You can add radarr, etc. to your existing stack without changing much. You mount your movies directory to /mnt/movies in the radarr container, shows to /mnt/tv in sonarr, and so on. And the qbittorrent downloads folder to /mnt/downloads. Then, when you add a new movie to radarr, it’ll add it to qbittorrent and get downloaded to the downloads folder, and when it’s done, radarr will move it from there to the movies folder. Then jellyfin sees the new file and adds it.
If you want to do everything the “radarr way”, you’ll also want to import the existing library to radarr so that it can identify the movie quality. There are buttons in radarr, etc. to rename and move files the way radarr wants them. Since nothing else cares about this, I let radarr handle the naming. During first setup there was a lot of re-downloading of media it couldn’t identify the quality of, but I didn’t care about that so I let it happen. But all of this is optional, you can disable quality upgrades and leave your existing library alone.
Thank you, this clears up some misconception i had about how the *arrs work!
Small note, the *arr stack (at least when running in docker) will prefer you mount qbittorrent’s download folder to
/config/Downloads(case sensitive). otherwise it whines about paths in the health menuReally? Mine doesn’t care about that.
Yeah it’s pretty much seamless. You just spin them up bare metal or docker (both are fine honestly) and follow any old tutorial for setup.
If using docker, ensure you mount the qbittorrent’s download folder to
/config/Downloadswith a capital D or you’ll get a warning about paths being set up wrong.Also, I assume this isn’t really an issue for you unless you mess with the downloads after the fact, but *arrs expect the torrented media inside a folder with the title of the media on it. It picks through torrent naming conventions fine, but when I migrated some movies yesterday I noticed it wouldnt pick up any video files that weren’t inside a directory.