You can also replace common or shared attributes between your configs using env files: https://docs.docker.com/compose/environment-variables/set-environment-variables/#use-the-env_file-attribute

Things I put there: UIDs, GIDs, TZ, shared mount paths, etc.

No, that is counter intuitive. It may appear more expensive at first, but on the long run it is a lot more cheaper. It avoid vendor lock-in, recurring increase of dev costs and licensing and lots of other plagues of closed proprietary development like blackbox development and justification of hidden complexity as a driving factor on costs. I worked with legacy closed proprietary sw development and lock-in combined with legacy complexity made man-hour costs exorbitant. These are partially solved by open-sourcing, as kicking out a team and putting a new one is easier, but most importantly transparency as a driving factor on quality of development.

This is the best: https://github.com/sahib/rmlint

First, good job. Looks very good and feels snappy. So high potential player in the ecosystem. So here a quick feedback after a few minutes using this:

1. when exiting the "home screen" that contains some albums, trying to get back to the same screen with same content is impossible. Getting back to "albums" screen has different content, alphabetically sorted.

2. I've been looking for a player with proper "preferred" management and features. This app let's us mark an album as preferred (heart), but not sure what it does here. But what I really miss in a player, is a way to mark preferred individual tracks and then to see all these tracks and be able to use that as a giant playlist that we can play in sequence or randomly. Please consider this feature and your app will be already more featured packed and useful than any other jellyfin player out there.

mine was really sluggish for a long time, then I saw someone in here explaining their similar issue and their fix. I don't have the post link, but it was related to DNS settings. Basically for some reason using my pihole dns made only nextcloud sluggish, the fix suggestion was to use 1.1.1.1, which worked. Now, it is a pretty fast nextcloud.

Awesome and thanks a lot for putting the time to explain it like this.

So for some reason I got side tracked with radarr and didn't see the need for trackt anywhere, but that seems the missing part on all this.

This also shows up that the Plex workflow is seamless (no Overseerr/jellyseerr need, no trackt need) than jellyfin right now.

Reading plex_debrid code, it seems it has some initial code on scanning current Jellyfin library, so finishing that code could remove the need of Trackt.

Now, one advantage of using Radarr ia that it will move and rename to a standad naming the incoming files, I think that only for this feature it is worth to keep it in the workflow.

So it seems like I'll need to fix plex_debrid to understand existing Jellyfin library and remove the need of trackt!

Thanks a lot!

Interesting. I installed here but I may be doing something wrong with my setup, because just using Jellyseerr is not triggering a RD download. First, using jellyseer required radarr anyway, the setup is like following:

1. Jellyseer request a movie/show, put that request via radarr
2. radarr will try search for a torrent via indexers (not working right now)
3. Download via Black Hole Torrent, which is basically monitoring plex_debrid folders

Still, not working as things aren't communicating with each other correctly and I didn't set any indexers (or set jackett).

My setup feels wrong or too complex, can you give a bit more of details on yours? How the parts communicate? :)

plex_debrid looks the way to go with. I was put off it because its name and I use jellyfin. Reading more closely it works with jellyfin, so cutting off the middleman (radarr) seems like a very good solution!

Is there anything special to know about it before trying myself? Any issues or roadblocks you had when setting this up?

Thanks for the answer!