Since I started using Jellyfin and arr, searching and watching have become separate activities. Gone is the experience where I discover something and am watching it immediately. And, search/discovery is done on separate apps than where I watch. So my behaviour has changed. I think ahead. I keep any eye out for what people are watching and download what seems good to me. The payoff is I’m not limited to what’s on any particular platform and I don’t pay subscription fees.

Depending on your download speed, you can manually download a TV show episode in seconds to minutes. By the time you watch that episode, at least the next one will be ready. It is quite rare to have to do this though, me and my family mostly add shows on Seer when we find them (recommendations, adverts, etc) and by the time we’ve sat down to watch it’ll be ready.

I did the whole lists thing others have mentioned but to be honest, we found there was too much choice, lots of crap and quickly ran out of space. Taking an active role in choosing shows and films works better for us and I’ll have a short list at any time to watch.

Anyone has a good tutorial on how to setup a complete are stack with docker on Linux?

Already one that quickly explains what arr does what would be helpful. I know there is radarr, sonarr, bazarr and loads more and I have no idea which system does what or how to connect them.

I’ve found tutorials about individual pieces, but those are of little help

And then the biggest one: I have jellyfin, I’d like to use it over the internet, bit I need to have that obviously VERY safe. How to do that? I know of a popular reverse proxy for those things (name escapes me for a sec) but again, the tutorials I’ve found were lacking at best.

I’m looking for something where I can write a bunch of docker files and start it all up from scratch on Linux, is that possible?

You’re right, it’s for building a private library, not a “what’s new to watch now” stream. There are other tools for that, like stremio or real-debrid.

Firstly, the core *arr suite is not a streaming service, whatsoever. It’s a media file manager meant to help with running a private server streaming app, like Plex or Jellyfin. With that out of the way:

I imagine sitting on coach, searching for show. Then you want to watch some, and then you have to wait half an hour for full episode (or even season?)

Download speeds depend on your own setup. IMO, a Usenet connection is the only way to use *arr. Downloads happen at the maximum speed and don’t rely on some other person’s seed rate. You, conversely, don’t need to worry about seeding.

When you manually add a show to Sonarr, you can select it to only pick up the pilot episode of the show, which could cut down on DL times by focusing bandwidth. You can also select a lower definition. With Usenet and something like a 720p quality, there’s no reason why this should take more than 5 minutes to be in your library.

I’ll also paste my comment I left below about connecting to lists:

You can connect your *arr profiles to monitor external lists of new titles by pointing the list manager to something like MDBList. They might not be as instantaneous as you might like, with a 24hr refresh period, but it’s pretty much a 0.99:1 Netflix replacement for me

I’ll also add that I’m not some CompSci nerd, either, so don’t be scared to give it a shot. My server runs off Plex on my Windows 10 desktop because I don’t know how to do any better but I’ve never had an issue watching what I want to watch

Any reason not using Streamio?

The whole point of the arrs is to download things in advance. Many users set up Import Lists… There are also apps like Boxarr.

Netflix’s model is to provide the cheapest, low-quality media you can bear to watch. The kind of browsing you are describing is your distressed search to find something watchable. If you populate a server with good TV or at least TV that interests you, you won’t want to hop between media like that.

Yes and no. You need to understand that no home service truly replaces Netflix, for a few reasons (the media might not be available on any of the services you’re using, for example).

It’s also not as simple as searching for a media in Jellyfin/Plex (or whatever other media frontend you choose, like Emby). There’s a fixed flow.

But let’s start by explaining the layers:

1. The frontend - Plex/Jellyfin/Emby/Kodi. This is what your users see, aka the “Netflix experience” - open the app, and all the media available on your storage device will be shown. Then they can click one and play it.

2. The request manager - Seerr (previously Overseerr/Jellyseerr). This is a separate interface where your users can request media. You still need to manually accept it (unless you set it up to automate things fully, but make sure you trust your users!). If something isn’t available, your users will come here and ask for it, then the manager will show the status (requested, accepted, downloading, available). Once available, your users can watch it through the frontend.

3. The media managers - Radarr/Sonarr/Lidarr/etc. This is the software responsible for keeping a list of media you want, regularly looking them up on torrent trackers, Usenet servers, etc., and matching your requirements (resolution, language, encoding, file size, and so on), then grabbing the release and passing it on to the download client.

When you accept a request in the request manager, it passes on the info to the media manager, which adds the requested media to its internal list and begins looking for it.

4. Download client - torrent/Usenet downloader (qBittorrent, sabnzbd, etc.) pretty straightforward, this thing takes an incoming download request from the media manager, and downloads the file according to protocol, then signals the manager that the download is ready.

At this point, control is passed back onto the media manager, which finds the freshly downloaded file, copies/moves it to the right place according to settings, renames it according to settings, marks it done then sends a signal to the request manager to indicate the request was fulfilled.

Finally, the media frontend, which is set up to watch the folder where the new media items are copied/moved and renamed, gets a notification that a new file is available, scans it, prepares metadata (poster, background image/music, description, actor and production lists, ratings, etc.), and makes it available in the search interface.

So the key differences with Netflix are:

• limited content compared to Netflix
• the ability to request new media
• no CDNs, so if you have lots of users and not much bandwidth/processing power (latter in case of transcoding), your users will struggle. A standard home server and internet connection can serve 3-4 users at the same time.
• limited language support. since these are pirated media, and most pirated media has at most 2-3 audio tracks, you’ll lose that Netflix perk of having 6-8-10 audio tracks available. subtitles can be supplemented though (audio tracks too but they rarely match perfectly to the video so it’s not as simple as downloading a file and call it a day).

That about covers all the functional differences between an arr stack and Netflix.

My media server has more shit I want to watch on it than Netflix does, it’s not even close. Yeah it took some time to build my library but it’s paid off. Even starting out just queue up a bunch of shit ahead of time and it will be ready to go when you go to watch it. I personally watch shows through the week as they take longer to get through and are more adaptable to the amount of free time I have. Then I queue up any new movies for the weekend and new shows as I hear about them. It’s not difficult to stay ahead of the curve with some minimal planning. My backlog will take me months to get through. Also your media library will never remove shit before you’re done and you’re not limited to just what the streaming services you have are currently offering.

Watch the trailers first ;)

Aside from the point that Jellyfin is meant to browse your own personal collection of files usually after the fact…

Some file formats like mkv do work even if partially downloaded, so if you’re downloading a torrent for a free libre open source movie, choose the option to download chunks in sequential order, and I think there’s a way that you can watch while downloading.

At least for me your argument is invalid.

1. I installed Wholphin on my TV and it’s both a jellyfin and an (experimental) jellyseer client
2. Here in Europe I have fast internet, so downloading the whole file is not really a problem. Also I configured my arr stack to prefer a certain size (on the smaller side) for each resolution.
3. I remapped my Netflix button to Wholphin

So, yes, you’re basically correct.

There are search layers that remove the need to access radarr / sonarr directly when searching for shows (someone mentioned jellyseer, for example), so that part of the process can be streamlined, and once you’re watching a show it’s generally very good at pulling new episodes as soon as they’re available, so you’re typically, at most, a day behind actual airing dates. But if you’re trying to just bounce around and try a bunch of different shows it wouldn’t be the best for that. The biggest constraint is generally the speed of your internet and the popularity of what you’re watching. With a high speed connection and a well seeded torrent it’s often only a a couple of minutes to download a pilot episode, and you could have the whole season done by the time you finish watching that.

The other question is one of storage. If you’ve got plenty of hard disk space then you can probably afford to just throw anything that sounds interesting on your pull queue and work your way through it when you actually have time to sit down and watch. Basically you sort of pre-emptively build your “Netflix at home” library and then do your bouncing around channel hopping stuff with the five or so vaguely interesting shows that you added while you were at work.

Is it a replacement for Netflix et al? Not strictly speaking, but if you don’t mind changing up your habits a little it’s probably close enough.

1. You can request in Seer and after downloading click the “Watch on Jellyfin”-Button, which then leads you to the movie. So its technically two separate services but feel like one. (With Seerr, Prowlarr, Radarr, a Downloading Client, Jellyfin and Bazarr its actually many selfhosted services but it doesnt feel like it.)

2. With my setup through Radarr I need to download the whole movie. For series it depends it also meeds to download one whole file, then it goes through sonarr and then gets sorted properly into the library with metadata and coverimage and so on.

So yes, I cannot sit down and then decide spontaniously what to watch. But I currently dont work like that: I hear about a good movie somewhere, add it to the library. Then, after 10’ to 2 days (depending on how popular it is) its downloaded. When I sit down because I have time to watch something, I then browse through the library, which I all know, that they are good and interesting movies.

If I have friends over, I ask, what they wanna watch, request it and we eat dinner and afterwards we watch the movie.

I imagine sitting on coach, searching for show. Then you want to watch some, and then you have to wait half an hour for full episode (or even season?) to download.

This is a fair take on how a locally hosted video server would go. It’s the same as someone who has a collection of disk media instead as well. Finding new media to watch is not instant, even with the best setups.

I actually consider this to be a feature, instead of a bug. The algorithms that Netflix (and YouTube and everybody else that serves content) have a lot of issues. The ability to find content, the act of discovery, is something I think is actually very valuable, and has been lost since we switched to online streaming.

I run a jellyfin server for my immediate family, and one of the benefits of not running an auto-download tool is that we all have a groupchat specifically for requesting new series/movies. I didn’t expect it at first, but it has been a great way to connect with my family over varied media we watch, as well as a way of sharing what’s new and interesting to them.

Of course, I switched from Spotify to a physical mp3 player with my own personal library, so maybe my perspective is a bit skewed. For sure there is a place for a lack of barriers (including skipping out on analytical thought) for consuming content. I just don’t think it should be the default.