Hi. I am currently pirating on a dedicated server (Debian 12) running i2p, jellyfin, qBittorrent-nox, and a vpn. I tried mullvad for a bit but need to get air VPN set up now. My current setup to actually get content to jellyfin is this:
Try to find it on i2p, if there skip to step 4
SSH into server and turn on vpn + qBittorrent
Torrent Linux ISOs
SSH again and copy the file into jellyfin (really space inefficient, would love a solution to this in particular without shitloads of symlinks.)
Rename files to work with jellyfin
Login to jellyfin and refresh libraries
If Linux ISO not on i2p, cross seed
Obviously this sucks. I know *arr would help, but I don’t know how to set up a VPN to not interfere with jellyfin connections to outside of LAN.
How would you go about automating this? Do you think I am a complete idiot going about it all wrong? (I know I do) Have any of you found a solution to vpns interfering with jellyfin?
I have Plex, Radarr, Prowlarr, and Qbittorrent all installed on the same dedicated server. I'm using a SOCKS5 proxy instead of a VPN, it works great because I set up Qbittorrent to use the proxy and I just leave it running 24/7. I also have Tailscale installed for remote access, setup for that is dead simple.
Here's my workflow if I'm away from home:
Turn on Tailscale on my phone.
Open my radar app (it's called LunaSea).
Search for and add the movie I want.
That's it. If I'm already at home, step 1 is not necessary.
Prowlarr and Radarr find the movie on my registered indexers, at the desired quality, and send the torrent to Qbittorrent. Then when the download is finished they automatically rename the files and move them to my Plex library (and they could do the same with Jellyfin). Roughly 10 minutes after I finish step 3 (more or less depending on seeds), the movie magically appears in my Plex library. I don't have to turn a VPN on or off.
It gets even more automated/complex when you add in something like overseerr which pairs up with sonarr and radarr to read your library and allows your users to search for a title and request it if it's not in your library. With the click of an approve button the automation will have their desired title on plex in a matter of minutes.
It's definitely not necessary, but damn it's convenient and easy now that it's set up. And my setup is relatively simple. Sonarr is for TV, Bazaar automates subtitles, there's Lidarr for music, and Readarr for books.... The list of 'arrs is long.
Can you elaborate on using a socks proxy? I assumed you would get an isp letter if you used one because people always recommended vpns.
Edit: also I don’t want to use tailscale or plex because proprietary. I’m not sure if it’s with mentioning because it looks like plex and plain router port forwarding will work for me.
Tailscale isn't necessary, it's just what I use for remote access. And you can use Jellyfin/Emby/Kodi with Radarr too, it's not specific to Plex.
SOCKS5 proxy keeps the letters away (I live in NYC). I've read that it's because ISP's don't bother actually monitoring torrent traffic. They only act when a copyright holder reports your IP for piracy. So if you hide your IP then they can't see you.
A proxy is not encrypted, to be clear. But it turns out encryption isn't actually necessary if you just want your ISP to stop bugging you. If laws change and torrenting becomes more dangerous, I'll probably switch to a proper VPN. But a proxy is faster and easier.
The arrs would be your best bet to reduce your input. If i'm not mistaken you can run them all through docker including a version of qbittorrentb that's bound to a vpn and the only way it access the internet is through that vpn. Or you could split tunnel your vpn and bind your qbittorrent to it and bipass your jellyfin instance.
I too have a similar debian setup with web UI running qbt, whilst also connected to a VPN. Would love to know how to connect to it outside of the local connection.