I only transfer 1080p WEB-DL quality from seedbox. Sonarr drives Usenet downloads for lesser qualities. When I implemented it this way a few years ago, I found usenet to be slow to get the 1080p WEB-DLs, so I started transferring those from the seedbox. The tracker I use doesn't allow duplicates / multiple release groups. Sonarr does on occasion reject duplicates, but only in the rare case where it downloaded the 1080p WEB-DL already from usenet. I check my drone factory folder every couple of months and find a handful of files to delete.
This sounds interesting, although I'm not convinced it would be as stable as my current setup. I have a high latency connection between seedbox and Sonarr server, with times during the day where connections time out. The less chatter between them the better. In my current setup, everything happens locally on the seedbox until download completed, and then uftp runs locally on the seedbox and attempts to transfer. If it fails to connect, sonarr eventually gets it from usenet.
- Is the ReleasePush deciding what gets downloaded by the download client? Or is it just notifying Sonarr of what was already matched to download by the client? I pretty much download every webdl release in my download client, Sonarr would need to decide to push to uftp, not to decide to send to the download client.
- If Sonarr is just pushing to uftp, how does it know when the download client has finished downloading?
As I understand your description, irssi on the seedbox would have to push a notification to Sonarr (chatter), then Sonarr decides if it gets downloaded, and notifies the download client of the download url (chatter), then presumably listens for a notification from the download client that the download has finished (chatter) and then remotely starts uftp on the server (chatter) to transfer the file to a local folder where Sonarr can import from (using the remote path mapping to find it based on the download location reported by the download client).
I'm probably misunderstanding some of it, but that's a lot of points of failure due to network timeouts. The seedbox shouldn't be relying on my computer at home to work.