I noticed that for some reason Sonarr was not picking up new releases until much later than they were posted on the tracker I used (30+ minutes). I narrowed it down to an issue where under heavy load, the tracker tends to not update the seeder/leecher count for quite a while.
So, for example a brand new TV show may be added, with 100+ seeders downloading, but they list it as 0 seeders/leechers. Sonarr seems to have a built in restriction to not download w/ 0 seeders (makes sense normally means a dead torrent).
However, I’d like some way to ignore this, at least for RSS feeds. Is there any way of overriding this restriction? Otherwise, would you ever consider a workaround with logic something like -
“If the torrent is < x hours old, allow download, regardless of if seeder count is 0.”
Or maybe just hardcode a seeder count of at least 1 on brand new torrents, so the logic further along will pass? I know this is really the fault of the tracker, I’m just trying to think of a safe workaround that could benefit all users of Sonarr. Any thoughts on this?
Sonarr does this for KAT already, if its less than 12 hours old we ignore the seeder count and grab it immediately. 1 seeder is the minimum for all other indexers and KAT after 12 hours.
Yeah, I just stumbled on that thread about KAT after searching a bit more. The tracker is TorrentLeech, but the problem is I’m using Jackett to get search functionality on the site. So, I have a feeling putting the fix in for that might be a bit messier than just one of the built-in trackers?
Maybe could do some check to see if the URL contains “/torznab/torrentleech” on a Torznab indexer, and then do the “KAT fix” (ignore seeders < 12 hours old) if so?
I don’t think we want to get that specific on something that is supposed to be generic. I’m not sure if there is a good solution here, having Jackett enforce a minimum seems broken as well.
One other piece of info I’ll add. I noticed that the RSS for TorrentLeech actually does show the correct number of seeders. The problem seems to be purely with the Website UI, which it turns out the Jackett indexer uses regardless of RSS or Search function.
So, for myself I was able to fix this (hopefully) by using Jackett for only “search”, and using the built-in TorrentLeech indexer in Sonarr for RSS. Still, wouldn’t hurt to add the option though, for future trackers that run into this, or people who are only using Jackett for whatever reason.