Paths for usenet and torrents

Information and Tutorials on features in Mylar and how to use it
Post Reply
Tippon
Posts: 17
Joined: Mon Sep 27, 2021 11:28 pm

Paths for usenet and torrents

Post by Tippon »

Probably a daft question, but I don't want to screw anything up.

I set up NZBGet yesterday, and overnight it's found a handful of my wanted comics. They got stuck on post processing, as I was getting the 'ComicRN: ImportError: No module named configparser' error. Rather than fight my Python installations, I switched to CDH and removed the ComicRN.py from NZBGet's scripts folder instead. I restarted both NZBGet and Mylar.

I told the errored files to try post processing again and they changed from error to success, but the files didn't move.

Do I need to be telling NZBGet to move them into the same place as my completed torrents once they've finished (my monitored folder), or should Mylar be picking them up from the NZBGet Download Directory under Download Settings?

I would just set everything so that they all end up in the same folder, but I'm sure I read somewhere that it can cause problems.
User avatar
evilhero
Site Admin
Posts: 2883
Joined: Sat Apr 20, 2013 3:43 pm
Contact:

Re: Paths for usenet and torrents

Post by evilhero »

No, keep your CDH location separate from your Folder Monitor location. Keeping them identical will result in a race condition and you'll have incomplete comics trying to be forcibly post-process.

Mylar won't pick up files that have already been downloaded - for issues that have already been downloaded, you'll need to move those to a separate directory and perform a manual post-processing against that directory (you could do it directly in the nzbget download directory, but I mean, try to play it safe at this point).

I'm not sure where you tried the post-processing again where they went from error to success - I'm guessing that's in NZBget, which has zero control over the CDH aspect in Mylar. In order for CDH to work, Mylar has to be the one to initiate the send to the download client, it can't be the other way around. So I'm guessing it just did something on NZBget's side.

So when you use CDH this is what happens:
Mylar sends the nzb to your NZBGet client. When your NZBGet client finishes a download, Mylar is already monitoring it for completion. Once it completes, Mylar knows it's completed and attempts to poll your NZBGet client for additional information about the download that it doesn't know (where it's located, file-size, actual completion status, etc). Since Mylar knows the final path from that query, it tries to post-process it directly from that specific location. Provided it has the proper access to the location where NZBGet downloads to, it can proceed with post-processing and move / metatag the files and then cleanup after itself.

Also worth noting, that if both Mylar and NZBget are on the same machine, there's no need for to specify a Download Location for NZBGet within Mylar. This is primarily used for machines that are remote to each other, or docker installations.

Hope that helps!
Tippon
Posts: 17
Joined: Mon Sep 27, 2021 11:28 pm

Re: Paths for usenet and torrents

Post by Tippon »

evilhero wrote: Thu Oct 07, 2021 2:16 am No, keep your CDH location separate from your Folder Monitor location. Keeping them identical will result in a race condition and you'll have incomplete comics trying to be forcibly post-process.
Thanks, I thought I'd heard something :)
evilhero wrote: Thu Oct 07, 2021 2:16 am Mylar won't pick up files that have already been downloaded - for issues that have already been downloaded, you'll need to move those to a separate directory and perform a manual post-processing against that directory (you could do it directly in the nzbget download directory, but I mean, try to play it safe at this point).

I'm not sure where you tried the post-processing again where they went from error to success - I'm guessing that's in NZBget, which has zero control over the CDH aspect in Mylar. In order for CDH to work, Mylar has to be the one to initiate the send to the download client, it can't be the other way around. So I'm guessing it just did something on NZBget's side.
That makes sense. I thought that telling NZBGet to reproccess them may have notified Mylar, but I'm still getting my head around anything to do with usenet at the moment. I tried post processing through the NZBGet interface, probably getting myself confused because I'd set up the script initially.
evilhero wrote: Thu Oct 07, 2021 2:16 amHope that helps!
It does, thank you :)

I'll move the existing files to my monitored folder and keep an eye on any future downloads :)
Post Reply