Projects / atrack


Atrack is an ntrack/Bittorrent tracker implementation designed to run on Google App Engine. The main goals in its design are minimalism and memory and CPU efficiency.

Operating Systems

Recent releases

  •  30 Oct 2009 00:13

    Release Notes: Optimizations and reliability fixes were made.

    •  21 Oct 2009 02:13

      Release Notes: The "compact" response format was implemented. If no bugs are found, this will be the last release before 1.0.

      •  08 Oct 2009 15:10

        Release Notes: This release finalizes core features, and begins preparing for the first stable release.

        •  29 Sep 2009 22:07

          Release Notes: Huge performance improvements were made by bypassing memcache's pickling. This should allow many more requests per CPU quota. Many other improvements were made.

          •  18 Sep 2009 14:18

            Release Notes: This release adds many performance and correctness improvements. It's starting to be usable as a production tracker.

            Recent comments

            16 Sep 2009 05:04 lostgoblin

            Betadoctor: Cool, added your tracker addresses to

            As for the html, like everything else it is public domain, you are free to do anything you like with it. (If your country doesn't recognize the public domain, you are free to take it all under the MIT license, which is basically the same).

            15 Sep 2009 09:05 betadoctor

            Uriel : Thanks for your fast response.

            I downloaded the latest code from the mercurial repo and replaced the old one with this newer version 3 :)

            As for the installation it runs on You can add to the main page as you wish.
            By the way, am i allowed to edit the static html ? .. or to which sort of copyright is this submitted ?

            Keep up the good stuff,


            14 Sep 2009 15:16 lostgoblin

            Betadoctor: Glad that you found it useful!

            As for scrape, I might implement it, but first I want to know where the bottlenecks are after some stress testing. I hope to find a way to make it optional so trackers that run short of quota can disable it and have the core functionality still intact. We will see...

            The latest code should all be available in the mercurial repo, there has been some polishing and optimization so you might want to update. I will make a new release in a few days after I'm done with stress testing.

            Also if you are running an installation I'd love to hear about your experience and will be happy to add it to the list of known installations in the atrack homepage.



            14 Sep 2009 08:50 betadoctor

            i also see in the live public tracker that you a re busy with the Alpha 3 ?
            Any idea when it is released ?
            Eager to test a newer version ..

            14 Sep 2009 08:32 betadoctor

            Very nicely done indeed .. been testing this baby for a couple of days now... have to say that the basic functionality works like a charm ... let me be so kind to give you a little user feedback.

            I would love to have the "scrape" function added thought i realize that this will increase the CPU and bandwidth times. But i think that this should also be considered basic functionality.

            The current size of the script ( about 10k) is really neatly done. *takes off his hat*

            Just my 2 cents.

            Greetz .. BetaDoctor


            Project Spotlight


            A Fluent OpenStack client API for Java.


            Project Spotlight

            TurnKey TWiki Appliance

            A TWiki appliance that is easy to use and lightweight.