10

This question already has an answer here:

Many pirated game/movie torrents these days have seed boxes which are basically dedicated servers for seeding. I was wondering if leech boxes can be set up by the publishers to delay the torrenting of games.

The speed of download of torrent is determined mostly by the seeder to leecher ratio. Lets say a hacked or custom made torrent client is set up which acts as a leecher. It would discard all the data as soon as it receives it and asks for it again. In this way it can easily soak up all the bandwidth of anyone trying to seed the torrent.

Well, eventually everybody else would be able to download it but it would take an entire month to do so. This would drive up day one sales.

marked as duplicate by M'vy, Steve, Community Oct 17 '17 at 11:23

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • 1
    I'm not an expert on BitTorrent clients, but I thought they gave preference to those users that also upload. Either way, to soak up all the available sources, the "attacker" should have enough clients. So I think this particular denial-of-service is academic. Possible, but not worth it. – S.L. Barth Oct 16 '17 at 11:27
  • Well, I dont know much about bittorrent clients either. I thought the hacked client could masquerade as multiple different clients to soak up bandwidth. Even if delays torrenting for most users for just a week, I think it would be worth it. – Souradeep Nanda Oct 16 '17 at 11:31
  • 3
    I suppose it's possible if the different "multiple" clients use different TCP/IP ports. And/or some IP spoofing may be necessary to pose as multiple clients. The attacker is still going to spend resources, for what I guess to be negligible gain. But I could very well be wrong. – S.L. Barth Oct 16 '17 at 11:37
  • 3
    @S.L.Barth - if you are spoofing large numbers of leechers I imagine you could also "fake seed" between them. – Hector Oct 16 '17 at 11:47
  • I think the MPAA tried this for a while but found that it wasn't very effective due to DHT and clients auto-banning repeatedly unresponsive IPs. – Polynomial Oct 16 '17 at 11:57
3

There are a variety of torrent poisoning techniques; however, BT has been developed to be resistant to these types of attack (up to a certain extent). I'm not sure about the counter-measures currently existing, but it's not difficult to identify uncooperative peers and block them. You might have missed a couple of chunks by then, but nothing dramatic. Unless you can fake a lot of peers.

See also How does torrent poisoning work?

10

You can poison a particular torrent with false IPs quite easily. The tracker is just a script that takes an announcement request over HTTP(S) with a torrent's infohash, the IP address and port of the peer, and some details about how much of the torrent has downloaded for that peer. The IP given is not required to match the source IP of the announcement.

There are a number of attacks beyond this, e.g. having peers at these address which return garbage data.

In terms of defense against this, modern BitTorrent clients are quite robust. For example, DHT allows the user to reach additional clients as long as it has contacted at least one legitimate peer. Clients also detect many common tricks to falsify availability and ban those peers for the remainder of the download. This means that for a torrent with any significant number of legitimate seeds and peers, the false peers would have to vastly (1000-to-1 or more) outweigh them in order to have any real effect. Of course the side-effect here is that the torrent jumps to the top of the list on search engines, because it appears to be very popular, which is the opposite of what most intellectual property holders want.

4

Most interestingly about the torrent system is that it is a real world application of game theory. Each client typically uses a tit-for-tat strategy.

BitTorrent peers have a limited number of upload slots to allocate to other peers. Consequently, when a peer's upload bandwidth is saturated, it will use a tit-for-tat strategy. Cooperation is achieved when upload bandwidth is exchanged for download bandwidth. Therefore, when a peer is not uploading in return to our own peer uploading, the BitTorrent program will choke the connection with the uncooperative peer and allocate this upload slot to a hopefully more cooperating peer.

Therefore, it should be secure against the attack vector in question by virtue of it's nature.

Not the answer you're looking for? Browse other questions tagged or ask your own question.