A BitTorrent DHT Crawler

BitTorrent is one of the largest source of data exchange over the Internet. The copyright and legal pressure from authorities systematically close torrent hosting sites, futilely trying to stop the damage done to the creative industry. As a consequence, the centralized portals allowing for the search of torrent files, are more and more replaced by the use of DHT (Distributed Hash Tables) in order to provide a decentralized means of content sharing. But this means that the discovery of torrent files is a much more complicated process. This paper presents an implementation which leverages the BitTorrent DHT Network in order to build an independent torrent search engine using both Sybil attack and BEP 51 protocol extension. Over half a million torrents have been captured in one hour and 61 315 IP addresses have been also tracked, downloading 7 733 torrents over 14 hours. These results indicate two things: the privacy of the users is at risk, as their actions can be retrieved by crawling, and, as a consequence, even if the adoption of distributed sharing will likely make illicit torrents harder to suppress, it will not help users hide their actions. So the presented solution can also be used to monitor pirates behavior in preparation for legal measures.