This is an excellent overview of the various issues you need to tackle. One of the things they mention here is that many of the port prediction techniques quickly run into problems. For example, if you're on a NAT with a lot of users behind it, ports will be assigned so quickly that it's almost impossible to determine the ports that will be assigned to specific clients. This is only relevant for NATs that have address/port dependent mapping and possibly filtering (I have to re-think that through every time...), but those are the only ones you'd need to use prediction for.
http://tools.ietf.org/html/draft-ietf-behave-p2p-state-05 -Adam On Nov 8, 2007 1:12 AM, Syzygy Dc <[EMAIL PROTECTED]> wrote: > > I am implementing udp nat traversal project and need to know how to do port > prediction. Can anyone provide me links to good literature on nat traversal > and advise from anyone experienced with nat traversal? > > Lots of thanks > > John > > __________________________________________________ > Do You Yahoo!? > Tired of spam? Yahoo! Mail has the best spam protection around > http://mail.yahoo.com > _______________________________________________ > p2p-hackers mailing list > [email protected] > http://lists.zooko.com/mailman/listinfo/p2p-hackers > > -- http://www.littleshoot.org Open Source, Open Standards, Open Data _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
