Attila is correct. rport is just a fancy way of saying that replies should be sent to the source port whence the request came, which is exactly how stateful connection tracking works in conventional source NAT gateway implementations.
Attila Sipos wrote: > STUN doesn't work for all NAT types, I believe rport does. > > rport is a very simple mechanism without very low overhead for achieving > simple NAT traversal without requiring a separate protocol such as STUN > which requires a STUN client and STUN server. > > rport's use cases are not unique but for UDP SIP signalling, it is the > most efficient way of NAT traversal. > > Regards > > Attila > > > > -----Original Message----- > From: [email protected] > [mailto:[email protected]] On Behalf Of KEERTHI > KUMAR > Sent: 24 November 2009 10:21 > To: [email protected] > Cc: [email protected] > Subject: [Sip-implementors] [RPort] Request to know unique use case of rport > > Hi All, > Myself Keerthi Kumar, currently planning to implement 'rport' > support in the SIP Stack using RFC 3581. But, now i am stuk at finding a > Valid use case that 'rport' solves and leads over STUN or other NAT traversal > techniques. > > So, i request you to please suggest me few use cases of 'rport'. > > Thanks, > T. Keerthi Kumar > > > The INTERNET now has a personality. YOURS! See your Yahoo! Homepage. > http://in.yahoo.com/ _______________________________________________ > Sip-implementors mailing list > [email protected] > https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors > > _______________________________________________ > Sip-implementors mailing list > [email protected] > https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors -- Alex Balashov - Principal Evariste Systems Web : http://www.evaristesys.com/ Tel : (+1) (678) 954-0670 Direct : (+1) (678) 954-0671 _______________________________________________ Sip-implementors mailing list [email protected] https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors
