wor...@ariadne.com (Dale R. Worley) writes: > "JC" <jc.hu...@tom.com> writes: >> There is one SIP call scenario that one SIP UA sends one reliable 183 to one >> SIP device, and the UA receives the PRACK, then, it sends 200OK of PRACK and >> one new reliable 180 immediately. All the messages are sent over UDP, due to >> the network issue, the SIP device receives the 2nd reliable 180 ahead of >> PRACK's 200OK. Now, I find the SIP device can't handle the case, and I have >> to prove that current case should be handled by it well. >> >> I looked through RFC3262 and not find it clarify the case clearly. So, is >> there any other RFC or SPEC which describe it? > > Section 4 of RFC 3262 is clear: The UAC receives the "2nd reliable 180" > and processes it normally. There is no rule that permits the UAC to > wait to receive the 200 for the PRACK before it will process the 180.
Hmmm, I'm not completely correct: The UAC can pretend that the 180 was lost in transmission and simply ignore it. There's no way the UAS can tell that it was not so. But it cannot plausibly ignore a series of retransmissions of the 180, waiting for the 200 for the PRACK to arrive. Dale _______________________________________________ Sip-implementors mailing list Sip-implementors@lists.cs.columbia.edu https://lists.cs.columbia.edu/mailman/listinfo/sip-implementors
