Hi,

The SIP RFC section "17.1.2.1 Overview of the non-INVITE Transaction" (quick
link http://tools.ietf.org/html/rfc3261#section-17.1.2.1) contains this
little snippet:

"For unreliable transports, requests are retransmitted at an interval which
starts at T1 and doubles until it hits T2."

Which would then imply for reliable transports that either retransmits
should not occur or possibly that a different retransmit timing scheme
should be used. I think it's safe to say the original authors intended the
former otherwise they would have included the different timing scheme.

The problem I've got is a UDP only SIP client that only communicates over
UDP is able to connect to TCP SIP clients through a stateless proxy.

UAC (UDP only) <=== UDP ==> Stateless Proxy <=== TCP ===> UAS

The UAC is going to retransmit any request, in this case a MESSAGE request,
if it doesn't get a response within T1. The stateless proxy is duly going to
send that retransmitted request to the UAS over the reliable TCP transport.

According to the SIP RFC the UAS seems to be entitled to treat the
re-transmitted MESSAGE request as a new request since it can rely on the
fact that SIP requests are never retransmitted over reliable transports. Has
anyone encountered this particular problem and come up with a pragmatic
solution?

Regards,
Aaron

_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to