Gao, I see, but RFC32618.2 UAS Behavior Note that request processing is atomic. If a request is accepted, all state changes associated with it MUST be performed. If it is rejected, all state changes MUST NOT be performed. I think we neet to consider the above sentence carefully. Regards, Shinji gao.yang2@xxxxxxxxxx>Dear Shinji>>The session state by my proposal is right.>The dialog state by my proposal is always c4.>>And I'd like to replace the words "a part of the original modification" >with "only refresh the precondtion state table",>replace "new modification" with "The others".>Beacuse currently RFCs, just "only refresh the precondtion state table" >can be treated as "a part of the original modification".>And it is words from the RFC.>>And thanks for your works.>>Gao>>OKUMURA Shinji <shin@xxxxxxxxxxxxxxx> >发件人: sipping-bounces@xxxxxxxx>2009-03-05 16:24>>收件人>sipping@xxxxxxxx>抄送>>主题> Summary of the rollback issue>>Hi, all>>This thread is too big and complicated.>I would summarize the discussion.>>For that purpose I make a simpler proposal.>It is the straw-man proposal, but it will work without a >problem, I think.>>signaling | O/A state | remote target>---------------------+-----------+-------------->init-INVITE/200/ACK | state1 | c1>re-INVITE/183-rel | state2 | c2>PRACK/200OK | [noSDP] | ->UPDATE/200OK | [noSDP] | c3>UPDATE/200OK | state3 | c4>4xx/ACK | state1 | c3 <- this record is my proposal.>>1. Main concept is full-rollback according to RFC3261.>2. UPDATE without SDP change remote target immediately.> it doesn't rollback.>3. The rollback is NOT depend on preconditions.> there is no need to check SDP attribute.>4. UAS should reject UPDATE received between 4xx and ACK.> UAC should NOT send UPDATE between 4xx and ACK.>5. UAC should ignore 200 response to UPDATE(with SDP) that send > before 4xx received.>>Now we have discussed three additional proposal.>>A. commit-latest-exchange(Liaison Statement ?)> 4xx/ACK | state3 | c4>>B. late-commit ( by Gao)> if UPDATE is "a part of the original modification"> 4xx/ACK | state1 | c3> else (UPDATE is "new modification")> 4xx/ACK | state3 | c4>> the decision depend on preconditions.>>C. Gonzaro's draft> 4xx/ACK | state1 | c1> and he says> C1. UAS should NOT send 4xx.> C1. UAC should send UPDATE after sending >ACK for 4xx.>>Is my summary up to this right ?>>Regards,>>Shinji_______________________________________________Sipping mailing list https://www.ietf.org/mailman/listinfo/sippingThis list is for NEW development of the application of SIPUse sip-implementors@xxxxxxxxxxxxxxx for questions on current sipUse sip@xxxxxxxx for new developments of core SIP
