Am 07.09.2010 00:40, schrieb Benny Prijono: > The behavior is expected, and is inline with the SIP spec. If call > gets established after CANCEL is sent, then we'll let it continue to > get established, and it is then user or app decision whether to > continue or not. If not, then it should call hangup() one more time. As pjsua (pj sip USER AGENT) API is a very high-level API I think the SIP stack itself should send BYE in such a race condition. regards Klaus > > We may change this behavior in the future though to make it do like > what you want, though there's no definite plan for this yet. Patches > would be welcome of course. > > Best regards, > Benny > > On Mon, Sep 6, 2010 at 11:29 PM, Viktor Krikun<v.krikun at gmail.com> wrote: >> Hello everyone, >> >> Writing some unit tests for my pjsua-based app I discovered an issue in call closing from CALLING state: Alice calls Bob, on switching to PJSIP_INV_STATE_CALLING Alice cancels the call (pjsua_call_hangup). As result the call never get canceled. >> >> That what I learned from Alice's logs: >> 1. Tx INVITE >> 2. pjsua_call_hangup() says "Delaying CANCEL since no provisional response is received yet" >> 3. Rx 1XX (100 or something, ready to send CANCEL) >> 4. Tx CANCEL >> 5. Rx 200 OK. Note: SDP included multiple codecs. >> 6. Tx ACK for the INVITE >> 7.PJSUA says "Got answer with multiple codecs, start updating media session to use only one codec.." reINVITE initiated >> 8. Rx 200 OK for the CANCEL >> 9. Rx 100 Ringing, Rx 200 OK, Tx ACK (for the new, reINVITE SIP dialog ) >> >> In result we have new SIP Session established (after re-invite to negotiate single codec) which wouldn't be canceled. >> >> PS: full logs are attached. >> >> Have I missed something? Thanks for your help in advance! >> >> >> _______________________________________________ >> Visit our blog: http://blog.pjsip.org >> >> pjsip mailing list >> pjsip at lists.pjsip.org >> http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org >> >> > > _______________________________________________ > Visit our blog: http://blog.pjsip.org > > pjsip mailing list > pjsip at lists.pjsip.org > http://lists.pjsip.org/mailman/listinfo/pjsip_lists.pjsip.org
