On Mon, Sep 29, 2008 at 10:54 AM, Dale Worley <[EMAIL PROTECTED]> wrote: > On Sun, 2008-09-28 at 23:58 -0400, M. Ranganathan wrote: >> 1. why does sipx proxy sending the call to voice mail in frame 59. >> 2. Why is it sending an INVITE to ~vm~~~vm~user3 ( what is that?) in Frame 65 >> 3. Why is sipxbridge seeing a 404 Not found in Frame 76 ( which it >> ignores since the OK has been sent to the ITSP already) > > There are a lot of things going on here that could be improved. In > regard to the UA at 192.168.5.236: > > It registers with a URI without a transport parameter, but it appears to > only receive UDP; it should thus use "transport=udp". > > It does not put a User-Agent header in responses. > > When it receives a BYE to an early dialog (which is valid), it responds > 200 to the BYE but doesn't respond to the INVITE. The INVITE still > needs a response. See section 15 of RFC 32761; section 15.1.2 > recommends a 487 response to an INVITE that is aborted by a BYE. > > Using BYE to terminate an INVITE that has not terminated is probably not > what you want. The BYE can only be sent in regard to a particular early > dialog, and it will cause the INVITE to continue forking. The usual > technique is to send a CANCEL, and then clean up any dialogs that got > established anyway with immediate BYEs.
OK I will try that. I thought that the proxy was issuing a CANCEL and hence I did not have to do so. However, you are correct and I should send a CANCEL and then a BYE. > > Because the UAS hasn't provided a final response to the INVITE at frame > 18, the proxy cancels it 20 seconds later at frame 54. The proxy then > continues forking the call starting at frame 59. > > The bizarre URI sip:[EMAIL PROTECTED] comes from the > redirector incorrectly constructing a voicemail URI for a voicemail URI > (compare frames 59 and 61). Diagnosing this will require a proper > sipx-snapshot, to see how the redirectors are configured. > > The 404 at frame 76 is the final response to the INVITE at frame 2. The > 404 is the chosen final response, ultimately from frame 67. > > Dale Will post a snapshot after I make the CANCEL correction. Thank you for looking at the trace! Ranga > > > -- M. Ranganathan _______________________________________________ sipx-dev mailing list [email protected] List Archive: http://list.sipfoundry.org/archive/sipx-dev Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev
