RE: [Serusers] Problems with ACK

Hello Jiri, hello Andrei. Thanks for your answers. I made a mistake posting the trace and I missed the ACK with problems. I'm attaching the full trace. (ack_full_trace.cap) Just for you to know, the setting is exactly as Jiri pointed. .100/UAC -----NAT/.248----> SER outbound @ .246 -----> GW @ .239 Can you guide me and tell me in which part of the RFC is this "3261 style transaction matching" rule? Thanks guys for your help. Best Regards, Ricardo Martinez RedVoiss Chile. -----Mensaje original----- De: Jiri Kuthan [mailto:jiri@iptel.org] Enviado el: martes, 27 de marzo de 2007 17:24 Para: Ricardo Martinez; serusers(a)iptel.org Asunto: RE: [Serusers] Problems with ACK I think the key message for problem analysis, client's ACK, is missing. This is the setting, right:? .100/UAC -----NAT/.248----> SER outbound @ .246 -----> UAS @ .239 The ACK in your message dump is only that one sent from proxy (246) to UAS (.239). The ACK which can possibly mismatch and produce the error message you mentioned must come from upstream. To be properly formatted, it must have Via identical to that of initial INVITE ( Via: SIP/2.0/UDP 192.168.1.100:5060;rport=46127;received=100.100.100.248;branch=z9hG4bK2893084087) -jiri At 17:50 27/03/2007, Ricardo Martinez wrote: No. Time Source Destination Protocol Info -- Jiri Kuthan http://iptel.org/~jiri/