[Asterisk-Users] Re: iax2-jitter-trunking?

[Rich Adamson]

> > Looking at sniffer traces of two simultanous calls, its apparent that two
> > calls use two different udp packets. However, since I'm routing the two
> > calls via simple extensions.conf entries without reference to an iax
> > context, I was wondering if that _might_ be the root cause for this.
> > (Or, could it be that jitterbuffer=yes disables trunking.)
> 
> For trunking to work, you will need to place the calls via a defined IAX 
> peer on the calling host (that has trunk=yes specified), receive the 
> calls via a defined IAX user on the receiving host (that also has 
> trunk=yes specified) and be using a codec that supports trunking (I 
> don't believe G.711 supports it, which is probably what you are using).

Been using strictly iax-gsm, no 711.

A test conducted with morning with an unattended remote * box indicated
that sending calls (via the context method) did in fact trunk two within
the same packet, but the responses came back in separate packets. Not 
sure I understand that piece as the return path should not be context
sensitive on that remote box. (Of coarse, maybe testing to a milliwatt
generator on the remote box isn't a valid test either. Will have to
wait for an on-site person for additional human testing.)

Given that two calls from A -> B are trunked using the context reference,
why would the return path for those two calls not be trunked?