2009/2/5 Andrew Wood <[email protected]>:
>
> In a call between two phones A & C through one proxy B if B inserts an
> RR into the invite C will rcho it back in the reslonses should or
> shouldnt this be removed when passibg it downstream to A?  The example
> im looming at in Johnstons book shows the proxy removing it from the
> 180 but leaving it in the 200

Imagine the following:

C                                              proxy
                            A

INVITE (require: 100rel) --------->
                                              INVITE (require: 100rel
+ RR) -------->

                                              <-------------------
180 (Contact + RR)
<--------------------  180 (Contact)

PRACK 
---------------------------------------------------------------------------->


If the 180 arriving to C has no RR, then the PRACK will be sent
directly to A instead of the proxy. this is not desirable in cases of
NAT and so.

In RFC3261, a 1XX response doesn't require to contain RR and Contact,
but for using some SIP extensions (as PRACK) it's needed.Other case
would be:

C                                              proxy
                            A

INVITE  ---------------------------->
                                              INVITE (RR)
------------------------------>

                                              <-------------------
180 (Contact + RR)
<--------------------  180 (Contact)

INFO 
---------------------------------------------------------------------------->


This is, C can send an in-dialog request during an early-dialog only
in case C has received a Contact header in the 1XX response. But as
above, in case the 1XX doesn't contain RR then the INFO will be sent
*directly* to the UAS instead of the proxy.

So, I see no reason for a proxy removing RR in 1XX responses. It has
no advantage at all, but some disavantages.

Regards.




-- 
Iñaki Baz Castillo
<[email protected]>

_______________________________________________
Sip-implementors mailing list
[email protected]
https://lists.cs.columbia.edu/cucslists/listinfo/sip-implementors

Reply via email to