[sipcore] IMS use of 305 Use Proxy response???

[Paul Kyzivat <[e-mail]>]

IMS people:

Can you please provide information on how 305 is being used in IMS 
implementations?

Does a recipient of the 305 insert the URI into the Route header and 
retry using the original R-URI? If so, is the new URI *added*, or does 
it replace some existing URI in the Route header? Where in the Route 
header does it go?

For context, if you haven't been following this discussion, the 305 
response semantics are underspecified in 3261. If this is starting to be 
used, then it is becoming more urgent to get some clarification so that 
all use it in consistent ways.

	Thanks,
	Paul

On 12/10/13 10:34 AM, Brett Tate wrote:
>>> As indicated within the following snippet from
>>> draft-rosenberg-sip-route-construct-01, the 305
>>> response is underspecified within RFC 3261.
>>>
>>> What is the expected behavior concerning how to
>>> handle a 305's Contact: replace Request-URI or
>>> utilized as a Route header?  The snippet indicates
>>> that the unstated assumption is that it populates
>>> the Request-URI.
>>
>> I don't know. AFAIK this was never satisfactorily
>> clarified.
>>
>> If we wanted to clarify this, I think the first thing
>> we would want to do is take a survey of any known
>> current uses of 305. (Hopefully there are none,
>> because nobody knows how to use it.)
>
> Someone has started sending 305 within IMS deployments.  I'm not sure if it was because of 3GPP TS 24.229, similar 3GPP specifications, or another reason.  The following is one of the snippets from 3GPP TS 24.229 V10.5.0.
>
> "If the HSS indicates to the S-CSCF that there is already another S-CSCF assigned for the user, the S-CSCF shall return a 305 (Use Proxy) response containing the SIP URI of the assigned S-CSCF received from the HSS in the Contact header field."
>
> Surveying the senders of 305 is potentially as important as how everyone is handling it.  For instance, what are they supplying within the Contact and how are they expecting the 305 to be handled?
>
> Thanks,
> Brett
>
> _______________________________________________
> sipcore mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/sipcore
>