On Tue, Nov 4, 2008 at 3:10 PM, Scott Lawrence
<[EMAIL PROTECTED]> wrote:
>
> On Tue, 2008-11-04 at 10:51 -0500, M. Ranganathan wrote:
>> Hello,
>>
>> I would like to discuss the implications of forwarding REFER to the
>> ITSP. There are some ITSPs that support REFER ( notably bt.com). If we
>> exploited this capability, that could result in a better user
>> experience. For example, this would have the effect of the caller
>> hearing Ringing on blind transfers assuming that the inbound INVITE
>> has an Allow: REFER.  Note that in general the ITSP may not
>> Allow:REFER as it may well be dealing with an endpoint that does not
>> handle REFER ( such as a PSTN phone ) so the user may not hear RINGING
>> on blind transfer anyway.
>>
>> Is XECS-1849 worthwhile? If you are calling pbx to pbx, I can see XECS
>> 1849  being worthwhile.
>
> If you look at INVITES from this and other ITSPs (setting aside the
> pbx-pbx case for a moment), is the presence or absence of 'Supported:
> refer' a reliable indication of whether or not it is supported?


I can experiment with this case. I am not sure about it at the moment.


>
> We probably should not use REFER to an ITSP unless they also support
> Refer with Replaces.


OK


>
>> Questions:
>>
>> Are we better off simulating RINGING using our park server for a
>> uniform user experience or should we forward REFER to the iTSP for
>> In-DIALOG refers received from the PBX?
>
> I don't think we should simulate ringing - there are many worms in that
> can...
>
>>  If so, we also need to be able
>> to handle inbound REFER from the ITSP (which adds more complexity to
>> the code).
>
> Why would we need to handle inbound REFER?  Have you seen one?

I am thinking about PBX to PBX connection via an ITSP  In that case, I
suppose, if I strip the Allow:REFER from the outbound INVITE from
sipxbridge, the ITSP cannot send an inbound REFER legally.


>
>> Is the current solution of turning off MOH on the phone and turning it
>> on for sipxbridge so that MOH plays till pickup for blind transfers
>> adequate for this release?
>
> Yes.
>
>
>



-- 
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

Reply via email to