[Freeswitch-dev] Bridge dialplan to sipx IVR services problem
Peter Olsson
peter.olsson at visionutveckling.se
Thu Oct 21 05:21:22 PDT 2010
Just call "deflect" in dialplan, with new destination as parameter (sip:xxx at domain.com)
/Peter
-----Ursprungligt meddelande-----
Från: freeswitch-dev-bounces at lists.freeswitch.org [mailto:freeswitch-dev-bounces at lists.freeswitch.org] För Joegen E. Baclor
Skickat: den 21 oktober 2010 14:12
Till: freeswitch-dev at lists.freeswitch.org
Ämne: Re: [Freeswitch-dev] Bridge dialplan to sipx IVR services problem
I just saw uuid_deflect in the API reference so i guess this is
doable. Is there a means to send the UUID of the bridged call as a uri
parameter by simply using the XML dialplan? If not I guess I have to
create a new socket app just the do the bridging.
On Thursday, 21 October, 2010 10:54 AM, Joegen E. Baclor wrote:
> Hi Anthony sorry for being vague. The call flow i am experimenting on
> looks like this
>
> 1. UA1 -> sipXproxy -> [FS (bridged-dial-plan)] -> [FS
> (auto-attendant-socket dial-plan)]
> 2. [FS (auto-attendant-socket dial-plan]) sends REFER to [FS
> (bridged-dial-plan)]
> 3. B-Leg of [FS (bridged-dial-plan)] consumes REFER and sends INVITE
> to Refer-to URI via sipXproxy
>
> The way sipX does this right now is REFER is actually processed by UA1.
> Having [FS (bridged-dial-plan)] breaks this.
>
> I have reason to believe that the ultimate answer to my previous
> question would be a no. Replicating authentication credentials would be
> costly as well as create a security hole since any UA which has a valid
> extension in the from URI can now make calls through the AA without
> being challenged since FS assumes authentication.
>
> I am thinking that we could still make the old behavior a possibility if
> there is a way to intsruct A Leg of the bridge call perform the REFER
> instead of the auto-attendant leg sending it. Is this possible?
>
> Thanks for you patience.
>
> Joegen
>
>
> On Wednesday, 20 October, 2010 11:07 PM, Anthony Minessale wrote:
>
>> I don't quite get what you are saying but..
>> Can't you define credentials in FS as noreg gateway and match it up to
>> the challenge realm?
>>
>>
>> On Tue, Oct 19, 2010 at 11:06 PM, Joegen E. Baclor
>> <joegen at opensipstack.org> wrote:
>>
>>
>>> Hi,
>>>
>>> Following Anthonys advise to configure a bridged dial plan infront of
>>> sipX socket based apps proves to be a reasonable approach to let
>>> attended transfer work for our apps. However, bridged dial plans for
>>> the auto-attendant has a side effect. sipX uses blind transfer to
>>> forward a caller to its intended destination. sipX proxy would then
>>> authenticate the user if it has proper permission to reach the resource
>>> as it receives to new INVITE as a result of the blind transfer. Since
>>> the call is bridged, b-leg will consume the REFER and send its own
>>> INVITE which doesn't have the notion how to authenticate against sipX.
>>> My question is is there a way to propagate the REFER to the caller
>>> instead of being consumed by the bridge?
>>>
>>> Thanks for any advise in advance.
>>>
>>> Joegen
>>>
>>> _______________________________________________
>>> FreeSWITCH-dev mailing list
>>> FreeSWITCH-dev at lists.freeswitch.org
>>> http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev
>>> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev
>>> http://www.freeswitch.org
>>>
>>>
>>>
>>
>>
>>
>
> _______________________________________________
> FreeSWITCH-dev mailing list
> FreeSWITCH-dev at lists.freeswitch.org
> http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev
> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev
> http://www.freeswitch.org
>
>
>
_______________________________________________
FreeSWITCH-dev mailing list
FreeSWITCH-dev at lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev
http://www.freeswitch.org
!DSPAM:4cc02ff332932087729767!
More information about the FreeSWITCH-dev
mailing list