On Sat, Sep 6, 2008 at 6:00 PM, Scott Lawrence
<[EMAIL PROTECTED]> wrote:
>
> On Sat, 2008-09-06 at 15:34 -0400, M. Ranganathan wrote:
>> Perhaps the issue is that the park server is only able to
>> handle a single dialog per call. Perhaps a solution would be to
>> provide the park server extension to sipxbridge as a configuration
>> parameter and avoid using the special id ~mh~~.  Then I can tear down
>> the original dialog with the park server for the given call ID and
>> re-establish it when I see the REFER. In this case, it would take a
>> little additional configuration information i.e. you MUST have the
>> park server enabled and provide its extension.
>>
>>
>> What is the best way to proceed on this?
>
> There can any number of arbitrary park orbits, so I don't think that
> providing sipXbridge with information about them is the answer.  More
> likely the fix is to get the park server to recognize that it's got two
> media sessions to one target.
>
>
>

There seems to be something else the matter. I turned off Music On
Hold for sipxbridge so there will only be a single dialog with the
park server and yet the Park Server will not wake up after the
timeout. Looking at a well behaved scenario ( ie. when the phone Parks
the call ) versus what sipxbridge is generate ( i.e. send INVITE to
Park Server extracting relevant information from the REFER as I cannot
transfer the REFER to the ITSP ), I notice only one apparent
difference:

sipxbridge sees an Inbound Refer-To as follows:

Refer-To: <sip:[EMAIL PROTECTED]>

It extracts the X-sipx-auth-identity and sends it forth in the
subsequent INVITE bound to the park server :

x-sipx-authidentity:
<sip:[EMAIL PROTECTED];signature=48C33E76%3A%3Ad433f73567f92a8e63347cab8bf8def8>

Note here that the header has been decoded. I am not sure if this is
the right thing to do.

In contrast, I took a look at a successful park, unpark scenario and
the corresponding header there looks like this:

X-Sipx-Authidentity:
<[EMAIL PROTECTED];signature=48C29A96%253A%253Ade81a79f34929caeaafabd84561f04f9>

Note that the stuff in the <> is not fully decoded. sip%3Auser2
versus sip:user2

Not sure if this has any relevance to the problem at hand.


Ranga







-- 
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