[Freeswitch-dev] get rid of mod_loopback for outbound calls

Tue Jul 28 07:58:54 PDT 2009

use mod_curl instead of mod_http....

On Tue, Jul 28, 2009 at 9:50 AM, Anthony Minessale <
anthony.minessale at gmail.com> wrote:

> you could try using a FS api call in your originate that uses mod_http to
> hit your routing logic.
>
> The other way is just not elegant at all and there is no way I can see to
> do it.
>
>
>
> On Tue, Jul 28, 2009 at 3:29 AM, Tamas Cseke <cstomi.levlist at gmail.com>wrote:
>
>> Hello,
>>
>> Thank you for your effort about this issue.
>> Our problem is that searching this routing data from database is quiet
>> painful.
>> Let me explain. we have manual calls from softphones and automatically
>> originated calls pushed into fifo.
>>
>> We have a quiet complicate routing logic setting certain variables (sip
>> headers, caller id number etc...),  select routes according to prefix,
>> time matching and so on.
>> For manual calls we only could accomplish these  routing in dialplan.
>> (we store the data in database  but we generate  xml dialplans from them).
>> For automatic calls  if  we can't use the dialplan we have to implement
>> this routing again (in C module). and every time when a manager or
>> customer would like a new  feature (eg. add
>> different sip headers for calls for different projects) we have to
>> implement it twice.
>>
>> I tried to make a patch to do a dialplan hunt before the real invite for
>> outgoing channels, but it's not very elegant...
>> We were thinking about to make a new endpoint module, but  the problem
>> is that outgoing channel routine create the session, that we need for
>> dialplan hunting, so I can't do it totally independent
>> Could you please advise me a way to do it elegant, or is it a very bad
>> idea?
>>
>> Here is the patch.
>> svn diff -r 14390:14396
>>
>> http://svn.freeswitch.org/svn/freeswitch/branches/cseket/src/mod/endpoints/mod_sofia
>>
>> Thanks in advance,
>> Tamas
>>
>> Anthony Minessale írta:
>> > yah, I looked into that and it's not possible to do.
>> >
>> > Your usage of the dialplan like a database is just not a good way to do
>> it.
>> > I suggest you use a database or something else to store that routing
>> data
>> > like others do.
>> >
>> >
>> >
>> > On Wed, Jul 22, 2009 at 10:55 AM, Tamas Cseke <cstomi.levlist at gmail.com
>> >wrote:
>> >
>> >
>> >> Hello,
>> >>
>> >> we are using mod_loopback for outbound calls to be able to do a
>> dialplan
>> >> hunt before the actual call origination.
>> >> change caller number, change routing according to the prefix and
>> >> something like that.
>> >> We have talked about a lot, that this is "not very" efficient  because
>> >> we have 2 extra channels (loopbacks) for every call.
>> >>
>> >> I diged into the state machine code and figured out that if I change
>> the
>> >> on_routing state handler functions for switch_ivr_originate to don't
>> >> veto the standard behavior I can perform a dilplan hunt.
>> >> But I still have to originate a channel which makes a dialplan hunt and
>> >> originate another. and this isn't what we really would like to...
>> >>
>> >> I've got the following tip from Anthony earlier:
>> >> 0416 16:39:28 <@anthm> maybe i could make an api command to pretend to
>> >> do a dialplan run and print out a list in inline format of what to do
>> >> 0416 16:39:48 <@anthm> that you could feed to originate inline dialplan
>> >>
>> >> But I don't know how to implement this. I need some explanation.
>> >>
>> >> 1) the dialplan pretender api returns applications in inline format
>> >> 2) we feed it to inline dialplan
>> >> but how could we originate the call, how should the whole process look
>> >> like?
>> >> could you please give me an example?
>> >>
>> >> I was thinking about someting like this, but it doesn't make sense I
>> quess.
>> >> ${originate(call_url inline ${pretend_dialplan_hunt(exten XML
>> context)})}
>> >>
>> >> Thanks in advance,
>> >> Tamas
>> >>
>> >>
>> >>
>> >> _______________________________________________
>> >> 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
>>
>
>
>
> --
> Anthony Minessale II
>
> FreeSWITCH http://www.freeswitch.org/
> ClueCon http://www.cluecon.com/
> Twitter: http://twitter.com/FreeSWITCH_wire
>
> AIM: anthm
> MSN:anthony_minessale at hotmail.com <MSN%3Aanthony_minessale at hotmail.com>
> GTALK/JABBER/PAYPAL:anthony.minessale at gmail.com<PAYPAL%3Aanthony.minessale at gmail.com>
> IRC: irc.freenode.net #freeswitch
>
> FreeSWITCH Developer Conference
> sip:888 at conference.freeswitch.org <sip%3A888 at conference.freeswitch.org>
> iax:guest at conference.freeswitch.org/888
> googletalk:conf+888 at conference.freeswitch.org<googletalk%3Aconf%2B888 at conference.freeswitch.org>
> pstn:213-799-1400
>
> _______________________________________________
> 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
>
>

-- 
-Rupa
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freeswitch.org/pipermail/freeswitch-dev/attachments/20090728/c0e243f4/attachment.html 