[Asterisk-Users] Fast AGI

Chris Stratton laststraw at gmail.com
Wed Oct 27 11:43:59 MST 2004


You are right that conventional tools for http don't make sense in
telephony. My thoughts were more along the lines of not wanting to
recreate a robust production worthy input/output stream handler that
has been done a million times.

Currently, I am writing a "Connector" for Tomcat that will handle the
"AGI" protocol. If the AGI protocol had been written to just follow
standard http I would not have to write the Connector. An option
instead of the connector is to just proxy the AGI request wrapping it
with http. Really, I could probably just get by with Prepending a "GET
/<somepath>"

Ultimately, I am just trying to pull it into a java servlet
environment that I am comfortable with.


On Tue, 26 Oct 2004 18:07:57 -0500, Steven Critchfield
<critch at basesys.com> wrote:
> On Tue, 2004-10-26 at 16:42 -0400, Chris Stratton wrote:
> 
> 
> > I have been working with FastAGI. I really like it and the idea. But I
> > was curious why the decision was made to not follow a standard http
> > protocol approach? If it had been http, I could then use all the
> > conventional tools that already exist to maintain state, grab name
> > value pairs or any number of things.
> 
> Probably because the conventional tools for http don't make sense in
> telephony? I do not see why I would care about sessioning on a phone
> call. Sessioning is to take care of lack of state, but we have state.
> 
> BTW, be more patient next time before reposting your message. Not all of
> the people here who could give you a definitive answer live on their
> computers.
> --
> Steven Critchfield <critch at basesys.com>
> 
>



More information about the asterisk-users mailing list