VoIP Mailing List Archives

[matt at hellohunter.com]

I apologize if this is a double post to -dev. I'm not sure why I don't see my message appearing, so I'm going to try again in the -user list (first timer posting here .

I have a situation where it would be useful for a caller not to be hungup, after finishing the "fifo in" execution (when the agent disconnects the call or the agent hangs-up). The caller is automatically hungup, in this situation. It would be preferable if the caller channel went further along the dial plan.  I thought I might get lucky implementing this setting with hangup_after_bridge to false, but fifo does not utilize this variable.

I tried looking thru the mod_fifo.c source, but my c skills are minimal. I also tried executing fifo in a lua app and setting setAutoHangup(false), but that also did not work. Any chance this could be done as a feature enhancement? Thanks.


--matt

[mattdfong at gmail.com]

I apologize if this is a double post to -dev. I'm not sure why I don't see my message appearing, so I'm going to try again in the -user list (first timer posting here .

I have a situation where it would be useful for a caller not to be hungup, after finishing the "fifo in" execution (when the agent disconnects the call or the agent hangs-up). The caller is automatically hungup, in this situation. It would be preferable if the caller channel went further along the dial plan.  I thought I might get lucky implementing this setting with hangup_after_bridge to false, but fifo does not utilize this variable.

I tried looking thru the mod_fifo.c source, but my c skills are minimal. I also tried executing fifo in a lua app and setting setAutoHangup(false), but that also did not work. Any chance this could be done as a feature enhancement? Thanks.


--matt

[anthony.minessale at g...]

there is a patch in jira that will implement this feature about to be added


2009/3/17 Matthew Fong <mattdfong@gmail.com (mattdfong@gmail.com)>

[mattdfong at gmail.com]

Hi Anthony, thanks for the reply.

I've searched thru jira, and didn't find anything when searching for fifo that was recently updated or related, except 


http://jira.freeswitch.org/browse/MODAPP-189


and I'm not sure if this does what I need. Was this what you were referring to? Thanks.


--matt

2009/3/17 Anthony Minessale <anthony.minessale@gmail.com (anthony.minessale@gmail.com)>

[mattdfong at gmail.com]

I upgraded to 

FreeSWITCH Version 1.0.trunk (12654M)


but caller is still being hungup (and not continuing on with dialplan) after agent disconnect with hangup_after_bridge=false


Is there a separate patch I need to apply? Thanks.


--matt

On Wed, Mar 18, 2009 at 11:39 AM, Matthew Fong <mattdfong@gmail.com (mattdfong@gmail.com)> wrote:

[anthony.minessale at g...]

This is the patch

http://jira.freeswitch.org/browse/MODAPP-237

it's not added yet.


2009/3/18 Matthew Fong <mattdfong@gmail.com (mattdfong@gmail.com)>

[mattdfong at gmail.com]

Hi Anthony,

I installed the patch, but I don't think it accomplishes what I want.


I want the opposite, I want the fifo caller to continue along with the dialplan after the agent (consumer) is finished with servicing the call. This might be useful in situations where there could be an IVR recording customer satisfaction results of the agent servicing the call. As is, FS ends the caller's channel after finishing up in the fifo (ie, agent (consumer) disconnects or hangsup)--there should be life after s/he has been serviced by an agent (preferably continuing on in the dialplan).


If I'm confused and missing something obvious, please correct me... Thanks


--matt




2009/3/19 Anthony Minessale <anthony.minessale@gmail.com (anthony.minessale@gmail.com)>

[anthony.minessale at g...]

The agent could transfer the caller to another extension.


2009/3/19 Matthew Fong <mattdfong@gmail.com (mattdfong@gmail.com)>

[mattdfong at gmail.com]

Hi Anthony,

I'm trying to use fifo in a different sense. Instead of using it for inbound call queing, I'd like to use it for outbound call making. So instead, my agents are "waiting" in the que, and once an outbound call is connected, the "caller" will take an agent out of the que.


So, in my case, the Fifo agent, would not be able to transfer the call because it's an outbound call, and the phone number on the other side is that of a non-employee.


Fifo works a little smoother this way, because in reality, for outbound call making to an agent, this is what's happening, not vica versa. How difficult would this be to implement? Thanks.


--matt

2009/3/20 Anthony Minessale <anthony.minessale@gmail.com (anthony.minessale@gmail.com)>

[mattdfong at gmail.com]

Also, I would not be able to have a hang-up script do it, because in this scenario, the fifo consumer could hang-up at any time without any prior warning--otherwise, I could just transfer the fifo caller out before the fifo agent hangsup...but there is no prior warning

--matt

On Fri, Mar 20, 2009 at 8:04 PM, Matthew Fong <mattdfong@gmail.com (mattdfong@gmail.com)> wrote:

[anthony.minessale at g...]

Even though it's an outbound call if your agent uses his sip phone to blind transfer the caller "customer",
The customer call will change the the routing state and hunt in your local
dialplan just like it was an inbound call.  That's how blind transfer was designed to work.

If your agent is not using a sip phone, you can use bind_meta_app to make *N (where N = 0-9) to trigger a software blind transfer.
 

2009/3/20 Matthew Fong <mattdfong@gmail.com (mattdfong@gmail.com)>

[mattdfong at gmail.com]

Hi Anthony,

Thanks for explaining blind transfer for me.


The issue is that the fifo caller (my employee/agent, fifo in), gets hung-up on when the fifo consumer (an outside line to another party, fifo out) hangs up. I think this is because fifo was written under the assumption that the first in first out would always be a caller, and the agent would consume a caller.


In my case, the roles are reversed, and there's no option to prevent the hangup of the caller.


If the fifo caller (my employee/agent) could somehow know when a fifo consumer (my outside line to another party) was going to hangup, s/he could blind transfer out to save his/her connection from being hung-up, but unfortunately people don't always tell you before hand they are going to hangup.


Right?!?!?! Thanks.


--matt

2009/3/20 Anthony Minessale <anthony.minessale@gmail.com (anthony.minessale@gmail.com)>

[anthony.minessale at g...]

I added a feature to latest trunk where you can set the variable transfer_after_bridge=xyz
where xyz is an extension you want to transfer to when the call ends assuming it's not hungup yet.

if you set this in your dialplan before entering fifo it will allow you to be transfered to the desired ext when you are done.

if you need to specify a dialplan and context you cana add that too

<extension>[:<dialplan>][:<context>]



2009/3/20 Matthew Fong <mattdfong@gmail.com (mattdfong@gmail.com)>