vicidial.org

First things first,

GoAutoDial CE 2.1 | Vicidial 2.4-309a Build 110430-1642 | Asterisk 1.4.39.1 | Single Server | No Digium/Sangoma Hardware | No Extra Software After Installation

I have recently installed, and everything is working fine. When the call connects, everything works. The calls are clear, coming through steady, everything's great.

My issue is that when I get:

Code: Select all

[Sep 19 15:18:53]     -- Executing [h@default:1] DeadAGI("Local/919053837588@default-0d2f,2", "agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----0-----CANCEL----------") in new stack
 [Sep 19 15:18:53]     -- AGI Script agi://127.0.0.1:4577/call_log--HVcauses--PRI-----NODEBUG-----0-----CANCEL---------- completed, returning 0

my calls are getting "stuck." The line isn't being reopened to dial the next call, and that is causing a severe problem with dial efficiency. The call can be seen in vicidial_auto_calls until what I assume is some sort of standard timeout. This eventually goes away, but can take more than a minute and probably averages around 60-70 seconds. The same thing happens with BUSY.

To clarify, any time I get BUSY or CANCEL as a result, the call remains as though it’s being dialed until it times out.

Specificically, in PHPmyadmin I can pull up vicidial_auto_calls and see the call remain despite the fact that asterisk has returned BUSY or CANCEL.

This is my dialplan (xxxxxx = my info, obviously)

exten => _91XXXXXXXXXX,1,Set(CALLERID(name)=xxxxxx)
exten => _91XXXXXXXXXX,2,AGI(agi://127.0.0.1:4577/call_log)
exten => _91XXXXXXXXXX,3,Dial(SIP/${EXTEN:1}@xxxxxx,,tTo)
exten => _91XXXXXXXXXX,4,Hangup

The only modification I have made is adding set CALLERID(name) which is working as I am in Canada. It has been confirmed to work, but I do not know if it may have some part in this problem. Frankly, I am quite new to the whole self-hosted dialer thing, and I have done everything I can think of to resolve this issue, so any help would be greatly appreciated.

I will be monitoring this thread fairly closely, so if I have left out any important information, let me know and I will respond ASAP.

from viewtopic.php?f=4&t=25881

imisner wrote:Hey DomeDan,

Would this apply to my issue here: viewtopic.php?f=7&t=25891

I need to resolve it as we are a small center with only 3 agents. The problem is more dramatic on a smaller group of agents, as a few built up in vicidial_auto_calls can happen more frequently on so few lines.

Thanks

EDIT: More specifically, does your patch include a fix for CANCEL? Because this issue is most prevalent for me when I'm using a dial timeout low enough to prevent excess answering machines, and the dial timeout returns CANCEL for me.

In short, no, my patch is just for CONGESTION.

And I have not seen them stuck as long as 60-70 seconds, only ca 5 seconds, try editing "drop timeout seconds" and see if that makes any difference

I have modified that a few times in trying to get it to work. My plan right now is applying both your patches, including the congestion one, and the one found here:
viewtopic.php?f=4&t=25060#p86411

I'm new to this whole thing, and it's a lot of trial and error for me. I don't think there could be any adverse effects, and I'm working on overall efficiency anyways.

After applying your patches I broke it.

Can anyone help me out with a bit of a walkthrough to have calls removed from the system after a CANCEL or BUSY result? I did a full reinstall thinking that perhaps this was an issue I had caused, but it's still going on.

To reiterate: if an agent receives a call everything is fine and the dialer immediately dials the next one upon disposition, but the busy or cancel results cause a call to get "stuck" for quite some time. This results in a severe dial efficiency loss.

You are using an 18-month old version of vicidial, why not upgrade to the latest svn/trunk?

Sounds like a plan. Thanks, I'll try it today.

I've tried a fresh install of Vicibox 4.0.0, completely wiped everything and started over and used the install from: viewtopic.php?f=8&t=25920

I'm still having the same issue. The CANCEL result is leaving calls in the vicidial_auto_calls which are timing out at the same rough length of time as before.

The cancel result is being returned whenever the DIAL TIMEOUT setting for the Campaign ends a call, or when the call is made to a disconnected number (the quality of my list is generally quite low right now)

Is there a way to have that table cleared more frequently so they don't get "stuck" or to have the cancel result end a call properly?

We have a patch that will clear cancels and congestion immediately. But that can be dangerous depending on your system. Does work nicely, though.