VICIdial Status and Hopper Issues

[inthegenes]

Good Morning,

I have two issues occuring with my company's VICIdial dialer platform that I am requesting assistance.

We have three VICIdial servers, all running the same version of asterisk, 1.4.30. Two are the same version of VICIdial and VERSION: 2.0.5-173
BUILD: 90320-0424 . And one is VERSION: 2.0.5-174 BUILD: 90522-0506. All installed using VICIdial Now CE 1.3. The latter being upgraded to the new version and build of VICIdial. We are currently using the interfaces for the 2.0.5-173 Version for both the agent interface and the admin interface in our production environment. All three servers are operating at load balance and are set to handle a variety of trunks. We also have a server for database purposes, MySQL version 5.0.77 running on Slackware version 10.1.

One issues that we are experiencing is that we have cases where the dialer, VDAD, is dispositioning calls as NEW. Here is an example record from our database:
7828569 NEW VDAD 3 2010-06-09 14:03:10

Above you will see an exert from our database, the lead id, the status, agent, call count and last local call time. In my experience once a lead is dialled it should no longer be NEW. This issue is widespread and has impacted mostly our calls being dialeld through a CAS trunk (Channelised T1). We have seen evidence of this issue on all of our projects. This issue is impacting our dialer production and our reports.
We are also experiencing hopper related issues on our CAS trunk campaign (local calling via T1), where leads were being duplicated in the hopper at least three times. Which caused leads to be redialled in very short intervals. This confirmed to not be caused by duplicate leads in the list, they were only duplciated in the hopper.

Another issue that we are also having is that the dialer is calling statuses that are not included in the Dial Statuses selected via campaign to be redialled. For example, we will have a disposition or status of JC and do not select it in the campaign's Dial Status listing, however the dialer is calling leads with a status of JC.

Assistance would be greatly appreciated.

[williamconley]

is this a cluster? and if so, what method did you use?

could you be running the clustered scripts incorrectly? (inside astguiclient.conf are specific instructions for which scripts should be run)

if not, you should troubleshoot with a test case and follow one lead through the system (with Debug On, perhaps even DebugX) and see where the error occurs.

[mflorell]

Dispositions of NEW for leads that have supposedly been dialed usually means the phone number cannot be dialed with your dialplan, look at those numbers and see if they fit your dialplan dialout rules.

As for JC, no clue on that one, it is not a default status.

I would suggest upgrading to 2.2.1.

[inthegenes]

William,

I am unable to give you a sound answer as to how they were clustered. this was executed by a consultant. The servers are clustered to one database and load balance configurations in regards to calls and or auto dialling are handled from the Admin interface (Admin/Servers). It is possible that they were not clustered correctly and we will look into that. We will also look into the use of DebugX.

Matt,

The phone numbers can be dialled by the dial plan. The issue only occurs when the campaign is on auto dial. When in manual mode the calls are dialled as intended. We have created different statuses for different campaigns and would only like specific statuses to be redialled after reset, however the Dial Status option on the campaign screen is not restricting what statuses are to be redialled.

[williamconley]

are you using lead filters on the campaigns? (a poorly executed lead filter can screw things up royally).

[rosary]

Good Day!

I am experiencing the same problem.
I have Vicidial Version 2.0.5 installed in CentOS 5.2 with asterisk version 1.2.30.2.

Some leads are disposed as NEW even though they are already dialed by the system.

I also checked the dialplan and made some test call (of the numbers that are already called but tagged as NEW), and it went through.

What might have caused this problem?

Your assistance is highly appreciated. Thank you.

[williamconley]

when you post, please post your entire configuration including (but not limited to) your installation method and OS with kernel or version, vicidial version and build, asterisk version, telephony hardware (model number is helpful here), cluster information if you have one, and whether any other software is installed in the box.

Similar to This:
Vicibox X.X from .iso | Vicidial X.X.X Build XXXX | Asterisk X.X.X | Single Server | No Digium/Sangoma Hardware | No Extra Software After Installation

this IS a requirement for posting along with reading the stickies (at the top of each forum) and the manager's manual (available on EFLO.net, both free and paid versions)
________

It could be any number of things from a bad install to an overloaded server ... can you reproduce the problem in any way?

if you scour your agi logs you may be able to locate one of the calls to one of those leads and find out what happened.

[rosary]

Hi, thank you for the reminders.

Here's my configuration:


CentOS 5.2 (Kernel 2.6.18-92) | Vicidial VERSION: 2.0.5-174 BUILD: 90522-0506 | Asterisk 1.2.30.2 | 1 Dialer, 1 WEB server, 1 Database Server| Wildcard TDM400P | No Extra Software After Installation


Also, i have read the stickies and the manager manual.

Thanks.

[williamconley]

Your installation method (.iso name and version? and perhaps a link to your installation instructions if you used an online instruction set ...) and a link to your cluster instructions (assuming you used instructions that are online) would be helpful.

Reason being that if others with a similar system see this post ... they could help (or if it is in the future, this thread could help them).

Have you been able to find out how often this occurs? (1 call in 100,000? 20 per day out of 10,000? every other call?)

Are they limited to calls on one server? (or are you ONLY using one dialer and not dialing with the other two at all?)

[rosary]

For the installation method, i made the SCRATCH_INSTALL included at the docs folder of the source as my reference. Similar to this one:
http://astguiclient.sourceforge.net/scr ... stall.html


Calls are limited to the DIALER. The other two servers are merely used as WEB and DATABASE servers.

As per my observation, more or less 25% percent of total leads that have been called are tagged as NEW (which are not supposed to be disposed as NEW since they have already been called).

Thank you for your reply.

[williamconley]

while i would love to help directly ... there is a reason CentOS is neither supported nor recommended. Very few people have succeeded in getting VICIdial completely "up and running" on CentOS (gardo is a very special guy!).

You should consider either GoAutoDial or Vicibox or Vicibox Redux.

OR you could try the Vicibox DEMO CD. It does not require installation. Capture the output of your asterisk debug and your agi debug (/var/log/asterisk, controlled by /etc/asterisk/logger.conf and /var/log/astguiclient controlled by admin->servers->logging) on your production server, and then compare to the Demo CD for a similar call.

MY Advice would be a stock install after spending more than an hour trying to resolve it (because stock install takes under an hour for us ... )

Also: If you install Vicibox Redux 3.0.6 ... you'll get the latest stock system with the latest VICIdial with a LOT more toys and gadgets (custom fields? call filtering? call menus?)

[rosary]

Thanks for the advises William, i really appreciate it.