vicidial.org

Hi Vici's

We've found on occasion the called_since_last_reset is blank for some records.
We've found no commonality for the reason that this field is blank.

It would seem that a script may be running to remove this value from the called_since_last_called field, but for what reason?

Check vicidial_list and see if you can find some commonality for this field called_since_last_called being blank.

Any insight would be appreciated.

Single server
VERSION: 2.6-387a
BUILD: 121130-1425

That's a new one for me. I don't think I've ever seen that. If you can replicate it please submit a bug report and we'll try to figure out why it's happening.

That's a new one for me. I don't think I've ever seen that. If you can replicate it please submit a bug report and we'll try to figure out why it's happening.

We've seem to narrow down the commonality to where called_since_last_reset reaches Y10

This file: VICIDIAL_last_local_call_time_UPDATE.pl appears to have the code to set the called_since_last_reset to blank.

This code below seem to validate that and then set the leads called_since_last_reset
In this section: foreach(@lead_id)

if ($calculate_recycle_counts > 0)
{
if ( ($Ncount[$b]==0) or ($Ncount[$b]=='') ) {$Ncount[$b] = 'N';}
if ($Ncount[$b]==1) {$Ncount[$b] = 'Y';}
if ( ($Ncount[$b]>1) && ($Ncount[$b]<11) ) {$Ncount[$b] = ($Ncount[$b] - 1); $Ncount[$b] = "Y$Ncount[$b]";}
if ($Ncount[$b]>10) {$Ncount[$b] = 'Y10';}
$called_since_last_resetSQL = ",called_since_last_reset='$Ncount[$b]'";
}
else
{$called_since_last_resetSQL = "";}

$stmtA = "UPDATE vicidial_list set last_local_call_time='$NEWcall_date' $called_since_last_resetSQL where lead_id='$lead_id[$b]';";

Do you actually use that script? I don't think it's been touched in over 5 years.

We don't use that script. Just grepped all perl scripts for the possibility of an update to that field called_since_last_reset.

What other script could possibly be setting that field to a blank value?

Out of the dozens of clients I checked(across hundreds of millions of leads) I only found a dozen or so examples. They all involved MANUAL dial outbound calls. I don't really have a way to test for this because it is incredibly rare on the systems we manage(less than one occurrence per one million leads), and I don't see any patterns in the log data of the few that I have been able to find.

An easy fix for this would be to just add a SQL to the end of day process that would turn these into called_since_last_reset='Y'.

update vicidial_list set called_since_last_reset='Y' where called_since_last_reset='' or called_since_last_reset is NULL;

I am hesitant to add a band-aid like this because it is such a rare thing, but if we can find out more about this then maybe a better solution will present itself.

How frequent is this happening on your systems?

13000 out of 1.8 million records.
The band aid is not ideal but functional. We'll engage that for the time being.

How many of the records did you find that were at a call count of greater than 10?
Should we pick this up on Mantis?
-Noah

About half were over 10 called count. Yes, this should be added as a ticket in Mantis, but I won't be fixing it until we can reproduce the problem on demand.

About half were over 10 called count. Yes, this should be added as a ticket in Mantis, but I won't be fixing it until we can reproduce the problem on demand.