vicidial.org

Asterisk 1.4 (installed from ViciBox Redux V 3.1.9 ISO)
VERSION: 2.4-324a
BUILD: 110707-1402
Single Server (HP XW8400)
No Digium/Sangoma Hardware
No Extra Software After Installation

Processors
4 (2 physical)
Model
Intel Xeon 5160
CPU Speed
3.0 GHz
Cache Size
4M Cache
-----------------------------


Hi guys,

One of my campaigns has leads in the hopper, call times 24hours, dial type set to Ratio with a factor of 2... but strangely enough it's just not dialing.

I have another campaign going on simultaneously that's auto-dialing fine and placing 16-20 calls at once... no issues there. But this one will only work with Manual dial.

Has anyone else encountered this?

How many "Ready" agents do you have on that campaign? How do you know it's not Dialing?

Asterisk CLI Output with ONLY that campaign live (shut down all other traffic)?

AGI output? (/var/log/astguiclient or agi debug in asterisk)

Just a possibility. I have not had time to check this out fully.

We have recently been having a problem with a campaign that has been running fine for 6 months and has started having similar issues.

It could be to do with trunks setting in Admin -- Servers part of the vicidial admin. It seems that if you have one campaign that is active with say a dial level of 4 and 10 agents l0gged in, vicidial sees that as needing 40 trunks and irrespective of if that campaign is using the 40 trunks because it may need 40 if your trunk setting in Admin -- Servers is 40 or below it will not allow any other campaigns to dial.

We have the campaign we have been using for 6 months and a new campaign which is a voice broadcast campaign which has a remote agent configured with 10 lines and a dial level of 13 therefore it will make 130 calls and try to put them through to agents. Now I have noticed that it seems that if that campaign is active, even if the dialable leeds have run out so it is not actually dialing anything, it will stop all other auto-dial campaigns on the server from working.

As soon as I made this campaign inactive it seemed to resolve the issue. I have not had time to properly test this but it is what appears to be happening.

Just an idea and may shed some light onto your issue,

Regards

fibres wrote:Just a possibility. I have not had time to check this out fully.

We have recently been having a problem with a campaign that has been running fine for 6 months and has started having similar issues.

It could be to do with trunks setting in Admin -- Servers part of the vicidial admin. It seems that if you have one campaign that is active with say a dial level of 4 and 10 agents l0gged in, vicidial sees that as needing 40 trunks and irrespective of if that campaign is using the 40 trunks because it may need 40 if your trunk setting in Admin -- Servers is 40 or below it will not allow any other campaigns to dial.

We have the campaign we have been using for 6 months and a new campaign which is a voice broadcast campaign which has a remote agent configured with 10 lines and a dial level of 13 therefore it will make 130 calls and try to put them through to agents. Now I have noticed that it seems that if that campaign is active, even if the dialable leeds have run out so it is not actually dialing anything, it will stop all other auto-dial campaigns on the server from working.

As soon as I made this campaign inactive it seemed to resolve the issue. I have not had time to properly test this but it is what appears to be happening.

Just an idea and may shed some light onto your issue,

Regards

Thank you for the reply fibres! That's exactly what the problem was.

For some reason the ViciBox Redux had by default limited the trunks to a maximum of 23.

Ya know, that actually sounds like a bug. So you're saying that if your system is trying but failing (no leads) to get 50 lines for a campaign ... that can block other campaigns from acquiring needed lines? Have you reproduced this with the latest release and/or SVN?

Thats what seems to be happening. It was a suspicion I had, though weirdly the first time I had the issue doing a myisamchk on the database and a reboot fixed the issue, though it is possible I had disabled the campaign causing the problem in the mean time and not realised.

I only had a suspicion so I thought I would post it and it seems to have been partly confirmed by space janitors comments.

I am not running quite the latest Trunk but fairly recent version. I will make some more tests later and post back my results. Including the output of some of the agi scripts when this is happening!

Regards

I have done a couple of things tonight.

Below is the output from VDADauto screen session.


With campaign disabled

loop counter: |5743448|
TIME DEBUG: 0.00|1|1| GMT: 18:41
2011-08-10 19:41:49|SERVER CALLS PER SECOND MAXIMUM SET TO: 100 |10||
2011-08-10 19:41:49|LIVE AGENTS LOGGED IN: 10 ACTIVE CALLS: 0|
2011-08-10 19:41:49|OLD TRUNK SHORTS CLEARED: 1 |'','Prefixin'||
2011-08-10 19:41:49|Prefixin 109.200.25.35: agents: 10 (READY: 10) dial_level: 30 (10|0|0) 0|
2011-08-10 19:41:49|Prefixin 109.200.25.35: Calls to place: 0 (0 - 0 [0 + 0||]) 0 |
2011-08-10 19:41:49|CAMPAIGN DIFFERENTIAL: 10 10 (10 - 0)|
2011-08-10 19:41:49|LOCAL TRUNK SHORTAGE: 0|0 (0 - 130)|
2011-08-10 19:41:49|Prefixin 109.200.25.35: CALLING|
TIME DEBUG: 0.00|1|1| GMT: 18:41
2011-08-10 19:41:49|| logindate UPDATED 1|'Prefixin'||


With Campaign Enabled, Remote agents logged in so based on dial level it would be making 130 calls if we had enough leads, but with no leads in the hopper.

loop counter: |5743408|
TIME DEBUG: 0.00|1|1| GMT: 18:43
2011-08-10 19:43:29|SERVER CALLS PER SECOND MAXIMUM SET TO: 100 |10||
2011-08-10 19:43:29|LIVE AGENTS LOGGED IN: 10 ACTIVE CALLS: 0|
2011-08-10 19:43:29|OLD TRUNK SHORTS CLEARED: 1 |'','Prefixin'||
2011-08-10 19:43:29|Prefixin 109.200.25.35: agents: 10 (READY: 10) dial_level: 30 (10|0|0) 0|
2011-08-10 19:43:29|Prefixin 109.200.25.35: Calls to place: 130 (300 - 0 [0 + 0||]) 130 MVT override: 130 |130 170| OTHER LIMIT override: 130|
2011-08-10 19:43:29|CAMPAIGN DIFFERENTIAL: 10 10 (10 - 0)|
2011-08-10 19:43:29|LOCAL TRUNK SHORTAGE: 170|170 (130 - 130)|
2011-08-10 19:43:29|Prefixin 109.200.25.35: CALLING|

I suspect it has todo with this local trunk shortage which to me says it is classing 130 trunks as being in use when the campaign is activated, even if the campaign is not making calls.

Regards

can you duplicate that with a single agent on another campaign making one call (until this shortage uses up all the trunks). Or does the other campaign continue with no effect? (Gotta test)

Test both ways: Try to CONTINUE another campaign and try to START another campaign while the shortage is in place.

Hi William

I am not sure exactly what your asking here.

Whats the difference between Continue another campaign and start another campaign?

Regards

Hi William

What exactly would you like me to test here?

I am eager to help resolve this issue if indeed it is a bug.

Regards

Try it with two campaigns running (and competing for the lines). Also try it with a single campaign running successfully, then start another campaign and see if competition begins.

Hi William

I have 2 campaigns.
No 1 is a live campaign with live agents. Today we have 15 agents online with a ADAPT_TAPERED method. With a current dial rate of 5.5 with Available only Tally switched on.

No 2 is a broadcast campaign which has a remote agent with 10 lines and a dial level of 13 using RATIO method.

If I have the No2 campaign active but with no loads to call and then start campaign No1 campaign No1 will fail to make any auto calls.

If I have the No1 campaign active with the above agents logged in and happily making calls everything is fine. If I then activate the No2 campaign it causes the No1 campaign to stop placing any more calls once the No2 campaign has been activated.

It seems for some reason the no2 campaign seems to take precidence and takes all lines.

Regards

10 lines and a dial level of 13

130 lines perpetually in use?

Yes when the campaign is actually active and has numbers in the hopper.

However this is causing issues even when there are no dialable leads on the campaign and there are no leads in its hopper.

Regards

so:

start campaign 1 & 2 and see what happens (you've done this, and campaign 2 wins)

start campaign 1. let it run ... then add campaign 2: does campaign 2 immediately shut down campaign 1 even if campaign 2 has NO leads at the moment it is started?

have you tried this in ratio AND adapt? (for both campaigns?)

Yes with campaign 1 running dialing numbers hapilly. Start campaign 2 with no leads in hopper and it shuts down campaign 1 from dialing.

I have not tried this on different methods. Just with Adapt on campaign 1 and Ratio on campaign 2.

I will try playing with switching methods later and post results.

Regards

ratio and adapt use completely different agi scripts. so it may be that one works where the other does not. seems like you've found a bug ... get precise details and post to the issue tracker. just the simplified version where you start a campaign with no leads and it shuts down a perfectly good campaign.

but first: upgrade to the latest code (not likely to fix this bug for you ... if it has already been fixed in the latest code, right?)

RATIO and ADAPT outbound campaigns use the exact same AGI scripts.

Hi guys.

I will do an upgrade to latest codebase over weekend. I am about due to upgrade as I try and do it every 2 months or so.

I will then run some more tests and post results. If I'm still having trouble I'll post it to the tracker.

Regards

mflorell wrote:RATIO and ADAPT outbound campaigns use the exact same AGI scripts.

whoa.

12799.ASTVDadapt (Detached)
2977.ASTVDauto (Detached)
AST_VDadapt.pl
AST_VDauto_dial.pl

Is that to imply that the two screens running and the two pl scripts are identical or that they are not agi scripts or that they "view only" and some other script is actually controlling?

Good to know.

Those are not AGI scripts, those are Perl scripts. Both of those perl scripts are used by both RATIO and ADAPT outbound campaigns in some similar and some different ways.

gotcha. so it was just that i used "agi" in my rant. i should have just said scripts. i hate when that happens. LOL

Hi Guys

I have been doing some further testing and this is defiantly an issue.

It happens on both Ratio and ADAPT

It also happens if the campaign is enabled with leads in it but with the call-time setting set to not work that hour. It is still disabling all other campaigns on the system.

How do we look to get this fixed?

Regards

Anyone got any ideas on this?
It is causing me some major issues.

Regards

Can someone please confirm if this is a bug or not? If so how do I go about reporting it in the bug tracker?

Regards

Are you using a filter?

Do you get any errors when you run the script that are in the screens manually with the --debugX flag?

Hi Matt

No there is no filter.

All scripts are running. If I disable the campaign that has a remote agent which allows 160 lines to be dialed if there is data there. Everything works.

However if I enable the campaign that has the ability to make 160 calls it blocks all channels and will not allow other campaigns to dial, even if the campaign has no leads in it. Because it could be making 160 calls if it had leads it blocks the whole system up.

Regards

Have you tried adding trunk reservations on your server? (you do this from the Server Modification screen in admin.php)

Hi Matt

I dont think this would help.

The 2 campaigns never run at the same time.

Campaign 1 is a normal campaign which has staff on it from 11am till 7pm.

Campaign 2 is a broadcast campaign which plays a recorded message on connection which runs in the morning from 9am-11am I want to be able to have this campaign set with a call time running from 9am and switch off at 11am. However when the calltime removes the leads bacause the time has finished vicidial still keeps the trunks locked as if campaign 2 had leads it would be making 160 calls. It therefore does not allow campaign 1 to work as it thinks all the trunks are in use.

Regards

Bump

Hi Matt

I dont think this would help.

but have you tried it?

No I have not tried it.

I dont understand how this would help. They are 2 campaigns that run at different times to eachother.

The issue seems to be that the first campaign is reserving the trunks it would need if it had and calls to make even though it does not. I have no such reservations set in trunk reservations.

Regards

Hi Fibres

I am facing the same issue if i start another campaign the autodailing stops but there are leads in hopper still not dailing
Did you got any solution
Please check

In the real time screen is "Trunk Short/Fill". Are there numbers other than Zero in these fields?