vicidial.org

I Have 10 Agents to dialing.
my agents are daily waiting 2+ minutes for one live call

Please any one help me to solve..!

1) Welcome to the Party!

2) As you are obviously new here, I have some suggestions to help us all help you:

When you post, please post your entire configuration including (but not limited to) your installation method (7.X.X?) and vicidial version with build (VERSION: 2.X-XXXx ... BUILD: #####-####).

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)

You should also post: 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. If your installation method is "manual/from scratch" you must post your operating system with version (and the .iso version from which you installed your original operating system) plus a link to the installation instructions you used. If your installation is "Hosted" list the site name of the host.

If this is a "Cloud" or "Virtual" server, please note the technology involved along with the version of that techology (ie: VMware Server Version 2.0.2). If it is not, merely stating the Motherboard model # and CPU would be helpful.

Similar to This:

Vicibox X.X from .iso | Vicidial X.X.X-XXX Build XXXXXX-XXXX | Asterisk X.X.X | Single Server | No Digium/Sangoma Hardware | No Extra Software After Installation | Intel DG35EC | Core2Quad Q6600

3) Of course you can merely increase the Auto Dial Level, but then you may have other factors that interfere such as max trunks in the dialer and max trunks from your carrier. Try it. See if 5 or 6 is better. But ...

4) This is not a magic pill moment. You should understand that every setting in the Campaign and System Settings and Admin->Servers is all about getting agents on the phone with prospects. They all have some sort of effect and each situation is different. So: Read The Vicidial Manager's Manual. Front to back. Don't skip anything. Go over every setting in the Campaign Detail view and click on the help buttons. Probably best to make a copy of the campaign before you make any changes to this one, so you can revert to the original settings after you compare results from any changes.

5) If AMD is on, turn it off (this is controlled by the Campaign Routing Extension in Campaign Detail view). Note that turning this off will likely cause less people to hang up immediately after answering (4 seconds of dead air pisses off the general public, and that's the basic result of using a program to listen to the front of the call and decide if it's an answering machine or a human: dead air!).

my viciBox and asterisk version

- ViciBox v.7.0.4-170113

vicibox7:~ # asterisk -rx "core show version"
Asterisk 11.25.1-vici built by abuild @ cloud107 on a x86_64 running Linux on 2017-01-08 07:00:49 UTC

vicibox7:~ # uname -a
Linux vicibox7 4.1.36-44-default #1 SMP PREEMPT Mon Dec 12 21:24:10 UTC 2016 (b1c0ac2) x86_64 x86_64 x86_64 GNU/Linux

Your Vicidial Version and Build are located on the bottom left corner of most Admin screens. That's the single most important bit you can post.

Your SVN revision level (also useful) is available in your System Settings page.

You posted your CPU type, but not the CPU count. There are 4 cores in each CPU ... but do you have 1 or 2 CPUs? (doesn't really make a difference, but is really the only thing of value in the cpu information). Also useful to know if hyperthreading is on.

Code: Select all

lscpu

CPU socket(s):
Thread(s) per core:

vicidial
VERSION: 2.14-583a
BUILD: 161226-2224
© 2016 ViciDial Group

Cpu info



and I am using vicidial via cloud

Which dial method are you using? How many lines? What do you see in the realtime screen?

I am Using RATIO dial method with 15 lines
200 in hopper, 20% Drop level and Next Agent Call is longest wait time
our agents getting call one by one every 2 mints.
the placed call count is avg 15-25 when 15 agents logged in.!

200 in hopper is a bit vague. Does the hopper ever become empty during your dialing?

no, while dialing time leads in hopper is above 500

what's your campaign routing extension?

It is 8369

change it to 8368

when I am set 8368 the answering machine calls will occur. so I was setting this 8369 for detecting/filtering the answering machine calls.

AMD uses A LOT of CPU power. If you have a lot of answering machines in your lists, the dial ratio will need to be increased to get calls past the AMD check to actual agents. But ...

If the dial ratio gets too high (like ... 15, which is very high) it's likely you'll overload your server.

You can't have it both ways. Either you want the agents busy (some will be clicking "answering machine" instead of talking to people) or you can have your server intercept those calls, but then you have to increase the dial level to keep the agents busy.

Check average server load here:

Code: Select all

uptime

It should also be available at the bottom of the Real Time Screen. Your "average server load" should NEVER exceed the number of cores your server has (EG: Average Server Load 4.0 / 4.0 / 4.0 for a 4 Core System is MAX). It's generally best to keep it at half that (2.0/2.0/2.0 for a 4 core system). If you exceed half, you should look at adding another server (multi-server vicidial is cool!) to allow sharing the load between servers.

In the end, however, even those who answer end up being very short calls because AMD takes a minimum of Two Extra Seconds of Silence (plus the existing 1 to 1.5 seconds inherent in Vicidial: Total up to 3.5 second of DEAD AIR). 3.5 seconds of Dead Air = pissed of prospects. And your agents have to deal with that at the onset of every call OR you can turn AMD off and avoid pissing off the prospects so your agents have a fighting chance of getting through their script. But then the agents will also have to push "A (Answering Machine)" a lot.

AMD = pissed off prospects, overloaded CPU, very high dial level.
NO AMD = push "A" a lot

And don't forget to check your CPU for overload.

the load avg of my cpu. Is it normal or bad?

Code: Select all

- ViciBox v.7.0.4-170113
vicibox7:~ # uptime
 16:15pm  up 20 days  7:49,  2 users,  load average: 0.07, 0.14, 0.16
vicibox7:~ # uptime
 16:18pm  up 20 days  7:53,  2 users,  load average: 0.35, 0.18, 0.17
vicibox7:~ # uptime
 16:19pm  up 20 days  7:53,  2 users,  load average: 0.36, 0.20, 0.18
vicibox7:~ # uptime
 16:21pm  up 20 days  7:55,  2 users,  load average: 0.20, 0.18, 0.18
vicibox7:~ # uptime
 16:22pm  up 20 days  7:56,  2 users,  load average: 0.24, 0.19, 0.18

Depends on what was happening while you took those readings. If you had all your agents on the server and were dialing at dial ratio 17, that's pretty good. If, however, you were not dialing at all it means absolutely ... nothing. lol

i took this reading when 3 out of 15 agents INCALL, 2 agents in Pause at dial level 7

Next check:

Reports => Real Time Report
Top Menu: "Choose Report Display Options"
"Show Carrier Stats": YES

What percentages do you get for cancel and congestion?

CARRIER STATS: HANGUP STATUS 24 HOURS 6 HOURS 1 HOUR 15 MIN 5 MIN 1 MIN
CANCEL 2739 16.3% 1217 15.4% 286 14.2% 88 13.6% 27 12.4% 7 13.2%
CONGESTION 1951 11.6% 860 10.9% 285 14.2% 67 10.3% 20 9.2% 5 9.4%

SBabludeva wrote:CARRIER STATS: HANGUP STATUS 24 HOURS 6 HOURS 1 HOUR 15 MIN 5 MIN 1 MIN
CANCEL 2739 16.3% 1217 15.4% 286 14.2% 88 13.6% 27 12.4% 7 13.2%
CONGESTION 1951 11.6% 860 10.9% 285 14.2% 67 10.3% 20 9.2% 5 9.4%

Actually safer to post all the statuses, but if your answer percent is over 30%, you're probably good there.

I note you were on dial ratio 7 instead of 15 as posted earlier. Is this because your agent wait time was better?

when I change the dial level it doesn't change the agent waiting time. that the reason I asking to solve the problem of reducing agent waiting time/increase the calling speed.

Which is why I asked the question. To find out why you are not at 15 presently. Thank you for answering! (Many don't respond to simple questions.)

So you have very good congestion on dial ratio 7, but I've not seen what the congestion percentage is when you're on dial ratio 15. If dial ratio 15 does not alter the wait time, when it obviously should, then something is blocking.

But without seeing the results of congestion when at ratio 15, I can't tell if your carrier is limiting your channels (many carriers will send back "congestion" if you request more channels than you are allowed). Alternately, if your TRUNK SHORT/FILL value increases when your dial ratio is at 15, then the limiting factor is Admin->Servers "Max Trunks" instead.

Check both of those and get back to us. Or ask questions if that was not clear.

SBabludeva wrote:I am Using RATIO dial method with 15 lines
200 in hopper, 20% Drop level and Next Agent Call is longest wait time
our agents getting call one by one every 2 mints.
the placed call count is avg 15-25 when 15 agents logged in.!

This is a bit strange. You are talking about 15 agents with a dial level of 7 at least, which means you should have about 7 calls placed for each READY (waiting) agent and about 100 calls in peak. What are your server settings for "Max Trunks" and "Max Calls per Second"?

@williamconley: great minds think alike. We have posted almost same time

I think we have found the issue. Somehow there was a change in the standard stock vicidial installation. In the past the max trunks value was 96 and now it's 23 (which nearly corresponds to his avg. 15-25).

Code: Select all

MySQL_AST_CREATE_tables.sql file:
max_vicidial_trunks SMALLINT(4) default '23',


blackbird2306 wrote:

SBabludeva wrote:I am Using RATIO dial method with 15 lines
200 in hopper, 20% Drop level and Next Agent Call is longest wait time
our agents getting call one by one every 2 mints.
the placed call count is avg 15-25 when 15 agents logged in.!

This is a bit strange. You are talking about 15 agents with a dial level of 7 at least, which means you should have about 7 calls placed for each READY (waiting) agent and about 100 calls in peak. What are your server settings for "Max Trunks" and "Max Calls per Second"?

max trunk is 23, Max calls per sec is 5

williamconley wrote:Which is why I asked the question. To find out why you are not at 15 presently. Thank you for answering! (Many don't respond to simple questions.)

So you have very good congestion on dial ratio 7, but I've not seen what the congestion percentage is when you're on dial ratio 15. If dial ratio 15 does not alter the wait time, when it obviously should, then something is blocking.

But without seeing the results of congestion when at ratio 15, I can't tell if your carrier is limiting your channels (many carriers will send back "congestion" if you request more channels than you are allowed). Alternately, if your TRUNK SHORT/FILL value increases when your dial ratio is at 15, then the limiting factor is Admin->Servers "Max Trunks" instead.

Check both of those and get back to us. Or ask questions if that was not clear.

their any possibility to cluster cache affect the calling speed?

Increase your max trunks. But watch your congestion! If congestion increases dramatically, you have likely exceeded your available trunks from your carrier.

Oh: And I'll now move this to "Support" where it belongs. lol

blackbird2306 wrote:In the past the max trunks value was 96 and now it's 23 (which nearly corresponds to his avg. 15-25).

Code: Select all

MySQL_AST_CREATE_tables.sql file:
max_vicidial_trunks SMALLINT(4) default '23',


23 was the normal average for a single PRI or T1.

But VOIP has no standards, no normal. So it's been set to a T1, and everyone else needs to know what they are doing and why they should change that number ... or play with it and see what happens.

williamconley wrote:23 was the normal average for a single PRI or T1.

So true! Though the help text has not been updated yet, still "Default is 96".

Ask your carrier how many concurrent calls are available and set your Max Calls per Second not higher than 2-4 (otherwise your congestion rate will be likely higher)

blackbird2306 wrote:Though the help text has not been updated yet, still "Default is 96".

Post it on the issue tracker. Very low importance, but they'll get it on a cleanup round.

I was set Max trunk is 96 and checking call flow last 3hr. It is pretty good now. out of 15, 10+ agents incall and agents get calls in every 30-40sec. this is the current readings of carrier stats

ANSWER 10003 55.5% 6007 58.5% 938 57.5% 345 59.0% 192 59.1% 42 64.6%
BUSY 289 1.6% 116 1.1% 16 1.0% 10 1.7% 6 1.8% 2 3.1%
CANCEL 2886 16.0% 1471 14.3% 242 14.8% 69 11.8% 41 12.6% 5 7.7%
CHANUNAVAIL 2249 12.5% 1442 14.0% 149 9.1% 57 9.7% 28 8.6% 6 9.2%
CONGESTION 2595 14.4% 1231 12.0% 285 17.5% 104 17.8% 58 17.8% 10 15.4%
NOANSWER 1 0.0% 0 0.0% 0 0.0% 0 0.0% 0 0.0% 0 0.0%
TOTALS 18023 10267 1630 585 325 65

Thank You guys

Good postback. ...

How long have you had this system, but haven't yet wandered through ALL the settings to see what they are used for?

i brought the system last nov 22 (Thanksgiving day).

Getting low call flow past 2 days.!

using same settings above this discussion.

Are you using the same leads? If so, they are over a year old and should probably be replaced with new ones. lol

Seriously, though: Settings need to be tweaked and results need to re-checked. Everything you checked that resulted in the previous settings needs to be checked again.

For older systems, there can be a buildup of NA calls (no answer autodial) if you don't set the called count limit to a reasonable number (zero is not reasonable number, that disables the feature). This way if you've called someone X times and not penetrated at all (no answer), the lead will be ignored in favor of other leads. This can sometimes make a HUGE difference, along with every other setting in the campaign.

There can also be a buildup of no-longer-needed log table entries. Those should be archived (there's a log archiver script in crontab -e). This helps if the DB is getting bogged down and is too slow to accomplish necessary tasks.