Ratio dialing 1:1 -- more than 20 calls placed for 6 agents

[lvish]

Hi

Recently our client went live with 120 seater with vicibox 4.0.1 Cluster , thanks to Kumba new VICIBOX is awesome!

Two issues... have come across

1. At times ratio dialing goes out of control. Though campaign is set for 1:1 or 1:1.5 dialing system places huge number of calls , for eg : for 6 agents waiting for calls , 30 calls will be placed. It normalizes on its own after some time and again goes out of control.
This is doesnt happen for all campaigns . Issue occurs randomly for one or two campaigns.

Not sure whether changing campaign settings during high traffic is triggering this issue....


2. On Restart of dialer servers vicidial doesnt start on boot . Manually we have to start by entering /etc/init.d/vicidial restart.

Intially 3 screens will be seen later it will reduce to 1 screen. (Screen -ls) After vicidial restart all 8 screens are displayed.


thanks in advance

lvish

-----------------------------

DB ,Web ,Asterisks-- Dell R720 -Hexa core 32GB, -- 60 Agents --64Bit version
Asterisks ,Web -- ML110 , 1 X Quad core 8GB RAM -- 25 Agents --- 4 PRis 32 bit
Asterisks ,Web -- ML110 , 1 X Quad core 8GB RAM -- 25 Agents -- 2 Pri's 32bit

VERSION: 2.6-375a
BUILD: 120831-1523

Dahdi 2.6 -PRI trunks X 6 no.s
Vicibox 4.0.1,
Balance dialing -- Yes in all servers.

---------------------------------------------


my.cnf
Basic ViciBox my.cnf for a quad-core 8-GB RAM or so
# Use the dedicate my-big.cnf for 16+GB RAM and 8+ Cores
#

# The following options will be passed to all MySQL clients
[client]
#password = your_password
port = 3306
socket = /var/run/mysql/mysql.sock

# Here follows entries for some specific programs

# The MySQL server
[mysqld]
port = 3306
socket = /var/run/mysql/mysql.sock
datadir = /var/lib/mysql
skip-external-locking
key_buffer_size = 640M
max_allowed_packet = 2M
table_open_cache = 8192
sort_buffer_size = 4M
net_buffer_length = 8K
read_buffer_size = 4M
read_rnd_buffer_size = 16M
myisam_sort_buffer_size = 128M #64M
thread_cache_size = 50
query_cache_size = 256M #32M
thread_concurrency=24 #8
skip-name-resolve
connect_timeout=60
long_query_time=3
log_slow_queries
max_connections=1768
open_files_limit=24576
max_heap_table_size=128M #32M
concurrent_insert=2
expire_logs_days=3
default-storage-engine=MyISAM
table_definition_cache=8192
table_cache=8192
myisam_recover
myisam_repair_threads=1


# This will disable networking
#skip-networking

# Replication Master Server (default)
# binary logging is required for replication
log-bin=mysql-bin

# binary logging format - mixed recommended
binlog_format=mixed

# required unique id between 1 and 2^32 - 1
# defaults to 1 if master-host is not set
# but will not function as a master if omitted
server-id = 1
# Replication Slave (comment out master section to use this)

[williamconley]

1) Welcome to the Party!

2) when you post, please post your entire configuration including (but not limited to) your installation method and vicidial version with 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 "from scratch" you must post your operating system and should also post the .iso version from which you installed your original operating system. 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) Did you change the IP address of any of these server after they were built? did you upgrade the OS after install? (or During install without rebooting before the installation!)

4) What is your dial method on the campaign?

5) Check your boot logs and find out why Vicidial crashes on restart.

[lvish]

Hi

it is
Vicibox 4.0.1 from ViciBox.x86_64-4.0.1.preload.iso | VERSION: 2.6-375a |BUILD: 120831-1523 (svn updated manually)|Asterisk 1.4.44-vici| Cluster
--DB-WEB-Asterisk on Dell R720 Hex Quad core -32G RAM | Dialer with Allo PRI hardware on ML110 -single quad core - 8GB RAM X 3 No.s.| No cloud/virtual machine used.

- yes IP address were changed after DB server was built - have executed IP address update script.

-After PRE load iso is installed Zypper update is done and rebooted , later vicibox-install is executed.

-Dial method is campaign.

- Will check the boot logs and revert.

Not sure --is it anything to do with mysql response? where at times DB response goes slow! and dialer ratios is calculated based on old data ..

regds

lvish

[lvish]

Dial method in campaign is Ratio.

[williamconley]

if your dial method is ratio, it should never exceed the stated calls per agent. do you have "available only tally" set to Y?

if you have your dial ratio set to 1.5:1, unless you have 20 agents logged in and available for a call it should never get to 30. Did you have 20 agents ready at any point? you know it will not kill those calls in progress immediately, it will wait for them to complete ...

[lvish]

Available only tally is N, Tried setting it to Y and threshold set to 2 call waiting agents but same result.

couple of campaigns has max 10 agents. At any given point around 6 will waiting for calls. It starts from 5 calls , every 4 to 10 seconds additional calls will be dialled . It goes upto 30 to 32 calls dialled for this campaign.

We were successful once in fixing the problem by logging off one of the agent who had logged in to different server , where as all other 9 were in one server.
But it happens even if all agents of the same campaign is logged in to same server.

[williamconley]

"for this campaign"? How many agents TOTAL on this server?

what is your server load (on each server) using "htop" or "uptime" from the command line? (these two commands will show 1 5 and 10 minute load averages).

And how many processors does each server have?

If you have a 4 processor system, you 1,5,10 minute averages should not exceed 2.0. If you have an 8 processor system, they should not exceed 4.0 (ie: 50% (half) the processors, represents "full load" which you should not exceed on average)

I do not recommend using any of the "threshold" settings if you are having problems such as this.

And you did not (apparently) mention whether you dial method is ratio?

[lvish]

two campaigns which has regular problem like this has 10 & 12 agents . Each server is loaded at max 28 agents.

-Let me observe load average today when problem occurs ( overall I have not seen the load greater than 12% on any servers in vicidial reports screen).

-Dialer server has 4 procs & DB server has 24 Procs.

- Thershold was set only for testing purpose and just to try my luck!! since there was no improvement in the situation settings were reverted .

--Dialing method is ratio!!!

--When issue is happening , campaign debug page in admin utilities ,shows the calculation & correct number of calls as per the ratio level set , but actually it is dialing more number calls resulting in drops.

[williamconley]

Are you certain you are not viewing multiple campaigns in your real time screen by accident? (instead of viewing only one at a time ...?)

[lvish]

Viewing only the problematic campaign -:)

[lvish]

One observation.

Issue is reproducible in below mentioned situation.

8 Agents of a campaign logs in to Server A & 2 agents of same campaign logs in to server B in the same cluster. ( Logs in means registers the SIP phone) Uses common web server for agent login. Dial method is 1:1 , Available Tally --Y.

Issue comes up when one of the two agent logged in to server B goes idle or waiting for calls. If these two agents are busy on call then ratio dialing works fine. If these two agents are waiting then automatically pacing will increase!!..

Below is the campaign debug output during the problem situation.
------------------------------------------------------------------------------------------------------------
Campaign Debug: XXXXX - Campaign_XXX _XX 2013-01-30 17:51:44

Total leads in hopper right now: 3004

Adapt Debug: 2013-01-30 17:51:42
XXXX hopper READY count: 3004
XXXX agents LOGGED-IN count: 10
4 |XXX_XX|2000|3004|24hours|15|60|
TIME CALL CHECK: 13595548491
CAMPAIGN DIFFERENTIAL: 10 -2.33333333333333 (0.0666666666666667 - 2.4)
CAMPAIGN ANSWERED STATUSES: XXX_XX|QCFAIL,CB1,CB2,CB3,CB4,SALE1,SALE2,NI1,NI2
XXX_XX|10.26|8.36| |177 2116 821| 4
XXX_XX|CallBacks|130|CB1,CB2,CB3,CB4,CB5|
XXX_XX|NotInterested|104|NI1,NI2,NI3,NI4,NI5,NI6,NI7,NI8
XXX_XX|Sale|134|Sale1,Sale2,Sale3,Sale4,Sale5,Sale6,Sale7
LIVE CALLS: 13|SELECT count(*) from vicidial_auto_calls where campaign_id=XXX_XX
AGENT TIME STATS: 31228 64189 102748 28835|SELECT sum(pause_sec),sum(wait_sec),sum(talk_sec) - sum(dead_sec) as custtalk,sum(dispo_sec) + sum(dead_sec) as acw from vicidial_agent_log where event_time > 2013-01-30 00:00:01 and campaign_id=XXX_XX and pause_sec < 65000 and wait_sec < 65000 and talk_sec < 65000 and dispo_sec < 65000 and dead_sec < 65000
AGENT CALLS: 840|SELECT count(*) from vicidial_agent_log where event_time > 2013-01-30 00:00:01 and campaign_id=XXX_XX and lead_id > 0

CAMPAIGN: XXXXXXX 4

192.168.1.5 Debug: 2013-01-30 17:51:41
XXX_XX 192.168.1.5: agents: 0 (READY: 0) dial_level: 1 (0|5|1) 0
XXX_XX 192.168.1.5: Calls to place: -2 (1 - 3 [0 + 3||]) 61
CAMPAIGN DIFFERENTIAL: 10 -0.1 (0.95 - 1.05)
LOCAL TRUNK SHORTAGE: 0|0 (59 - 120)


192.168.1.8 Debug: 2013-01-30 17:51:43
XXX_XX 192.168.1.8: agents: 0 (READY: 0) dial_level: 1 (0|2|0) 0
XXX_XX 192.168.1.8: Calls to place: 0 (0 - 0 [0 + 0||]) 0
CAMPAIGN DIFFERENTIAL: 10 -0.4 (0.85 - 1.25)
LOCAL TRUNK SHORTAGE: 0|0 (0 - 0)

[williamconley]

Can you check to see if (perhaps) both dialers are dialing for the available agents?

Turn off Balance Dialing.

[lvish]

only 192.168.1.5 has the trunks. Have enabled balance dialing in both.

Let me try by disabling balance dial in both.

[lvish]

Hi

did mysql repair & hard fixed the NIC to 1000 MB on switch side -- looks like things are fine .. touch wood!!

regds

[racosta79]

Hello,

I am trying to install Vicidial using vicibox server 4.0.3 with a Dell PE R720 as my DB, Intel Servers as Asterisk and HP Proliant ML110 G6 as web. My problem is that I want to use Dell's PERC to make a RAID 10, but when I boot with my burned CD (64bit) no hard disk is detected.

How did you do it? Did you use Software RAID instead?

Thank you!

[williamconley]

1) The question you posted has nothing to do with this forum topic. This is commonly referred to as Hijacking and is pretty much "bad karma" as well.

2) Software RAID is completely useless with Vicidial (FYI)

3) If you set up your RAID and the OS does not recognize it, it is entirely possible that your RAID controller does not yet have drivers available for the OS you tried to boot from. Using a motherboard or RAID controller that is "bleeding edge" (or even Really Recent or New) is not a great idea. Linux drivers can often lag by up to a year (and sometimes beyond).

4) Best Bet: Whichever .iso you are using, try another one. If you are not entirely certain that you properly set up your RAID, TEST it. Try a "stock" OpenSuSE, Ubuntu or CentOS or other Linux OS and see if any OS .iso will recognize it. You can also try GoAutoDial and hope (GoAuto 2.1 is fairly old, so it isn't all that likely to be an improvement, but since it IS free, why not try it?)

If it were me, I'd try a stock Ubuntu latest OS and verify the validity of the RAID setup. Perhaps even using the ubuntu install to partition the drive for the Vicibox .iso on the next attempt.

[racosta79]

1) I am desperate, what should I do then? I found someone that has already solved my problem as I understand. Don't be mad!

2) What if hardware RAID is not possible because of drivers like this case?

3) Ivish did it, so there should be a driver or he used soft raid... ?

4) I want to use ViciBox Server as installer

[williamconley]

Don't use RAID. Just use a regular HD. Software RAID is countrproductive.

[lvish]

hi

we configured hw raid using cd supplied along with the server by dell. Then used vicibox 4 64bit preload for installation.
I faced similar problem of hdd not detecting when trying to install using vicibox 3.1.15!!

regds
lvish

[racosta79]

Thank you Ivish!

My server came with no cd out of the box and I dont know where and what to download? Do you remember what is the name printed on the CD?

Regards!

[lvish]

i dont remember what was written on cd.

u can check out this link http://en.community.dell.com/support-fo ... 55205.aspx

[fibres]

Hi ivish

I have never used a r720 but have had experience of plenty of r210 and r4xx series dell servers all which had varying versions of the perc raid controller. On al the ones I have tried I have never used a dd to configure. If you watch the server on boot you will see the perc card initialise and give the option of pressing a key combination to enter into the raid control,er setup, usually ctrl - F or something like that. This will load the controller bios school should let you configure it.

I then suggest following Williams advice of trying latest Ubuntu and see if that picks up the raid 10 that way you know the array is live!

On a final note as William also said highjacking a thread is bad karma, also it makes it much more difficult for others to find if facing similar issues.

I would suggest creating a new post specifying your issues, and if you get it working without further help just post a quick follow up on how you ot it working for others, if your still having issues we will gladly help you in the new topic!

Regards

[racosta79]

The sad thing is that the RAID is working well, I tried installing Ubuntu and Centos and both picked up the raid setup and installed succesfully, but with the OpenSuse that comes with the Vicibox Server it does not "see" the RAID

I will create a post asking this as general question or hardware specific question.

Thank you!

[williamconley]

Scratch install will resolve your issue. If you'd like, you could join the Alpha testing of GnuDial (Ubuntu 12.04 with Vicidial SVN .iso install which assumes DHCP).