vicidial.org

by **kpanik** » Sat Sep 30, 2006 1:52 pm

someone knows gives me of the suggestions for having a good result in the zttest ? Yesterday they are successful to carry it a high level , the peak more low was 99.987 the problem is that as soon as leaves 10 agent, the level is lowered to the point to make to fall the lines.

by **mflorell** » Sat Sep 30, 2006 9:31 pm

what is the loadavg on the server when you have probelms?

by **kpanik** » Sun Oct 01, 2006 11:33 am

the my load average is max 12,8 %

by **kpanik** » Sun Oct 01, 2006 3:32 pm

this is my zttest

Opened pseudo zap interface, measuring accuracy...
99.951172% 99.951172% 99.951172% 99.963379% 99.963379% 99.951172% 99.963379%
99.951172% 99.963379% 99.963379% 99.975586% 99.951172% 99.938965% 99.963379% 99.987793%
99.926758% 99.951172% 99.938965% 99.963379% 99.963379% 99.987793% 99.926758% 99.987793%
99.926758% 99.951172% 99.951172% 99.951172% 99.963379% 99.951172% 99.963379% 99.963379%
99.951172% 99.951172% 100.000000% 99.938965% 100.000000% 99.914551% 99.938965% 99.963379%
99.951172% 99.963379% 99.963379% 99.987793% 99.963379% 99.926758% 99.963379% 99.963379%
99.975586% 99.951172% 99.951172% 99.951172% 99.951172% 99.951172% 99.963379% 99.963379%
99.951172% 99.963379%
--- Results after 57 passes ---
Best: 100.000000 -- Worst: 99.914551 -- Average: 99.957382

by **kpanik** » Sun Oct 01, 2006 8:24 pm

how many passes of the zttest ???

by **mflorell** » Mon Oct 02, 2006 9:56 am

Is that a loadavg of 12.0 or 0.12?

You should run zttest for several hours to get a good result, not a couple minutes.

by **kpanik** » Mon Oct 02, 2006 11:51 am

hi matt,

the loadvg is the 12.75

...

randomly i see in the clie this message NOTICE[16200]: app_dial.c:1056 dial_exec_full: Unable to create channel of type 'Zap' (cause 34 - Circuit/channel congestion)

and before or after ....

!! Got reject for frame 7, but we have nothing -- resetting!

why ??

suggestion for me .optimizing my kernel or configuration ????

i've a dual xeon cpu nocona EMT64 HT enable 3,2 Ghz the supermicro mainboard chipset e7520 intel a digium te410p whith module echo cancellation ... 2 hdd sata 250 gb on raid 1 3 gb ram ecc registered ..

on digium board i attached 4 PRI E1 .... (ITALY EUROISDN)...

my zaptel conf ::

span=1,0,2,ccs,hdb3,crc4
span=2,4,2,ccs,hdb3,crc4
span=3,4,2,ccs,hdb3,crc4
span=4,4,2,ccs,hdb3,crc4
bchan=1-15
dchan=16
bchan=17-31
bchan=32-46
dchan=47
bchan=48-62
bchan=63-77
dchan=78
bchan=79-93
bchan=94-108
dchan=109
bchan=110-124
loadzone=it
defaultzone=it

my zapata.conf is

[channels]
language = it
resetinterval = never
progzone = it
context=default
switchtype=euroisdn
pridialplan=unknown
prilocaldialplan=unknown
priindication = outofband
echotraining = yes
immediate = no
internationalprefix = 00
nationalprefix = 0
answeronpolarityswitch = yes
hanguponpolarityswitch = yes
signalling = pri_cpe
echocancel = yes
callgroup=1
pickupgroup=1

; card 0 - span 1
switchtype = euroisdn
signalling = pri_cpe
group = 1
context = incoming
channel => 1-15,17-31

; card 0 - span 2
switchtype = euroisdn
signalling = pri_cpe
group = 1
context = incoming
channel => 32-46,48-62

; card 0 - span 3
switchtype = euroisdn
signalling = pri_cpe
group = 1

; card 0 - span 4
switchtype = euroisdn
signalling = pri_cpe
group = 1
context = incoming
channel => 94-108,110-124

my zaptelrevision is 1.2.9.1 asterisk 1.2.12.1 slackware current

and kernel 2.6.17

that's all

by **enjay** » Mon Oct 02, 2006 1:44 pm

A load of 12.8 is extremely high you need to get that load down or you will continually have miscellaneous problems

1. Dropped Calls
2. Bad sounding calls
3. Overlapping Hangup Issues
4. Bad Recordings

among many others..

-enjay

by **kpanik** » Tue Oct 03, 2006 4:36 am

ok. I've 120 line used with IVR (gsm) and I've 50 IVR active always. as I can reduce the workload of the server ?

by **mflorell** » Tue Oct 03, 2006 9:03 am

I would recommend getting another server.

VICIDIAL load is not like IVR or inbound load. The constant starting and hanging up of calls generates a lot of load that you do not see in systems that have the same number of minutes of talk time, but much fewer calls in a day.

by **kpanik** » Tue Oct 03, 2006 6:31 pm

Hi matt I've a 2 server. 1 dual xeon cpu nocona EMT64 HT enable 3,2 Ghz the supermicro mainboard chipset e7520 intel a digium te410p whith module echo cancellation 2 hdd sata 250 gb on raid 1 3 gb and another, dual xeon cpu nocona EMT64 HT enable 3,2 Ghz the supermicro mainboard chipset e7520 intel 2 hdd SCSI 73 gb on raid 1 3 gb.

it would be better to put asterisk in the server with hd SCSI and vicidial e mysql, apache and php in the second server ?

by **mflorell** » Tue Oct 03, 2006 9:21 pm

Yes, moving MySQL/Apache/PHP on the non-SCSI server would be best. Even with this you may still need a second Asterisk/VICIDIAL server to handle the load.

by **kpanik** » Fri Oct 06, 2006 9:40 pm

hi matt ...

if I want install mysql, apache and php in the server scsi and asterisk in the another server how configure this ?

by **mflorell** » Sat Oct 07, 2006 1:43 pm

well, install them. Then just run astguiclient installation on there and your Asterisk Server and use the Database server IP as the machine that has the MySQL server on it.

by **kpanik** » Fri Oct 13, 2006 4:58 pm

Hi Matt,

if I install asterisk and vicidial in a server, how configure my astguiclient.conf ? The my asterisk.conf is :

PATHhome => /usr/share/astguiclient
PATHlogs => /var/log/astguiclient
PATHagi => /var/lib/asterisk/agi-bin
PATHweb => /var/www/htdocs
PATHsounds => /var/lib/asterisk/sounds
PATHmonitor => /var/spool/asterisk/monitor

VARserver_ip => 192.168.1.249 --- This is my asterisk/vicidial
VARDB_server => 192.168.2.250 --- This is my mysql/apache/php
VARDB_database => asterisk
VARDB_user => cron
VARDB_pass => 1234
VARDB_port => 3306

by **mflorell** » Sat Oct 14, 2006 9:11 pm

That should work for your Asterisk/VICIDIAL server.

vicidial.org

zaptel zttest

zaptel zttest

Who is online