load jump

[rambo]

Hello every one,

The load on our asterisk/vicidial which is our dialer is jumping from 3.00 to 20.00.

First thing in the morning it is working fine but then after about 5 hours later calls stop coming in because the load is getting so high.

we are using g729, IAX, this is a Australian voip coming.
We are doing all recordings but they are done on another server.
there is only 15 agents.
one server Asterisk/vicidial server Xeon 2.8, 4GB ram, 200GB SATA
one DB server AMD, 1GB Ram, 200GB IED
one Asterisk/vicidail server Xeon 2.4, 512 RAM, 80GB SATA
one Asterisk/vicidial server Xeon 2.4, 1GB Ram, 80GB SATA

We are running at night with really know problems because they are just doing outbound calls with no recordings.

Let me know if any one has any thoughts about this

[mflorell]

Please explain exactly how you are doing recording on a server other than the VICIDIAL asterisk server.

can you do a "top -c" at the point that you have very high load and post it?

[rambo]

Hello Matt

here is a picture

top - 15:01:48 up 5 min, 3 users, load average: 8.06, 4.36, 1.76
Tasks: 68 total, 1 running, 67 sleeping, 0 stopped, 0 zombie
Cpu(s): 27.2% us, 4.6% sy, 10.3% ni, 56.3% id, 0.0% wa, 0.7% hi, 1.0% si
Mem: 2074488k total, 192092k used, 1882396k free, 3648k buffers
Swap: 2097136k total, 0k used, 2097136k free, 114004k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3135 root 15 0 45948 15m 5016 S 14.1 0.8 1:20.68 /usr/sbin/asterisk -vvvvvvvvvvvvvvvvvvvvvgc
3733 root 15 0 5560 4408 1512 S 5.9 0.2 0:12.40 /usr/bin/perl /home/cron/AST_update.pl
6485 root 22 2 5020 3788 1492 S 4.9 0.2 0:00.15 /usr/bin/perl /home/cron/AST_send_action_child.pl --data1=3268031
6492 root 21 2 5024 3792 1492 S 4.6 0.2 0:00.14 /usr/bin/perl /home/cron/AST_send_action_child.pl --data1=3268032
3086 root 16 0 4776 3564 1508 S 0.7 0.2 0:01.54 /usr/bin/perl /home/cron/AST_VDauto_dial.pl
3163 root 17 2 5316 4128 1508 S 0.3 0.2 0:01.08 /usr/bin/perl /home/cron/AST_manager_listen.pl
3173 root 17 2 4524 3280 1508 S 0.3 0.2 0:00.77 /usr/bin/perl /home/cron/AST_manager_send.pl
3898 root 16 0 2064 1092 836 R 0.3 0.1 0:00.39 top -c
5250 root 17 0 2064 1092 836 S 0.3 0.1 0:00.21 top -c
1 root 16 0 684 252 216 S 0.0 0.0 0:00.19 init [3]
2 root 34 19 0 0 0 S 0.0 0.0 0:00.00 [ksoftirqd/0]
3 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 [events/0]
4 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 [khelper]
5 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 [kthread]
7 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 [kblockd/0]
8 root 20 -5 0 0 0 S 0.0 0.0 0:00.00 [kacpid]
116 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 [khubd]
187 root 20 0 0 0 0 S 0.0 0.0 0:00.00 [pdflush]
188 root 15 0 0 0 0 S 0.0 0.0 0:00.00 [pdflush]
190 root 17 -5 0 0 0 S 0.0 0.0 0:00.00 [aio/0]
782 root 10 -5 0 0 0 S 0.0 0.0 0:00.01 [kseriod]
189 root 25 0 0 0 0 S 0.0 0.0 0:00.00 [kswapd0]
820 root 10 -5 0 0 0 S 0.0 0.0 0:00.00 [ata/0]
822 root 11 -5 0 0 0 S 0.0 0.0 0:00.00 [scsi_eh_0]
823 root 11 -5 0 0 0 S 0.0 0.0 0:00.00 [scsi_eh_1]
834 root 16 0 0 0 0 S 0.0 0.0 0:00.00 [khpsbpkt]
918 root 19 -4 1488 484 408 S 0.0 0.0 0:00.01 udevd
1242 root 16 0 1524 556 464 S 0.0 0.0 0:00.01 /usr/sbin/syslogd
1291 root 16 0 1488 364 296 S 0.0 0.0 0:00.00 /usr/sbin/klogd -c 3 -x
3007 root 16 0 1516 480 412 S 0.0 0.0 0:00.00 /usr/sbin/inetd
3012 root 16 0 3376 992 724 S 0.0 0.0 0:00.09 /usr/sbin/sshd
3024 root 16 0 1680 600 500 S 0.0 0.0 0:00.00 /usr/sbin/crond -l10
3026 daemon 15 0 1680 412 312 S 0.0 0.0 0:00.00 /usr/sbin/atd -b 15 -l 1
3029 root 16 0 5324 1688 944 S 0.0 0.1 0:00.00 sendmail: accepting connections
3032 smmsp 21 0 5112 1256 732 S 0.0 0.1 0:00.00 sendmail: Queue runner@00:25:00 for /var/spool/clientmqueue
3035 root 18 0 1480 460 388 S 0.0 0.0 0:00.00 /usr/sbin/acpid
3043 root 21 0 1532 344 280 S 0.0 0.0 0:00.00 /usr/sbin/gpm -m /dev/mouse -t ps2





Let me know your thoughts, I'm not sure how are we doing the recordings other then it is all done on one server

thanks again

[mflorell]

A load of 8.06 is very high, what hardware are you on?
(CPU/RAM/drives)

[rambo]

Hello Matt,

Asterisk the main dialer
we are on a XEON 2.8, 4GB ram and 200GB Sata

DB
AMD 3.0, 1GB ram, 200IED

I had 4GB of Ram but I took one of the 2GB cards out to see if one of them is bad, we are testing it now and it still looks about the same.

thanks for your help

[enjay]

Per Matt, how are you doing recording on another server other than the Asterisk/VICIDIAL server? Additionally are you using record_delay?

[rambo]

Hello Enjay,
I'm still waiting on my guy to get back to me on how the recordings are set up. We are not use the recording delay yet but we will have it by next week end.

Thank you for the quick response.

We are doing some inbound calls and they are working great but when the agent gets an inbound call and they try to transfer it to the TPV it calls but does not connect. When we make a manual dial from the closer campaign with a client on the phone with an agent to the TPV it works great.

Here is what we have in the extensions.conf for inbound is there something we would need to add.

; inbound VICIDIAL call with prompt for 4-digit fronter code
exten => _X.,1,Answer ; Answer the line
exten => _X.,2,AGI(agi-VDADcloser_inboundCID.agi,CL_UNIQUE-----XXXXXXXXXX-----Closer-----park----------888-----1)
exten => _X.,3,Hangup

Thanks for everyone’s help.

[enjay]

The record_delay will definately help you with system load you gotta figure without the delay even unanswered calls are being recorded this is huge overhead on the system. I use a 25 second delay and that brough my load down significantly.

-enjay

[mflorell]

What is "the TPV" that you are transferring calls to?

How exactly are the agents transferring these calls?

[rambo]

Hello Matt,

we followed everything on the manual to set this up, we can transfer to the TPV on a manual outbound call but when a inbound call comes in and we click transfer and then click on D1 the number shows up and then we click transfer with customer.



this what it say when an inbound call comes in. not working
STATUS: Calling: UID: DCagcW1159620042dmin Waiting for Ring... 23 seconds

When you are making a manual outbound call, which is working
STATUS: Called 3rd party: 40017274507312 UID: DCagcW1159622668dmin

[mflorell]

What exactly do you have in your D1 preset?

[rambo]

Hello Matt,

what Im not sure what you mean by D1 preset to, can you tell me were I can find that at.

thanks

[mflorell]

"then click on D1 the number..."

I was assuming you knew what it was since you used it, but the DTMF and number presets are in the Campaign Detail screen.

[rambo]

Hello Matt,

I have the number of the TPV in the box for the D1 I understand what DTMF is. I have it set up for the closer campaign and it is working so I set it up in the ingroups to the same way I set it up as the out bound.
Thanks

[rambo]

Hello Matt,

We got the TPV to work on inbound calls but then when we try to make a call on the TPV it is saying Dial time out contact administrator.

I changed the dial time out on the closer campaign, any thoughts.

thanks

[mflorell]

" the number of the TPV in the box for the D1"

What is this number?

Are you trying to do a consultative transfer?

Are you using "CXFER..." in the transer-conf number field?