vicidial.org

This appears to just be a warning but it spams continuously in the log. I've heared there is a solutionf or this, any ideas?

-enjay

that error occassionally is OK, but continuously menas there is something wrong with your VOIP connection. It means that you might not have a good enough connection to your provider to handle the number of calls you are trying to push through.

Are you having any audio quality issues?

How many voice channels do you have going out?

What kind of network connection is between you and your provider?

yea Im receiving in excess of 25-30 per second.. I dont have a connection to a voip provider.

I have an IAX trunk to another Asterisk server which has 2 PRIs... I have 47 available B's on those PRI's

When I dial out of that server directly I have no audio issues it sounds good I dont have any echo.

The servers reside on the same switch at gig rate speeds.

Are you recording the calls?

What is the loadavg on both servers?

are you showing any dropped packets doing an "ifconfig"?

Not recording calls YET whats wierd is that it has stopped Im thinking it was a configuration issue with the registration between the servers..

Strange things are afoot at the circle k..

-enjay

I moved the T1's to the VICIDial server and there is echo funny thing is its not on the agent side the customer heres it but they dont hear the agent echoing they hear themselves echoing..

What do you have your echocanel set to in your zapata.conf file?

What kind of phone are your agents using?

sure do here is my config

Code: Select all

;
; Zapata telephony interface
;
; Configuration file

[trunkgroups]
trunkgroup => 1,48
spanmap => 1,1,2
spanmap => 2,1,0

[channels]

language=en
context=from-pstn
rxwink=300              ; Atlas seems to use long (250ms) winks
;
; Whether or not to do distinctive ring detection on FXO lines
;
;usedistinctiveringdetection=yes

usecallerid=yes
hidecallerid=no
callwaiting=yes
usecallingpres=yes
callwaitingcallerid=yes
threewaycalling=yes
transfer=yes
cancallforward=yes
callreturn=yes
echocancel=yes
echocancelwhenbridged=yes
echotraining=800
rxgain=0.0
txgain=0.0
group=0
callgroup=1
pickupgroup=1
immediate=no

;faxdetect=both
faxdetect=incoming
;faxdetect=outgoing
;faxdetect=no

Group=1
signalling=pri_cpe
switchtype=dms100
channel => 1-24


Group=2
signalling=pri_cpe
switchtype=dms100
channel => 25-47


Currently all agents are using IDEFisk as phones..

-enjay

I would suggest removing the echotraining line and changing echocancel from yes to 128 and see if that helps things at all after a reboot

Still get echo.. everything on the agent side is great its the customer side that hears themselves.. sound like mr roboto..

What kind of T1 cards are you using?

What codec are you using on the agent phone side?

What are the zaptel.conf settings?

Have you tried another type of softphone for the agents?

Im using Digium TE411P quad span with echo cancellation

currently using ulaw I've tried with g729 but I get the compatibility error


heres my zaptel.conf

Code: Select all

span=1,1,2,esf,b8zs
span=2,0,0,esf,b8zs
bchan=1-24
bchan=25-47
dchan=48
loadzone=us
defaultzone=us


I've tried several other softphones though they get echo especially firefly which gets echo on both sides..

OK, well that answers it. Your problem is a non-documented problem with the Digium Echo-can cards, their failure to detect dtmf properly resulting in audio loss. I ran into this as well, the only solution is to disable dtmf detection on hardware in the driver:

in zaptel/wct4xxp.c file change this line:
static int vpmdtmfsupport = 1;

to this line:
static int vpmdtmfsupport = 0;

recompile zaptel (make clean; make; make install)
reboot your machine (turn off power and unplug for 10 seconds)
then see if the audio problem is still there.

yea still experiencing the same issue.. left unplugged for 60 seconds to be sure..

try turning vpmsupport off and recompile and reboot:
static int vpmsupport = 0;

same dealio.. I called Digium support and they basically told me to disable hyper threading as well as moving some stuff (usb, sata etc) off the same interrupt as the digium card, I have done this and still have echo. Been on hold (in the queue) for 25 minutes now hopefully they have something to say about it.

Nothing good back from Digium.. can you think of anything in the meetme that could be causing echo. If I move the T1's to the freepbx system I dont get the echo so Im wondering if the "conferencing/meetme" has something to do with it.. This is highly speculated at this point Im just getting to the point where I dont even know where to investigate..

-enjay

The issue is most likely with the TE411P card. Digium is upgrading their chipset to Octasic(same chip Sangoma uses)for the new TE412P cards which tells you how much faith they have in their OKI chipset that is in the TE411P. My guess is that they might offer you a TE412P card to test and see if it removes your echo.

We are testing a TE412P card right now from Digium as a beta test and I am not supposed to release any details about my tests until the card is released, but I can tell you it works well.

We also have a few Sangoma a104d cards in production and they remove echo very well.

well thats good to know.. I just hope I get a resolution soon. May start from scratch again since I get echo on both of the VICIDIAL/Asterisk servers.. Any OS recommendations?

-enjay

We always use Slackware and a custom kernel, mostly use Linux 2.4.31 but have run a couple servers on 2.6.15-17 recently with good results.

You aren't by chance using CentOS are you?

Are there any known issues with CentOS?

That's basically what we use and so far seems to work well (except to one server but that seemed more to be H/W related).

Im using Fedora FC4 on Hp Proliant DL 360's with 2gb RAM Xeon 3.0ghz with Hyperthreading

on the Database/Apache server im using a HP Prolian DL 380 with Dual 3.4ghz Xeons and 4gb RAM also running Fedora FC4

We have just seen some issues happen more on CentOS for some reason. depending on how you have installed CentOS, you may have a process-priority application that will mess with the execution priority of scripts running on your system and it will really mess up VICIDIAL

I have a dialplan configured so that if dialing any local prefixes it goes out the local PRI's If its Long Distance it goes via an IAX2 trunk to my secondary VICIDial/Asterisk Server to use the long distance PRI's..

when I attempt to dial local numbers it works great when I try to dial long distance numbers I get flooded with

app_meetme.c:1523 conf_run: Unable to write frame to channel: Success

we visited this before (obviously) however I did not have the LD PRI's in at that time so testing wasnt possible..

My IAX2 Peers are registered on both sides..

any ideas?

What codecs are you using?

we usually do "disallow=all" and "allow=ulaw" to ensure that only ulaw will be used in local IAX connections.

yea Im using dissallow all and allow ulaw..

These two servers are using different verisons of Asterisk correct?

There were quite a few security changes made to IAX in the last release(1.2.9) that may cause some issues I've heard.

1.2.7.1 on both servers..

Could you post the iax conf settings for both servers?

Also the Dial strings from extensions.conf for dialing each other?

NOTE: Making Long Distance Phone calls, Locals work fine.

Server 1 [no Long Distance PRI only local]

IAX.conf

Code: Select all

[general]
bindport=4569
disallow=all                    ; same as bandwidth=high
allow=ulaw
allow=g729

register=cobastsrv01:XXX@192.168.0.6
register=cobastsrv01:XXX@192.168.0.7

[cobastsrv02]
type=friend
username=cobastsrv02
secret=XXX
host=dynamic
disallow=all
allow=ulaw
qualify=yes
trunk=yes
;context=dialer

[cobastsrv03]
type=friend
username=cobastsrv03
secret=XXX
host=dynamic
disallow=all
allow=ulaw
qualify=yes
trunk=yes
context=dialer


extensions.conf

Code: Select all

exten => _1602NXXXXXX,1,Dial(Zap/g1/${EXTEN}||o)
exten => _1480NXXXXXX,1,Dial(Zap/g1/${EXTEN}||o)
exten => _1623NXXXXXX,1,Dial(Zap/g1/${EXTEN}||o)
exten => _91NXXNXXXXXX,1,Dial(IAX2/cobastsrv01@cobastsrv02/${EXTEN}|1)
exten => _1NXXNXXXXXX,1,Dial(IAX2/cobastsrv01@cobastsrv02/${EXTEN}|1)


Server 2 [PRI]

IAX.conf

Code: Select all

[general]
bindport=4569
disallow=all                    ; same as bandwidth=high
allow=ulaw

register=cobastsrv02:XXX@192.168.0.5
register=cobastsrv02:XXX@192.168.0.7

[cobastsrv01]
type=friend
username=cobastsrv01
secret=XXX
host=dynamic
disallow=all
allow=ulaw
qualify=yes
trunk=yes
context=dialer

[cobastsrv03]
type=friend
username=cobastsrv03
secret=XXX
host=dynamic
disallow=all
allow=ulaw
qualify=yes
trunk=yes
context=dialer


extensions.conf

Code: Select all

exten => _1NXXNXXXXXX,1,Dial(Zap/g1/${EXTEN}||o)
exten => _91NXXNXXXXXX,1,Dial(Zap/g1/${EXTEN}||o)
exten => _602NXXXXXX,1,Dial(IAX2/cobastsrv02@cobastsrv01/${EXTEN}||o)
exten => _480NXXXXXX,1,Dial(IAX2/cobastsrv02@cobastsrv01/${EXTEN}||o)
exten => _623NXXXXXX,1,Dial(IAX2/cobastsrv02@cobastsrv01/${EXTEN}||o)


here is the debug output on server 1

Code: Select all

    -- Executing MeetMe("Local/8600056@default-2b3c,2", "8600056") in new stack
       > Channel Local/8600056@default-2b3c,1 was answered.
    -- Executing Dial("Local/8600056@default-2b3c,1", "IAX2/cobastsrv01@cobastsrv02/19256943102|1") in new stack
    -- Called cobastsrv01@cobastsrv02/19256943102
    -- Call accepted by 192.168.0.6 (format ulaw)
    -- Format for call is ulaw
    -- Nobody picked up in 1000 ms
    -- Hungup 'IAX2/cobastsrv02-16384'
  == Manager 'sendcron' logged off from 127.0.0.1
  == Parsing '/etc/asterisk/manager.conf': Found
  == Manager 'sendcron' logged on from 127.0.0.1
  == Manager 'sendcron' logged off from 127.0.0.1
Jul  5 10:41:52 WARNING[4954]: pbx.c:2412 __ast_pbx_run: Timeout, but no rule 't' in context 'default'
    -- Executing DeadAGI("Local/8600056@default-2b3c,1", "call_log.agi|h") in new stack
    -- Launched AGI Script /var/lib/asterisk/agi-bin/call_log.agi
    -- AGI Script call_log.agi completed, returning 0
  == Spawn extension (default, 8600056, 1) exited non-zero on 'Local/8600056@default-2b3c,2'
    -- Executing DeadAGI("Local/8600056@default-2b3c,2", "call_log.agi|h") in new stack
    -- Launched AGI Script /var/lib/asterisk/agi-bin/call_log.agi
    -- AGI Script call_log.agi completed, returning 0


Here is the debug output on server 2

Code: Select all

    -- Accepting AUTHENTICATED call from 192.168.0.5:
       > requested format = ulaw,
       > requested prefs = (),
       > actual format = ulaw,
       > host prefs = (ulaw),
       > priority = mine
    -- Executing Dial("IAX2/cobastsrv01-3", "Zap/g1/19256943102||o") in new stack
    -- Called g1/19256943102
    -- Hungup 'Zap/1-1'
  == Spawn extension (dialer, 19256943102, 1) exited non-zero on 'IAX2/cobastsrv01-3'
    -- Executing DeadAGI("IAX2/cobastsrv01-3", "call_log.agi|h") in new stack
    -- Launched AGI Script /var/lib/asterisk/agi-bin/call_log.agi
    -- AGI Script call_log.agi completed, returning 0
    -- Hungup 'IAX2/cobastsrv01-3'


exten => _91NXXNXXXXXX,1,Dial(IAX2/cobastsrv01@cobastsrv02/${EXTEN}|1)
exten => _1NXXNXXXXXX,1,Dial(IAX2/cobastsrv01@cobastsrv02/${EXTEN}|1)

Is there a reason you have a 1 second timeout on these?

That seems to be your problem. Try removing the 1 at the end and make sure you have the "o" flag on that as well:

exten => _91NXXNXXXXXX,1,Dial(IAX2/cobastsrv01@cobastsrv02/${EXTEN}||o)
exten => _1NXXNXXXXXX,1,Dial(IAX2/cobastsrv01@cobastsrv02/${EXTEN}||o)

That did indeed resolve my problem thankyou sir!

-enjay

So yay all that works fine and well but Dialing from a Trixbox through the VICIDial Asterisk server is still proving to be a pain and yea I've posted several times on trixbox.org with 0 response (kinda sad)..

Basically it comes down to "No Authority Found" on my VICIDial side my context is [dialer]

on the trixbox side my context is [from-internal]

any reason this would cause a problem?

those two servers iax2 show registry show as registered as well..

-enjay

I have been tinkering with trixbox a little lately and it's a mess. It could be any number of dozens of problems in the configurations it generates that causes the issues you are having.

I have never had a problem with IAX calls if both sides show as registered so I would blame the trixbox side.

Ohh I ENTIRELY blame trixbox too anyone any good with IAX trunks and have an understanding of the clusterfunk we call trixbox?

guys,

im also having problem on this..

it seems that when im trying to login on vicidial.. using ZAP, Manual dialing here is the error and it keeps on flooding

Sep 7 04:54:35 WARNING[4002]: app_meetme.c:1524 conf_run: Unable to write frame to channel: Resource temporarily unavailable

you said that this is for our voip provider right? we are using IAX, binfone. but we are not dialing yet, only logging in vicidial.

What zaptel timer are you using?

im using Digium TE205P

my zaptel.conf

span=1,1,0,esf,b8zs
fxoks=1-16
loadzone=us
defaultzone=us

What kind of agent phones are you using? If SIP or IAX then what codec?

Zap for my Agents ulaw

IAX in telco ulaw