vicidial.org

Hello,

I'm currently having problems configuring a Netborder Call Analyzer, and have reached to a point where I got stuck.

This is the server build:

VERSION: 2.7-401a
BUILD: 130508-2256

This is the paraxip config:

Code: Select all

[paraxip]
disallow=all
allow=ulaw
type=peer
username=paraxip
fromuser=paraxip
secret=test
fromdomain=192.168.1.118
host=192.168.1.118
insecure=port,invite
outboundproxy=192.168.1.118:5062
context=vicidial-auto-external
enable-netborder-cpd=true


Code: Select all

TESTSIPTRUNKP = SIP/paraxip


Code: Select all

exten => _7X.,1,AGI(agi://127.0.0.1:4577/call_log)
exten => _7X.,2,Dial(${TESTSIPTRUNKP}/8${EXTEN:1},,To)
exten => _7X.,3,Hangup


The above causes that any call with prefix 7 to be routed through the proxy, and then the call will be made through our carrier, which uses prefix 8.

This part is working ok. The call reaches the Netborder proxy, and it's established. However, there's about a 6 second delay before the call is passed to the agent after it has been answered, and I see the following in the Netborder engine logs just before it's passed to the agent:

This is what I see on the callanalyzer enging logs:

Code: Select all

2014-05-21 11:56:53:581 [47388934801728] WARN - netborder.cpa.CPAEngineImpl : No audio received for this call. Make sure RTP packets are allowed to reach NCA engine.


Code: Select all

1400662510-632142-635065990-316,2014-05-21,,,818008472911,Unknown,11:55:10.686,11:55:13.995,11:55:23.997,,11:55:24.027,Unknown
1400662599-861594-198593003-320,2014-05-21,,,818008472911,Unknown,11:56:39.914,11:56:43.577,11:56:53.579,,11:56:53.610,Unknown


I turned off the firewalls, and didn't do any difference.


Any help would be highly appreciated.

Something else that's happening, after the call is passed to the agent, it's being terminated around 30 seconds later.

For what is worth, my issue is fixed.

Updated the RTP port range to match what was configured on the asterisk rtp.conf file (10000 to 20000) and ensure firewall was open for those ports.

Everything is working fine now, with the exception that I see now a lot of PDROPS. Is that normal?

PDROP's are caused when the remote end answers then hangs up straight away before ViciDIAL has chance to route the call. You should be able to see in the AGI log file exactly what the CPD is saying about these calls. You can also check on the CPD side too.