SMDR challenge

[Ceaserx]

Hi guys, I have quite a complicated site I support that have 5 controllers (2 x 3300 MXE's and 3 x AX's).

All of these controllers are connected in a cluster and they all dial out of the main controller that has SIP trunks.The TMS program they make use of now has an issue where it bills calls made on one of the MXE systems twice. They tap into each controllers SMDR feed separately. I have done some tests and the calls made from a IP phone show correctly in the SMDR string but an Analog phone is missing an "A" in the string. That is the only difference I can detect. Please have a look at the two examples and let me know what you think.

Record 1 is the Controller with the SIP trunks and 2 is the local controller where the extension resides.

This is a call from an IP phone 1:
1) 06/12 15:51:53 0000:00:06 X9999 27828740000 A 001 5182199 W0051056 C <-EOR
2) 06/12 15:51:52 0000:00:04 182199 00828740000 A X9999 005 5182199 W0051056 A 5 <-EOR

Now this is a call from an Analog phone
1) 06/12 15:44:22 0000:00:06 X9999 0012 00732040000 001 5182229 I0051896 B <-EOR
2) 06/12 15:44:21 0000:00:06 182229 00732040000 A X9999 949821 005 5182229 I0051896 A 5 <-EOR

 

[DoubleUT]

The situation within a cluster where calls can breakout via other nodes should use a unique identifier per call. This can and should be set in the SMDR options. This TAG is maintained throughout all systems that call is routed. Your SMDR programm should be able to determen that this is one call based on that.

 

[Ceaserx]

Hi DoubleUT

I can confirm that all systems have a unique identifier programmed in the SMDR

 

[LoopyLou]

The the issue is likely with the CDR product not able to track the calls properly.

An apple a day keeps the doctor away. Anyone else and you need to throw it harder.

 

[becca1jane]

There should be 3 SMDR logs from the call made from the IP phone (you are currently showing logs ending with A and C):

1) 06/12 15:51:53 0000:00:06 X9999 27828740000 A 001 5182199 W0051056 C <-EOR
2) 06/12 15:51:52 0000:00:04 182199 00828740000 A X9999 005 5182199 W0051056 A 5 <-EOR

So there are possibly 3 call records for the analog phone as well (you have displayed logs ending with A and B):
1) 06/12 15:44:22 0000:00:06 X9999 0012 00732040000 001 5182229 I0051896 B <-EOR
2) 06/12 15:44:21 0000:00:06 182229 00732040000 A X9999 949821 005 5182229 I0051896 A 5 <-EOR

Is it possible log C from the analog phone has the "missing A" you are looking for, to show the call was Answered?