<div class="gmail_quote">On Wed, Mar 17, 2010 at 8:34 PM, Michael
Collins <span dir="ltr"><<a href="mailto:msc@freeswitch.org">msc@freeswitch.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<br>
<div class="gmail_quote"><div>
If you *really* want to get fancy you can use mod_xml_cdr and get a true
3D picture of your call. You can even use it to post the XML CDRs to an
external web server and do all your billing offline. :D<br>
</div></div></blockquote></div><br><br>
Yes, I was planning on using mod_xml_cdr to pass the info to our webserver to
update the database, once I know that mod_cdr_csv is getting the right
data.... which it still isn't :-(<br>
<br>
Much the same problem as before - in mod_cdr_csv's logs the start-time
and answered-time aren't what I expect. I'm gradually getting a better
handle on whats happening...<br>
<br>
The 4 significant events in my dialplan are:<br>
<br>
ivr-started (I dial a switchboard number, and enter the extn I want to be
connected to)<br>call-started (I'm connected to that extn)<br>call-ended (the call with the extn ends)<br>ivr-ended (I may return to the extn, or end at the same time as the call)<br>
<br>
Master.csv contains 2 rows, both with from and to numbers matching the caller and callee appropriately.<br><br>
It seems like the first/blind_transfer row is not picking up the info
from the call that it ought to, and some of it is being recorded against the second/main
ivr row and then lost when it gets new data.<br><br>First row has:<br>start = call-started <br>answer = empty<br>end = call-started (probably actually call-connected)<br>hangup_cause = BLIND_TRANSFER<br>bleg uuid = empty<br>
<br>Second row (also appears in the separate csv file for the caller's extension) has<br>
start = ivr-started <br>
answer = ivr-started<br>
end = ivr-ended<br>
hangup_cause = NORMAL_CLEARING<br>
bleg uuid = present<br><br>
Cheers,<br>
Fraser<br>
<br>