Jim--<br><br><br><div class="gmail_quote">On Thu, Jun 4, 2009 at 1:40 AM, Jim Boykin <span dir="ltr"><<a href="mailto:boykinjim@gmail.com">boykinjim@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi,<br>
<br>
Asterisk does not post CDR when dial status is CHANUNAVAIL.</blockquote><div><br>CDR's are, at the current time, and always have been attached to the channel struct;<br>so, if you don't create a channel, then there is nowhere to attach a CDR, and no way<br>
to process that.<br> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
<br>
Can someone tell me what are the conditions under which CDR is not posted?</blockquote><div><br>We try to filter the CDR if a channel were created, but did nothing; an example is<br>where a Dahdi device is taken off hook, and then hung up again. But getting all the<br>
conditions right has been tricky to filter this sort of event sequence.<br><br>I think you'll find that CDR's are one of the least solid parts of Asterisk at the moment.<br>There's brave and creative folks working on fixing the current implementation, but<br>
as far as I'm concerned, it's got some fundamental problems, and needs to be <br>overhauled. If you are interested, you can read my spec for a new approach by:<br><br>svn co <a href="http://svn.digium.com/svn/asterisk/team/murf/RFCs">http://svn.digium.com/svn/asterisk/team/murf/RFCs</a><br>
<br>and then looking at the pdf in that dir, for my spec for the CEL->CDR proposal.<br><br>While I have abominated the complexity of the ForkCDR/NoCDR/etc mechanisms<br>of the current solution, I have considered making the spec include them for backward<br>
compatibility... Current implementations based on the current mechanisms shouldn't<br>have to be made obsolete, although they usually do depend on a great deal of undocumented<br>behavior, that may be tricky to imitate.<br>
<br>murf<br> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
<br>
Thanks<br>
Jim<br clear="all"><br></blockquote></div><br>-- <br>Steve Murphy<br>ParseTree Corp<br><br>