Even in the case like this where they don't map the same front and back (there are some of these in the spec), on a sip call, it is supposed to always track the code too, and re-use the number from the other leg.  If it is not doing this, please file a jira for this issue.  We used the q931/qsig sip interop specs to set those tables of how they map each way, I can't recall the exact spec but it is commented in the code next to those case statements.  Take a look in there and see if the mapping is wrong or not.

Mike

On Oct 9, 2013, at 12:14 PM, Jeff Ullman <togullu@gmail.com> wrote:

I see following 480 in entering state : 

2013-10-05 23:39:04.281717 [DEBUG] sofia.c:6282 Channel sofia/external/14106686531 entering state [terminated][480]


And it is set as NO_USER_RESPONSE .

I used following table to matchup cause codes : 
http://wiki.freeswitch.org/wiki/Hangup_Causes

So I used NO_USER_RESPONSE   as  408. 

Ideally I should have seen   NORMAL_UNSPECIFIED    which is defined Q.850  ( cause 31)  and matches with wireshark capture. 

Am I right? 

Thanks 






On Tue, Oct 8, 2013 at 8:38 AM, Brian West <brian@freeswitch.org> wrote:
Please provide a complete sip trace, 408 is nowhere in either the log or the info you've provided.

--