[SIPForum-techwg] comments on Interoperability Draft 3

Chris Sibley Chris.Sibley at cbeyond.net
Wed Jan 11 15:01:24 EST 2006


> While I can't speak to the larger issue here, I strongly recommend
> "604" be replaced by "404".  In general, the 6xx codes are a bad idea.
> Whereas a 4xx code indicates that a particular agent has determined
> a particular fork of a request cannot be serviced, a 6xx code
> that the agent has determined that all forks of a request cannot be
> serviced, and tears down all the forks.  But the recipient of one fork
> can not know what URI the UAC originally requested, and so can never
> know if *all* forks of the request are useless.



You definitely bring up an interesting point. I was originally thinking
that if the terminating switch on the SS7 network returned an
'Unallocated (unassigned) Number' value in the 'Cause Value' field of an
SS7 ACM message, then the number shouldn't be accessible via any other
entity's PSTN connection either (i.e. the number is either
in-service/allocated on the terminating switch, or it isn't.)

However, I neglected to consider that other, non-SS7, routes might be
available (e.g. via SIP, using an ENUM lookup to determine the proxy
server's address). Therefore, it is possible that another fork might
actually be capable of reaching the dialed number. 

So, I agree with you that we should keep the response code 404 (as
opposed to changing it to 604).



This email may contain confidential information. If you are not
the intended recipient, please advise by return email and delete
immediately without reading or forwarding to others.
 - Cbeyond 

More information about the techwg mailing list