On Fri, 2009-12-04 at 11:13 -0500, Dale Worley wrote:
> On Fri, 2009-12-04 at 10:58 -0500, Scott Lawrence wrote:
> > On Fri, 2009-12-04 at 10:46 -0500, Dale Worley wrote:
> > > 
> > > As for how to handle it, the bridge should reflect the authentication
> > > challenge back to its incoming side, so the caller can provide
> > > authentication (if possible).
> > 
> > No... not for ITSP connections at least - there's no way that a call
> > from the PSTN is going to have credentials.  The call should fail,
> > probably with a 403 response that has text explaining that authorization
> > is required by the internal proxy.
> 
> It seems to me that sending a 401/407 back is a good strategy:  (1) It
> unambiguously shows that authorization is required by the proxy (or some
> other component of the terminating system).  (2) The bridge doesn't have
> to figure out whether the originating system is an ITSP (to be given a
> 403) or an external SIP system (to be given the 401/407).
> 
> A 403 response would be better if we know that there are originating
> systems that handle 403 failures in a better way than they handle
> 401/407 failures.

This has to be put in context: that the mission of sipXbridge is to
interface to ITSPs and other lame systems that require hacks like
registration for the domain and intercepting REFER, _not_ to be a
general purpose interface to other SIP systems (we want those to go
direct to sipXproxy, which _would_ just return the challenge responses).

_______________________________________________
sipx-dev mailing list [email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev
sipXecs IP PBX -- http://www.sipfoundry.org/

Reply via email to