On Mon, 2009-03-02 at 15:20 -0500, Robert Joly wrote:
> > 
> > On Mon, 02 Mar 2009 14:08:51 -0500, Joly, Robert (CAR:9D30) 
> > <[email protected]> wrote:
> > 
> > > 
> > >
> > >> -----Original Message-----
> > >> From: Spitzer, Andy (BL60:9D30)
> > >> Sent: Monday, March 02, 2009 2:00 PM
> > >> To: Joly, Robert (CAR:9D30); [email protected]
> > >> Subject: Re: [sipX-dev] Fixing 2233 - Caller ID is not 
> > displayed on 
> > >> the Gateway.
> > >>
> > >> Woof!
> > >>
> > >> On Mon, 02 Mar 2009 13:37:09 -0500, Robert Joly <[email protected]> 
> > >> wrote:
> > >>
> > >> > Solution #2: strip off the P-Asserted-Identity as the
> > >> request leaves
> > >> > sipXproxy after is has completed its spiraling.
> > >>
> > >> Instead of striping it as it leaves sipXproxy, set it to the
> > >> From: header so they match?
> > >
> > > That would still leave you with drawback 'b' of solution #1.  The 
> > > identity that was asserted by sipXproxy is the user's 
> > actual identity 
> > > - not its caller ID alias.  As an example, consider 100 and 200 two 
> > > valid users on SIP server 'example.com'.  User 100 configures its 
> > > caller ID to be '200'.  With this proposal, the sipXproxy 
> > will assert 
> > > that the identity of the requester is [email protected] when 
> > in fact it 
> > > is [email protected].
> > > 
> > 
> > Well, perhaps I'm being obtuse here, but isn't the reason we 
> > "fake" the outbound caller ID is so that it looks as if you 
> > are someone you are not?
> > 
> > In the real world, jamming the "from" field is to tell the 
> > gateway to set the calling line ID to a known DID number 
> > (either the caller's individual one, or the pilot number of 
> > the hunt group).  It isn't really meant to mascarade one 
> > registered user as another.  But so what?  That's what the 
> > admin wants to do, so why would it be "wrong" to change PAI?  
> > It's just as "wrong" to change "from" but we allow that (now).
> 
> I understand what you are describing and you may well be right.  I was
> approaching it from the point of view that the semantics of PAI is to
> 'assert' the actual identity of the request originator within a trust
> domain.  Given the intent of the PAI, my natural reflex is not fudge it
> with admin-definable data as it could denature its purpose but I realize
> that is all very debatable.  Scott, I would like to hear your take on
> this.

I'd rather strip it than change it.


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

Reply via email to