-----Original Message-----
From:
"Weiter Leiter" <bp4mls@googlemail.com>
To: "Roger Lewau"
<roger.lewau@serverhallen.com>
Cc: serusers@lists.iptel.org
Date: Mon, 11 Sep 2006 03:55:25 +0200
Subject: Re: Handling 302
responses
Hi,
On 9/11/06, Roger Lewau <roger.lewau@serverhallen.com>
wrote:
> Hello list,
>
> I have searched this list for an
answer to the 302 response handling problem
> but found no real
solution. It seems no one actually has an aswer for this
> so I
studied the RFC 3261 and found the following statement:
>
> The
requesting client SHOULD retry the request at the new address(es) given
> by the contact header field.
>
> In my mind that statement
is completely off the wall, it is not the
> requesting client that
should be responsible for establishing the forwarded
> call, it never
is in the rest of the telecom industry so why should it be
> the case
for SIP?
SIP, indeed, moved some inteligence towards the edge of the
network,
into the clients, compared to older protocols.
On the other
hand, this helps to protocol's scalability (and this
characteristic can
be observed with dns or http or most of the scaling
protocols).
> Instead, this should ofcourse be the responsibility of the
>
forwarding client or the service provider on behalf of the forwarding
> client. But as the RFC is not crafted that way I need to find a way
to
> handle call forwarding in a proper way so that the cost for the
forwarded
> call ends up on the forwarding clients bill. As call
forwarding is a basic
> requirement in any phone network there must be
some one reading this list
> who has solved this issue that can share
there insight.
Normally the forwarding is handled by the registrar
responsible for
the callee (because it offers the callee greater
flexibility with his
forwarding settings).
But if the 'final' proxy is
missing this feature and a 3xx is replied,
what would prevent you to bill
your client which presumably makes a
new request, probably still through
your proxy, to some other
destination?
WL.
>
> Any
help on this issue is highly apreciated.
>
> Kind regards
> Roger Lewau
>
>
>