Re: WGLC: draft-ietf-sip-connect-reuse-08.txt

Vijay K. Gurbani wrote:
> Paul Kyzivat wrote:
>> Interestingly, doing the DNS lookup for blue.com,
and recognizing that 
>> you already have a connection from that address
would work in this case.
> 
> Precisely, and therein lies the problem, right? 
example.com
> is using one connection to the virtual server to
deliver requests
> destined for another domain on the same virtual
server.
> 
> Or are we talking past each other?

Perhaps. I guess it depends on your assumptions about how
these things 
are demultiplexed. If at the connection level then it is a
problem. If 
at the message level then it isn't a problem.

	Paul


_______________________________________________
Sip mailing list  https://ww
w1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use sip-implementorscs.columbia.edu for questions on current
sip
Use sippingietf.org for new developments on the application of
sip

   From: Paul Kyzivat <pkyzivatcisco.com>

   Perhaps. I guess it depends on your assumptions about how
these things 
   are demultiplexed. If at the connection level then it is
a problem. If 
   at the message level then it isn't a problem.

My impression has always been that demultiplexing is done at
the
message level.  It seems implicit in the RFC 3263 processes
that once
one has resolved the address/port that any connection (with
the right
protocol) to that address/port is OK.  In any case, the
recipient
would have to see the first message before committing to
service only
one domain, as there isn't any other way to tell which
domain a new
connection is for.

Which has a derivative problem when setting up TLS that the
recipient
doesn't know which domain it needs to provide a certificate
for when
the connection is established.  IIRC, someone has proposed a
solution
where the recipient presents a certificate that specifies
all of the
domains it is authoritative for.

Dale


_______________________________________________
Sip mailing list  https://ww
w1.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use sip-implementorscs.columbia.edu for questions on current
sip
Use sippingietf.org for new developments on the application of
sip