Question on a NAT traversal requirement

[Paul E. Jones]

Roni,

This is a requirement statement, not a statement tied to a particular
solution.  Interpreting the words, I think what I said would be correct
irrespective of any particular solution.  However, I delved into H.323v1 and
H.323v2 discussions, as I think any person's immediate question might be
"why not v1 or v2".

You said in a separate reply that using an ALG or a proxy solution, even
H.323v1 might be supported.  That's entirely right, which makes the
requirement even more confusing.  This requirement just says that, whatever
we do to address NAT/FW issues, we need to ensure it works H.323v3 or higher
and it's OK if the final solution will not work with H.323v1 or v2.

Paul

> -----Original Message-----
> From: Even, Roni [mailto:roni.even at polycom.co.il]
> Sent: Wednesday, August 23, 2006 2:45 AM
> To: Paul E. Jones; OKUBO Sakae; itu-sg16 at external.cisco.com
> Subject: RE: Question on a NAT traversal requirement
> 
> Paul,
> I would like to emphasis that this is true only for Nat traversal based
> on H.460 (17, 18, 19) and not to other means of NAT traversal
> Roni Even
> 
> > -----Original Message-----
> > From: Paul E. Jones [mailto:paulej at PACKETIZER.COM]
> > Sent: Wednesday, August 23, 2006 7:18 AM
> > To: 'OKUBO Sakae'; itu-sg16 at external.cisco.com
> > Subject: RE: Question on a NAT traversal requirement
> >
> > Mr. Okubo,
> >
> > Given this wording, I would say the answer to your question is "it MAY
> > work
> > with H.323v1 or H.323v2, but SHALL work with H.323v3, v4, v5, v6, ..."
> >
> > In reality, I don't believe v1 systems can be supported, since H.323v1
> > systems were required to use H.245v2.  H.323v2 lifted this
> restriction, so
> > devices could implement later versions of H.245 approved independently
> of
> > H.323.
> >
> > Paul
> >
> > > -----Original Message-----
> > > From: OKUBO Sakae [mailto:okubo at MXZ.MESH.NE.JP]
> > > Sent: Tuesday, August 22, 2006 4:10 AM
> > > To: itu-sg16 at external.cisco.com
> > > Subject: Question on a NAT traversal requirement
> > >
> > > Dear NAT traversal experts,
> > >
> > > TTC (The Telecommunication Technology Committee) is now working on
> > > Technical Paper "Requirements for Network Address Translator and
> > > Firewall Traversal of H.323 Multimedia Systems" to convert it to a
> > > Technical Report in Japanese language.
> > >
> > > I have a question on how to interpret the following part:
> > >
> > > ----------
> > > 9.3	Requirements on Signalling and Media Streams
> > > b)	NAT traversal mechanisms for H.323 multimedia systems shall
> > > support no less than ITU-T H.323v3, H.245v7, H.225.0v4, and H.235v3.
> > > All later versions of H.323, H.245, H.225.0 and H.235 may be
> > > supported.
> > > ----------
> > >
> > > Take an example of H.323 that has now v6. Which of the following
> > > statements are true?
> > >
> > > NAT traversal
> > >
> > > 1/ shall not support v1 or v2.
> > > 2/ may support v1 or v2.
> > > 3/ shall support v3.
> > > 4/ shall support v3 and in addition may support any combinations of
> > > v4, v5 and v6.
> > > 5/ shall support v3 and may support v4, or (v4 and v5), or (v4, v5
> and
> > > v6).
> > >
> > > In other words, NAT traversal with the following support conforms to
> > > this requirement:
> > >
> > > a/ v3
> > > b/ v4
> > > c/ v5
> > > d/ v6
> > > e/ v3, v4
> > > f/ v3, v5
> > > g/ v3, v6
> > > h/ v3, v4, v5
> > > i/ v3, v4, v5, v6
> > > j/ v4, v5
> > > k/ v4, v6
> > > l/ v4, v5, v6
> > > m/ v5, v6
> > >
> > > --
> > > Best regards,
> > >
> > > OKUBO Sakae
> > > e-mail: okubo at aoni.waseda.jp
> > >          (<== sokubo at waseda.jp, valid until 31 March 2007)
> > > Visiting Professor
> > > Global Information and Telecommunication Institute (GITI)
> > > Waseda University
> > > ******************************************************************
> > > Waseda University, YRP Ichibankan 312        Tel: +81 46 847 5406
> > > 3-4 Hikarinooka, Yokosuka-shi, Kanagawa-ken  Fax: +81 46 847 5413
> > > 239-0847 Japan
> > >                        H.323 videoconferencing: arranged by advice
> > > ******************************************************************
> > >
>