Francis Dupont wrote:
>
> In your previous mail you wrote:
>
> There is no proposal to reveal port/protocol numbers
> in the flow label field, for the simple reason that it's impossible;
> there is no reversible hash of all those bits into 19. So the
> idea fails even without the security argument (which is a
> strong one too).
>
> => so you give up the appendix A of draft-conta-ipv6-flow-label-02.txt?
> The diffserv flow label in proposition c) will be an extension of
> the DS field with same kind of properties?
Sure, it was included as FYI.
>
> As for whether people believe in 5-tuple classification (for
> non-encrypted traffic), it surely isn't an issue for this WG.
>
> => I'd like to get the head of the 5-tuple classification in the core (:-)!
>
> What Alex and I came here to discuss was doing better than
> that with IPv6.
>
> => I understand but I shan't support the c) option until the 5-tuple
> re-classifier monster is killed.
That is not the business of this WG and irrelevant to the issue of
using the flow label. It's also running code and I think out
of the hands of the IETF and in the hands of the market to decide.
Brian
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
