Brian E Carpenter wrote:
> [EMAIL PROTECTED] wrote:
> >
> > Robert,
> >
> > Very well put! Just some additional thoughts:
> >
> > - If diffserv WG so wants, it can even define a specific
> (standard) DSCP
> > value to be used between domains to signify that the flow
> label in the same
> > IP header has "diffserv semantics". This way the
> classification could be
> > done the same for all source addresses (otherwise ingress
> router would need
> > to know which source addresses use diffserv semantics for
> the flow label?).
>
> That's an interesting thought, but I think it doesn't work.
> The egress router
> has to *know* that the flow label is in diffserv format, and
> I think that
> requires some magic.
>
Wouldn't the absence of intserv state for the flow at the domain egress be
the magic for this determination?
A related question: Flow label field is presumably end-to-end, i.e.
non-mutable. How can some random diffserv domain egress router know if the
flow label value should be honored or not (even if it had a specific format
for the PHB-ID)?
> >
> > - If there is a diffserv relationship between a client and
> an ISP, the
> > client could still use intserv for select flows by doing
> intserv signaling,
> > which would explicitly tell the ISP that the associated
> flow label does not
> > have any diffserv semantics.
>
> Only for unencrypted traffic I think? The problem case is
> encrypted traffic,
> where all you can "see" are the IP addresses and the flow
> label. I think the
> intserv flow spec will be insufficient in that case.
>
I thought the current definition of the MF-classifier in intserv would allow
usage of the flow label field for classification. At least this has been the
whole function of the flow label: to enable flow classification based on the
IPv6 header data only (e.g. even in the non-encrypted case IPv6 transport
headers should NEVER be used for flow classification). If the flow label is
not part of the intserv classifier specs, then maybe we need to revisit RSVP
(RSVPv6?). Or perhaps the starting NSIS work will make this happen?
Jarno
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
