Michael Thomas wrote:
>
> Brian E Carpenter writes:
> > The point is that when a packet crosses an administrative domain boundary,
> > the downstream ISP typically wants to reclassify the packet all over again,
> > i.e. does not accept the incoming DSCP as definitive. This was a very
> > clearly stated ISP requirement at the start of diffserv and is fundamental
> > in the diffserv architecture.
>
> Right... I must be missing something. You copy the inner DSCP
> to the tunneled packet header, it traverses the network just like
> normal with whatever re-writing rules apply at the administrative
> edges, and once the packet is decapsulated, you copy the outer
> DSCP back into the inner and goes along its merry way.
>
> What am I missing?
The tunnel aspect is discussed in RFC 2983 in some detail. The problem I mention
is when the packet is inside a transport mode ESP header when it crosses
an ISP boundary, and the SLA in place requires reclassification (i.e.the
downstream ISP has not contracted to honor the upstream ISP's DSCP
values). I think you're missing the distinction in diffserv between
BA and MF classifiers (see RFC 2474 and 2475).
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]
--------------------------------------------------------------------
