Jarno Rajahalme wrote:
> Why would any subsequent provider give a S on what the origin
> intended?
Because they do that now by snooping the immutable
protocol / port bits. If they had a standard set of
DSCP values to work with they could use those.
> There is difference in "muting" to only locally understood
> value and to
> globally standardized values.
Yes, but if the original meaning is lost one of the
downstream providers would have no way to resurect it.
> All this discussion is based on an assumption that a diffserv
> domain egress
> cannot map from a local (yes, it's own) DSCP->PHB mapping to
> a standard
> recommended DSCP value for the same (or equivalent) PHB. If
> this mapping CAN
> be done, that is what NEEDS to be done, and we can forget this whole
> discussion.
I agree completely.
> This far Brian has maintained that this cannot be
> forced on The Operators.
He is correct it can't be forced, but the standard could say
MUST and a BCP could be written to enforce it. What he is
concerned about is that if those actions were taken several
operators have said they will not deploy. Which is worse,
having them deploy something that is useless in the end, or
refusing to deploy what they will reluctantly have to do in
another form?
Tony
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
