Re: Private draft submission

[Ben Black <[e-mail]>]

Anyone have copies of the DPA drafts around?

Destination Preference Attribute for BGP
        <draft-ietf-idr-bgp-dpa-05.txt>

Application of the BGP Destination Preference Attribute in Implementing
Symmetric Routing
        <draft-ietf-idr-dpa-application-02.txt>

Current Practice of Implementing Symmetric Routing and Load Sharing in the
Multi-Provider Internet
        <draft-ietf-idr-symm-multi-prov-02.txt>

They may offer an option which can be implemented entirely within the
existing framework.

On Thu, Nov 04, 1999 at 02:55:20PM -0500, Vijay Gill wrote:
> On Thu, 4 Nov 1999, Paul Traina wrote:
> 
> > Simple facts of life wrt. bgp metrics:
> > 
> > (a) we don't change the tie breaking algorithm
> > (b) any "metric" is taken as advisory, and said metric
> >     should become a modifier to the LOCAL_PREF attribute
> >     generated by the ingress router to an AS.
> > 
> > 
> > So, if we were to do Vadim's old proposal of a vector of preference
> > metrics, we'd get:
> > 
> > 	LOCAL_PREF += SUM(k1p1, k2p2, k3p3, k4p4, k5p5)
> > 
> > where the Ks are scaler multipliers looked up in a table based upon
> > AS number's.
> > 
> > The big thing to realize is that all of this stuff just changes
> > local_pref.  Changing the tie breaking logic at this point is
> > insanity.
> 
> No one is disagreeing on that one.
> 
> We just need to push something that is workable and consisent and I think
> vadim's draft is about as useful as anything I've seen in here. Thats for
> sure. 
> 
> /vijay
> 
> 

-- 
 --b