Re: Proxy: comments on section 9.1.3

["'Mathew Richardson'" <[e-mail]>]

> Cambria, Mike <[e-mail]> [Fri, Sep 06, 2002 at 11:53:53AM -0400]:
>
>(Pretending that I'm) reading the draft for the first time...
>
>Your comment:
>
>Somewhere in section 9.1.x, Do we want to mention the fact that the best 
>Loc-RIB BGP route for a given destination is still advertised to outgoing 
>peers via the Adj-RIB-out after it passes the outbound policy filtering,
>even though it is not really used for forwarding?
>
>and the text in 9.1.3 can be read as a contraction of section 2 (when only
>considering the base document, no route reflectors, route servers etc.):
>
>"... one must focus on the rule that a BGP speaker advertises to its peers
>(other BGP speakers which it communicates with) in neighboring ASs only
>those routes that it itself uses."

I thought that that was in there somewhere :)  I guess I've just gotten
in to the bad habit of skipping the intro and going straight to the
"good stuff" :)

>If, in the scope of an implementation of only what is specified in this
>draft, one can advertise to a peer what is in Loc-RIB but not the Route
>Table, the text in section 2 should be clarified.  Otherwise, the text in
>9.1.3 (and your comment) should be clarified, as it doesn't mention anything
>about the route table.

How about changing the paragraph after c) in 9.1.2 to:

  The local speaker SHALL then install that route in the Loc-RIB,
  replacing any route to the same destination that is currently being
  held in the Loc-RIB.  This route SHALL then also be installed in
  the BGP speakers forwarding table.

<snip>

I personally much prefer this.  It helps keep the base spec focused
on describing the base protocol.

mrr