[75494] in North American Network Operators' Group

home help back first fref pref prev next nref lref last post

RE: The Cidr Report

daemon@ATHENA.MIT.EDU (Geoff Huston)
Sat Nov 13 19:36:02 2004

Date: Sun, 14 Nov 2004 11:35:15 +1100
To: "Roldan, Brad" <broldan@Covad.COM>,
	"Christopher L. Morrow" <christopher.morrow@mci.com>,
	"Randy Bush" <randy@psg.com>
From: Geoff Huston <cidr-report@potaroo.net>
Cc: <nanog@merit.edu>, <eof-list@ripe.net>, <apops@apops.net>,
	<routing-wg@ripe.net>, <afnog@afnog.org>, <swinog@swinog.ch>
In-Reply-To: <AF696B16DAC33B46B1911BAD0CCECF7F024E484E@ZANEVS03.cc-ntd1.
 covad.com>
Errors-To: owner-nanog-outgoing@merit.edu


Interestingly enough what Covad appears to be saying is:

If we had a way to announce two things

1 - here are the advertisements for covering aggregates for Covad

AND

2 - do not believe any more specifics for these address blocks, as they are 
NOT part of Covad's routing policy for these prefixes

then we would not be seeing this unfortunate case of unauthorized route 
leakage being resolved in a way that seems to have unfortunate bgp 
implications in terms of more specifics appearing.

So its an interesting question. How could Covad achieve a routing policy 
announcement of the form as stated in 2 above?

regards,

Geoff



home help back first fref pref prev next nref lref last post