[8178] in North American Network Operators' Group
Re: consistent policy != consistent announcements
daemon@ATHENA.MIT.EDU (John Scudder)
Fri Mar 14 12:43:43 1997
From: John Scudder <jgs@ieng.com>
To: apb@iafrica.com (Alan Barrett)
Date: Fri, 14 Mar 1997 12:14:44 -0500 (EST)
Cc: randy@psg.com, nanog@merit.edu
In-Reply-To: <Pine.NEB.3.95.970314182414.162i-100000@apb.iafrica.com> from "Alan Barrett" at Mar 14, 97 06:57:51 pm
> > > M
> > > / \
> > > A B * Peer link
> > > | * | Customer link
> > > RRRRRRR
> > > Point1 * * Point2
> > > VVVVVVV
[...]
> R could request A to provide it with a list of ASes for indirect
> customers behind A. (R probably already does that.) That would be
> sufficient information for R's router at the R/B interconnection to tag
> M's routes as customer routes. Essentially, when R's router at the R/B
> interconnection receives a route with path "B M", it could use the fact
> "M is an indirect customer" rather than "B is a non-customer" to tag the
> route appropriately.
Furthermore, R could provide sufficient incentive for A to provide a
list of indirect customers by accepting only registered routes (or AS
paths). (This should sound familiar.) E.g. (A) and (A M) routes would
be accepted but all other (A *) would not be in this example.
Alternately, R could audit routing tables at Point1 and Point2 from
time to time, as I mentioned earlier. It ought to be rather simple to
find routes which are in Point1 as "exportable" and Point 2 as
"non-exportable", or vice-versa. The rest follows.
Regards,
--John
