[155887] in North American Network Operators' Group
RE: Level 3 BGP Advertisements
daemon@ATHENA.MIT.EDU (Paul Vinciguerra)
Wed Aug 29 16:09:27 2012
From: Paul Vinciguerra <pvinci@VinciConsulting.com>
To: Blake Dunlap <ikiris@gmail.com>, "nick@flhsi.com" <nick@flhsi.com>
Date: Wed, 29 Aug 2012 20:08:52 +0000
In-Reply-To: <CAJvB4tmMqUVdsr6+czZ7dO_ULy-m95qoJgYT-W5QV2jfCteVDg@mail.gmail.com>
Cc: "nanog@nanog.org" <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
-----Original Message-----
From: Blake Dunlap [mailto:ikiris@gmail.com]=20
Sent: Wednesday, August 29, 2012 4:00 PM
To: nick@flhsi.com
Cc: nanog@nanog.org
Subject: Re: Level 3 BGP Advertisements
On Wed, Aug 29, 2012 at 2:56 PM, Nick Olsen <nick@flhsi.com> wrote:
> I hear you guys, It's done that way for a bit of traffic steering.
>
> If I could get away with just the aggregates I would, Trust me.
>
> Nick Olsen
> Network Operations (855) FLSPEED x106
>
> ----------------------------------------
> From: "Berry Mobley" <berry@gadsdenst.org>
> Sent: Wednesday, August 29, 2012 3:45 PM
> To: nanog@nanog.org
> Subject: Re: Level 3 BGP Advertisements
>
> [...]
>
> >Please, unless you really know why you need to do otherwise, just=20
> >originate your aggregates.
>
> +1
>
>
That should be unnessecary, the local prefs should already be winning as a =
customer vs transit/peer for equal prefix length.
As an aside, generally inbound traffic steering as a reason for disaggregat=
ion is fairly frowned upon by the community at large as it effectively make=
s everyone else pay more in additional hardware cost for your savings.
-Blake
If you have provided addressing from your aggregate to your customer and th=
ey have indicated that they are multi-homing, you need to preserve their pr=
efix-length in your outbound advertisements, or the redundant provider carr=
ies the inbound traffic. Is this also frowned on? To me, this is the mult=
ihoming tax we all pay for.
Paul
