[73529] in North American Network Operators' Group
Re: BGP Homing Question
daemon@ATHENA.MIT.EDU (Patrick W Gilmore)
Fri Aug 27 11:18:48 2004
In-Reply-To: <D9214A92-F828-11D8-9397-000D93B24C7A@isc.org>
Cc: Patrick W Gilmore <patrick@ianai.net>
From: Patrick W Gilmore <patrick@ianai.net>
Date: Fri, 27 Aug 2004 11:16:40 -0400
To: nanog@merit.edu
Errors-To: owner-nanog-outgoing@merit.edu
On Aug 27, 2004, at 8:58 AM, Joe Abley wrote:
> On 27 Aug 2004, at 08:13, Rick Lowery wrote:
>> I know=A0they would not be=A0good Internet citizen, but=A0if they =
needed to=20
>> do this for a temp basis does anyone see an issue?
>
> There's not much bad citizenry in what you are suggesting: the=20
> assigning-RIR problem is a non-problem, and your two sites are still=20=
> only going to originate one prefix each (which they would presumably=20=
> do even if you had a separate LIR assignment for the European node).
There is zero "bad citizenry" in this, and don't let anyone tell you=20
differently. It is your netblock, you get to use it as needed. This=20
is much better than getting another /20 for an EU site that only needs=20=
a /24.
Also, filtering will not be an issue, if you are careful. Anyone who=20
does not hear the /24 will hear the /20. Packets for the /24 will go=20
to your US upstream. As long as your US upstream peers with your EU=20
upstream, and does not filter the /24 being announced over that peering=20=
link, they will send the bits where they belong. Since this is much=20
more common than the alternative, you will likely have full=20
connectivity.
Anyone knows who filters these days? Sprint stopped when Sean left. =20
Verio stopped when Randy left. I don't know anyone beating that drum=20
any more. (Kinda nice, actually.) I've heard some Asian ISPs do, but=20=
don't remember who.
--=20
TTFN,
patrick
