[60230] in North American Network Operators' Group

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

RE: AS announcement question (easy)

daemon@ATHENA.MIT.EDU (Pete Templin)
Mon Aug 4 14:56:11 2003

Date: Mon, 4 Aug 2003 13:50:56 -0500
From: "Pete Templin" <pete.templin@texlink.com>
To: "Mike Donahue" <mdonahue@WATG.com>, <nanog@merit.edu>
Errors-To: owner-nanog-outgoing@merit.edu


Is it a class C, or a /24? =20

Regardless, longest match wins.  If old provider is announcing your =
space as part of a larger announcement, you'll be a bad netizen for a =
few days because you're violating the RFCs concerning inconsistent AS, =
but your traffic will swing almost entirely over to new provider, and =
rather quickly.  Traffic originating in old provider's network will =
likely still come in on your old link.  (I've done this for three new =
customers in the past month.)  If your old provider is specifically =
announcing your space as a /24, standard BGP path selection processes =
will take place all over the 'Net, and YMMV on the traffic balance.  On =
or before disconnecting the link to old provider (if they are =
specifically announcing your /24), be sure to have them stop announcing =
that /24.

Pete Templin
Senior Staff Engineer
TexLink Communications
(210) 892-4183
pete.templin@texlink.com


-----Original Message-----
From: Mike Donahue [mailto:mdonahue@WATG.com]
Sent: Monday, August 04, 2003 1:22 PM
To: 'nanog@merit.edu'
Subject: AS announcement question (easy)


Hi, sorry for this low-level question, but at least I'm not posting in =
html.
:)

My company owns a class C, and we're switching ISPs.  The "new" provider =
is
telling us that they can start announcing, without us having to tell the =
old
provider to stop announcing.  I.e., a 2-3 day period where both are
announcing our class C.

This conflicts with my extremely (admittedly)limited knowledge.  Can =
someone
let me know if this is OK/not OK?

Thanks in advance,

Mike

Michael Donahue
WATG
(949) 574-8500 x261

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