[126141] in North American Network Operators' Group
Re: MikroTik strikes again ?
daemon@ATHENA.MIT.EDU (Alexander Harrowell)
Mon May 3 06:49:40 2010
From: Alexander Harrowell <a.harrowell@gmail.com>
To: nanog@nanog.org
Date: Mon, 3 May 2010 11:48:42 +0100
In-Reply-To: <4BDEA4A9.9020904@spectraaccess.com>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
--nextPart2692073.sSqVaudT2N
Content-Type: Text/Plain;
charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
On Monday 03 May 2010 11:25:45 Bret Clark wrote:
> Uhm....okay...but why does anyone prepend their ASN that much? Are you=20
> saying the Mikrotik did that on purpose?
>=20
There was a well-known routing incident last year in which a difference=20
between the Mikrotik and Cisco CLIs caused the propagation of extremely lon=
g=20
AS-PATH attributes, which caused certain Cisco routers to crash.
Basically, someone remembered their Cisco IOS syntax and typed "bgp-prepend=
=20
47868" into a Mikrotik; the correct syntax would have been "bgp-prepend x=20
47868" where x is an integer between 0 and 16 representing the desired numb=
er=20
of prepends. The Mikrotik correctly tried to prepend 47868 47868 times, but=
=20
had only one byte to store this value and therefore produced 255 prepends.
Some Cisco machines, it turned out, had a bug that caused path lengths clos=
e=20
to 255 to crash them. Fun and games ensued.
The Renesys blog has much, much more:=20
http://www.renesys.com/blog/2009/02/longer-is-not-better.shtml
=2D-=20
The only thing worse than e-mail disclaimers...is people who send e-mail to=
=20
lists complaining about them
--nextPart2692073.sSqVaudT2N
Content-Type: application/pgp-signature; name=signature.asc
Content-Description: This is a digitally signed message part.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.12 (GNU/Linux)
iEYEABECAAYFAkveqhwACgkQ0c69vkueJcQz3wCdFpcHlNmdVQ8A0d5/1ukLozJC
MIwAn3RvG+LQcYBRqU1Ixlr5nOt4pqeZ
=H7g7
-----END PGP SIGNATURE-----
--nextPart2692073.sSqVaudT2N--
