[147570] in North American Network Operators' Group
Re: Time Warner Routing Issue
daemon@ATHENA.MIT.EDU (Courtney Smith)
Wed Dec 14 17:41:26 2011
From: Courtney Smith <courtneysmith@comcast.net>
Date: Wed, 14 Dec 2011 17:40:14 -0500
In-Reply-To: <mailman.16842.1323896954.1873.nanog@nanog.org>
To: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Dec 14, 2011, at 4:09 PM, nanog-request@nanog.org wrote:
>=20
> Message: 3
> Date: Wed, 14 Dec 2011 15:46:52 -0500
> From: Brian Christopher Raaen <mailing-lists@brianraaen.com>
> To: nanog@nanog.org
> Subject: Time Warner Routing Issue
> Message-ID:
> =
<CAEV7mMFS7pAmxwB_SLNpYEq9m3OwAL792k0hpSx5Nse2H=3DuHtg@mail.gmail.com>
> Content-Type: text/plain; charset=3DISO-8859-1
>=20
> They said the reason was they didn't
> have an LOA (which they had gotten back in October) and the ip blocks =
were
> not in the Level3 Radb list. I could still see announcement to some =
peers
> (Shaw Cable in Canada) in a few looking glasses and BGP routers. =
However
> my network blocks were not showing for the larger US carriers like =
Qwest
> and AT&T. One of their techs just called me back and said that Level3
> should be advertising it, but I still do not see the routes on the =
AT&T
> route server. After noting that the tech said that it may take =
another day
> for BGP to "propagate" to other peers as they update their radb =
tables. In
> my experience I've never seen anything where I had to wait for a route =
to
> propagate other than standard routing table updates which usually take =
less
> then an hour, and I'd really not expect this many problems between =
Tier1
> and Tier2 providers. I need to know if this matches other's =
experience and
> wanting to know what other people were seeing with traceroutes and =
"show ip
> bgp". The networks in question at the following 4 /24's
>=20
> 68.68.176.0/24
> 68.68.177.0/24
> 68.68.178.0/24
> 68.68.179.0/24
>=20
> the serial ip address is 72.43.84.254
>=20
> Thanks for your assistance.
>=20
> ---
> Brian Raaen
> Zcorum
> Network Architect
>=20
I believe the TWC tech is referring to prefix filter updates and not =
routing table updates. Some of TWC's peers most likely use IRR based =
filters on their BGP sessions. Most networks only check once a day for =
IRR changes and apply those changes to filters. It looks TWC put in =
proxy route objects today for your blocks so I am guessing someone at =
TWC forgot to make the updates in advance. Work around is for them to =
reach out to the peers with IRR based filters and request a manual =
update. Sounds like that is what they are doing.
route: 68.68.176.0/24
descr: RR-RC-Princetown Cable Company, Inc.-Albany
origin: AS11351
notify: ipaddreg@rr.com
mnt-by: MAINT-RR
changed: ipaddreg@rr.com 20111214
source: RADB
route: 68.68.177.0/24
descr: RR-RC-Princetown Cable Company, Inc.-Albany
origin: AS11351
notify: ipaddreg@rr.com
mnt-by: MAINT-RR
changed: ipaddreg@rr.com 20111214
source: RADB
Run the below on 12/15 and it should return your blocks.
$ whois -h filtergen.level3.net radb::as11351 | grep 68.6
$=20
Courtney Smith
courtneysmith@comcast.net
() ascii ribbon campaign - against html e-mail=20
/\ www.asciiribbon.org - against proprietary attachments
