[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Unaggregated prefixes in BGP4+ cloud

Hi Rob,

so far I have always tried to avoid prefix filtering on my
IPv6 routers. The problem is that this practice may hide
some of the router bugs or configuration mistakes that 
we are trying to fix as part of the 6bone effort.
Anyway I think that at present we do need some kind of
mechanism to enforce the 6bone hardening rules outlined in your
draft and prefix filtering seems to be an effective way to deal
with the matter.

So I will do as you suggest.
In addition, in order to help debugging, I'm going to make 
available an html page showing the prefix filters configured 
at the CSELT pTLA.


> ----------
> From: 	Robert J. Rockell[SMTP:[email protected]]
> Sent: 	Wednesday, November 03, 1999 3:22 PM
> To: 	Guardini Ivano
> Cc: 	'6bone'
> Subject: 	Re: Unaggregated prefixes in BGP4+ cloud
> Give them a week, and then filter them?   :)
> Thanks
> Rob Rockell
> Sprintlink Internet Service Center
> Operations Engineering
> 703-689-6322
> 1-800-724-3329, PIN 385-8833
> Ines|e gnyne qh vagr bz s|e Ino ngg una {e hgr bpu plxyne?
> On Wed, 3 Nov 1999, Guardini Ivano wrote:
> ->Hi all,
> ->
> ->during the last two weeks the number of unaggregated IPv6 prefixes
> ->advertised 
> ->within the BGP4+ cloud is increased a lot (look at
> ->http://carmen.cselt.it/ipv6/bgp/odd-routes.html 
> ->and http://carmen.cselt.it/ipv6/bgp/graphs/index.html) and I think that
> this
> ->is a
> ->very undesirable thing that should be fixed especially if we really
> consider
> ->the 6bone as a
> ->gymnasium for production use of IPv6.
> ->The ASs that are currently generating most of the unaggregated prefixes
> are:
> ->
> ->- AS7680 with 22 unaggregated prefixes
> ->- AS4556 with 20 unaggregated prefixes
> ->- AS2852 (CESNET) with 8 unaggregated prefixes
> ->- AS11008 (CENTAURI-AR) with 8 unaggregated prefixes
> ->
> ->Unfortunately I was not able to locate contact persons for AS7680 and
> AS4556
> ->in that they
> ->seem not to be registered in the 6bone database. 
> ->The poor use of the 6bone registry is another critical issue that we
> should
> ->try
> ->to address especially for the sites participating in the BGP4+ cloud.
> For
> ->example
> ->I think that any pTLA or pNLA should make sure that any new downstream
> BGP4+
> ->peer
> ->is correctly registered in the 6bone database before setting up the
> ->connection.
> ->
> ->Bye
> ->Ivano
> ->
> ->
> ->