As long as this is used in tandem with a default route, this is a
workable solution. While this solution is less than desirable, its
better than a default route alone.  At least with this solution, one
with two to three peers can still receive the aggregate routes from the
various upstreams (I do agree that there may be many problems cause with
this solution.

-----Original Message-----
From: Daniel Roesen [mailto:dr@...Sent: Saturday, April 12, 2003 8:13 PM
To: isp-bgp@isp-bgp.com
Subject: [isp-bgp] RE: Prefix length distribution statistics


On Fri, Mar 07, 2003 at 08:48:33AM +1000, David J. Hughes wrote:
> Excellent.  So, by dropping anything worse than a /22 the table size 
> is reduced to less than 45,000 entries.

And you won't reach a lot of places anymore.


Regards,
Daniel

-- 
Daniel Roesen                                  email: dr@...
Network Engineer                               phone: +49 89 92699 0
Cable & Wireless IP GSOC Europe                  fax: +49 89 92699 809
Landsberger Str. 155, 80687 Munich, Germany   comnet: 49 9412

------------------------ANNOUNCEMENT---------------------------------
---------------------------------------------------------------------
       >> Solving Spam With Sender Accountability << Blacklists,
whitelists, spam trap networks, baysian filters and more. Be part of the
solution. ISPcon's Roundtable discussion is April 23.
   http://www.ispcon.com/spring2003/attend-specevents.asp
----------------------------------------------------------------------
----------------------------------------------------------------------









------------------------ANNOUNCEMENT---------------------------------
---------------------------------------------------------------------
       >> Solving Spam With Sender Accountability <<
Blacklists, whitelists, spam trap networks, baysian filters and more.
Be part of the solution. ISPcon's Roundtable discussion is April 23.
   http://www.ispcon.com/spring2003/attend-specevents.asp
----------------------------------------------------------------------
----------------------------------------------------------------------

Replies

[isp-bgp] RE: Prefix length distribution statistics, Daniel Roesen