[9393] in North American Network Operators' Group

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

Re: Warning, CDP interaction at Mae-EAST

daemon@ATHENA.MIT.EDU (Brett D. Watson)
Tue May 20 18:21:23 1997

To: Golan Ben-Oni <bnite@tremere.ios.com>
cc: Owen DeLong <owen@delong.sj.ca.us>, nanog@merit.edu
From: "Brett D. Watson" <bwatson@genuity.net>
Reply-To: bwatson@genuity.net
Date: Tue, 20 May 1997 15:03:58 -0700

> > Why on EARTH?!? would _ANYONE_ run CDP at a MAE or any other exchange point?

> 
> Given that its on by default cdp needs to either be turned off completely,
> or disabled for a particular interface, folks probably just leave it on.
> 
> I highly doubt CDP contributes to congestion problems with the little
> amount of traffic it generates - Although if you've ever looked at some of
> the traffic at the Mae's you'd be sickened by all of the other things
> floating around.  

  debug ip udp sometime on a nap router.  there are many wonderful 
bits of traffic floating around:

peer1.wdc1#
May 20 14:53:03.681 MST: UDP: rcvd src=192.41.177.98(520), dst=192.41.177.255(520)
May 20 14:53:07.968 MST: UDP: rcvd src=207.240.0.1(123), dst=207.240.56.2(123), length=56
May 20 14:53:08.396 MST: UDP: rcvd src=192.41.177.70(123), dst=255.255.255.255(123), length=56no
May 20 14:53:18.152 MST: UDP: rcvd src=207.240.0.25(123), dst=207.240.56.2(123), length=56 debug all

  cool, broadcast ntp, broadcast rip.  i'm even seeing bootp from 
various peers here and at mae-west.  i noticed this last year 
sometime and posted about it.  i wonder how much all this extraneous 
garbage really contributes to total traffic at the naps.

-brett



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