[10565] in North American Network Operators' Group
Re: Keynote/Boardwatch Results
daemon@ATHENA.MIT.EDU (Golan Ben-Oni)
Wed Jul 9 18:44:28 1997
Date: Wed, 9 Jul 1997 18:00:26 -0400 (EDT)
From: Golan Ben-Oni <bnite@tremere.ios.com>
To: Eric Germann <ekgermann@cctec.com>
cc: leong@inversenet.com, Robert Laughlin <robert@portal.dx.net>,
Jack Rickard <jack.rickard@boardwatch.com>, nanog@merit.edu
In-Reply-To: <3.0.32.19970709172805.0076d99c@207.87.243.20>
On Wed, 9 Jul 1997, Eric Germann wrote:
> If you really want to test raw performance, why not just enable the chargen
> service on a router in the core and let him suck that down and measure
> throughput. No more effective DOS attach than 4 simultaneous chargen
> sessions on a T1. I suspect a 7513 should be able to spit a few characters
> per second :)
The tests that I've run in the past wouldn't support your statement. In
most cases, people leave those ports disabled, and appropriately so since
the routers have better things to do, and don't echo or generate traffic
nearly as quickly as the line rate would support.
