[29779] in North American Network Operators' Group
Re: bad idea?
daemon@ATHENA.MIT.EDU (Matt Zimmerman)
Fri Jul 7 00:05:07 2000
Date: Thu, 6 Jul 2000 22:07:52 -0400
From: Matt Zimmerman <mdz@csh.rit.edu>
To: Jamie Rishaw <jamie.rishaw@mypotential.com>
Cc: nanog@merit.edu
Message-ID: <20000706220752.A2992@csh.rit.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <B12B27355515D411B82300D0B7445683073637@DHARMA>; from jamie.rishaw@mypotential.com on Wed, Jul 05, 2000 at 11:20:50AM -0700
Errors-To: owner-nanog-outgoing@merit.edu
On Wed, Jul 05, 2000 at 11:20:50AM -0700, Jamie Rishaw wrote:
> True, but in the world of what-would-we-use-this-practice-for,
> it mostly applies to web, which for most intents is a one shot
> connect-get-drop; I doubt a single TCP RST would really irk too
> many users.
However, more and more sites are maintaining server-side state (even sites
that don't have a traditional "shopping cart"). An alarming number of these
seem to break in terms of navigation if that data is lost. Some, myself
included, would argue that these sites are poorly designed, but they do
exist in large numbers.
So, there is still a penalty in many cases for failing-over unnecessarily.
--
- mdz
