Re: [Asrg] Ban the bounce; improved challenge-response systems

[J C Lawrence <claw@kanga.nu>]

On 10 Apr 2003 13:48:00 +0200 
list-ietf-antispam  <list-ietf-antispam@faerber.muc.de> wrote:

> However, it might be possible to check for a connection abort and to
> abort the handling if that happens.

There's a race condition there.  Did the abort happen before or after
the far end received the 2250?  This is the core problem in protocol
design, and it is insolvable.  There is no way to establish perfect
knowledge of each other's state between two systems via messages over a
non-instantaneous link.  There are many things you can do to cut the
window of risk down, but it can't be zero.

-- 
J C Lawrence                
---------(*)                Satan, oscillate my metallic sonatas. 
claw@kanga.nu               He lived as a devil, eh?		  
http://www.kanga.nu/~claw/  Evil is a name of a foeman, as I live.
_______________________________________________
Asrg mailing list
Asrg@ietf.org
https://www1.ietf.org/mailman/listinfo/asrg