Great Circle Associates Firewalls
(October 1997)
 

Indexed By Date: [Previous] [Next] Indexed By Thread: [Previous] [Next]

Subject: Re: sex,lies, and application proxy based fw vs Check Point
From: "Paul D. Robertson" <proberts @ clark . net>
Date: Fri, 31 Oct 1997 08:12:37 -0500 (EST)
To: Michael S Hines <mshines @ purdue . edu>
Cc: Akhila Sinha <ccf15429 @ cc . iitd . ernet . in>, firewalls @ GreatCircle . COM
In-reply-to: <3457363745d0002 @ scribe . cc . purdue . edu>

On Wed, 29 Oct 1997, Michael S Hines wrote:

> OOB = out of bounds - such as an undefined packet type presented to TCP/IP - 
> which it should discard, but which - depending on implementation (ie the 
> programmers perogative) may misbehave.

Actually 'Out Of Band', which is a perfectly well-defined packet which 
should be, if definined in the application layer protocol, processed 
immediately (hence out of band), rather than in the order it was received 
in the TCP stream.  OOB data is indicated by the URG flag set in the 
packet.  As Darren has pointed out, the applications programmer of an 
application receiving OOB data must specificily ask to receive such 
data.  It's important to understand that this is a perfectly 
legitimate, well-defined TCP packet which was being handled incorrectly by 
Microsoft's TCP implementations.  Hence my assertion that packet filters 
(with or without state) don't protected from Internetwork or lower 
transport layer problems that they don't know about.

Paul
-----------------------------------------------------------------------------
Paul D. Robertson      "My statements in this message are personal opinions
proberts @
 clark .
 net      which may have no basis whatsoever in fact."
                                                                     PSB#9280



Follow-Ups:
References:
Indexed By Date Previous: Re: Linux et al PFs
From: Darren Reed <avalon @ coombs . anu . edu . au>
Next: Re: FIREWALL: Encryption round up?
From: "Paul D. Robertson" <proberts @ clark . net>
Indexed By Thread Previous: Re: sex,lies, and application proxy based fw vs Check Point
From: "Michael S Hines" <mshines @ purdue . edu>
Next: Re: sex,lies, and application proxy based fw vs Check Point
From: "Gary R. Wolfe" <Gary . Wolfe @ Eng . Sun . COM>

Google
 
Search Internet Search www.greatcircle.com