RE: Firewall-1 and ISA D.o.S.

From: Jim Harrison (SPG) (jmharr@microsoft.com)
Date: 02/18/02


Date: Mon, 18 Feb 2002 08:53:50 -0800
From: "Jim Harrison (SPG)" <jmharr@microsoft.com>
To: <overclocking_a_la_abuela@hotmail.com>, <vuln-dev@securityfocus.com>

Interesting DoS (similar in concept to the UDP flood that thor@hammerofgod.com reported a few months ago), but how would you have the developers deal with it?
Every packet that is seen by any firewall takes some CPU time to examine and decide what to do with it.
Granted, under normal circumstances, this processing overhead is "assumed" and the performance specs for the device take that into account.
<rant>
Under situations where there is some jerk in the LAN that has decided to dump his job and leaves such a bomb lying in wait (really stupid to do it while he's still there), it's easily blocked at the network level so that the firewall doesn't have to deal with it. Tracking down this sort of game is comparatively simple and I'd personally take great pleasure in defenestrating that particular jackass.
</rant>

* Jim Harrison
MCP(NT4, 2K), A+, Network+
Services Platform Group

Never be afraid to try something new. Remember that amateurs built the Ark. Professionals built the Titanic.

-----Original Message-----
From: overclocking_a_la_abuela@hotmail.com [mailto:overclocking_a_la_abuela@hotmail.com]
Sent: Monday, February 18, 2002 04:43
To: vuln-dev@securityfocus.com
Subject: Re: Firewall-1 and ISA D.o.S.

In-Reply-To: <3.0.5.32.20020218085949.012f4100@192.228.128.13>

When you stop the attack, the firewall recovers, but

think that in the case of ISA D.o.S. I´m sending

spoofed packets so it will be more difficult to find the

attacker ( if you have not IDS or similar ).

Suppose the length of the D.o.S. is 1 hour... nobody

can surf the web, you can not access the ISA...,

probably no VPN,...

Think about it.

Hugo Vázquez Caramés

Security Consultant

>Received: (qmail 19118 invoked from network); 18

Feb 2002 06:09:16 -0000

>Received: from outgoing3.securityfocus.com

(HELO outgoing.securityfocus.com) (66.38.151.27)

> by mail.securityfocus.com with SMTP; 18 Feb

2002 06:09:16 -0000

>Received: from lists.securityfocus.com

(lists.securityfocus.com [66.38.151.19])

> by outgoing.securityfocus.com (Postfix)

with QMQP

> id 1EBEAA44EF; Sun, 17 Feb 2002

21:25:10 -0700 (MST)

>Mailing-List: contact vuln-dev-

help@securityfocus.com; run by ezmlm

>Precedence: bulk

>List-Id: <vuln-dev.list-id.securityfocus.com>

>List-Post: <mailto:vuln-dev@securityfocus.com>

>List-Help: <mailto:vuln-dev-

help@securityfocus.com>

>List-Unsubscribe: <mailto:vuln-dev-

unsubscribe@securityfocus.com>

>List-Subscribe: <mailto:vuln-dev-

subscribe@securityfocus.com>

>Delivered-To: mailing list vuln-

dev@securityfocus.com

>Delivered-To: moderator for vuln-

dev@securityfocus.com

>Received: (qmail 24253 invoked from network); 18

Feb 2002 00:53:21 -0000

>Message-Id: <3.0.5.32.20020218085949.012f410



Relevant Pages

  • Re: Stealth vs. Blocked
    ... > firewall logs)? ... > really viable DOS attack would be a flood of packets. ... Most good packet filtering personal firewalls intercept traffic at the ...
    (alt.computer.security)
  • Firewall-1 and ISA D.o.S.
    ... Check Point was not able to reproduce this attack ... a special situation: a firewall that accepts ... packets to port 80 with the SYN flag. ... In the case of Microsoft ISA Server I have been ...
    (Vuln-Dev)
  • Re: Application Layer packet inspection???
    ... packet to IIS causing a buffer overflow. ... the firewall itself: witness the recent attack on Snort: the ... attacker would send packets designed to attack the IDS and not any existing ...
    (comp.security.firewalls)
  • Re: [Full-Disclosure] A new TCP/IP blind data injection technique?
    ... For example the BorderWare Firewall will not accept fragmented packets, ... Then pass or drop the packet. ... > should be fairly easy to turn this into a practical attack. ... The other fragment of Bob's packet carry the ...
    (Full-Disclosure)
  • Re: Kerio PFW 2.14 - Safe?
    ... >> down user interface. ... Then consider the fact that most packet ... If Kerio 'X' says it's stateful it most ... >> way to know for sure would be to stand between the firewall and the ...
    (comp.security.firewalls)