Full Disclosure mailing list archives

RE: Secure OWA


From: "Renshaw, Rick \(C.\)" <rrenshaw () ford com>
Date: Wed, 30 Aug 2006 11:00:18 -0400

-----Original Message-----
From: Brendan Dolan-Gavitt [mailto:mooyix () gmail com] 
Sent: Wednesday, August 30, 2006 9:58 AM
To: Renshaw, Rick (C.)
Cc: Dude VanWinkle; Adriel Desautels; full-disclosure () lists grok org uk
Subject: Re: [Full-disclosure] Secure OWA

On 8/30/06, Renshaw, Rick (C.) <rrenshaw () ford com> wrote:


-----Original Message-----
From: full-disclosure-bounces () lists grok org uk
[mailto:full-disclosure-bounces () lists grok org uk] On Behalf Of Dude 
VanWinkle
Sent: Saturday, August 26, 2006 2:30 PM
To: Adriel Desautels
Cc: full-disclosure () lists grok org uk
Subject: Re: [Full-disclosure] Secure OWA

The only real fault I know about is the fact that you can guess 
passwords
eternally without locking out user accounts.

There's two sides to this risk.  If you allow OWA logins to lock out 
accounts, and your OWA page is available from anywhere on the 
Internet, you are handing an easy DOS tool to anyone that knows the 
account names for people on your server.


Perhaps. But a temporary lockout period would deter brute-force attempts
while still making an attacker do some work to keep the accounts locked
(eg, if you have a lockout of 5 minutes, brute forcing is no longer 
practical, but at the same time, if you want to DoS someone's account 
you have to keep coming back every 5 minutes. And that increases the 
risk you'll get caught.)

-Brendan

My point was not matter which way you go on this issue, there is some risk.
The only thing that you can do is balance one risk against the other and
find the point where you feel comfortable with the risks.  You could
implement something like an exponential backoff wait between failed logins
without lockouts, which would make it more difficult to brute-force the
account, but there are ways around that too.  At the end of the day, you
have to pick which risk you are more comfortable dealing with, brute-force
attacks or DOS attacks.  Personally, I'd take the DOS, because it's better
than allowing passwords to be brute-forced (in my mind).

Rick

Attachment: smime.p7s
Description:

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/

Current thread: