WebApp Sec mailing list archives

Re: [WEB SECURITY] Fundamental error in Corsaire's paper?


From: Achim Hoffmann <kirke11 () securenet de>
Date: Sat, 29 Apr 2006 21:09:17 +0200 (MEST)

On Fri, 28 Apr 2006, Amit Klein (AKsecurity) wrote:

!! On 28 Apr 2006 at 11:58, Armag wrote:
!!
!! >
!! > What is the final verdict, the original topic of this thread?
!! > The Corsaire article - is there a fundamental error in the
!! > recommendation part of it?
!!
!! Well, if you ask me, then yes, there is a problem in the Corsaire paper,
!! since it doesn't mention that in almost all of the cases, the cookie
!! path is useless for improving security. The only case wherein this recommendation actually
!! adds
!! security is the very rare situation of a JS-turned-off (indeed...) non-MSIE browser
!! going to a non-IIS (or more likely non-Windows) website, and even that
!! is not guaranteed (who knows what other tricks work for the various servers out there). So
!! up to this negligible situation, I hold to my original claim - "There is no such thing as
!! path security".

agree with Amit here, path in cookies are unreliable, at least for the
application.
Please see also my inquiry here Subject: sequence of cookies in a request
I got only 2 responses, and they pointed me back to the RFCs. Sigh.

If we don't blame the browser vendors, call on to fix the flaws, and we also
do not blame application programers to avoid insecure cookies for their
session handling, where's the culprit then? Ahh, I see, it's the user again
accessing the wrong application with the wrong browser at the wrong time ...

Well, my post is a bit off-topic to the initial subject, but the question
and my other question "sequence of cookies in a request" show again that
cookies are a fundametal threat in todays web applications.
I claim too "There is no path security".
(cookie2 with encrypted values are a different story, however ...)

!! -Amit
!!
!! PS - which isn't to say I'm against using cookie path. Go ahead, use it - it will probably
!! save you from cookie namespace collisions. But don't think you buy any security this way
!! (well, up to the aforementioned bizarre scenario, that is...)


{-: Achim



-------------------------------------------------------------------------
Sponsored by: Watchfire

Watchfire's AppScan is the industry's first and leading web application 
security testing suite, and the only solution to provide comprehensive 
remediation tasks at every level of the application. Change the way you 
think about application security testing - See for yourself. 
Download a Free Trial of AppScan 6.0 today!

https://www.watchfire.com/securearea/appscansix.aspx?id=701300000007kaF
--------------------------------------------------------------------------


Current thread: