[38738] in bugtraq
Re: ASP.NET __VIEWSTATE crypto validation prone to replay attacks
daemon@ATHENA.MIT.EDU (Michal Zalewski)
Thu May 5 17:39:45 2005
Message-ID: <1e89feb30505051306458c1e4@mail.gmail.com>
Date: Thu, 5 May 2005 22:06:21 +0200
From: Michal Zalewski <lcamtuf@gmail.com>
Reply-To: Michal Zalewski <lcamtuf@gmail.com>
To: bugtraq@securityfocus.com
In-Reply-To: <1e89feb305050513002386241b@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
> Microsoft has addressed your issues 1-a, 1-b and 1-c by adding a
> property "ViewStateUserKey" to the System.Web.UI.Page class in
> .NET Framework 1.1. The documentation for this property is here:
No, I think this statement is simply not true. The key is intended to
solve a wholly different class of problems, and has nothing to do with
replay attacks I reported.
Reported issue 1-a: same user moves __VIEWSTATE from one page to
another. Adding unique user identifier to __VIEWSTATE does not help,
because in both cases, it's precisely the same user within the same
system.
Reported issue 1-b: same user retains __VIEWSTATE and reuses it at a
later date. Again, it's precisely the same person and the same unique
ID.
Reported issue 1-c: this is the least relevant of all reported
problems, and indeed can be in some situations mitigated by binding
__VIEWSTATE data to a specific user.
/mz
