--- Advisory RFP9903 --------------------------------- rfp.labs ------------

        Attacker can execute remote programms under debugger context

                           (AeDebug vulnerability)

----------------------------------- rain forest puppy / rfp@wiretrip.net ---

Table of contents:
        - 1. Scope of problem
        - 2. Solution
        - 3. Miscellaneous Updates

----------------------------------------------------------------------------

October is Octoberfest Advisory month: one rfp.labs release planned each week
for the whole month of October! (Now, let's see if I can pull it off....)

-----------------------------------------------------------------------------

----[ 1. Scope of problem

        Let me start off with the mechanism has been discussed before. In
light of the recent RASMAN remote registry fiasco, I took a quick check
and found another similar issue. In all my NT SP5 installs, plus various
other occasions (installation of Visual Studio 5 or 6, etc), the following
registry key holds the program to execute as a debugger:

\HKEY_LOCAL_MACHINE\Software\Microsoft\Windows NT\CurrentVersion
        \AeDebug\Debugger

...as well as a key that indicates whether or not to prompt the user to run
the debugger on system crash:

\HKEY_LOCAL_MACHINE\Software\Microsoft\Windows NT\CurrentVersion\AeDebug\auto

        Now, the problem is very simple. First, also by default, the winreg
AllowedPaths includes \HKEY_LOCAL_MACHINE\Software\Microsoft\Windows
NT\CurrentVersion\. This means any keys under it, including AeDebug, are
accessible remotely, providing the right ACLs on the keys allow so. Well,
just so happens that Everyone has Special Access to Debugger and Auto
under AeDebug. Included in this Special Access is the permission to Set Value.

        This means these keys are remotely accessible and allow anyone to
change their values. By changing their values, an attacker can set a
command (or string of shell delimited commands) to execute upon a
application crash/fault automatically. Now, I have not confirmed this, so
I will disclaim THIS IS JUST MY THEORY, but I would think the debugger would
execute with a few more priveleges than the normal user, so these commands
may be run with elevated priveleges. Of course, the actual attack
wouldn't commence until an application crash occured. Only if we had a
way to make something crash remotely..... >:)

----[ 2. Solution

        There has been previous discussion on this type of vulnerability--all
the way back to 1997 (found on NTBugtraq). The solution consists of two
parts. First, remove

\HKEY_LOCAL_MACHINE\Software\Microsoft\Windows NT\CurrentVersion\

from the winreg AllowedPaths key, found at:

\HKEY_LOCAL_MACHINE\System\CurrentControlSet\Control\SecurePipeServers
        \winreg\AllowedPaths

This will prevent remote modification of these keys. Next, remove the Set
Value and Create Subkey permission from Everyone's Special Access. This will
prevent local users from modifying the keys.

        That's it. Very simple.

----[ 3. Miscellaneous Updates

        - Like I mentioned before, October is Octoberfest Advisory month.
I'm going to attempt to release something every week. Hopefully you'll
find the stuff worthwhile...I think there's some planned goodies. A
little NT, a little unix, a little web, all good. :)

        - My website is narrowing completion! Many people have been
wondering if I have archives of my releases, programs, research, etc. The
answer is 'yes', and soon it will be made public. More information on
this in the next release (next week :)

        - Many people have emailed me wondering about the release of version
two of the RDS script. This is one of the planned releases. Gimme time
to finish coding it.

        - You may have noticed no humor, sarcasm, or snide remarks in this
advisory. Yeah, so?

        - Why the formal advisory numbering system and format? Well, it's
really to better organize my own personal filing, really. And you'll see
how it fits into my website design in a little bit. In case you're
wondering, RFP9901 was the ODBC article posted May 25th, and RFP9902 was
the RDS posted June 23rd? (or July 23rd).

        - Phrack 55 is out--good stuff. www.phrack.com Packetstorm is back.
packetstorm.securify.com. Technotronic still rules. www.technotronic.com.

        - Practical application of one of the perl problems I talked about in
Phrack 55 landed JFS $1,000 for hacking securelinux.hackpcweek.com.
Congrats, JFS.

--- rain forest puppy / rfp@wiretrip.net ---------------- ADM / wiretrip ---

 Support your local security professional. Buy him/her sparcs and ciscos.