RE: Webwasher CSM Appliance Script Security Restriction Bypass

["Frank Berzau" <frank () cyberguard com>]

As discribed in our first response on Dec 23, we meanwhile ran
additional tests against older versions of Webwasher CSM because the
initial posting of ".v0rt3x" does not specify a version number.

Our first test results are confirmed:
The script mitigation of Webwasher CSM 5.1 and newer cannot be bypassed
by changing the capitalization of the Run method as described in the
posting.

Since Dec 23, we've been trying to contact ".v0rt3x", asking for further
information, for his PoC example etc. We are still waiting for a
response.
In our email archives we could not find the initial notifications
mentioned by ".v0rt3x" in his initial posting in the timeline section. 

Based on our test results and the lack of detailed input from ".v0rt3x"
we declare the reported issue disproved and consider this case resolved.

Thanks, Frank


-----------------------------------------------------
Frank Berzau
Director, European Support
CyberGuard Corp.
  

-----Original Message-----
From: d0t v0rt3x [mailto:d0tv0rt3x () gmail com] 
Sent: Thursday, December 22, 2005 9:42 PM
To: full-disclosure () lists grok org uk; bugtraq () securityfocus com
Subject: Webwasher CSM Appliance Script Security Restriction Bypass

Vendor: Webwasher (http://www.webwasher.com/)
Product: Webwasher CSM Appliance
Affected versions: CSM Suite 5.x
Author: .v0rt3x (d0tv0rt3x[at]gmail[d0t]com)
Date: 2005-Dec-22

....Background....
"...Webwasher appliances provide high-performance "Proactive
Filtering" of bidirectional SMTP, HTTP, HTTPS, and FTP traffic to
detect and cleanse all forms of malware. The result is a security
appliance that delivers the Blended Protection you need to protect
against malicious content and unwanted email..."

....Description....
Webwasher CSM includes an encapsulation script mechanism with the aim
of filtering malicious scripts.
The encapsulation script makes use of specific potentially malicious
tokens in order to detect and neutralize the malicious script.
The detection of the tokens is case sensitive. However, some of the
tokens can be executed whether they are written in lower case or upper
case letters.
In other words, by creating a specially crafted script, an attacker
can bypass the filtering mechanism and execute malicious scripts.

....Proof.of.Concept....
1) Create a malicious script by using an object which executes ".Run"
method (e.g. one of the many WScript.Shell exploits).
2) Replace ".Run" with ".run".
3) Execute the malicious script "safely" through Webwasher CSM.

....Timeline....
2005-May-15: Vendor was notified by mail.
2005-Aug-15: Vendor was notified again via contact form.
2005-Dec-22: No response from the vendor - public disclosure.