Guest

Cisco Security Agent

Cisco Security Agent and the Microsoft Word Exploit

  • Viewing Options

  • PDF (362.7 KB)
  • Feedback

Product Bulletin No. 3501

SUMMARY

A critical "day zero" vulnerability was announced on May 22, 2006 for Microsoft Word versions 2003 and Word XP (2002). [1] This vulnerability is actively being exploited. Microsoft has announced it will release a fix for this vulnerability in its already-scheduled security update on June 13, 2006.
This vulnerability has already been exploited in several attacks. Cisco Systems ® has obtained exploit files, and has confirmed that Cisco ® Security Agent is effective in stopping these exploits, using the default security policy configuration. No changes to the default configuration or policy updates were required to receive this protection. Current supported versions of Cisco Security Agent of 4.0.3.x, 4.5.1.x, 5.0.0.x, and 5.1.0.x are effective in stopping the exploits seen to date.

DETAILS OF THE VULNERABILITY

Opening a specially crafted Word document (including documents hosted on Websites or delivered as e-mail attachments) could trigger the vulnerability. Other Microsoft Office document types, such as PowerPoint, could also be used as attack vectors. Microsoft Office documents can contain embedded objects; for example, a malicious Word document could be embedded in an Excel or PowerPoint document.
The CERT advisory [2] states that by convincing a user to open a specially crafted Word document, an attacker could execute arbitrary code with the privileges of the user running Word. If the user is logged in with administrative privileges, the attacker could take complete control of a vulnerable system.

HOW CISCO SECURITY AGENT STOPS THE EXPLOIT

Cisco Security Agent default policies contain six different rules that stopped the exploit from doing any damage. No changes to the Cisco Security Agent binaries or default configuration are required to get this protection.
The following actions have been observed being blocked by Cisco Security Agent running the default security policies:

• Modification of system files by a suspicious (downloaded) application

• Read/write of a Word document by a suspicious (downloaded) application

• Modification of Cisco Security Agent protected registry keys

• Memory injection into running processes

• Suspicious (downloaded) application opening a command shell

• Execution of a system function from a buffer, via a buffer overflow

This testing is shown in Figures 1 and 2.
The exploit was tested at Cisco with Cisco Security Agent in Test mode, which does not block malicious behavior. This allows the agent to report all rules that would be applied if the agent was in protect mode, to observe all possible ways that the Cisco Security Agent default policies would stop the exploit. When the agent is in protect mode (the typical operational configuration), the first rule would kill the exploit. No subsequent events would be seen, since the exploit would be terminated before it could perform any malicious actions.
Testing was performed against the Cisco Security Agent default policies. No binary or policy update was needed for Cisco Security Agents to be effective. In short, this was a true test of "day zero" protection. This is similar to what we have seen with earlier exploits and worms-the default Cisco Security Agent configuration stopped the exploit, with no binary or policy updates required. The following is a partial list of prior worms and exploits that Cisco Security Agent has stopped using the default security policy settings:

Bagle

E-mail worm

SQL Snake

Network worm

Blaster

Network worm

JPEG/GDI+

Malware downloader

Bugbear

E-mail worm

MyDoom

E-mail worm

Code Red

Network worm

Nimda

Network worm

Debploit

Network worm

Pentagone/Gonner

E-mail worm

Fizzer

E-mail worm

Sasser

Network worm

Gator/Gain

Spyware

Sircam

E-mail worm

Hotbar

Spyware

Sobig

E-mail worm

SQL Slammer

Network worm

Zotob

Network worm

This exploit is only the latest example of new and mutating attacks that can seriously impact organization's computing and network environments. The key to stopping these new attacks is the ability to stop the attack without requiring any changes to the default configuration, and to have multiple rules in the default policies that provide defense in depth.
Figures 1 and 2 show the entire Cisco Security Agent event log and details on how it would have prevented the attack.

Figure 1. Page 1 of Event Log - Cisco Security Agent 5.1 Default Configuration Stops the Microsoft Word Buffer Overflow Exploit

Figure 2. Page 2 of Event Log - Cisco Security Agent 5.1 Default Configuration Stops the Microsoft Word Buffer Overflow Exploit

REFERENCES:

[1] Microsoft Security Advisory 919637: http://www.microsoft.com/technet/security/advisory/919637.mspx
[2] CERT Vulnerability Note VU#446012: http://www.kb.cert.org/vuls/id/446012