More IE security woes: Attacks appear imminent as exploit is improved

Hackers make the Metasploit exploit more effective, and thus more appealing to cybercriminals

Hackers working on the open-source Metasploit project have spiffed up a zero-day attack on Microsoft's Internet Explorer, making it more reliable -- and more likely to be used by criminals.

Security experts have been worried about the flaw since it was first disclosed on the Bugtraq mailing list Friday. But the original demonstration code was unreliable and has not been used in real-world attacks.

[ Discover what's new in business applications with InfoWorld's Technology: Applications newsletter and Killer Apps blog. ]

"The Metasploit exploit that was released last night will be more reliable against certain attacks than the initial exploit," said Ben Greenbaum, senior research manager with Symantec, in an interview Wednesday.

As of Wednesday morning, Symantec had not seen the exploit used in Internet-based attacks, but security experts say this type of code is for a very popular hacking technique called a drive-by attack. Victims are tricked into visiting Web sites that contain malicious code where they are then infected via the browser vulnerability. Criminals also place this type of code on hacked Web sites in order to spread their attacks.

On Monday, Microsoft published a Security Advisory on the flaw, offering some workarounds for the issue. It affects IE version 6 and version 7.

Microsoft's latest IE 8 browser is not affected by the bug, which has to do with the way that IE retrieves certain Cascading Style Sheet (CSS) objects, used to create a standardized layout on Web pages.

Concerned IE users can upgrade their browser or disable JavaScript in order to avoid an attack.

To improve the exploit, the Metasploit team used a technique borrowed from two well-known security researchers, Greenbaum said. "The initial exploit used heap-spraying technology," he said. "It's kind of like a shotgun attack, where you try a lot of things at once and hope one of them catches."

The latest attack uses a .net dlll memory technique developed by Alexander Sotirov and Marc Dowd. "This will be much more reliable than the heap-spraying technology," he added. "There's really no question about it."