Several independent coding groups have posted code on the Internet that can allow hackers to exploit a previously disclosed vulnerability in Microsoft's Windows operating system.
The Windows flaw, which was rated "critical" by Microsoft when it was disclosed earlier this month, allows a hacker to gain control of a Windows system through a security hole in the DCOM (Distributed Component Object Model) interface. Microsoft released a patch for Windows NT, Windows 2000, Windows XP and Windows Server 2003 in security bulletin MS03-026 (http://www.microsoft.com/technet/treeview/default.asp?url=/technet/security/bulletin/MS03-026.asp).
At least three different versions of exploit code have been posted on the Internet over the last few days, said Gunter Ollmann, manager of X-Force security assessment services, EMEA (Europe, Middle East, and Africa) at Internet Security Systems Inc. (ISS), which is based in Atlanta. Some of the code is "quite elegant" and can be run by just about anyone with a compiler and some programming savvy, he said. Versions have been posted for both Linux and Windows users, he said.
The greatest threat to networks comes from individuals who will take this code and use it to create mass-mailer worms, the likes of which have created havoc on the Internet several times in recent years, Ollmann said. ISS had not detected any worms as of Monday morning, but had detected several attacks by hackers running the exploit code on their own machines, and expects a worm to appear shortly, he said.
Security administrators who are following the recommended guidelines for defending their networks against attacks should have nothing to worry about, said Marty Lindner, team leader for incident handling at the CERT Coordination Center in Pittsburgh.
Some of the largest worms to exploit disclosed vulnerabilities, including the Nimda worm and the SQL Slammer worm, had patches that had been available for several weeks, Lindner said. The worms had such a large effect because many administrators don't follow best practices, such as downloading security patches as soon as they are made available and blocking ports at the firewall, among other things, he said.
Microsoft strongly urges all customers to download the patch, which will protect them against all kinds of attacks, said Stephen Toulouse, security program manager with Microsoft's Security Response Center.
If a worm is released, corporations will be mostly threatened by code that is executed via e-mail attachments, or telecommuters who use VPN (virtual private network) connections without firewalls to protect their home PCs, said Scott Blake, vice president of information security and international technical services for Bindview Corp., which is based in Houston.
Blake also is the chairman of the communications committee at the Organization for Internet Security (OIS). This group of vendors, including Microsoft and Oracle Corp., works to establish vulnerability reporting guidelines to minimize the posting of exploit code before software vendors have a chance to develop a patch, he said.
In this case, the group that discovered the vulnerability, The Last Stage of Delirium Research Group, followed the OIS guidelines for reporting the vulnerability by holding onto its exploit code, Blake said. But since this vulnerability was particularly "juicy," several groups of coders rushed to discover their own exploit code, he said.
"The OIS is under no illusion that we can stop people from finding the holes," he said.
While there is nothing Microsoft can do to stop organizations from posting exploit code either, Toulouse noted that using that code to damage a company's network is a criminal act. "We continue to believe that the publication of exploit code is just not good for customers," he said.
Versions of the code have been posted to Metasploit.com, Xfocus.org, and the Full Disclosure mailing list, Ollmann said.