Security pros have uncovered a new technique for detecting PCs infected by the Conficker worm.
The technique is based on a discovery by members of the Honeynet Project, which found that Conficker's attempts to cloak itself from network administrators may have backfired. As part of its defenses, Conficker deploys a fake patch for the Microsoft vulnerability it exploits that researchers have speculated is meant both to fool admins into thinking systems are patched and to keep other types of malware from exploiting the vulnerability.
However, Honeynet Project researchers Felix Leder and Tillmann Werner noted there are flaws in the way Conficker "patches" these compromised systems that can be used to detect if a PC is infected with Conficker. The discovery prompted members of the Conficker Cabal-a group of researchers and vendors fighting Conficker-to begin working on ways to use the discovery to help network scanners detect the worm.
"What we've found is pretty cool: Conficker actually changes what Windows looks like on the network, and this change can be detected remotely, anonymously and very, very quickly," Dan Kaminsky, director of penetration testing for IOActive, wrote on a blog. "You can literally ask a server if it's infected with Conficker, and it will tell you."
He explained that the worm "makes NetpwPathCanonicalize() work quite a bit differently than either the unpatched or the patched MS08-067 version." Leder and Werner are slated to come out with a paper later the week of March 30 that describes the situation in more detail, and have released a proof-of-concept scanner of their own that can detect the differences in the patch.
Conficker first appeared late in 2008 targeting a flaw in Microsoft's Windows Server service. The latest variant, dubbed Conficker C but also known as Downadup and Conficker D, is programmed to begin contacting 500 domains from a list of 50,000 domain names starting April 1.