Original URL: http://www.theregister.co.uk/2006/03/13/virtual_rootkit/
Virtual rootkits create stealth risk
MS researchers grapple with phantom malware
Security researchers have uncovered new techniques to hide the presence of malware on infected systems. By hiding rootkit software in virtual machine environments, hackers have the potential to avoid detection by security software, boffins at Microsoft Research and the University of Michigan warn.
To validate their concerns the researchers created proof-of-concept code, called SubVirt, which takes advantage of security vulnerabilities to load a VMM (virtual machine monitor) beneath either a Windows (Windows/VirtualPC) or Linux (Linux/VMWare)installation, eWeek reports. SubVirt loads four malicious service onto compromised machines including a phishing web server, a keystroke logger and anti-security software countermeasures.
In one attack scenario, security vulnerabilities might be used to gain administrative rights to a target machine. This access might then be used to manipulate a system boot sequence to load a virtual machine rootkit. Thereafter, attackers are free to deploy other malicious software that hides from detection.
"Any code running within an attack OS is effectively invisible. The ability to run invisible malicious services in an attack OS gives intruders the freedom to use user-mode code with less fear of detection," the researchers said.
Existing anti-rootkit tools commonly rely on comparing file system and API discrepancies to check for the presence of rootkits, a technique that wouldn't be able to unearth virtual machine malware. The researchers hope their work will help security firms adapt their technology in order to combat the new class of threat.
Hardware detection, low-level security software and booting from a safe medium are all possible countermeasures. The team is due to present its findings at an IEEE Symposium on Security and Privacy later this year. ®