Those findings come from a new study, conducted by Imperva, of more than 10 million Web application attacks targeting the websites of 30 large businesses and government agencies, launched between January 2011 to May 2011. The study also assessed traffic that flowed via the onion router, better known as TOR, which helps anonymize Web traffic.
Attackers often employed those techniques in combination, whether to steal data, surreptitiously install malware on servers, or simply create a denial of service. "For example, a hacker may use directory traversal during a reconnaissance phase of the combined attack to identify the directory structure of an attacked server before sending an additional effective exploit vector, such as an RFI," according to the report.
Interestingly, the LulzSec hacking group employed three of those techniques, sometimes in combination. But LulzSec's exploits, which largely occurred in June, fell outside the scope of the report. "Consequently, we didn't directly witness any attacks from Lulzsec," according to Imperva's report. However, the Imperva researchers did see an "incredible similarity" between the most prevalent Web application hacking techniques, and the techniques used by LulzSec's members.
Imperva's research presents an interesting contrast with other vulnerability information, such as the Open Web Application Security Project (OWASP) list of the top 10 worst Web application vulnerabilities. According to Amichai Shulman, CTO of Imperva, when it comes to the OWASP top 10, "RFI and directory traversal were not identified as top vulnerabilities, yet our research shows that these are two of the most common attacks used by hackers to steal data."
The difference comes from assessing vulnerabilities, versus what's actually being attacked, he said via email. "The shortcoming of OWASP Top 10 is that they concentrate on the most prevalent vulnerabilities. And while this is important, it does not concentrate on what hackers are actually hacking."
According to Imperva's research, attackers largely pursue the easiest exploits, rather than the most prevalent vulnerabilities. "Our report shows that if there is a vulnerability out there--even overlooked by Web application developers, not appearing in OWASP top 10, though easily exploitable--then hackers will go after it," said Shulman.
Beyond attack type, the Imperva report also assessed attack origin. Overall, most Web application attacks are launched from botnets involving exploited PCs located in the United States (for 61% of attacks), followed by China (9%), Sweden (4%), and France (2%).
But the identity of whoever's behind those attacks, and where they might be based, isn't clear. "Our data shows that it is increasingly difficult to trace attacks to specific entities or organizations," said Rob Rachwald, director of security strategy at Imperva, in a blog post. "This complicates any effort to retaliate, shut down cybercriminal gangs or identify potential acts of war."
Black Hat USA 2011 presents a unique opportunity for members of the security industry to gather and discuss the latest in cutting-edge research. It happens July 30-Aug. 4 in Las Vegas. Find out more and register.