Imperva, which released the data in its first biannual "Web Application Attack Report," published today, gathered information from more than 10 million Web application attacks that occurred between December 2010 through May 2011 targeting enterprise and government websites, as well as via TOR. The U.S. leads the world in originating Web application attacks, and 61 percent of these attacks come from bots in the U.S. The botnet command-and-control servers, however, aren't necessarily from the U.S., according to the report.
Rob Rachwald, director of security at Imperva, says attack automation is becoming prevalent as attackers increasingly deploy automated tools to try to break into websites.
Interestingly, the pervasive SQL injection was not the most common attack technique in Web app attacks, even though it's the most publicized. The directory traversal attack -- an exploit where an attacker gets access to restricted directories -- accounted for 37 percent of attacks during the six-month period, followed by cross-site scripting (XSS), with 36 percent of the attacks. Around 23 percent of the attacks were SQL injection ones, and 4 percent were remote-file include attacks.
"RFI and directory traversal attack [numbers] were kind of a surprise," Rachwald says. "This mimics the kinds of attacks we saw from LulzSec."
Rachwald pointed out that the Anonymous spin-off group Lulz Security had exploited thousands of servers using RFI attacks, often using RFI as a first stage and then following up with SQL injection. "I don't think people think about RFI," he says. "RFI and directory traversal don't even appear on the OWASP Top 10. We focus a lot on vulnerabilities in the industry, and this shows how important it is to match that up against actual attack data."
A full copy of the report is available here (PDF) here for download.
Have a comment on this story? Please click "Add Your Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message.