Editor’s Note: This story is excerpted from Computerworld. For more Mac coverage, visit Computerworld’s Macintosh Knowledge Center.
Security researchers warned Friday that a new class of vulnerabilities dubbed “clickjacking” puts users of every major browser at risk from attack.
Details of the multiple flaws—six different types by one count—are sketchy, as the researchers, who presented some of their findings at a security conference earlier this week, have purposefully kept their information confidential as at least one vendor works on a fix.
Although the clickjacking problem has been associated with browsers—users of Internet Explorer, Firefox, Safari, Opera, Google Chrome and others are all vulnerable to the attack—the problem is actually much deeper, said Robert Hansen, founder and chief executive of SecTheory LLC, and one of the two researchers who discussed the bug in a semi-closed session at OWASP AppSec 2008 on Wednesday.
In an interview on Friday, he called clickjacking similar to cross-site request forgery, a known type of vulnerability and attack that sometimes goes by “CRSF” or “sidejacking.” But clickjacking is different enough that the current anti-CRSF security provisions built into browsers, sites and Web applications are worthless.
“At a high level, almost everyone is affected by it,” Hansen said. “The problem is that a lot of people who spent a lot of time defending [against cross-site request forgery] didn’t see this coming. This works completely differently, and has much wider-reaching issues. [Attackers] can get users to click a button [in clickjacking] where they may not be able to get them to click a button in JavaScript.”
Hansen’s research partner, Jeremiah Grossman, chief technology officer at WhiteHat Security Inc., explained how attackers could exploit clickjacking vulnerabilities.
“Think of any button on any Web site, internal or external, that you can get to appear between the browser walls,” Grossman said in an e-mail on Friday. “Wire transfers on banks, Digg buttons, CPC advertising banners, Netflix queue, etc. The list is virtually endless and these are relatively harmless examples. Next, consider that an attack can invisibly hover these buttons below the users’ mouse, so that when they click on something they visually see, they actually are clicking on something the attacker wants them to.”
Hansen seconded Grossman’s example with one of his own. “Say you have a home wireless router that you had authenticated prior to going to a [legitimate] Web site. “[The attacker] could place a tag under your mouse that frames in a single button an order to the router to, for example, delete all firewall rules. That would give them an advantage in an attack.”
Hackers would not need to compromise a legitimate site in order to conduct a clickjacking attack underneath it, Hansen added.
There are several possible solutions to the clickjacking problem, but only one makes sense. “The only people who can fix this in a scalable way are the browser vendors,” Hansen said.
He and Grossman have been in contact with Microsoft, Mozilla, and Apple, the makers of Internet Explorer, Firefox and Safari, respectively. Together those companies’ programs account for more than 98 percent of all browsers used last month, according to data from Net Applications.
It’s not clear how serious the browser makers have taken the warnings by Hansen and Grossman, however, or how soon they will update their applications. “All are working on solutions,” said Hansen. “But no one said that they were necessarily putting something in the next version.”
For the moment, the best defense against clickjacking attacks is to use Firefox with the NoScript add-on installed. Users running that combination will be safe, said Hansen, against “a very good chunk of the issues, 99.99 percent at this point.”
In the next breath, however, he called the Firefox-NoScript solution a stop-gap fix suitable only for technical users. “If my Mom was using NoScript, I’d be taking all kinds of technical support calls,” he said. “It’s not the right solution.”
In the meantime, people shouldn’t panic. “Truthfully, there’s a very small number of companies that can do something about this,” he said.
Hansen and Grossman plan to release virtually all of their research, including proof-of-concept code, when Adobe wraps up a patch. Last week, the two promised Adobe that they would withhold most of their information after showing the vendor attack code that exploited a bug in its software.
On Friday, Hansen declined to confirm that the affected Adobe software was Flash, the ubiquitous multimedia content player that most users run as plug-in to their browser.