The Black Hat conference and its post-event, DefCon, promise to be a security funhouse in the coming week, as experts in Las Vegas seek to shock and amaze by poking holes in today's network technologies.
The Black Hat conference and its post-event, DefCon, promise to be a security funhouse in the coming week, as experts in Las Vegas seek to shock and amaze by poking holes in today's network technologies. The Web, wireless LANs, routers and desktop software may all look different reflected in the Black Hat/DefCon hall of mirrors, where security vendors will be revealing their hacker sides.
"We're showing malware we created called Jinx," says Itzik Kotler, manager of the security operations center at Radware and a presenter at Black Hat, which runs through Aug. 7. Kotler describes Jinx as attack code that can be used to take over the machines of victims using versions of Mozilla's Firefox browser that pre-date Firefox 3, Mozilla's latest release. (You might want to upgrade now if you haven't already.)
"It's the first proof-of-concept of such malware, with no code injection, no interfering with the kernel," says Kotler, adding the Jinx exploit code will be published for all to see. He hinted Radware is working on similar Jinx-like malware aimed at Microsoft Internet Explorer.
Why all the effort? "We believe people need to be prepared for this. There's a popular demand for Web 2.0, but it's a bad situation in that we've given huge power to browsers, but these browsers often have logic flaws that allow these attacks," Kotler says.
For vendor AirTight Networks, which makes wireless LAN (WLAN) intrusion-prevention systems, its focus is how some wireless LAN vendors may not be implementing the IEEE's new 802.11w security standard correctly. (Compare WLAN Security products.)
The 802.11w standard (Cisco calls it "management frame protection") is supposed to make WLANs resistant to denial-of-service (DoS) attacks. But AirTight will show how it's possible with some implementations of 802.11w in vendor equipment to conjure up an attack that hits WLAN access points with malformed packets, not bringing them down but triggering a disconnection response in their WLAN clients.
"This attack involves a special packet which has the effect of disconnecting the endpoint," says Pravin Bhagwat, chief technology officer at AirTight, which dubs this the "autoimmunity disorder in WLANs."
The WLAN DoS attack, which involves tampering with the MAC address at Layer 7 by sending a continuous stream of injected packets at intervals of about 30 seconds, basically results in the WLAN access point being exploited as the vector for disabling WLAN endpoints.
Some of the WLAN equipment that will be shown to be vulnerable to this attack includes that of D-Link, Cisco, Buffalo and open source Madwifi. Either these vendors aren't implementing 802.11w correctly or the standard will need to be improved to prevent the "autoimmune disorder" in WLANs, according to AirTight.
All about the rootkit
Cisco gear will also get pounded in another session with Core Security Technologies, which is expected to show how it's possible to install a rootkit on the Cisco IOS. A rootkit is code designed to hide from detection so someone can control processes without being noticed.
"This does assume you have access to the Cisco device because you are the administrator or somehow broke in," says Ivan Arce, chief technology officer at Core Security.
The Cisco IOS rootkit would give an attacker the ability to do things such as change how traffic passes through a Cisco device. "People don’t understand it's possible to have a rootkit on IOS," says Arce, adding that Cisco is aware of the research and earlier this year issued an advisory on it.
Rootkits will be a hot topic at Black Hat as some of the world's foremost researchers on the subject reveal new discoveries they've made about subverting software.
Researcher Joanna Rutkowska, whose devastating insights into Microsoft software and rootkits impressed Black Hat audiences in the past, is expected to take on the Xen hypervisor, this time with help from colleagues.
But it doesn't stop there.
Google Gadgets, those small Web applications that allow users to customize Web pages, will be in for the Black Hat treatment, too.
"The current architecture in the security model around Google Gadgets is highly insecure," says Tom Stracener, senior security analyst at Cenzic. The Web application security assessment provider says it will prove how it's possible for Google Gadgets to take control over each another and steal information from each other.
Still, Black Hat isn't all about deconstructing security. Some experts will show how to take preventative measures to shore up perceived vulnerabilities -- such as the "cold boot" encryption hack.
When Princeton University researchers earlier this year pointed out how it's possible for an attacker to swipe cryptographic keys off a computer through the cold boot technique, it sparked a debate over the safety of stored encryption keys and how they could be grabbed in memory when a machine is being turned off, particularly if subjected to cold temperatures.
"Information can take minutes or even hours to fade out on a computer," says BitArmor CEO Patrick McGregor. "There can be small pieces of information floating around." The Princeton University research generated a lot of concern that "full-disk encryption was useless," McGregor says.
But while the cold boot attack method is not particularly difficult to accomplish -- "you could plug a USB drive into a laptop" to carry it out, says McgGegor -- the situation isn’t as dire as some think. BitArmor claims to have a few basic defenses, including leveraging temperature sensors in Dell and HP computers, and a way to design a "secure enclave to protect full-disk encryption keys."
BitArmor says it is using these techniques effectively in its own products today and will share what they are at Black Hat.