Robert Hansen (a.k.a. "RSnake") discussed the newly discovered vulnerabilities in a blog published Saturday and in presentations in Las Vegas and Sweden last week. Hansen and other security experts advised enterprises to move swiftly to mitigate the possibility of attacks that exploit the flaws.
In a nutshell, Hansen is warning enterprises about the use of "nonroutable" IP addresses, particularly as described in the Internet Engineering Task Force's RFC 1918 standard. These addresses, sometimes called "private IP addresses," are frequently used in corporate networks to name systems and devices that are used only internally and have no need to be routed over the Internet. RFC 1918 is used widely in large enterprise networks, where an organization may need to preserve a finite number of public IP addresses.
The problem, Hansen observes, is that some enterprises and technologies use private IP addresses as a means of securing themselves -- they assume that because RFC 1918 addresses are used only internally, an external attacker would not be able to take advantage of them. But Hansen points out that the spectrum of RFC 1918 addresses is so limited that a hacker might be able to create parallel environments that also use RFC 1918, and then exploit IP address collisions between the networks to compromise the enterprise's internal environment.
In a series of scenarios, Hansen describes a variety of ways in which RFC 1918 vulnerabilities could be exploited to allow an attacker to interrupt service or gain access to a company's internal network. Some of these attacks exploit the browser's ability to retain IP addressing information in its cache, as well as virtual private network routing and addressing flaws that might allow the compromise of a business partner's or home office user's networks.
Enterprises can take steps to mitigate the threats, Hansen says. "The first three attacks rely on the fact that VPNs can be told what to route," he explains in his blog. "If the VPN can be limited to only route the IP spaces that both parties agree upon, this attack would quickly fall down, or at minimum would only be effective against the IP addresses that were allowed to be routed. All of these attacks require that the browser caches content, and that that content persists beyond the initial request.
HD Moore, creator of Metasploit and director of security architecture for BreakingPoint Systems, says companies should consider changing the way they use RFC 1918. "The core problem is that the browser needs to have a different profile or cache for each network location," he says. "The mobile aspect of laptops and smartphones undermines any privacy or security feature based on control of an IP address or DNS name. Cache poisoning is just one method of exploiting this -- many other attacks become possible when the attacker can impersonate a trusted host."
"The bits that require more research are identifying common Web applications deployed internally -- such as OWA and SharePoint -- enumerating common host names and IP addresses where these systems are located, and leveraging these applications to either steal data or run code on the user's system," Moore says. "An easier attack would be to embed a signed Java applet into the Web page of a trusted internal site, tricking the user into loading this code when they access the server."
Have a comment on this story? Please click "Discuss" below. If you'd like to contact Dark Reading's editors directly, send us a message.