Java has a nasty habit of getting you owned. This latest 0day is no exception to the long-lived trend of reliable Java-based exploitation. Here’s what you need to know:
How does it work?
The current exploitation method being employed in the wild right now leverages two zero day flaws in Java. The first flaw leverages an implementation issue (logic bug) within ClassFinder.findClass(), which is only present in Java 7. The method findClass() allows Java code to access restricted classes, when it should disallow such access. By using findClass() to gain access to restricted classes, the exploit is then able to use a reference for sun.awt.SunToolkit to access the getField() method. The second flaw, found in getField(), allows Java code to modify private fields (logic bug), which should normally be immutable when accessed externally (from outside the class it was declared in). The exploit uses getField() in order to modify the permissions of java.beans.Statement, which is then used to disable the Java Security Manager. At this point, the exploit is able to run code arbitrarily. No memory corruption occurs in the exploitation of these vulnerabilities. This makes the exploit 100% reliable for Java 7, across multiple platforms and through multiple browser-based attack vectors.
What can we do to mitigate this threat?
- Install the patch from Oracle for CVE-2012-4681.
- Audit your environment for systems that have Java. Determine what systems actually need Java and remove it according to your needs. Once you have reduced your Java footprint, you can further reduce your attack surface by restricting how Java runs; determine if a system needs Java for Internet facing applications or if it needs Java for desktop and internal applications. Configure systems that need Java granularly – more information on how to do this can be found in the links below.a. Controlling Java<Applet> Tags http://blogs.msdn.com/b/ieinternals/archive/wp-content/uploads/15/controlling-java-in-internet-explorer.aspxb. Controlling Java <Object> Tags (ActiveX) http://support.microsoft.com/kb/2751647

Scott Lang, Sr. Director, Product Marketing at BeyondTrust
Scott Lang has nearly 20 years of experience in technology product marketing, currently guiding the product marketing strategy for BeyondTrust’s privileged account management solutions and vulnerability management solutions. Prior to joining BeyondTrust, Scott was director of security solution marketing at Dell, formerly Quest Software, where he was responsible for global security campaigns, product marketing for identity and access management and Windows server management.