Perimeter
5/31/2011
07:19 PM
Tom Parker
Tom Parker
Commentary
Connect Directly
RSS
E-Mail
50%
50%
Repost This

A Tale Of Two Hacks

The similarities and differences in the Lockheed and RSA attacks

The U.S. government has been collaborating with its largest contractors (often called the defense industrial base, or DIB) for some years now to establish operating procedures and technical countermeasures in an attempt to fend off the growing number of seeming targeted attacks. These attacks seek to steal sensitive data from the networks of the Defense Department’s largest private-sector contributors.

Many of the threat actors responsible for such attacks have been categorized as APTs. APT isn’t a term I like to use too readily due to its lack of clarity, but let’s be clear: While many technological aspects of so-called APT attacks are up for discussion, one thing remains true -- we’re talking about China.

On March 22, Lockheed Martin (LMT) shut down one of its networks in response to an unidentified intrusion impacting what appeared to be its corporate, unclassified network. According to various sources, Lockheed employees were instructed to change their network passwords and told that their RSA SecurID tokens would be replaced. As news of this broke, so did the speculation drawing parallels between this and the RSA hack earlier this year -- which had already been categorized by RSA executive chairman Art Coviello as an APT (China). Naturally, this has led to some unfounded assumptions that this, too, must be the handiwork of the Chinese.

As someone who strongly believes that cyber-attribution is possible (today), I’m extremely hesitant to apply a buzzword like APT (and therefore implicating a specific state actor) without some sort of sound evidence -- and you should be, too. The media has been instrumental in its coverage of many such incidents over the past two or three years, and we can safely assume that it is no longer (if it ever was) a secret that techniques, such as leveraging client-side vulnerabilities and social engineering, are far more effective for penetrating an Internet perimeter than the infrastructure attacks that have been preferred in years gone by.

While exact details regarding what data was stolen from RSA earlier this year aren't known, RSA has released a number of blog posts and customer advisories that allow us to get a pretty good idea of the possible impact to customers based on what was stolen. Notably, instructions on monitoring RSA audit logs for signs of compromise attempts were circulated, which specifically advise for the identification of bad authentication PINs (the something you know) with a good token code (the something you have). Lots of these in a log file is a sure-fire sign that someone is trying to brute-force his way in and has possession of both a good username and stolen (or cloned) token. This supports earlier speculation that it was the seed values that were stolen.

Seed values are a sensitive, key component to the algorithm used for authentication and are linked to the serial numbers that are inscribed on the rear of most RSA hard tokens. Assuming that it was the seed value/serial data sets from RSA’s token manufacturing database that were stolen, a successful attack against a SecurID infrastructure would still require knowledge of both a valid username -- and the something you know.

If we consider this in the context of the Lockheed compromise, I find it highly unlikely that an organization like Lockheed waited three months to determine it wanted to switch out its population of fobs throughout its employee base. Many organizations in similar positions made risk-based decisions back in March that as long as the sanctity of their users IDs and PINs were preserved and that sufficient monitoring was in place to identify possible attempts to brute-force both missing factors (the ID and PIN), that it was an unnecessary move to switch out the physical fobs.

If Lockheed has indeed been hit by an attack similar to many of the others targeting the defense industrial base (many of which deploy key loggers capable of capturing a user’s remote authentication credentials and PINs), then it is more than reasonable for Lockheed to reassess the risk and determine the high likelihood that someone now possesses the factors required to compromise its SecurID protected environments.

The take-home here is that while the two attacks might not necessarily be directly associated with the same threat actor(s), the risk that they are might have been enough to trigger a countermeasure on the part of Lockheed. This reaction, however, should not be seen as an indicator of attribution, and far more details, which are unlikely ever to publicly surface, are needed before ever trying to make such a determination.

Tom Parker is director of security consulting services at Securicon.

Comment  | 
Print  | 
More Insights
Register for Dark Reading Newsletters
White Papers
Flash Poll
Current Issue
Video
Slideshows
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2011-0460
Published: 2014-04-16
The init script in kbd, possibly 1.14.1 and earlier, allows local users to overwrite arbitrary files via a symlink attack on /dev/shm/defkeymap.map.

CVE-2011-0993
Published: 2014-04-16
SUSE Lifecycle Management Server before 1.1 uses world readable postgres credentials, which allows local users to obtain sensitive information via unspecified vectors.

CVE-2011-3180
Published: 2014-04-16
kiwi before 4.98.08, as used in SUSE Studio Onsite 1.2 before 1.2.1 and SUSE Studio Extension for System z 1.2 before 1.2.1, allows attackers to execute arbitrary commands via shell metacharacters in the path of an overlay file, related to chown.

CVE-2011-4089
Published: 2014-04-16
The bzexe command in bzip2 1.0.5 and earlier generates compressed executables that do not properly handle temporary files during extraction, which allows local users to execute arbitrary code by precreating a temporary directory.

CVE-2011-4192
Published: 2014-04-16
kiwi before 4.85.1, as used in SUSE Studio Onsite 1.2 before 1.2.1 and SUSE Studio Extension for System z 1.2 before 1.2.1, allows attackers to execute arbitrary commands as demonstrated by "double quotes in kiwi_oemtitle of .profile."

Best of the Web