The Kerberos Checksum Vulnerability disclosed last month was just the latest in a string of massive vulnerabilities that were discovered over the course of only eight months last year. Did I say massive? More like earth-shattering, and potentially destructive (in computer terms) on a Biblical scale. Not for nothing have we taken to referring to them as the Four Horsemen of the Apocalypse. Although they may not be heralding the actual Judgment Day, these four vulnerabilities should serve as a wakeup call to anyone even remotely concerned about information security.
The four were disclosed between April and November of last year. Individually, each was referred to as a “black swan event,” or a “unicorn.” We question this attempt to emphasize the rarity and unpredictability of such vulnerabilities. Four in less than a year? This would seem like a flock of black swans or a herd of unicorns. What if these types of potentially devastating flaws are not as rare as we thought? What if these four, as with the actual horsemen, are the harbingers of far, far worse?
By way of reminder, the four vulnerabilities, in order of disclosure, are:
● Heartbleed (CVE-2014-0160)
● Shellshock (CVE-2014-6271 and CVE-2014-7169)
● Winshock (CVE-2014-6332) (aka “Windows OLE Automation Array Remote Code Execution Vulnerability”)
● Kerberos Checksum Vulnerability (CVE-2014-6324)
Why are these four so significant? Let’s take a look at what makes them stand out.
Wildly massive scope
If it has a processor and is connected to the Internet, one of these four can (and probably did) directly and significantly affect it:
● Heartbleed. Impacts encrypted Web communications to over 66 % of Web servers
● Shellshock. Impacts any UNIX/Linux server
● Winshock. Impacts any Windows workstation
● Kerberos Checksum. Impacts any Windows-based managed network
This means that a hacker with prior knowledge of these vulnerabilities could easily:
● Gain access to any Web server’s private certificate and use it to eavesdrop on encrypted Web traffic or perform man-in-the-middle (MITM) attacks
● Perform remote code execution (meaning “do whatever he or she wants”) on any UNIX/Linux internet servers
● Run any code with the highest privileges on any Windows workstation just by having it surf to a specially crafted Web page (a.k.a. a “drive-by attack”)
● Gain unlimited (“domain admin”) privileges over managed corporate networks (which represent over 90% of corporate networks today)
Unlike malware or various recently-damaging and sophisticated cyberattacks, which were created and discovered in the modern era, some of these vulnerabilities have been around since many security admins were in diapers.
● Shellshock: existent for 25 years
● Winshock: existent for 19 years
● Kerberos Checksum Vulnerability: existent for 14 years
● Heartbleed: existent for “only” two years
This means that these high-impact vulnerabilities remained unnoticed and were hiding in plain sight for an average of 15 years. Even with all the super-talented security researchers looking for weaknesses, performing code reviews and more, these vulnerabilities continued to provide “god mode” access to 90% of the Internet.
Covering all sources
Our four vulnerabilities equally covered open and closed source systems -- none were immune:
● Heartbleed -- open source
● Shellshock -- open source
● Winshock - -closed source
● Kerberos Checksum -- closed source
What else is lurking
Companies should initiate targeted efforts to expose and mitigate the hidden breaches that lurk in their networks, especially since we now know that some of the core IT system infrastructure that we’ve trusted and used for decades is vulnerable. Security experts everywhere should understand that the (figurative) apocalypse is already upon us. Our networks are already breached.
Security practitioners also need to pursue attackers actively and continuously throughout the internal network in order to find those that have circumvented traditional threat prevention systems by exploiting the unknown vulnerabilities like those listed above. We need to identify the behavior of these attackers as early as possible by identifying the subtle changes in user behavior that can indicate malicious attack behavior. Then, even if attackers are able to find these vulnerabilities, we can minimize or even eliminate the potential damage if we find them early in the attack lifecycle. That should be the goal in this new reality that we live in.