How the New Science of Vulnerability Management Can Help Struggling Federal Networks


IT teams often have the tools to find network vulnerabilities. What they don't have is the manpower to fix them all.

John Breeden II is an award-winning journalist and reviewer with over 20 years of experience covering technology. He is the CEO of the Tech Writers Bureau, a group that creates technological thought leadership content for organizations of all sizes. Twitter: @LabGuys

Computers and networking aren’t getting any simpler. Every time a new application, technology, client, server, cloud, device or almost anything else is added to a network, the number of potential vulnerabilities that an adversary could use to successfully attack it grows. And most of the time, each additional item added brings with it multiple vulnerabilities, so the attack footprint grows much faster than the network. Even older devices and programs can hide previously unknown vulnerabilities, which means no part of a network is truly safe ground in terms of cybersecurity.

When networks were smaller, IT teams simply tried to find and fix vulnerabilities as soon as possible, generally performing that task chronologically as problems were discovered. This gave rise to vulnerability and penetration testing to unmask as many vulnerabilities as possible with the goal of enabling the fixing of problems before an attacker could exploit them. The problem today—especially in federal IT where manpower shortages are a big issue—is not finding the vulnerabilities, it’s figuring out when to fix them all.

As part of a review series I have been researching and conducting for CSO Magazine, I learned just how big a problem fixing known vulnerabilities has become. Scanning several moderately sized networks—think the equivalent of a small governmental agency or a branch of a larger organization—popular vulnerability assessment tools were easily able to locate between 500,000 and 4 million vulnerabilities, with more being found or created every day. In that environment, even an aggressively funded IT team with many members would take months or years to get to them all. And they would be swimming upstream the whole time, as new potential threats are continually located. Smaller teams with just a few people have almost no hope of fixing every problem on their network.

The biggest danger, however, is not the time it would take to get through the complete list of vulnerabilities—assuming the number somehow remained static—but the fact that IT teams working either chronologically or randomly, or using any method without true context, will curse them to spend valuable time fixing vulnerabilities with minimal impact if exploited at the expense of critical, highly dangerous ones that are likely being actively exploited by hackers and malware. It’s not like missing the forest because of the trees, it’s more like missing the forest because of all the other forests.

Enter the newest type of cybersecurity defense program: vulnerability managers. They work in a variety of ways, but generally combine threat assessment used to find vulnerabilities with some form of threat intelligence used to prioritize the millions of vulnerabilities hiding inside networks. Most also have an auditing function to help track the efforts of IT teams eliminating problems and to check to ensure each fix is properly completed.

On the one hand, the rise of vulnerability managers is, in a sense, an admission that things are so bad in cybersecurity that we can no longer concentrate on just fixing problems. We must resort to managing the problems to keep our networks alive and our data safe enough to eke out daily operations. Imagine if we faced the same situation with human health care? But then on the other hand, a vulnerability is not a breach. It’s a potential problem that could, perhaps, be exploited by attackers if they find it before you do. Looked at along those lines, I have been highly impressed with the vulnerability manager programs that I’ve evaluated so far.

One of the most comprehensive is Risk Fabric by Bay Dynamics. It uses any existing vulnerability scanner or SIEM, simply collecting its data and thus never impacting network traffic. It then compiles all the vulnerabilities and intelligently ranks them by severity. It uses several factors to make its determinations including user behavior, threat intelligence, the importance of the asset, any alerts that might indicate that the vulnerability is already being exploited, and even the potential money lost if the asset is compromised along with many other data points.

Once collected, users can parse the data from Risk Fabric however they like, prioritizing risk factors to skew the results, or simply accepting the program’s decision. As users fix vulnerabilities, Risk Fabric will check to ensure that the vulnerability has been quashed, and that nothing new has taken its place.

Taking vulnerability management to new heights, Crossbow by SCYTHE is likely one of the most dangerous defensive programs ever created. Crossbow goes beyond just identifying network and device vulnerabilities, also focusing on potential threats related to employees, IT staffers and other security programs. It’s more akin to a do-it-yourself malware kit than any cybersecurity program, allowing for the loading of real-world attacks that have been successful elsewhere or the creation of new threats from scratch using a toolbox. And these are real threats with only the payload neutered—though that is optional. Deploying Crossbow is much more akin to a live fire exercise in the military than a simulation because the virtual threats it fires are real.

The idea is that Crossbow can be used to identify unknown vulnerabilities and run what-if scenarios to see how users, IT teams and existing defenses would react. For example, it could create a malware strain to hide on keydrives, which could then be spread out in an employee parking lot to see how the staff would react. Another attack we crafted relied on tricking users into visiting a compromised website. Crossbow was extremely helpful in setting that site up, and loading it with the proper malware for an advanced drive-by attack. Federal agencies that already have aggressive “red teams” tasked with attacking their own networks would find Crossbow to be an incredibly powerful weapon for their arsenal.

If you can wrap your head around the fact that it’s almost impossible to fix everything these days in cybersecurity, especially in a timely manner, then perhaps the fact that there are many vulnerability management programs being developed and deployed will provide some comfort. Given the size and complexity of most federal networks, vulnerability management tools are something that should be considered as part of a layered defense for an increasingly dangerous world.