Patch Management: How to Prioritize an Underserved Vulnerability

Why is one of the biggest problems in cybersecurity also one that CISOs largely ignore? Here are three reasons and a road map to a modern approach.

John Bock, Senior Research Scientist

December 19, 2019

6 Min Read
Dark Reading logo in a gray background | Dark Reading

Optiv Security recently conducted a survey of 200 chief information security officers (CISOs) across the US and UK and, despite their geographical differences, their view of patch management is the same: It's on the bottom of their list of priorities. This result is startling because unpatched vulnerabilities are one of the most common causes of data breaches — 57% of all breaches in fact, according to Ponemon Research.

Survey respondents were asked: If you could stop the business for six months and have the luxury of time to execute any security priorities, which of the following areas would you choose to focus on? CISOs cited employee education, infrastructure simplification, and the creation of DevSecOps models as their top priorities. Catch up on basic functions like patching and vulnerability scanning finished dead last.

These results raise this question: Why is it that one of the biggest problems in cybersecurity is also one that is least paid attention to by CISOs? There are a variety of factors at play here, but the three most commonly encountered are:

Factor 1: For CISOs, patch management is likely a long-term friction point between security and IT (and operations and engineering). Typically, security informs IT of the need to patch, and then IT implements the patch. This relationship has been in place for two decades now and is one of the oldest security workflows. A consistent attribute of that relationship is also that IT will not have the velocity that security wants when it comes to applying patches, so it ends up being a sore spot between the two groups. Compared with other security challenges and new security technologies, it can seem both mundane and a point of conflict to be avoided.

Factor 2: Since patching is an administrative function, a CISO's responsibility is to inform that a patch is needed versus dictate how patches are applied and when. This is generally more of a political or psychological factor, as security needs cooperation with IT to be successful and respecting boundaries of either group's responsibilities is a part of that. Although the role of security can sometimes be seen as "job done" when the owner of the asset that needs to be patched is informed, that's not always the most effective approach.

Factor 3: The final, and perhaps toughest, factor to modify is the culture and set of policies that restrict patch and change management. Organizations that have been around for long enough to experience a bad patch rollout and deal with the consequences will have a substantial amount of rigor that security remediation activity needs to pass through. Many of these policies were drafted before automation and orchestration existed, and often, best practices for change management will reflect a cautionary approach that may not factor in capabilities enabled by current technologies. Also, security doesn't want to be seen as the one removing current safeguards in the name of risk reduction. More often than not, the prime mover in improving speed here is going to be the business adopting a broader change.

So, what else can companies do to help prioritize patch management in this landscape? Here's a road map that security teams can follow:

Discover and analyze the full patch management cycle. This means security should have a full understanding of exactly how every asset is being patched, down to the specific tools and revisions. The goal here is to map the steps in the process and also identify points where security can be an instrument for visibility. If it takes over a month on average to patch a vulnerability, a company should be able to see that workflow and break out where slowdowns are occurring. Does the capability exist to patch a vulnerability immediately, but it needs to wait until the next change window? Does the patch need to go through multiple levels of a change management process that takes weeks? Does the organization lack automation for patch management? Without breaking down the workflow into discrete steps, it is difficult to give a precise diagnosis.

Perform an independent tools analysis. When it comes to tooling for patch and configuration management, it is almost always going to be a decision from IT and engineering. That said, as a CISO, it is worthwhile to research and understand what capabilities exist in that market versus what the organization is currently using. The goal here isn't to call out IT if the department is behind the times, but to be able to go to IT staff with information that will help them be more successful overall, and provide an additional voice for the resources they would need to bring those capabilities online. The older an organization is, the more value you will tend to gain from this exercise.

Attack surface reduction. Have a conversation about attack surface reduction with IT and work down two primary tracks: first, reduction as a factor of reduced attack surface from the start of an asset's life cycle, and second, how IT determines when an asset is retired from the network. Modern configuration management and orchestration platforms can reduce the need for repeated manual work during hardening of hardened images but still requires more up-front development than all-in-one-style builds. At the other end of the life cycle, you have assets that are still live in the environment but not in active use, and often not under active management. These forgotten systems are often standouts in vulnerability management reports because the vulnerabilities present are an attractive target for attackers.

While patch management isn't the most dynamic topic on a CISO's agenda, it is still a foundational component of a security program. Being one of the most long-lived topics also can lead to a mindset of "everything that can be done has been done," and a challenge without room for improvement. However, given the capabilities that are now widely available, it should be worth a re-evaluation of current processes and seen as an opportunity where security can achieve its goals while still keeping up with the velocity of the business using a modern approach.

Related Content:

Check out The Edge, Dark Reading's new section for features, threat data, and in-depth perspectives. Today's top story: "The Next Security Silicon Valley: Coming to a City Near You?"

About the Author

John Bock

Senior Research Scientist

John Bock is senior research scientist at Optiv. Prior to this role, John was vice president of threat research, and before that he was the leader of Optiv's application security practice, which provided application pen testing and other software security services. With more than 15 years of application security and penetration testing experience, he is able to provide practical strategies for addressing security challenges and employing advanced capabilities to enable security assessment and defense.  Prior to joining Optiv, he held consulting and engineering positions at Casaba Security, Foundstone, and Internet Security Systems. He is also a contributing author and technical editor to multiple security publications, including the Hacking Exposed series.

Keep up with the latest cybersecurity threats, newly discovered vulnerabilities, data breach information, and emerging trends. Delivered daily or weekly right to your email inbox.

You May Also Like


More Insights