Since April 2014, a new trend in security has experienced a meteoric rise, with headlines grabbed in both mainstream media and the tech press. Vulnerabilities, once the preserve of the researcher and defender, were suddenly thrust into the limelight when an OpenSSL bug was announced. Previously, the chances of a big security bug being discussed outside of our industry were low, the media being more concerned about the outcomes and impacts rather than the methods. But with a catchy name and a simple vector logo, Heartbleed went viral—and not in the sense we’re used to.
This new found fame certainly has some advantages; for example, it pushes the plight of plucky patchers onto the agenda of Boards everywhere. But for many, the fire drill caused by a new logo vulnerability can be driving the wrong behaviour. A good example is POODLE, which was announced around the same time as MS14-062; but due to the catchy name, POODLE got far more attention when MS14-062 was arguably the more troublesome flaw.
The fire drill caused by a new logo vulnerability can be driving the wrong behaviour
This knee-jerk reaction, driven by disclosure and popularisation of bugs by slapping a name and logo on them, is frustrating when bigger security issues garner far less attention. With that in mind, I decided to create my own security issues that I think actually deserve a cool logo and memorable names in the small hope that these are also propelled in front of the people who set business priorities.
Over the coming weeks I’ll be discussing vulnerabilities such as Invader, Stutter, Bandit, EagerBeaver and Subversion. But for now, I give you Glimpse.
Like many others in the industry, one of my first jobs was in IT management. Tasked with building out the network to support the growing business and dealing with the far too frequent outages caused by over utilised systems, I’d occasionally provision and release new platforms. This was some time ago when the default server build was more physical in nature: virtualisation was something only a few were playing with on the desktop and cloud services were just hosting providers. With only a handful of systems spun up and released per year, it was low enough to have a naming convention based on famous characters from your favourite novel; I can still remember the IP addresses and names of many of my systems, which read like a who’s who of the literary world.
But that was 15 years ago, and boy have things changed since then. We’ve moved from a few physical servers running many services, towards many virtual systems running one or part of the application stack. When you’re bursting the amount of IP addresses communicating on the network by 50% per day, it’s no wonder traditional tools are failing to give clarity in complexity.
Glimpse, the lack of an accurate understanding of systems and data owned by an organisation
This has led to the security issue that almost every company faces today: Glimpse, the lack of an accurate understanding of systems and data owned by an organisation. Glimpse impacts almost everyone and has ripple effects throughout the operational environment. With a lack of visibility, data driven decisions and answers to simple questions are almost impossible to make. When Heartbleed hit, the major issue wasn’t that OpenSSL 1.0.1b had a huge flaw in it; it was that many struggled to answer the simple question, “Where are we vulnerable to this?”
Answering “Where are we vulnerable to a major flaw?" should take seconds and have a high level of accuracy. But unfortunately, due to Glimpse, the response is often more of a guesstimate rather than fact. The security team that answers with, “We have 1254 systems affected by Heartbleed ± 3%” would be trusted by the business far more than another team that simply said “I don’t know, let me find out for you.”
I recently spoke to the security director of a large retailer, and I asked what he felt his biggest issue was. He replied, “I can tell you exactly how many tins of beans I have in every store around the world and in every warehouse, but I can’t tell you how many servers are currently running.” This is a real issue that deserves a logo, a catchy name, and maybe even a theme song. Glimpse is a real problem that must be replaced with complete discovery and visibility.
Glimpse is a real problem that must be replaced with complete discovery and visibility
Find out about other security issues that also deserve the Heartbleed treatment; in my next blog, I’ll introduce you to Subversion.