Data, Data Everywhere, nor Any Time to Think

I remember when I first started trying to solve network security problems, using fancy network analytics.  I applied the classic suspension of disbelief that’s necessary to work on any emerging technology – first, you assume all the hard problems will be easy, and second, you assume the impossible ones will just go away.  Happily, much of this is true – it’s funny how well it works.  Only later do you learn which problems are the truly hard ones.

What’s hard about network security analytics?  Well, not the security, and not the analytics – we’ve found we can do plenty on both of those that pays off really well, given the data.  The pesky data, now that’s a different kettle of enchiladas.

data-everywhereAt first, I didn’t want to talk about data gaps – that sounded like a challenge to good analytics.  I was half right.  Eventually, enough CISO’s got it through my skull that uncovering data gaps may be pointing to reasons why analytics will be held back, but it’s also major value, in and of itself.  I was being dense – if we try to analyze security data, and we find it’s got holes in it, well, this means the security team didn’t know what was going on to start with!  Turning up these gaps is one of those inconvenient truths.  These days we’ve gotten pretty good at it.

But then what?  Typical security organizations are drowning in data, so how can I complain about needing more?  Well, facts are just facts; useful information, or better yet, actionable intelligence is something else altogether.  We stockpile data from sensors, but we struggle to find useful signal in there.  We deploy automated signal reduction engines, but they just turn mountains of alerts into hills of alerts, and we still don’t have time or people enough to climb those.  And along come these network security analytics people saying “what you need is more data”.  Hmmm.

Of course, what we need is the RIGHT data, processed the right way, at the right time.

Inside the mind of an attacker

This morning, I woke up, walked downstairs, and performed my morning rituals, including a review of OmniFocus on my iPad to see what was on tap for today. I looked at my list of projects, my next actions, and those items that are due in the next few days. Then, I went to work.

In many homes across the world, days began in similar fashion. Some of those reviewing their projects, however, had a decidedly different thematic thread: their projects have the goals of breaking into the networks and servers of key government and industry organizations for purposes of espionage, theft, or disruption. And they get paid to do it.

Some of us remember the earliest days of the Internet when servers were open to all. In fact, anyone could log onto the root account at Richard Stallman’s server and create their own personal account. My, how far we’ve come when breaking into networks and systems is a career path!

In the early days of people breaking into systems and networks, most actors were solo and focused on showing their own skills while demonstrating the weakness of those they attacked. Early viruses and worms (like the Morris Worm) were often the result of bugs in the target systems and mistakes in the attacking code.

hackerToday, governments across the world are applying their resources investing in full-time staff to break into systems and networks in other parts of the world. From the Syrian Electronic Army to the People’s Army, the US Government, and organized crime, attacks come from many different sources looking for a variety of results. This means the mentality is professional, organized, and coordinated, and the attackers are motivated by a variety of results, from financial to patriotic.the early days of people breaking into systems and networks, most actors were solo and focused on showing their own skills while demonstrating the weakness of those they attacked. Early viruses and worms (like the Morris Worm) were often the result of bugs in the target systems and mistakes in the attacking code.

Knowing this, it’s essential that you determine the best way for you to defend against these attackers. They aren’t going to give up, so you need to be diligent and focused on your defenses. And we’ll talk more about that next time.

Negative Unemployment

I recently attended a gathering of Wall St CISOs, one of whom referred to the “negative unemployment” in our industry.  I thought this was a great phrase, and I’ve found it’s a quick way to get across some quite deep points about current security.

At first, it just sounds cute, but in practice, it’s about as cute as the Oil Crisis.  Bad guys have figured out how to make money by attacking our weak defenses.  We’re scrambling to catch up.  The C-Suite and the board are more accommodating than they have ever been – something to do with the recent dismissal of the Target CEO, I shouldn’t wonder.  We know we need people, so we go to hire them, and what do we find?  Bad resumes.

knowledgegapHave you found it easy to hire the talent you need?  If so, lucky you – feel free to drop hints in the comments section (or just gloat – your peers tell me they aren’t having it so easy).

It makes for an ugly choice.  Do we hold standards high, waiting for people with the right skills to come along?  Or do we hope to train people new to the field?  As I look around, I can see our discipline soaking up some people of – how should I put it? – marginal aptitude.  I’ve seen this before – I remember the go-go days of the late 90’s, when Silicon Valley start-ups sucked in all kinds of people with no business working in such environments.  When that went all pear-shaped, it wasn’t so bad – sure, some stock options suddenly lost a zero or two in value, but it’s not really fair to whine about that.  Watching the same thing happen in corporate IT security is a much scarier proposition.

Defending Against Botnets

Botnets have been around for many years, but Distil Networks’ recently-released research shows that their use not only continues to grow dramatically, but that use is becoming more sophisticated. In having the bots focus their attacks during off-hours, the attackers may have a greater window of opportunity for damage before discovery.

This underscores the need to expand security analytics beyond the reactive focus of IPS/IDS to also include complete proactive analysis of what could happen. For example, analyzing all of the possible paths into and through an enterprise network–including from vendors and partners–within the overall context of the complete, complex network, allows the enterprise to ensure limited access before any paths are probed by a bot.

The botnets are a primary contributor to the distributed denial of service attacks, for instance, which are reported to have volumes up to 300Gbps.

As we have seen from widespread and newsworthy breaches over the past few years, it is very difficult to react quickly to an attack in progress. While such defenses are critical, equally vital are analytics that determine and monitor the effectiveness of the entire network as a system including all of its security controls and system vulnerabilities in context. This is one of the reasons RedSeal’s analytics include the complete set of possible network paths and not simply flows currently active in the network.

The key to winning the game is leverage. Knowing more, being more proactive, being certain that your intentions are realized by technology. How can you know?

Recently, I have seen firewall configuration files containing well over 150,000 lines of configuration. These devices live within networks with thousands of other devices that forward packets according to a variety of rules 250px-whackamole(routing, access control, load balancing, and more). The only way to know what’s really going on is to perform an in-context analysis of the  network. This is very difficult to do well, and impossible to do without automation. Furthermore, if you don’t  do it, you are relegated to playing Whack-A-Mole with the probes and attacks that are being launched against  you, probably at the rate of thousands per day.

Use automation as a proactive offense against what could be launched even as you continue to deploy reactive systems to respond to attacks that make it through your defenses.