Culture

Risk management and security theatre

Bruce Schneier is often outrageous, these days, but generally worth reading.  In a piece for Forbes in late August, he made the point that, due to fear and the extra trouble casued by TSA regulations, more people were driving rather than flying, and, thus, more people were dying.

“The inconvenience of extra passenger screening and added costs at airports after 9/11 cause many short-haul passengers to drive to their destination instead, and, since airline travel is far safer than car travel, this has led to an increase of 500 U.S. traffic fatalities per year.”

So, by six years after the event, the TSA had killed more US citizens than had the terrorists.  And continues to kill them.

Given the recent NSA revelations, I suppose this will sound like more US-bashing, but I don’t see it that way.  It’s another example of the importance of *real* risk management, taking all factors into account.

Google’s “Shared Endorsements”

A lot of people are concerned about Google’s new “Shared Endorsements” scheme.

However, one should give credit where credit is due.  This is not one of Facebook’s functions, where, regardless of what you’ve set or unset in the past, every time they add a new feature it defaults to “wide open.”  If you have been careful with your Google account in the past, you will probably find yourself still protected.  I’m pretty paranoid, but when I checked the Shared Endorsements setting page on my accounts, and the “Based upon my activity, Google may show my name and profile photo in shared endorsements that appear in ads” box is unchecked on all of them.  I can only assume that it is because I’ve been circumspect in my settings in the past.

Click on everything?

You clicked on that link, didn’t you?  I’m writing a posting about malicious links in postings and email, and you click on a link in my posting.  How silly is that?

(No, it wouldn’t have been dangerous, in this case.  I disabled the URL by “x”ing out the “tt” in http;” (which is pretty standard practice in malware circles), and further “x”ed out a couple of the letters in the URL.)

The Biggest Gap in Information Security is…?

As a person who’s committed to helping raise awareness in the security community as a whole, I’ve often found myself asking this question. While there are several issues that I think contribute to the state of information security today, I’m going to outline a few of the major ones.

One major problem that spans every industry group from government to finance, all the way over to retail, is the massive amounts of data stored, the large number of devices to manage and frankly, not enough people to do it all. Or not enough people with the appropriate level of security skills to do it. I recently had a student in an Ethical Hacking class who asked me if I would be open to discussing some things in private with him concerning some issues he had at work. During dinner he confided in me that he sees his job as becoming more and more impossible with all the security requirements. He let me know that he had recently completed a penetration test within his company and felt he didn’t really get anything out of it. My first question was how many nodes were in the scope of the test. His response was 20,000. So naturally my next question was how big was his pen test team. To that he looked at me blankly and said “It was just me”. My next question was how long did he have to complete the test. And to that his reply was 3 days. This shocked me greatly and I candidly let this individual know that with a scope that big it will usually take one person more than three days to do proper discovery and recon and wouldn’t even give you time to even start vulnerability discovery, mapping, and exploitation testing/development.  I also informed him that for a job like that I usually deploy 3 people and usually contract a time of 2 to 4 weeks. Keep in mind this young man was a very intelligent and skilled person, but he lacked the skills to pull this off. After more conversation I realized that he himself was responsible for scoping the 3 day time to complete the test.

This brings me to the first main point; I see a trend of corporations and entities placing more security responsibility on individuals without giving them enough resources or training. This person admitted he really didn’t even have the skills to know how long it would take him and he based his time estimate off something he found on the web using google, which was why he was in the class. After the class he emailed me and thanked me for finally giving him the understanding to realize what it would take to successfully complete his internal testing. He drafted a plan for a 4 week test and put in a request to have temporary help for the 4 week duration. 2 months later he sent me another email and a redacted copy of the penetration test (after I signed a NDA of course). I was impressed with his work and let him know that. This demonstrated that even the most intelligent people can become overwhelmed if put into an impossible situation with no tools.

Second is the increasingly swift changing threat models. What would be considered a very secure computer 10 years ago (basic firewall, and up to date anti-virus) would be considered a joke today. I can remember when OS patches were mostly just non-security related bug fixes. If the bug didn’t affect you, you didn’t worry about the patch since it often broke other things. This way of thinking became the norm, and still exists in some places today. Add to that the web based attack vectors and client side attacks, it gets even more detrimental. I watched as Dan Kaminsky wrote himself into the infosec history books with his DNS attack. At the same time I saw one pen test customer after the other totally ignore it. Once we were able to exploit this in their environment we usually got responses like “i thought this mostly affected public/root dns servers”. The bottom line is DNS is DNS, internal or external. While Dans’ demonstration was impressive, thorough and concise, it left the average IT admin lost in the weeds. As humans when we don’t truly understand things we typically either do nothing, or do the wrong things. A lot of the media coverage of this vulnerability mostly focused on the public side threat. So from a surface look, it appeared to be something for “others” to worry about. Within weeks of that presentation there were new mobile device threats identified, new adobe reader threats, and many other common application vulnerabilities were identified. With all these “critical” things identified and disclosed within weeks of each other, it is apparent why some security professionals feel overwhelmed and behind the curve! Throw in the fact that I’m learning from clients and students alike that they’re now expected to be able to perform forensics investigations, and the weeds get deeper.

The last thing I want to point out is a trend I’ve noticed in recent years. The gap between what I like to call the “elite” of the information security world and the average IT admin or average whitehat/security professional is bigger than it’s ever been. Comments I’ve heard is “I went to blackhat and I was impressed with all of what I witnessed, but I don’t truly understand how it works and what to really do about it”. I think part of this is due to the fact that some in the information security community assume their audience should have a certain level of knowledge and refuse to back off that stance.

Overall I think the true gap is in knowledge. Often times individuals are not even sure what knowledge is required to perform their job.  Check back soon as I’ll be sharing some ideas as to how to address this problem.

Keatron Evans, one of the two lead authors of “Chained Exploits: Advanced Hacking Attacks From Start to Finish”, is a Senior Instructor and Training Services Director at the InfoSec Institute.