Ever since we learned that last autumn's massive Target data breach was accomplished with the use of access credentials stolen from a third-party vendor, I've been concerned about similar threats at my company. We use lots of vendors, many of which have access to our network. I've spent a lot of money, time and energy fortifying my network and its perimeter. But what if one of the vendors gets compromised? Could hackers sneak into my network through the side door, posing as a legitimate service employee?
Of course, this is really nothing new. I've written a few columns in the past about problem vendors and some of the things I've done to deal with the consequences of business managers signing contracts with third parties without involving my team. I've also mentioned in the past that I try to review third-party SSAE16 (previously SAS70) reports on our vendors that audit firms have produced, and I hope those reports are accurate and unbiased.
But if you look at what happened to Target, there isn't much in an SSAE16 report that really addresses the threat of stolen credentials. Target's vendor was compromised by simple phishing emails that tricked its employees into executing malware that stole usernames and passwords, and that's what the attackers used to get into Target's systems. That's a kind of social engineering -- using human nature to get what they wanted. When performing an SSAE16 review, audit firms look at the policies and controls specified by the vendor for basic security practices. But in every organization, there are employees who will be tricked into clicking malicious links and attachments. It's certainly true at my company, where a couple of people a week routinely get fooled by claims that they have a UPS package waiting for them, or an IRS audit or refund they must see immediately, or an invoice for something they didn't order. No matter how many emails I send reminding people of these scams, the posters on the walls, or the security awareness training, somehow the human brain doesn't kick in soon enough to stop the finger from clicking.
If I can't stop my own people from falling for malicious email scams, how can I expect a small air-conditioning contractor, HR recruiting firm, or food-service company to stop their people from doing the same?
Let's be realistic. The real question is not "How can we stop the phishing or the click-happy recipients?" but "What can we do to contain the damage?" We must assume that people will continue to fall for these scams, regardless of how much training and reinforcement we do. And the same is true for our vendors. I can't prevent all credential theft, but I can monitor the use of those credentials along with the traffic going into and out of my network, and I do. With my SIEM, I look for unusual patterns of account usage and network traffic. I also push hard on IT support staff to respond to virus alerts quickly and take fast action to contain infections, since that's the first step in many compromises. Incidentally, this is also where Target fell short. From what I've read in the news, Target's IT staff ignored repeated virus warnings from their monitoring systems, including those directly responsible for the breach.
But how can I be sure I am aware of every bit of malicious traffic on my network, or whether the activities of users on my network are 100% legitimate? It's the things I don't see that keep me up at night.
This week's journal is written by a real security manager, "J.F. Rice," whose name and employer have been disguised for obvious reasons. Contact him at firstname.lastname@example.org.
To join in the discussions about security, go to blogs.computerworld.com/security.