Speeding Incident Response With 'Indicators' Of A Compromise

Advanced persistent threat: I like the term -- it sounds evil, and it is...well, at least I think it is. There has been a lot of news, opinions, and genuine FUD on APT since Google went public with news of its breach several weeks ago. Until then, I really don't think anyone ever paid much attention to what APT was, even though well-respected people, like Richard Bejtlich and the folks at Mandiant, have been talking about it for a while.I'd like to avoid rehashing some of the statements and arguments that people have made and look at something more interesting. Sure, some of the blogs and articles have been really good and insightful, but others were so off the mark that I wonder if they can even spell APT.

One area I've found particularly interesting is the topic of detection of these threats. I'm a big fan of actionable intelligence, and I really like the direction of Mandiant's "Indicators of Compromise," or IOCs.

Matt Frazier covered the topic of IOCs in the M-unition blog, "Combat the APT by Sharing Indicators of Compromise." He points out that the traditional distribution method of compromise data is exchanged in CSV or PDFs that are full of "known bad" information. The known bads include information like "name, size, MD5 has values and paragraphs of imprecise descriptions supplemented by ad-hoc exchanges between targets." If you've ever spent much time doing malware research, then you know exactly what he means. Gathering useful, actionable information can be painful.

Being on the frontline of fighting malware for a large network, I'm excited to see an effort to produce more actionable information in an easily digestible format (XML). From the network intrusion detection side of things, I've had many situations where events in the IDS showed a system was definitely compromised even though the antivirus product installed on the endpoint said all was well. How could IOC help?

First off, the XML could easily be processed and put into a database that could then be queried to quickly produce a one-sheet document on how to determine whether a host was indeed compromised. The sheet would show different indicators in a Boolean decision tree that a first responder could follow to confirm the compromise. That would be a much easier process than giving the first responder links to five different Web sites with varying levels of information and analysis that they'll have to weed through to figure out what's relevant to this particular case.

I'm getting a little ahead of myself since the details and tools of IOC haven't been published to the M-unition blog yet, but it sounds promising. Since it sounds like it's going to be relatively open and accessible for anyone to use, I can envision tying it in with the Emerging Threats Snort ruleset. Snort rules have a "reference" field that could point to the unique ID of a specific IOC to make it easier for detecting the compromise at the host level. Ah...and a security pro can wish, right?

Mandiant's Intelligent Response product is very impressive, and the company has produced some incredibly powerful free tools (i.e.m Memoryze & Audit Viewer), so I expect good things to come from its Indicators of Compromise effort. Keep an eye on the M-unition blog for more info, and I'll post more about my testing as the tools and IOCs are released.

John H. Sawyer is a senior security engineer on the IT Security Team at the University of Florida. The views and opinions expressed in this blog are his own and do not represent the views and opinions of the UF IT Security Team or the University of Florida. When John's not fighting flaming, malware-infested machines or performing autopsies on blitzed boxes, he can usually be found hanging with his family, bouncing a baby on one knee and balancing a laptop on the other. Special to Dark Reading.