The New Attack Pattern

Roger Thompson, who built PestPatrol and is now doing other work in testing and development, recognized the pattern a year ago with the WMF vulnerability: Black hat researchers develop new vulnerabilities and sell them to exploiters with an embargo date, meaning they cant be used before that date, which is probably on or shortly after patch day.

To quote Roger:

...what really happened in 2005 was that the Bad Guys got down to business. They dont want to bring the Internet to its knees anymore, they just want to farm it and expand their networks. I think the cleverest of them, the CoolWebSearch gang, have stopped combing the security mailing lists for vulnerability announcements, and have started advertising to buy 0-days "Why give your 0-days up for free? Sell them to us and make money."

In 2006 the most prominent of these vulnerabilities have been in Microsoft Office applications. There have been scores of other vulnerabilities in less-important programs, and perhaps some of these were for hire as well. If youre going to target a small number of networks, an obscure program might be adequate.

There are some important caveats about these vulnerabilities: They are usually addressable by anti-virus programs without a patch of the underlying vulnerability, and they almost always require a social engineering breach to get through.

The anti-virus protection is generally not valuable to the victims of the "very limited, targeted attacks" this is why so many experts have been grumbling about the "death" of the signature-based AV approach lately. But it is valuable to the rest of us because it limits the effective life span of the attacks. And there are many products that attempt to block unknown attacks through a variety of approaches.

The social engineering aspect is an unfortunate constant in security. If you can fool people into performing dangerous actions, you can gain at least some level of access. And when youre targeting a small number of networks, its not hard to imagine an e-mail spoofed to look like its coming from some senior exec to other addresses scraped off of the companys Web site. Its probably not hard to get through to at least a few users. Even so, its still not clear how many of the attacks are successful.

Many experts think its too late to stop botnets. Click here to read more.

So there are still problems, but your likelihood of being attacked has gone down. Your ability to defend yourself has also gone up. You can expect things to change over the next year, as they always do, to adjust to conditions. For instance, if the "patch day=zero day" pattern continues, Microsoft may have to adjust its strategies.

Perhaps patches could move to a more frequent schedule. My moneys on a general tightening of policies to block attacks.

Security Center Editor Larry Seltzer has worked in and written about the computer industry since 1983. He can be reached at larryseltzer@ziffdavis.com.

Check out eWEEK.coms Security Center for the latest security news, reviews and analysis. And for insights on security coverage around the Web, take a look at Ryan Naraines eWEEK Security Watch blog.