The Storm Worm botnet (thought to be the largest network of compromised machines in the world) has begun to figure out which security researchers are trying to disrupt its command-and-control systems and knock them offline with unmanagable crapfloods from its zillions of zombie machines.
The worm can figure out which users are trying to probe its command-and-control servers, and it retaliates by launching DDoS attacks against them, shutting down their Internet access for days, says Josh Korman, host-protection architect for IBM/ISS, who led a session on network threats.
"As you try to investigate [Storm], it knows, and it punishes," he says. "It fights back."
As a result, researchers who have managed to glean facts about the worm are reluctant to publish their findings. "They're afraid. I've never seen this before," Korman says. "They find these things but never say anything about them."
And not without good reason, he says. Some who have managed to reverse engineer Storm in an effort to figure out how to thwart it have suffered DDoS attacks that have knocked them off the Internet for days, he says.
As researchers test their versions of Storm by connecting to Storm command-and-control servers, the servers seem to recognize these attempts as threatening. Then either the worm itself or the people behind it seem to knock them off the Internet by flooding them with traffic from Storm's botnet, Korman says.