Patience Pays Off for Hackers in Web Security War

LAS VEGAS—The creators of the Coreflood Trojan have managed to stick their digital hands into the pockets of victims for years. And they have done it largely under the radar, according to research revealed at the Black Hat conference here Aug. 6.

Some may remember the Coreflood Trojan from the well-publicized case of Joe Lopez, a Miami businessman who sued his bank for failing to protect his account after attackers wired more than $90,000 to a bank in Latvia. The masterminds behind that attack were the makers of the Coreflood Trojan, according to Joe Stewart, director of malware research at SecureWorks.

But since that time, the minds behind Coreflood have been busy. By getting access to a Coreflood command-and-control server—since shut down—Stewart and The Spamhaus Project uncovered 50GB of mostly compressed data stolen from infected users. As much as four times that amount is believed to have been previously harvested and deleted, according to scripts left behind by the gang.

The stolen data included 463,582 user names and passwords to more than 35,000 domains. Nearly 8,500 passwords were for banks and credit unions in the United States and overseas. In one directory, Stewart found numerous subdirectories left over from checking account validity. In one subdirectory, 740 stolen accounts were found in the configuration files. Those the hackers tested for validity had an average of $4,553.74 in savings and $2,096.31 in checking—meaning the hackers could have had access to more than $2.5 million in a single financial institution if the averages held true for the other accounts.

So just how did the hackers manage to go seemingly undetected for so long? According to Stewart, it was by not targeting things like instant messaging or e-mail, which get a lot of attention from security vendors. Instead, the hackers relied on drive-by attacks, and would pick a hosting provider and do a mass hack of every single Web page on that particular server. Then they would wait for users—particularly domain administrators with high-level rights.

Their approach is different from many gangs of cyber-criminals who use massive e-mail campaigns to spread their malware, as in the Storm botnet.

“The key I think is not to spread through things that are in your face—like instant messaging, worms or e-mail worms …. [That] gets the anti-virus companies’ attention real quick,” Stewart said. “It’s much better to put out an infected Web site somewhere and let people find it.  They infect one corporate user, they wait until a domain administrator logs into that workstation … and then once that malware has privileges, it just rolls out its own code just like an update to all the others on the network.”

The group did not rely on zero-day attacks, just standard exploits that one can get from various underground forums. Its victims include hospitals, colleges, law firms and a U.S. state police agency that SecureWorks declined to identify.

"Their trick is not in getting that initial infection—their trick is being patient and waiting for the right person to log into that workstation and then (taking) over that whole network," he said.