No one ever said policing the Internet was easy. Bot herders control networks of compromised computers sometimes numbering into the thousands. Already, an FBI-led initiative dubbed "Operation Bot Roast" has identified 1 million compromised computers.
On Wednesday, FBI officials laid out charges against three men—Robert Alan Soloway of Seattle, James C. Brewer of Arlington, Texas, and Jason Michael Downey of Covington, Ky.—as part of Operation Bot Roast. But security professionals say bot herders are growing increasingly sophisticated as they search for ways to thwart their opponents.
Officials at Sunnyvale, Calif.-based Mi5 Networks reported seeing bots that connect to multiple command and control servers as well as bots that scan internal networks for different vulnerabilities and then only deliver the exploit payload for which the specific machine is vulnerable. Battling botnets, said Mi5 CEO Doug Camplejohn, has officially turned into a "game of Whack-a-mole."
"Our findings show that weve entered the second phase of botnet evolution in that theres no longer just a single C&C [command and control] head to cut off," he said. "Even if you do cut off all the C&C heads, bots keep collecting data and distributing it via peer-to-peer networks."
Finjan Chief Technology Officer Yuval Ben-Itzhak said botnet operators are utilizing a new technique he called "evasive attack" to infect users while keeping their profiles low.
"Basically, the hacker stores the IP address of search engine crawlers and URL filtering crawlers in their databases, so when they visit the hackers site for classification, the hacker server presents legitimate content," he said.
As a result, malicious sites are misclassified as normal, Ben-Itzhak explained. But when users visit the site, malicious code is served.
"We also found that malicious code is served only once to each user, to prohibit any forensic activity to detect it later," he said. "The combination [of] evasive attacks with code obfuscation is the recent blend hackers are using to control the visibility of their malicious code while having more time to infect many PCs and be undetected by traditional anti-virus products."
As famed boxer Muhammad Ali once said of an opponent: "His hands cant hit what his eyes cant see." In other words, finding sites that host the malicious code infecting computers as well as the command and control centers for botnets makes disabling them that more difficult. Dealing with botnets then takes the combined efforts of users, vendors, police and Internet Service Providers, some said.
"ISPs are far more technically capable [than the typical end user] of detecting abusive traffic coming from unsuspecting users, and should develop and implement policies that allow them to collect and act on data pertaining to abuse," said Adam ODonnell, director of emerging technologies at Cloudmark, headquartered in San Francisco. "While this sort of activity may be a challenge, ISPs should engage their anti-abuse software providers and build partnerships to detect and implement remediation strategies."
"This may involve shunting the user off to a walled garden that informs them that their system is infected to blocking certain kinds of traffic using both network and content filtration," he added.
Meanwhile, Ben-Itzhak urged businesses and home users to utilize products that can provide real-time content inspection as opposed to signatures or reputation attributes as another layer of security.
"I would encourage the lawmakers to add companies that host malicious content to be responsible as well," he said. "If a hosting company was notified of hosting malicious code and that company did not take any action to remove it in a reasonable time, that hosting company should be responsible for serving the cybercrime as well."
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 eWEEKs Security Watch blog.