Security researchers at Herndon, Va.-based NetWitness Corp. have unearthed a massive botnet affecting at least 75,000 computers at 2,500 companies and government agencies worldwide.
The Kneber botnet, named for the username linking the affected machines worldwide, has been used to gather login credentials to online financial systems, social networking sites and e-mail systems for the past 18 months, according to NetWitness.
A 75GB cache of stolen data discovered by NetWitness included 68,000 corporate login credentials, login data for user accounts at Facebook, Yahoo and Hotmail, 2,000 SSL certificate files and a large amount of highly detailed “dossier-level” identity information.
In addition, systems compromised by the botnet also give attackers remote access inside the compromised network, the company said.
“Disturbingly, the data was only a one-month snapshot of data from a campaign that has been in operation for more than a year,” NetWitness said in a statement announcing the discovery of the botnet late yesterday.
NetWitness did not release the names of the companies compromised in the attacks, which it described as being highly targeted and well coordinated.
But a story Wednesday in the Wall Street Journal identified pharmaceutical company Merck & Co., Cardinal Health Inc., Paramount Pictures and Juniper Networks Inc. as some of U.S. firms that had been infiltrated. Systems belonging to 10 government agencies were also penetrated in the attacks.
According to the Journal, the attacks started in late 2008 and appeared to originate in Europe and China.
Computers in as many as 196 countries have been affected, with many systems compromised after users clicked on phishing e-mails with links to sites containing malicious code.
Most of the compromised systems appeared to be in Egypt, Mexico, Saudi Arabia, Turkey and the U.S., the Journal reported, quoting an unnamed source with information on the attacks.
NetWitness, which provides a range of network monitoring and forensics services for companies and government agencies, discovered the botnet in January during a routine engagement with one of its clients.
According to the company, the botnet is a variant of the ZeuS botnet, which is known primarily for stealing banking credentials.
More than half of the infected systems in the Kneber botnet also contained the competing Waledac Trojan, probably because those behind the attacks wanted to build some redundancy into their attacks, NetWitness said.
“The coexistence of ZeuS and Waledac suggests the goals of resilience and survivability and potential deeper cross-crew collaboration in the criminal underground,” the company noted.
NetWitness’ discovery comes just weeks after Google disclosed that it and several other high-tech firms had been victims of organized cyberattacks originating from China.
Both incidents underscore what analysts are calling the Advanced Persistent Threat (APT) faced by a growing number of financial, commercial and government entities.
The term has been used for some time in government and military domains to describe targeted cyberattacks carried out by highly organized state-sponsored groups and organized cybergangs with deep technical skills and computing resources.
Such attacks are typically highly targeted, stealthy, customized and persistent. They also often involve intensive surveillance and advanced social engineering.
In many cases, the attacks target highly placed individuals within organizations, who are tricked into visiting malicious sites or downloading malicious software onto their systems.
In a blog post last month , security vendor McAfee’s chief technology officer George Kurtz noted that APTs had begun to change the threat landscape.
“These attacks have demonstrated that companies of all sectors are very lucrative targets,” Kurtz said. APTs, he noted, have become “the equivalent of the modern drone on the battlefield. With pinpoint accuracy they deliver their deadly payload and once discovered – it is too late.”
Confronting the threat does not always require the implementation of new technologies. But it does require rethinking some of the strategies that companies may be adopting to protecting data, Kurtz and others said. Among the steps:
1. Your adversaries are not just organized crime any more
Given the enormous growth in organized cybercrime over the past few years, most companies have implemented defenses for protecting personal and financial data from theft.
While that’s important, it’s also essential for companies to think about protecting their intellectual property data, said Ed Skoudis, co-founder of InGuardians, a Washington-based security consultancy.
“The threat has shifted,” Skoudis said. “If you go back over 10 years, the primary threats we faced were from hobbyists. Then the landscape changed, and the primary threat we most had to deal with was organized crime. And now it has shifted again,” he said.
Many cyberattackers are more are interested in corporate espionage and in stealing intellectual property than they are in going after credit card numbers or patient health data.
“There’s still the concern about people stealing 27 million cards,” Skoudis said. But a failure to adequately protect IP and corporate secrets against similar theft could result in far more long-term damage, he said. “You got to realize what your threat is and how to look for it,” he said.
2. Add network monitoring to the task list
Advanced persistent threats by definition are designed to get around firewalls, antivirus software, intrusion detection systems and other controls a company might have in place for blocking illegal access to data.
So companies need to have tools for monitoring anomalous behavior on their network, and for detecting unusual long-term persistent network connections and other “outlier behavior” Skoudis said.
Companies might also want to consider using more “white-listing” approaches to block all but a very narrow and specific set of “known good” activities on their systems and networks , Kurtz said speaking with Computerworld on Friday.
“A lot of infections and zero-day threats can be eliminated using white listing because only trusted code can run on the system,” he said.
Also vital is the need for companies to monitor their logs closely, Skoudis said. Looking at firewall logs, network based IDS alerts and Web proxy server logs can help companies identify suspicious activity on their networks, he said.
Companies need to establish a baseline for normal behavior on their networks and then regularly compare log data against this baseline to detect malicious activity, he said.
“The APT tries to be the needle in the haystack. You need to go through the information you already are gathering to find it,” he said. “You’ve got to be looking for those outliers,” he said.
3. Most Web attacks still require human intervention to succeed
Targeted attacks depend on humans clicking on something or browsing over to a malicious Web site, Kurtz said.
McAfee’s analysis of the attacks against Google and other companies showed that intruders gained access to an organization by sending a tailored attack to one or a few targeted individuals.
The attacks were probably designed to look like they came from a trusted source, leading the target to click on a link or file.
“There’s much more upfront reconnaissance taking place these days,” with intruders lurking in social networking sites and elsewhere to gather information on targets, he said. Strong user authentication and access control measures can help mitigate this issue, Kurtz said. Tools are also available that can help companies verify the authenticity of links that users might click on, to help prevent them from browsing to a malicious Web site or downloading malware, he said.
Lastly, companies need to continue emphasizing user education and training, he said. “A lot of these attacks involve a human element. There is no patch” against that.