Category: #IT/CYBER/TECHNOLOGY

Built off the Tinba banking Trojan and distributed through the elusive HanJuan exploit kit, Fobber info-stealer defies researchers with layers upon layers of encryption.

A stealthy new info-stealing browser injection malware aims to make security researchers’ job very difficult. Fobber evades detection and defies anaylsis by sliding from one program to another, using randomly generated filenames, encrypting command-and-control communications with a custom algorithm, and encrypting individual pieces of code within the payload, so that each function must be separately, painstakingly decrypted before it can be run.

Researchers at Malwarebytes discovered Fobber, and Fox-IT researchers have confirmed that it is based off of the Tinba banking Trojan. So far, Malwarebytes has not witnessed Fobber stealing banking credentials, but that may just be a matter of time, according to Malwarebytes senior security researcher Jerome Segura.

“I think they’re testing the waters,” he said. All infections, thusfar, have been in the Netherlands, so Segura believes the Fobber authors are still testing out the tool before rolling out operations on a larger scale.

Malwarebytes found Fobber by accident when they stumbled across activity by the elusive HanJuan exploit kit. Opportunities to study HanJuan are rare, because it usually takes great pains to hide itself. Malwarebytes simply referred to it as the “Unknown exploit kit” when they first wrote about it in August 2014.

“It’s a very discreet exploit kit,” said Segura, “so that’s what caught our attention.”

Considering its usual discretion, the researchers discovered HanJuan acting in a way that seemed out of character. It was being hosted on a legitimate Dutch website that had been compromised, and was being distributed through a malvertising campaign. An embedded ad within the Adf.ly URL shortener service directed victims to the compromised site.

Once researchers had a look at the payload HanJuan was delivering, they saw “we have something new on our hands,” says Segura. “It’s very well encrypted. A lot of attention to detail in there.”

Written for both Flash and Windows Explorer, Fobber uses a memory stack pivoting exploit. As Segura wrote in a blog post “Unlike a normal Windows program, Fobber makes it a habit to ‘hop’ between different programs.” Fobber.exe itself will eventually terminate, and the malware execution will continue in Verify Class ID, until that terminates and picks up again in Windows Explorer, until that terminates and picks up again in a web browser.

Beginning with the Verify Class ID process, Fobber really frustrates any security researcher’s attempts to analyze it. The code for each function must be decrypted before it can be executed; then it re-encrypts itself after completion.

It also encrypts all communication with the command-and-control server, using a custom algorithm. According to Segura’s blog “Content sent by the server is signed by its RSA1 key (to prevent botnet hijacking) while the Fobber code has the public key embedded within, notifying the signature before processing the content.”

The malware then performs browser injection (it works on Internet Explorer, Google Chrome, and Mozilla Firefox), hooks into certain functions (InternetCloseHandle and HttpSendRequest in IE), and waits to see when interesting credentials are being requested.

Fobber could then act like a man-in-the-middle and lift those credentials, and then use them for a variety of attacks — including fraudulent banking transactions that would appear to the bank to be completely legitimate requests coming made from a customer’s own machine with their valid credentials.

All of these techniques make it difficult for security companies to discover malware, put a name to it, and develop effective countermeasures.

“If you don’t make the headlines,” says Segura, “you have less scrutiny, and you can keep using” the tool for longer.

Malwarebytes has passed on its information about Fobber, HanJuan, the malvertising campaign and the compromised website to Dutch law enforcement.

 

[Dark Reading]

Internet pioneer and DNS expert Paul Vixie says ‘passive DNS’ is way to shut down malicious servers and infrastructure without affecting innocent users.

Botnet and bad-actor IP hosting service takedowns by law enforcement and industry contingents have been all the rage for the past few years as the good guys have taken a more aggressive tack against the bad guys.

These efforts typically serve as an effective yet short-term disruption for the most determined cybercriminal operations, but they also sometimes inadvertently harm innocent users and providers, a problem Internet pioneer and DNS expert Paul Vixie says can be solved by employing a more targeted takedown method.

Vixie, CEO of FarSight Security, which detects potentially malicious new domain names and other DNS malicious traffic trends, says using a passive DNS approach would reduce or even eliminate the chance of collateral damage when cybercriminal infrastructure is wrested from the attackers’ control. Vixie will drill down on this topic during his presentation at Black Hat USA in August.

Takedowns typically include seizing domains, sinkholing IPs, and sometimes physically removing equipment, to derail a botnet or other malicious operation.

Perhaps the most infamous case of collateral damage from a takedown was Microsoft’s Digital Crimes Unit’s takeover of 22 dynamic DNS domains from provider No-IP a year ago. The move did some damage to Syrian Electronic Army and cybercrime groups, but innocent users were also knocked offline. Microsoft said a “technical error” led to the legitimate No-IP users losing their service as well, and No-IP maintained that millions of its users were affected.

The issue was eventually resolved, but not after some posturing in hearings on Capitol Hill, and debate over whether Microsoft was getting to heavy-handed in its takedown operations.

Vixie says the key to ensuring innocent users and organizations don’t get swept up in the law enforcement cyber-sweep is get a more accurate picture of just what is attached to and relying on the infrastructure in question. “There is a tool that you can use to find out [whether] the Net infrastructure belongs to bad guys so you don’t target anything else” that shares that infrastructure and is not malicious, Vixie says.

Passive DNS is a way to do that, says Vixie. With passive DNS, DNS messages among DNS servers are captured by sensors and then analyzed. While Vixie’s company does run a Passive DNS database, he says he’s advocating that investigators and task forces doing botnet or domain takedowns use any passive DNS tool or service.

Vixie says the two-part challenge in takedowns to date has been ensuring law enforcement “got it all” while not inadvertently cutting off innocent users and operations in the process.

Passive DNS not only can help spot critical DNS name servers, popular websites, shared hosting environments, and other legit operations so they aren’t hit in a takedown operation, he says, but it can also help spot related malicious domains that might otherwise get missed. That helps investigators drill down to the malicious tentacles of the operation, according to Vixie.

Vixie in his talk at Black Hat also plans to lobby for researchers and service providers to contribute data to passive DNS efforts.

Meanwhile, it’s unclear what long-term effects takedowns have had on the cybercrime underground. “I’m involved in the same [volume] of [takedown] cases than I ever was. The trend of bad guys is on an upward swing,” Vixie says.

 

[Dark Reading]

Since May 2014, the Chinese government has been amassing a ‘Facebook for human intelligence.’ Here’s what it’s doing with the info.

Leading into 2015, the cybersecurity community was still reeling from the impact of a destructive attack unlike any other we have seen in terms of visibility, scale, and impact. Already halfway into 2015, there is no shortage of breaches. We have already witnessed major compromises in healthcare, the US government, the Bundestag, and media being attacked by sophisticated adversaries, in most cases, roaming freely on networks for months at a time.

Attackers from China, Russia, North Korea, ISIS, and even potentially friendly governments have dominated the headlines. In case you have your head in the sand, this is not going away anytime soon. Compared to traditional espionage, “cyber espionage,” or CNE as the military likes to designate it, has a lower cost of entry, less risk if you are caught or compromised, and can often yield equivalent intelligence to feed an ever-growing set of interested consumers. For criminals, the use of e-commerce systems and vulnerable payment mechanisms provides an avenue for rapid monetization and prosperity. Activists or hacktivists as they present themselves on the Internet are able to use electronic mediums to disseminate messaging from banal greets to truly meaningful causes that impact people’s lives across the globe.

Since May of 2014, the Chinese government has been amassing what can only be described as the “Facebook for human intelligence targeting” from the databases lifted from some of our most fundamental and essential systems. Why would anyone want healthcare records? If you take a step back, these records are part of a bigger picture, used in concert with the personnel records of US government workers and any other databases that have been stolen over the years. The beneficiary of that data can build an interesting picture detailing the confidential history, preferences, behavioral patterns, and more, of millions of potential intelligence targets.

The point that most people miss is that “cyber” data doesn’t just get used for cyber attacks, or cyber bullying, or cyber theft. The People’s Republic of China doesn’t only conduct network-based espionage, they are a major government on the world stage. They have human intelligence collectors whose job is to identify people with access to interesting or useful information and to collect that information. MICE is a common acronym we use in the information security industry — Money, Ideology, Compromise, and Ego – a simple set of motivations that can be used to entice or coerce a target to provide continued or temporary access to data.

Using stolen healthcare data, these human collectors can identify someone with access to sensitive information who unfortunately has a sick relative. As the healthcare bills pile up and they become increasingly despondent to help their sick relative get the medical treatment they need, an opening begins to emerge. The human collector, if they are able to identify this opening, can approach the target and begin to sow the seed for access, a simple trade of money for information, information that may seem insignificant to the target, but in aggregate across many different sources becomes quite valuable.

[Learn more from Adam about how to consume, operationalize and integrate threat intel during his training session on the fundamentals of intelligence-driven security, Black Hat 2015 Las Vegas August 1-2 & 3-4.]

It has been said that the network defender must be right 100 percent of the time, while the attacker need only be lucky once. The asymmetry of this is terrifying! Your network defenders should be in front of 10 monitors with an intravenous drip of caffeine and sugar twitching at every packet surging across your enterprise. The reality is that this is true, but we have systems and tools to help deter and detect these attackers.

These tools out of the box, while capable, don’t necessarily have all the smarts they need to root out these attackers:  these tools need intelligence. Intelligence-driven security means learning from previous attacks whether successful or not, and incorporating what you have learned into your defense posture. The military, in dealing with asymmetry encountered in Latin America in the 1980’s pioneered a process for incorporating intelligence into their targeting processes that has been continuously improved upon in the past 10 years.

This process involves taking the intelligence gleaned from every action, operation, or encounter and feeding it into the next operation to rapidly adapt to the changing environment. This same process introduced into security operations, what I call intelligence-driven security, can drive the cost of protecting the enterprise down, while simultaneously allowing the Security Operations Center (SOC) to have meaningful conversations with the business owners, the C-Suite, and the Board. Enterprise security isn’t just about blocking malware anymore, it’s about protecting the business and against dedicated and sophisticated threat actors.

Adam Meyers has over a decade of experience within the information security industry. He has authored numerous papers that have appeared at peer reviewed industry venues and has received awards for his dedication to the field. At CrowdStrike, Adam serves as the VP of Intelligence. Within this role it is Adam’s responsibility to oversee all of CrowdStrike’s intelligence gathering and cyber-adversarial monitoring activities. Adam’s Global Intelligence Team supports both the Product and Services divisions at CrowdStrike and Adam manages these endeavors and expectations. Prior to joining CrowdStrike, Adam was the Director of Cyber Security Intelligence with the National Products and Offerings Division of SRA International.

[Dark Reading]