Back in 440 BCE, a Spartan king named Demaratus secretly warned an ally of an imminent attack by writing a message on the wooden back of a wax tablet, and then applying a beeswax surface to hide it. Not only did this clever move help thwart the attack, but it’s an early example of what would later be called steganography: the practice of concealing a message within another message. Well, fast forward about two and a half millennia -- and swap a world full of wax tablets to one full of digital tablets -- and cyber criminals have borrowed a tactic from the Ancient Spartan Royalty playbook by unleashing Stegoloader: a nasty new malware that hides within a PNG image file, so that it can covertly infect endpoints and raid businesses of their confidential data.
While Stegoloader first reared its malicious head in 2013, it wasn’t until last year things really started to pick up with successful attacks on organizations in several industries, including healthcare, financial, education and manufacturing. While Stegoloader has the potential to spread through various vectors, the only verified method so far is via a software piracy website, where it is bundled with key generation programs. Because of this, researchers from Dell SecureWorks who recently analyzed Stegoloader say that it’s more similar to “mass market” commodity malware than it is to threats used in targeted attacks. Yet with that being said, they also add there are probably several undiscovered Stegoloader modules, and that the malware’s potent information-stealing capabilities (described below) would certainly fit a campaign against a targeted victim.
Stegoloader: Stage One
Stegoloader carries out its attack in two stages. The first is through a deployment module, which checks to see if it’s being analyzed by monitoring mouse cursor movements. If the malware detects a lack of movement or too much movement, it terminates. It also terminates if it spots any processes associated with popular security tools. Once it confirms that the coast is clear, the deployment module downloads a seemingly harmless PNG image file from a legitimate website, and uses the gdiplus library to decompress bits from each pixel. The data stream is then decrypted via a hard-coded key and the RC4 algorithm.
Stegoloader: Stage Two
Having safely deposited Stegoloader on its victim’s endpoint, the deployment module terminates and gives way to the main module. This is when the malware receives encrypted instructions via its Command and Control (C&C) server, which could be to gather information on system details, web browser history, installed applications, and so on. Based on this intelligence, bad guys can tell Stegoloader to install additional modules to inflict additional damage, such as to steal Interactive Disassembler (IDA) instances, determine its host’s location, install Pony Loader malware to steal application passwords, and install Vundu malware to display ads and download even more threats. And since sometimes the best offense is a good defense, Stegoloader can also be ordered to stop execution, execute shellcode, or go to sleep.
To make things even trickier, while the attack is unfolding traditional signature-based products aren’t sounding any alarm bells, because the PNG image and the decrypted code reside in the endpoint’s memory -- rather than being saved to the disk. In addition, static malware analysis is slowed down because of the dynamic construction of strings in the binary.Read more:Serious Business: Cyber Security and Brand Survival
Like most advanced malware, stopping Stegoloader is a “good news, bad news” scenario. First the bad news. There’s no bulletproof way to stop Stegoloader attacks from slipping past network defense systems. This doesn’t mean that the floodgates are open. It just means that businesses that haven’t yet had a brush with Stegoloader should brace for impact. It’s probably more a matter of when, not if. The good news, however, is there is a way to mitigate Stegoloader and make sure it does not spread throughout your organization.
How Container Security Solutions Stop Stegoloader
Containment security solutions (also known as Application Isolation) use a lightweight virtual container on the endpoint to isolate the application environment. Since the container is continuous, Stegoloader’s standard sandbox evasion techniques are not effective. As a result, if Stegoloader downloads and attempts to execute, it will be trapped in the virtual container. It will also be unable to access the network and spread to other endpoints and servers. Since the virtual container is wiped clean at least once a day, the multi-stage exploit would not be able to complete its mission and continue to exfiltrate data from the endpoint.
Adapting to a Worsening Threat Environment
Stegoloader isn’t the first – and certainly won’t be the last – malware family that uses digital steganography to carry out successful attacks. Containment solutions help organizations stay a step ahead of this innovative cyber attack approach, which is the only place they can be given the massive costs of a data breach.
Israel Levy is the CEO of BUFFERZONE, an advanced endpoint security company that protects enterprises from advanced threats, including zero-day, drive-by downloads, phishing scams and APTs.