How Cryptocurrency Mining Malware Evades Detection

How Cryptocurrency Mining Malware Evades Detection

The recent rise of cryptocurrency mining malware is driven not just because of its high-profit potential, but also because of its ability to remain undetected in a compromised system.

Early this month, Nova Scotia-based St. Francis Xavier University announced that it purposely shut down all its network systems in response to a cryptojacking attack, whereby attackers attempted to illicitly mine the cryptocurrency Bitcoin using the university’s collective computing power. 

The price of Bitcoin as of November 13, 2018 (8:30 AM GMT+7) is $6,370, way below the all-time high of nearly $20,000 in December 2017, but still way above the $317 price of Bitcoin way back in January 2015.

Mining Bitcoin, in order to be profitable, one needs to invest in a reasonable number of powerful computers and high electricity cost. Illicit cryptocurrency mining, also known as cryptojacking, hijacks the computing power of someone else’s without their consent to mine cryptocurrency such as Bitcoin.

Cryptocurrency Mining Malware Evasion Techniques

Many organizations continue with their usual IT operations without even realizing that the organization’s computers are illicitly used by cyberattackers for cryptocurrency mining.

Based on the combined data of several Cyber Threat Alliance (CTA)members, from 2017 to September 2018, illicit cryptocurrency mining increased by 459%.

Researchers at Trend Microrecently discovered a cryptocurrency mining malware that uses multiple techniques to evade detection. The cryptocurrency mining malware discovered by researchers at Trend Micro as Coinminer.Win32.MALXMR.TIAOODAM uses the following evasion techniques to make it harder for detection tools to discover it:

First, Coinminer.Win32.MALXMR.TIAOODAM evades detection as it arrives on the victim’s computer as Microsoft Windows Installer MSI file. Windows Installer is a software program that’s used for installing and uninstalling software. Using a real Windows component, researchers at Trend Micro said, makes the cryptocurrency mining malware looks less suspicious and potentially allows it to bypass certain security filters.

Second, Coinminer.Win32.MALXMR.TIAOODAM evades detection as it creates copies of the kernel file ntdll.dll and the Windows USER component user32.dll in %AppData%\Roaming\Microsoft\Windows\Template\FileZilla Server\{Random Numbers}. Researchers at Trend Micro theorized that this is done to prevent detection of the cryptocurrency mining malware’s application programming interface (API) – a set of programming instructions and standards for accessing a software application.

Third, Coinminer.Win32.MALXMR.TIAOODAM evades detection as it comes with a self-destruct mechanism, a feature that deletes every file under its installation directory and removes any trace of installation in the system.

Fourth, Coinminer.Win32.MALXMR.TIAOODAM evades detection as it comes with WiX, which stands for Windows Installer XML, a free software toolset for building Windows Installer packages from XML. Researchers at Trend Micro said that WiX is an added layer developed by the attackers to evade detection.

Initial infection of Coinminer.Win32.MALXMR.TIAOODAM could come from emails laden with malicious attachments and malicious URLs.

Early this year, attackers used multiple techniques in their cryptocurrency mining malware to evade detection, this time, in utilizing Tesla’s computing power to mine a cryptocurrency.

In February this year, RedLockreported that Tesla’s Amazon Web Services (AWS) cloud account was compromised by attackers through Kubernetes – an open source software used to deploy and manage cloud-based applications and resources. RedLock said that Tesla's Kubernetes wasn’t password protected, enabling attackers to execute from the Kubernetes a cryptomining command.

According to RedLock, the cryptocurrency mining malware used in the Tesla cryptojacking incident wasn’t initially detected as the attackers used the following evasion techniques:

First, the attackers configured the cryptomining malware to keep the CPU usage low to evade detection. CPU, which stands for central processing unit, is the component that performs most of the processing inside a computer. Unusual high CPU usage is a sign that computers are being used for cryptocurrency mining.

Illicit Bitcoin mining is typically discovered through unusual high CPU usage. It’s, however, not enough to check the CPU usage as guage for cryptocurrency mining as other cryptocurrencies have less CPU usage.

Aside from the illicit mining of Bitcoin, attackers are also drawn in hijacking the computing power of others to mine the cryptocurrency Monero (valued $105 as of November 13, 2018 at 10 AM GMT+7) as this cryptocurrency uses less computing power and anonymous compared to Bitcoin.

Second, attackers in the Tesla cryptojacking incident didn’t use a well-known public “mining pool” – a means by which cyrptominers share their computing power over a network and split the reward equally. By using an “unlisted” or semi-public mining pool, the attackers in the Tesla cyrptojacking incident was initially unnoticed.

Third, in addition to using an unlisted mining pool, the attackers in the Tesla cyrptojacking incident hid the true IP address of their mining pool server behind CloudFlare, a free content delivery network (CDN) service, making detection even more difficult.

Fourth, the attackers in the Tesla cryptojacking incident configured their cryptomining malware to listen on a non-standard port, making it difficult to detect unusual activity based on port traffic.

Prevention

Protecting your organization’s computers or network from cryptocurrency mining malware is important as this malware can damage your organization’s computers, negatively impact business operations and can lead to further cyberattacks.

Here are some measures to prevent attackers from using your organization’s computing power for cryptocurrency mining:

• Use email inspection tool that can detect malicious attachments and URLs.

• Monitor configurations. By monitoring configurations, exposed or unprotected software can be identified.

• Monitor traffic. By monitoring traffic, suspicious traffic can early on be identified.