NetWire RAT Steals Payment Card DataThreat actors used a remote access trojan with keylogging capabilities rather than traditional point-of-sale malware By: Incident Response Team
During an incident response engagement in September 2016, SecureWorks® incident response analysts observed payment card data being collected by a generic remote access trojan (RAT) rather than typical memory-scraping malware.
In many payment card data breaches, a point-of-sale (POS) system is infected with malware that searches for specific processes in memory known to store card data in plain text. The malware copies card data from the running processes, a technique known as memory scraping, to encoded files on disk. These files are then transmitted to a threat actor, often over commonly open ports 80 and 443 (HTTP and HTTPS). The threat actor sells the card data or uses it for fraudulent purchases.
In the September 2016 incident, SecureWorks analysts observed card data being collected by the NetWire RAT instead of traditional POS malware. NetWire has a built-in keylogger that can capture inputs from peripheral devices such as USB card readers. The attack methodology is very similar to traditional POS malware. A threat actor sends a phishing email with a malicious attachment to an employee working on a POS computer. If the employee opens the attachment, malware to harvest card data is downloaded or installed. Without proper security protections in place, these infections can remain undetected for months or years.
Keyloggers expose more than just card data; credentials for online accounts and applications such as email, property management systems (PMS), and Internet browsers are at risk. Other sensitive information typed by the user, including Social Security numbers, phone numbers, addresses, and birthdates, can also be compromised. Using a RAT with keylogging capabilities, a threat actor could gather necessary information to commit identify theft and further compromise an organization’s network. The generic NetWire RAT variant used in this incident did not contain specific capabilities to target POS systems.
The NetWire RAT variant observed by SecureWorks analysts is a .NET compiled binary. The binary file description is “TeamViewer 10” (see Table 1); however, this binary is not associated with the legitimate TeamViewer application.
|Filename||MD5 Hash||Compile Date||File Description||File Version|
|May 30, 2016
Table 1. NetWire RAT binary details.
Upon execution, the “Windows Folder.exe” file copies itself to C:\Users\<username>\AppData\Roaming and creates a Windows shortcut (LNK) file in the victim’s Startup directory as a persistence mechanism. The Microsoft Autoruns for Windows tool reveals the autostart mechanism that the NetWire RAT creates and uses (see Figure 1). This persistence mechanism ensures that NetWire launches automatically when the victim logs into the system.
Figure 1. NetWire persistence mechanism. (Source: SecureWorks)
The “Windows Folder.exe” executable spawns and injects code into the legitimate notepad.exe Windows process (see Figure 2). Process injection helps the malware avoid detection; however, review of active network connections show notepad.exe communicating to 185 . 35 . 138 . 227 over port 4588. The notepad.exe process should never have an active network connection.
Figure 2. Child process created by the NetWire RAT. (Source: SecureWorks)
NetWire logs keystrokes and peripheral inputs into encoded files in the C:\Users\<username>
Figure 3. Example of decoded keylogger output. (Source: SecureWorks)
Payment card data breaches can cause significant financial and reputational damages for an organization, and can lead to restrictions imposed by compliance bodies and loss of future business. Prevention and detection is critical to ensure that threats to customer data are prevented or detected. Traditional antivirus software and other systems that rely on low-level indicators do not effectively detect and block common and pervasive malware. Organizations should apply endpoint detection mechanisms that apply behavioral analysis and human intelligence to detect threat actor activity.