
The Nitol botnet was a massive and notorious cyber threat that made headlines in 2013. It's estimated to have infected over 12 million devices worldwide.
The botnet was discovered by Microsoft in 2013, and it was found to be using a unique tactic to evade detection.
Nitol was designed to be a highly adaptable and resilient botnet, capable of spreading through various means, including infected apps and compromised websites.
Microsoft Action
Microsoft took action against the Nitol botnet, blocking over 500 variants of the malware that were secretly installed in counterfeit versions of Windows.
The company obtained a court order to take control of the domain 3322.org and over 70,000 subdomains that hosted the malware, which had been active since 2008.
This operation is the second of its kind in six months, following a similar action against the Zeus botnet in March.
Microsoft's investigation began in August 2011, when they analyzed the security of the supply chain and discovered that counterfeit PCs sold on the black market in China were infected with malware.
The company purchased 20 PCs in China and found that 20% of them were infected with the Nitol malware, which could spread to others through shared files and devices.
Over 4,000 Windows machines were found to be infected with Nitol, including some in Fairfax, Virginia.
Microsoft was able to block the communication between infected machines and the botnet's command and control servers, effectively crippling the malware's ability to spread.
The company also obtained a restraining order against Peng Yong and other suspects, allowing them to take control of the 3322.org domain and block the malware's activities.
Analysis and Detection
Analyzing the Nitol botnet requires a deep understanding of its evasion techniques. The malware uses a novel technique to bypass sandbox analysis, making it challenging for automated analysis technologies to detect.
One method the malware uses is password protection. By protecting the malicious macro-based documents with a password, the malware can bypass sandbox analysis entirely. This is because the process to enter the password is a complex user interaction event that automated analysis technologies struggle to emulate.
If this caught your attention, see: Spectrum Internet Password
The other method involves delayed execution. The malware uses the "ping" utility to delay the execution of the malicious code, making it difficult for sandboxes to analyze. This is because the ping process can take a long time to complete, often exceeding the time threshold set by sandboxes.
To detect this malware, security professionals should look for the following indicators:
- Password protection: Check for password prompts when opening documents.
- Delayed execution: Look for ping commands or other methods used to stall execution.
Command Servers
Command Servers are often used to issue commands to compromised systems, and they can be identified by their domain names.
We found two command server domain names: zwx5060.3322.org and ylddos.3322.org.
The domain name ylddos.3322.org is associated with an American IP address, 67.215.65.132, which is registered to opendns.com.
Most hardcoded domain names are resolved by Chinese dynamic DNS services.
Curious to learn more? Check out: Domain Name System Blocklist
Bypass Malware Sandbox Analysis
Malware authors have been trying to evade sandbox analysis for a while, and this strain of malware uses a novel technique to do so.
One way they bypass sandbox analysis is by using password protection, making it difficult for automated analysis technologies to emulate the user interaction event required to enter the password.
Password protection is a simple yet effective way to bypass sandbox analysis, as shown in Figure 1, which displays a password prompt while opening a malicious macro-based document.
Delayed execution is another technique used by this malware to evade analysis. They use the "ping" utility to delay the execution, invoking the command "ping 8.8.8.8 -n 250" and waiting for the ping process to complete, which can take up to 5 minutes.
This delay is often enough to bypass sandbox analysis, as most sandboxes are configured with a smaller time threshold for executing samples.
The use of the ping command to delay execution is a novel technique, and it's a good reminder that malware authors are constantly evolving their tactics to evade detection.
Here are the two methods used by this malware to bypass sandbox analysis:
- Password Protection: The documents are password protected, making it difficult for automated analysis technologies to emulate the user interaction event required to enter the password.
- Delayed Execution: The malware uses the "ping" utility to delay execution, which can take up to 5 minutes to complete.
Malicious Document Execution Analysis
The code checks for active debugging using GetTickCount, which is a common technique used by malware to evade detection.
This code then searches for the default browser in the http://shell/open//command registry, which is a specific location where browser settings are stored.
After finding the default browser, the code creates a browser process in suspended mode, essentially freezing the browser's activity.
The code then unmasks and writes the browser process memory with a UPX compressed file, which is a type of compression used to pack malware into a small executable.
Netskope Active Threat Protection detects these malicious macro-based documents as W97m.Downloader, providing a specific name for this type of malware.
Customers who have deployed the Netskope Active Platform and Netskope Introspection can set the respective malware & threat detection policies to detect and remediate against this malware.
Analysis of VBScript
The VBScript code in the malicious document drops and executes a VBScript file, which is responsible for downloading and executing the second stage payload.
The VBScript file was obfuscated, but the debug feature of the VBScript editor was used to deobfuscate the file content, revealing its true intentions.
The VBScript file launches the "ping" utility to delay execution and then connects to the "http://doktrine.fr/mg.txt" domain to download the second stage payload.
The payload in the ".qsb" file is xor encoded, which the VBScript file decodes and writes to another file with the ".fyn" extension before executing it.
The ".fyn" file is a Windows-executable file that executes in the code section before jumping to a marked region, indicating the execution of unpacked code.
The code in the ".fyn" file checks if the execution environment is VMware using process enumeration, suggesting an attempt to evade detection.
The VBScript file's behavior is a clever attempt to evade detection, but its use of a "ping" utility to delay execution is a telltale sign of malicious activity.
Report and Details
The Nitol botnet was a massive network of compromised computers that was dismantled in 2013.
It's estimated that the botnet had over 3.2 million infected devices.
The malware behind the botnet was a type of Android malware that infected devices through third-party app stores.
The malware was designed to steal sensitive information and send it to the botnet's command and control servers.
The botnet's operators were using the stolen data to make money through various schemes.
The Nitol botnet was eventually taken down by Microsoft in collaboration with the authorities in China.
Featured Images: pexels.com

