Security Experts:

Qbot Infects Thousands in New Campaign

A recent distribution campaign resulted in thousands of machines being infected with the Qbot malware, Cylance security researchers warn.

Qbot, which is also known as Qakbot or Quakbot, has been around since 2009, but multiple layers of obfuscation, server-side polymorphism and periodic improvements allow it to remain a persistent threat.

The malware is known for its credential stealing functionality and the ability to spread through network shares, but also includes backdoor capabilities. For two weeks in February last year, the threat managed to ensnare over 50,000 computers worldwide into a botnet. In July, a SentinelOne report on the Furtim-related SFG malware tied Qbot to a fast-flux proxy-based network called Dark Cloud or Fluxxy.

What’s unclear regarding the newly observed Qbot outbreak is how the malware managed to infect such a large number of machines in a short period of time. Most probably, Cylance says, updated exploit kits helped with the distribution.

The core functionality of Qbot has remained fairly consistent over the years, and the polymorphic nature of the threat helped it evade detection. Focusing on this aspect allowed the researchers to discover how often the executable code is modified.

The same as with previously observed samples, the malware continues to configure a scheduled task to request updates, with one command set to run on a weekly basis. The payload received from the server is encrypted, and “the first 20 bytes serve as the RC4 key to decrypt the data,” the security researchers say.

By creating a script to send HTTP requests to each of the three URLs the malware itself receives updates from, the security researchers discovered that files with a unique hash would be supplied every 10 minutes. They also managed to collect a total of 140 unique files supplied by the server over a period of 24 hours.

“All 141 downloaded files were 32-bit Windows executables. Across the 141 files, all have unique compile timestamps, and the earliest one occurred on May 15, 2017,” the researchers say.

Analyzing two files with the same import hash but with different file hashes revealed that, of nine PE sections each of them contains (.text, .code, .rdata, .data, .CRT, .exp, .code (again), .rsc, and .reloc), all section hashes match except those for .text, .rdata, and .data.

Different .text sections could reveal a change in executable code, and initial analysis revealed that all 27 functions identified matched 100%. Following deobfuscation, however, the security researchers discovered that nine functions had received some changes, albeit the overall Qbot functionality remained the same.

“Qakbot continues to be a significant threat due to its credential collection capabilities and polymorphic features. Unhindered, this malware family can rapidly propagate through network shares and create an enterprise-wide incident,” Cylance notes.

In an emailed comment to SecurityWeek, Michael Patterson, CEO of Plixer, pointed out that there is no shortage of vulnerabilities that malicious applications can exploit and that threats will continue to evolve. Thus, defense systems should adapt to ensure more efficient detection.

“Qakbot’s dynamic polymorphic abilities make it particularly evasive to antivirus systems. This means the virus can more easily maintain its presence without being detected," Patterson said. "It does however need to communicate on the network in order to carry out its dastardly deeds. In the case of Qakbot, it uses HTTPS to communicate with command-and-control (C&C) and FTP to upload stolen data. Network Traffic Analytics can be leveraged against flow data to watch for this one-two punch combination especially where odd FQDNs patterns are detected.” 

Related: Improved Qbot Worm Targets Public Institutions

view counter