Security researchers observed recent Qbot attack campaigns using a new persistence mechanism that helps the banking Trojan avoid detection.
In early April, Cisco Talos observed a new Qbot campaign that infected users’ machines with a dropper. The campaign used the infected machine to create a scheduled task that executed a JavaScript downloader. This asset, in turn, made a request from one of several hijacked domains.
Specifically, the downloader requested the uniform resource identifier (URI) /datacollectionservice[.]php3 from the domains, which were XOR-encrypted at the beginning of the JavaScript. A successful communication attempt yielded obfuscated data that the campaign saved in two files: the first 1,000 characters in (randalpha)_1.zzz and the remainder in (randalpha)_2.zzz.
At that point, the campaign created a scheduled task designed to execute a batch file. This process used the two .zzz files to assemble a Qbot executable before deleting them. Finally, the campaign ran the malware payload, enabling it to target financial information on the infected machine.
Tracing the Attack Trail of Qbot
Qbot has gotten up to all kinds of trouble over the past few years. Back in 2017, IBM X-Force observed a campaign in which the malware (also known as Qakbot) locked hundreds of thousands of Active Directory users out of their company’s domain, preventing them from accessing their employer’s servers or network assets.
Fast-forward to 2019: In March, Varonis spotted an operation leveraging a new variant of the malware that compromised and took over thousands of victims around the world. That same month, the SANS Internet Storm Center (ISC) discovered a malspam campaign in which Emotet served up Qbot as its follow-up payload.
Use UEM and AI to Defend Against Sophisticated Malware
Security professionals can help their organizations defend against sophisticated malware like Qbot by using a unified endpoint management (UEM) solution to monitor how devices report to the environment and take the necessary precautions if anything appears to be malicious in nature. Organizations should also consider enlisting the help of artificial intelligence (AI) to help fill the defense gaps created by rule-based security tools.