A recent malicious email campaign delivered an Ursnif Trojan variant. The variant is different, in that it uses...
mouse movements for decryption and evasion. How does this work?
As security researchers and vendors improve the security within their products, malicious actors are continually looking for ways to bypass them and continue their efforts. This cat and mouse game continues to play out, and is best seen in how malware authors are continually developing creative ways to create new attacks or workarounds. Many times, these techniques are very creative and, with a new variant of the Ursnif Trojan, we saw attackers use mouse movements to decrypt and evade sandbox detection.
Sandboxes are used to validate that downloaded files from the internet are safe to run on the endpoint. They're sent to the sandbox and executed on a virtual machine to determine their intended purpose. Since this can detect malware, attackers are continually looking for ways to bypass this security layer.
There have been multiple methods used in the past to detect sandboxes, such as searching for VMware registry keys, virtual adapters, low CPU and RAM, and doing nothing for hours to determine if a file is on a VM.
In this case, the malware would sit idle. This is also a way to avoid sandboxes, since the scans don't last hours, and users don't perform the malicious actions if they are tipped off to these variables. This would allow the files to enter your network where, like a Trojan horse, they'd wreak havoc.
The Ursnif Trojan's spin on sandbox detection is to use the previous and current mouse point locations to validate that it's not sitting in a sandbox. The technique, discovered by Forcepoint Security Labs, looks for the delta between these pointer locations and uses these variables to create a base seed that can assist with decryption.
The Ursnif Trojan goes through the base seeds to decipher the key, and once it matches the proper checksum, which can essentially take a brute force-like combination to achieve, the malware executes the remainder of the code. It does this because the D-value of the mouse movement is always zero, and it will never be able to decipher the proper decoded code at this starting point. Since this is the case, it will never execute within a sandboxed environment.
It's this type of creativity that enables malware authors to always find a unique way to pivot against today's latest technology. It's also a reason to never rely on a single layer to defend yourself against today's modern threats. By only placing your trust in sandboxing to prevent malware, you leave yourself open to creative attackers that will evade your best defenses.
As always, security needs to be performed in layers, and the Ursnif Trojan shows just how important a layered approach really is.
Ask the expert:
Want to ask Matt Pascucci a question about security? Submit your question now via email. (All questions are anonymous.)
Learn how to identify malicious emails before they are opened
Discover how to detect malware than can evade sandboxes
Read about what to do when you click on malicious links
Dig Deeper on Hacker tools and techniques: Underground hacking sites
Related Q&A from Matthew Pascucci
Troubleshooting VPN session timeout and lockout issues should focus first on isolating where the root of the problem lies -- be it the internet ... Continue Reading
What sets web roles and worker roles apart in Microsoft's Azure Cloud Services? Here's a look at how they are different. Continue Reading
Container security continues to be a pressing issue as containers and hosts are being used more frequently. Learn how to keep your enterprise safe ... Continue Reading