Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Analytics //

Security Monitoring

8/12/2013
07:39 AM
50%
50%

Attackers' Toolbox Makes Malware Detection More Difficult

From virtual-machine detection to taking a 30-minute nap, the array of techniques used by attackers to stymie malware analysis is growing

Sometimes the simplest techniques can foil the complex systems created by security firms and large enterprises to detect malicious programs and files. Putting malware to sleep, waiting for a user to click, or looking for the hallmarks of a virtual machine can set off warning bells and cause a malicious program to cease running, making analysis difficult at best.

Click here for more of Dark Reading's Black Hat articles.

At the Black Hat security conference earlier this month, two researchers from digital defense firm FireEye showed off the menagerie of techniques developed by attackers to detect whether their programs were being run by analysts in a virtual machine or file-based sandboxes. Software security firms and defenders must do better to detect and allow the inspection of malicious programs that use such anti-analysis techniques, says Zheng Bu, senior director of security research for FireEye and a co-presenter of the session.

"File-based sandboxes alone are not effective in detecting malware," Bu says. "The tools are only as good as the person that uses it or the system that leverages the tools."

Since 2005, the number of variants of malware that security firms need to analyze and recognize with their software has skyrocketed. To cull the truly new malware from the known variants, antivirus firms monitor files for malicious behavior and then use automated analysis systems and cloud-distribution platforms to analyze the new threats and then pass the pattern files back down to the customers' anti-malware defenses.

If attackers efforts make automated analysis, which handles the vast majority of malware, more difficult, then they break the model, says Dean De Beer, chief technology officer for ThreatGRID, a malware-analysis service.

"If the attacker can force the workload back onto the human analyst, then they are succeeding because that means that the more malware that can be produced in that manner, the less threats a security firm can detect," he says.

The techniques identified by Bu and his colleague, Abhishek Singh, also of FireEye, include pausing execution for a specific period of time or waiting for human interaction, attempting to detect if the current system on which the malware is running is a virtual machine, and running on only certain types of systems with specific attributes to rule out virtual machine systems.

[Malware writers go low-tech in their latest attempt to escape detection, waiting for human input -- a mouse click -- before running their code. See Automated Malware Analysis Under Attack.]

The simplest techniques can be the best. Because analysis systems need to analyze malware quickly, for example, they typically look for malicious behavior in the first few minutes. That means that techniques that pause execution for a certain amount of time or that wait for a certain action -- such as a mouse click -- can be extremely effective.

Yet security firms have their own ways to deal with those techniques. Anything from hooking into a sleep function and, essentially, accelerating time, to assigning a higher level of suspicion to any program that appears to sleep following installation can help defeat the techniques, De Beer says.

Moreover, these techniques are not yet in wide use because they do require more advanced understanding of analysis techniques and programming, says Liam O'Murchu, manager of security response for Symantec's North American operations.

"There are only a few people that are really innovating, and there are only a few that are following what's out there," he says. "It's among those people where we really will see a large pickup of these types of things being used."

While there are examples of the technique being used to escape detection for some time, such as in the case of Trojan.Upclicker, it requires a lot of effort to continually defeat the latest defenses, he says.

Yet the same is true for defenders, FireEye's Bu says. Any type of sandbox requires an expert hand to be truly effective against modern malware, he says.

"It is a technology, it is a tool, and it's only as good as the system or people who use it," Bu says.

Have a comment on this story? Please click "Add Your Comment" below. If you'd like to contact Dark Reading's editors directly, send us a message. Veteran technology journalist of more than 20 years. Former research engineer. Written for more than two dozen publications, including CNET News.com, Dark Reading, MIT's Technology Review, Popular Science, and Wired News. Five awards for journalism, including Best Deadline ... View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Windows 10 Migration: Getting It Right
Kevin Alexandra, Principal Solutions Engineer at BeyondTrust,  5/15/2019
Artist Uses Malware in Installation
Dark Reading Staff 5/17/2019
Baltimore Ransomware Attack Takes Strange Twist
Kelly Jackson Higgins, Executive Editor at Dark Reading,  5/14/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win a Starbucks Card! Click Here
Latest Comment: This comment is waiting for review by our moderators.
Current Issue
Building and Managing an IT Security Operations Program
As cyber threats grow, many organizations are building security operations centers (SOCs) to improve their defenses. In this Tech Digest you will learn tips on how to get the most out of a SOC in your organization - and what to do if you can't afford to build one.
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2019-12198
PUBLISHED: 2019-05-20
In GoHttp through 2017-07-25, there is a stack-based buffer over-read via a long User-Agent header.
CVE-2019-12185
PUBLISHED: 2019-05-20
eLabFTW 1.8.5 is vulnerable to arbitrary file uploads via the /app/controllers/EntityController.php component. This may result in remote command execution. An attacker can use a user account to fully compromise the system using a POST request. This will allow for PHP files to be written to the web r...
CVE-2019-12184
PUBLISHED: 2019-05-19
There is XSS in browser/components/MarkdownPreview.js in BoostIO Boostnote 0.11.15 via a label named flowchart, sequence, gallery, or chart, as demonstrated by a crafted SRC attribute of an IFRAME element, a different vulnerability than CVE-2019-12136.
CVE-2019-12173
PUBLISHED: 2019-05-18
MacDown 0.7.1 (870) allows remote code execution via a file:\\\ URI, with a .app pathname, in the HREF attribute of an A element. This is different from CVE-2019-12138.
CVE-2019-12172
PUBLISHED: 2019-05-17
Typora 0.9.9.21.1 (1913) allows arbitrary code execution via a modified file: URL syntax in the HREF attribute of an AREA element, as demonstrated by file:\\\ on macOS or Linux, or file://C| on Windows. This is different from CVE-2019-12137.