Android Malware Uses Blog Posts as C&C

Newer and more complicated Android malware variants are expected to emerge, along with the rising number of malicious Android apps. A new backdoor that we were able to analyze proves just that. Malware targeting the Android platform are continuously improving in performance as well as using new techniques to thwart analysis and to avoid detection.

This Android malware, which Trend Micro detects as ANDROIDS_ANSERVER.A, arrives as an e-book reader app and can be downloaded from a third-party Chinese app store. It asks for the following permissions upon installation:

Based on the permissions requested alone, it is easy to see that this particular malware has a lot of capabilities. Once granted, the permissions can be used to execute the following:

• Access network settings
• Access the Internet
• Control the vibrate alert
• Disable key locks
• Make a call
• Read low-level log files
• Read and write contact details
• Restart apps
• Wake the device
• Write, read, receive, and send SMS

For more information on how cybercriminals utilize permissions in conducting malicious routines, check out our e-book, “When Android Apps Want More Than They Need.”

From our analysis, we found that this malware has two hardcoded C&C servers to which it connects in order to receive commands and to deliver payloads. The first server is just like the usual remote site to which the malware posts information to and gets commands from. The second C&C server, however, caught our attention more. This is a blog site with encrypted content, which based on our research, is the first time Android malware implemented this kind of technique to communicate.

Below is a diagram of how ANDROIDOS_ANSERVERBOT.A uses the blog site as a C&C server:

Further analysis of the blog content revealed six encrypted posts containing backup C&C server URLs:



In addition, 18 binaries have been uploaded to the blog from July 23 to just last September 26. It should also be noted that one of the updates is named _test, which suggests that this malware is still being further developed.



Decrypting the posts and analyzing the binaries, we found out that the files are just different versions of one file. Comparing them, one difference we found is that the newer versions had the capability to display notifications that attempt to trick users into approving the download of an update.



Another addition to later versions is the capability to terminate four security-related apps:

• com.qihoo360.mobilesafe
• com.tencent.qqpimsecure
• com.ijinshan.mguard
• com.lbe.security

The use of blog platforms in malware activities is not unheard of. In fact, early this year, a botnet was found using Twitter for issuing commands to infected systems. If anything, this recent adaptation of mobile malware is another sign of continued development and proliferation.