This is another quick post going over the process to acquire memory from a Linux system, but instead of using LiME, I’m going to use AVML which stands for Acquire Volatile Memory for Linux, and could be found here. The tool has been developed by Brian Casewell for Microsoft and is a “userland volatile memory acquisition tool”.
AVML tries to acquire memory from the following memory sources:
The installation is straight forward and well documented on the Github page. I used the build on Ubuntu, which is really just “copy & paste” no super power required there, thanks to Brian! One note is there are two builds, one will provide an upload feature to upload the images to Azure and the other build without that. The size is really small, mine with full features was 5.5MB. After finishing the build you will find the binary (at least on my system) under:
When creating a forensic image, I also create a list of files and directories within that image, as seen in Figure 1, just for further checking and verification purposes. So, as usual, was doing the image to share and I noticed the following:
Figure 1: List of files found in a Forensic Image
One of my current students asked if using Stealth Alternate Data Streams (ADS), could bypass AVs? Therefore, I wanted to prove that for the student by doing a simple experiment. What was done is the following:
1. Turned off Windows Defender on my Windows System (used for testing)
2. Created a malicious reverse shell (reverse meterpreter) and copied it over to my Windows system. It was named rev.exe.
Contents of the directory I copied the rev.exe to:
3. Created a reverse shell listener (multi-handler) on my attacking system (Kali) and was waiting for the victim machine to connect back to it.
4. Used the commands we know to hide the reverse shell named “rev.exe” in LPT1.txt and then checked the contents of the temp directory (location of files) using FTK Imager Continue reading →
I know it seems that the zone has been abandoned for a year, and that is why I didn’t want the year to end without posting anything. Anyway, this presentation has been covered in Feb-2016, and thought why not share it with the DFIR community, maybe it will be useful to someone out there.