This class discusses stealth malware techniques used on Windows. Rootkits are a type of stealth malware that try to hide their presence, and this class shows the data structures they manipulate to achieve this.
Rootkits are a class of malware which are dedicated to hiding the attacker’s presence on a compromised system. This class will focus on understanding how rootkits work, and what tools can be used to help find them.
This will be a very hands-on class where we talk about specific techniques which rootkits use, and then do labs where we show how a proof of concept rootkit is able to hide things from a defender. Example techniques include
•Import Address Table (IAT) hooking
•System Call Table/System Service Descriptor Table (SSDT) hooking
•Interrupt Descriptor Table (IDT) hooking
•Direct Kernel Object Manipulation (DKOM)
•Kernel Object Hooking (KOH)
•IO Request Packet (IRP) filtering
•Hiding files/processes/open ports
•Compromising the Master Boot Record (MBR) to install a “bootkit”
The class will help the student learn which tools to use to look for rootkits on Windows systems, how to evaluate the breadth of a tool’s detection capabilities, and how to interpret tool results.
This class is structured so that students are given a homework to detect rootkits *before* they have taken the class. This homework is given in the context of the following scenario:
“You, being the only ‘security person’ in the area, have been called in to
examine a running Windows server because "it's acting funny." They don't
care that you like Mac/Linux/BSD/Plan9 better, you need to look at it! You
are solemnly informed that this is system is mission critical and can only
be rebooted if absolutely necessary. You must investigate whether any sort
of compromise has taken place on the system, with minimal impact to the
mission. What do you do? What DO you DO?”
The homework is then for the student to use any means at their disposal to write up answers to the following questions: “What malicious changes were made to the system?”, “What tools did you use to detect the changes?”, “How can you remove the changes?”. The students’ answers are then anonymized and shared with the rest of the class afterwards, so that they can see how others approached the problem, and learn from their techniques. The anonymization of the homework before distribution is important so that students know that even though they don’t know, and aren’t expected to know, anything about the area yet, their entry will not be judged by other students.