6.858 Fall 2012 Lab 1: Buffer overflows

Handed out:	Wednesday, September 5, 2012
Part 1 due:	Friday, September 14, 2012 (5:00pm)
All parts due:	Friday, September 21, 2012 (5:00pm)

Introduction

This lab will introduce you to buffer overflow vulnerabilities, in the context of a web server called zookws. The zookws web server is running a simple python web application, zoobar, where users transfer "zoobars" (credits) between each other. You will find buffer overflows in the zookws web server code, write exploits for the buffer overflows to inject code into the server, figure out how to bypass non-executable stack protection, and finally look for other potential problems in the web server implementation.

Exploiting buffer overflows requires precise control over the execution environment. A small change in the compiler, environment variables, or the way the program is executed can result in slightly different memory layout and code structure, thus requiring a different exploit. For this reason, this lab uses a VMware virtual machine to run the vulnerable web server code.

To start working on this lab assignment, you should download the VMware Player, which can run virtual machines on Linux and Windows systems. For Mac users, MIT has a site license for VMware Fusion. You can download VMware Fusion from this web site.

Once you have VMware installed on your machine, you should download the course VM image, and unpack it on your computer. This virtual machine contains an installation of Ubuntu 10.04 Linux, and the following accounts have been created inside the VM.

Username	Password	Description
`root`	6858	You can use the `root` account to install new software packages into the VM, if you find something missing, using apt-get install pkgname.
`httpd`	6858	The `httpd` account is used to execute the web server, and contains the source code you will need for this lab assignment, in `/home/httpd/lab`.

For Linux users, we've also tested running the course VM on KVM, which is built into the Linux kernel and should be much easier to get working than VMware. KVM should be available through your distribution, and is preinstalled on Athena cluster computers; on Debian or Ubuntu, try apt-get install qemu-kvm. Once installed, you should be able to run a command like kvm -m 512 -net nic -net user,hostfwd=tcp:127.0.0.1:2222-:22,hostfwd=tcp:127.0.0.1:8080-:8080 vm-6858.vmdk to run the VM and forward the relevant ports.

You can either log into the virtual machine using its console, or you can use ssh to log into the virtual machine over the (virtual) network. To determine the virtual machine's IP address, log in as root on the console and run /sbin/ifconfig eth0. (If using KVM with the command above, then ssh -p 2222 httpd@localhost should work.)

The files you will need for this and subsequent lab assignments in this course is distributed using the Git version control system. You can also use Git to keep track of any changes you make to the initial source code. Here's an overview of Git and the Git user's manual, which you may find useful.

The course Git repository is available at git://g.csail.mit.edu/6.858-lab-2012. To begin with, log into the VM using the httpd account and clone the source code for lab 1 as follows.

Before you proceed with this lab assignment, make sure you can compile the zookws web server:

In order to run the web server in a predictable fashion---so that its stack and memory layout is the same every time---you will use the clean-env.sh script. This is the same way in which we will run the web server during grading, so make sure all of your exploits work on this configuration!

The reference binaries of zookws are provided in bin.tar.gz, which we will use for grading. Make sure your exploits work on those binaries.

Now, make sure you can run the zookws web server and access the zoobar web application from a browser running on your machine, as follows:

The /sbin/ifconfig command will give you the virtual machine's IP address. In this particular example, you would want to open your browser and go to the URL http://172.16.91.143:8080/. (If you're using KVM with the command above, just access http://localhost:8080/ on your host.) If something doesn't seem to be working, try to figure out what went wrong, or contact the course staff, before proceeding further.

Part 1: Finding buffer overflows

In the first part of this lab assignment, you will find buffer overflows in the provided web server. Read Aleph One's article, Smashing the Stack for Fun and Profit, as well as this paper, to figure out how buffer overflows work.

Exercise 1. Study the web server's code, and find examples of code vulnerable to memory corruption through a buffer overflow. Write down a description of each vulnerability in the file /home/httpd/lab/bugs.txt; use the format described in that file. For each vulnerability, describe the buffer which may overflow, how you would structure the input to the web server (i.e., the HTTP request) to overflow the buffer, and whether the vulnerability can be prevented using stack canaries. Locate at least 5 different vulnerabilities.

You can use the command make check-bugs to check if your bugs.txt file matches the required format, although the command will not check whether the bugs you listed are actual bugs or whether your analysis of them is correct.

Now, you will start developing exploits to take advantage of the buffer overflows you have found above. We have provided template Python code for an exploit in /home/httpd/lab/exploit-template.py, which issues an HTTP request. The exploit template takes two arguments, the server name and port number, so you might run it as follows to issue a request to zookws running on localhost:

You are free to use this template, or write your own exploit code from scratch. Note, however, that if you choose to write your own exploit, the exploit must run correctly inside the provided virtual machine.

You find gdb useful in building your exploits. As zookws forks off many processes, it can be difficult to debug the correct one. The easiest way to do this is to run the web server ahead of time with clean-env.sh and then attaching gdb to an already-running process with the -p flag. For instance, to attach to zookd-exstack, start the server and, in another shell, run

When a process being debugged by gdb forks, by default gdb continues to debug the parent process and does not attach to the child. Since zookfs forks a child process to service each request, you may find it helpful to have gdb attach to the child on fork, using the command set follow-fork-mode child. We have added that command to /home/httpd/lab/.gdbinit, which will take effect if you start gdb in that directory.

Exercise 2. Pick two buffer overflows out of what you have found for later exercises (although you can change your mind later, if you find your choices are particularly difficult to exploit). The first must overwrite a return address on the stack, and the second must overwrite some other data structure that you will use to take over the control flow of the program.

Write exploits that trigger them. You do not need to inject code or do anything other than corrupt memory past the end of the buffer, at this point. Verify that your exploit actually corrupts memory, by either checking the last few lines of dmesg | tail, using gdb, or observing that the web server crashes.

Provide the code for the exploits in files called exploit-2a.py and exploit-2b.py, and indicate in answers.txt which buffer overflow each exploit triggers. If you believe some of the vulnerabilities you have identified in Exercise 1 cannot be exploited, choose a different vulnerability.

Submit your answers to the first part of the lab assignment by running make submit. Alternatively, run make handin and upload the resulting lab1-handin.tar.gz file to the submission web site.

Part 2: Code injection

In this part, you will use your buffer overflow exploits to inject code into the web server. The goal of the injected code will be to unlink (remove) a sensitive file on the server, namely /home/httpd/grades.txt. Use the *-exstack binaries, since they have an executable stack that makes it easier to inject code. The zookws web server should be started as follows.

We have provided Aleph One's shell code for you to use in /home/httpd/lab/shellcode.S, along with Makefile rules that produce /home/httpd/lab/shellcode.bin, a compiled version of the shell code, when you run make. Aleph One's exploit is intended to exploit setuid-root binaries, and thus it runs a shell. You will need to modify this shell code to instead unlink /home/httpd/grades.txt.

Exercise 3. Starting from one of your exploits from Exercise 2, construct an exploit that hijacks control flow of the web server and unlinks /home/httpd/grades.txt. Save this exploit in a file called exploit-3.py.

Explain in answers.txt whether or not the other buffer overflow vulnerabilities you found in Exercise 1 can be exploited in this manner.

Verify that your exploit works; you will need to re-create /home/httpd/grades.txt after each successful exploit run.

Suggestion: first focus on obtaining control of the program counter. Sketch out the stack layout that you expect the program to have at the point when you overflow the buffer, and use gdb to verify that your overflow data ends up where you expect it to. Step through the execution of the function to the return instruction to make sure you can control what address the program returns to. The next, stepi, info reg, and disassemble commands in gdb should prove helpful.

Once you can reliably hijack the control flow of the program, find a suitable address that will contain the code you want to execute, and focus on placing the correct code at that address---e.g. a derivative of Aleph One's shell code.

Note: SYS_unlink, the number of the unlink syscall, is 10 or '\n' (newline). Why does this complicate matters? How can you get around it?

The test either prints "PASS" or fails. We will grade your exploits in this way. If you use another name for the exploit script, change Makefile accordingly.

The standard C compiler used on Linux, gcc, implements a version of stack canaries (called SSP). You can explore whether GCC's version of stack canaries would or would not prevent a given vulnerability by using the SSP-enabled versions of the web server binaries (zookd-ssp and zookfs-ssp), by using the zook-ssp.conf config file when starting zookld.

Part 3: Return-to-libc attacks

Many modern operating systems mark the stack non-executable in an attempt to make it more difficult to exploit buffer overflows. In this part, you will explore how this protection mechanism can be circumvented. Run the web server configured with binaries that have a non-executable stack, as follows.

The key observation to exploiting buffer overflows with a non-executable stack is that you still control the program counter, after a RET instruction jumps to an address that you placed on the stack. Even though you cannot jump to the address of the overflowed buffer (it will not be executable), there's usually enough code in the vulnerable server's address space to perform the operation you want.

Thus, to bypass a non-executable stack, you need to first find the code you want to execute. This is often a function in the standard library, called libc, such as execl, system, or unlink. Then, you need to arrange for the stack to look like a call to that function with the desired arguments, such as system("/bin/sh"). Finally, you need to arrange for the RET instruction to jump to the function you found in the first step. This attack is often called a return-to-libc attack. This article contains a more detailed description of this style of attack.

Exercise 4. Starting from your two exploits in Exercise 2, construct two exploits that take advantage of those vulnerabilities to unlink /home/httpd/grades.txt when run on the binaries that have a non-executable stack. Name these new exploits exploit-4a.py and exploit-4b.py.

Although in principle you could use shellcode that's not located on the stack, for this exercise you should not inject any shellcode into the vulnerable process. You should use a return-to-libc (or at least a call-to-libc) attack where you vector control flow directly into code that existed before your attack.

In answers.txt, explain whether or not the other buffer overflow vulnerabilities you found in Exercise 1 can be exploited in this same manner.

The test either prints two "PASS" messages or fails. We will grade your exploits in this way. If you use other names for the exploit scripts, change Makefile accordingly.

Part 4: Fixing buffer overflows and other bugs

Now that you have figured out how to exploit buffer overflows, you will try to find other kinds of vulnerabilities in the same code. As with many real-world applications, the "security" of our web server is not well-defined. Thus, you will need to use your imagination to think of a plausible threat model and policy for the web server.

Exercise 5. Look through the source code and try to find more vulnerabilities that can allow an attacker to compromise the security of the web server. Describe the attacks you have found in answers.txt, along with an explanation of the limitations of the attack, what an attacker can accomplish, why it works, and how you might go about fixing or preventing it. You can ignore bugs in zoobar's code. They will be addressed in future labs.

One approach for finding vulnerabilities is to trace the flow of inputs controlled by the attacker through the server code. At each point that the attacker's input is used, consider all the possible values the attacker might have provided at that point, and what the attacker can achieve in that manner.

You should find at least two vulnerabilities for this exercise.

Finally, you will explore fixing some of the vulnerabilities you have found in this lab assignment.

Exercise 6. For each buffer overflow vulnerability you have found in Exercise 1, fix the web server's code to prevent the vulnerability in the first place. Do not rely on compile-time or runtime mechanisms such as stack canaries, removing -fno-stack-protector, baggy bounds checking, XFI, etc.

You are done! Submit your answers to the lab assignment by running make submit. Alternatively, run make handin and upload the resulting lab1-handin.tar.gz file to the submission web site.