Lecture Schedule

• [Aug 30: Lecture 1] Lecturer: Jonathan Katz
  Private-key encryption; CBC-mode encryption; plaintext padding
  
  Homework (ungraded): implement CBC-mode encryption and decryption (using AES). Your program should read the plaintext from an ASCII text file, and output the ciphertext (in hex) to another file. You may assume the plaintext has length a multiple of 128 bits. You may also hardcode an IV; you do not have to worry about generating a (pseudo)random IV.
  
  
• [Sept 4: Lecture 2] Lecturer: Jonathan Katz
  In-class lab: Implementing padding-oracle attacks. Useful files are here.
  
  References (optional):
  • Wikipedia article on padding-oracle attacks
  • Security Flaws Induced by CBC Padding Applications to SSL, IPSEC, WTLS, ..., by Vaudenay
  • Padding Oracle Attacks on the ISO CBC Mode Encryption Standard, by Paterson and Yau
  • Cryptography in the Web: The Case of Cryptographic Design Flaws in ASP.NET, by Duong and Rizzo (video here)
  • Practical Padding Oracle Attacks, by Rizzo and Duong
  • Efficient Padding Oracle Attacks on Cryptographic Hardware, by Bardou et al.
  
  Homework (graded): Complete your implementation of the padding-oracle attack, and submit the program along with a Makefile. (You can assume your program will be compiled in a directory containing all the files linked from the class website.) Name your program attack.c, and the executable attack. Your program should take two command-line arguments, the first specifying an input file (containing a ciphertext, encrypted under some unknown key, formatted as in challenge-ciphertext.txt) and the second specifying an output file. Your program may call a function cbcdec which implements a "padding oracle" for the same unknown key, and should output (in hex) the resulting plaintext to the specified output file.
  
  You should not exploit buffer-overflow or similar attacks in the padding-oracle code, or attempt to extract the key from the compiled padding oracle. You may assume the challenge ciphertext is exactly 3 blocks long (as is challenge-ciphertext.txt), for simplicity.
  
  Submit your homework via the submit server
  
  Also: please install BackTrack 5 (any release) on your laptop in advance of lecture 4, and bring your laptop to class that day. (Note: the easiest is just to run BackTrack within a VM on your machine.)
  
  
• [Sept 6: Lecture 3] Lecturer: Jonathan Katz
  Wireless security; attacking WEP
  
  References:
  • Intercepting Mobile Communications: The Insecurity of 802.11, by Borisov et al.
  • (optional) More on the Fluhrer-Mantin-Shamir attack (their paper is available as well)
  
  
• [Sept 11: Lecture 4] Lecturer: Jonathan Katz
  In-class lab: Wireless security; attacking WEP
  
  
• [Sept 13: Lecture 5] Lecturer: Mike Hicks
  Introduction to the web; web-app penetration testing (slides)
  
  References
  • W3Schools tutorials on
    • HTML
    • Javascript
    • SQL
    • PHP
    • ... and many more
  • Aspect Security on-line trainings: an indexed PDF will take to all the ones available to us. You can use the account information you received in e-mail to access these. Relevant modules include 211-216, 411-412 (among others). Many thanks to Aspect Security for giving us free access to this on-line material!
  • Zap proxy on-line videos for setup and basic use, and the wiki documentation
  • BadStore manual, which contains links to useful penetration testing materials
  
  
• [Sept 18: Lecture 6] Lecturer: Mike Hicks
  In-class lab: Penetration testing on Bad Store
  
  References
  • SQL injection cheat sheet
  
  
• [Sept 20: Lecture 7] Lecturer: Mike Hicks
  Static analysis for finding vulnerabilities
  Lecture slides (3 slide decks): PQL; LAPSE; pointer analysis (bug in the last slide deck corrected 10/15/12, 9:30am)
  
  Required reading:
  • LAPSE: A Security Auditing Tool for Java (poster)
  • Finding Security Errors in Java Applications using Lightweight Static Analysis
  • Finding Security Vulnerabilities in Java Applications with Static Analysis
  References:
  • A few billion lines of code later: using static analysis to find bugs in the real world (experiences of the Coverity static analysis company)
  • Parameterized Object Sensitivity for Points-to and Side-Effect Analyses for Java (primer on points-to analysis)
  • Finding Application Errors and Security Flaws Using PQL: a Program Query Language (more on PQL)
  
  
• [Sept 25: Lecture 8] Lecturer: Mike Hicks
  In-class lab: Finding security vulnerabilities using LAPSE
  
  References
  • MITRE's list of the top 25 most important software errors
  
  Homework (graded, due October 15, 9am): Complete the following labs by exploiting the vulnerabilities in WebGOAT:
  • Numeric SQL injection
  • String SQL injection
  • Reflected XSS attack
  • Bypass client-side Javascript validation
  You do not have to do the extra credit parts. You will need to use a web proxy to complete some of these (e.g., Zap). You are free to try using LAPSE to find the potential source of these vulnerabilities.
  
  Please submit a document describing how you completed the attacks. In particular, say what you typed into a form field, how you changed in Zap (if at all), and what the outcome was. Feel free to collect screenshots to explain. It's OK for you to view the solution videos if you get stuck.
  
  
• [Sept 27: Lecture 9] Lecturer: Jeff Foster
  Symbolic execution (slides, lecture material)
  
  References:
  • Godefroid, Klarlund, and Sen. DART: Directed Automated Random Testing, PLDI 2005.
  • Cadar, Dunbar, and Engler. KLEE: Unassisted and Automatic Generation of High-Coverage Tests for Complex Systems Programs
  
  
• [Oct 2: Lecture 10] Lecturer: Jeff Foster
  In-class lab: Using KLEE to find security vulnerabilities. (Lab notes: Symbolic Execution Lab)
  
  Homework (graded, Due Oct 25, 2012): Finish part 3 of the symbolic execution lab, and write up your experiences. In particular, discuss the following:
  • Were you able to reproduce each bug in Figure 7 of the paper?
  • What symbolic inputs did you use?
  • How long did KLEE take to find the bugs with the symbolic inputs you chose? What timeout did you use? (I suggest sticking to something between 1 minute and 1 hour.)
  • If you run KLEE multiple times with the same program and symbolic input, what is the distribution of the time KLEE takes to find the bug? Is there a high degree of variation?
  • Do the results vary if you choose different search strategies? (See KLEE Options.)
  Submit your homework as a pdf or plain text file by email to Prof. Foster.
  
  
• [Oct 4: Lecture 11] Lecturer: Jeff Foster
  Software vulnerabilities (slides: Software vulnerabilities and Return-oriented programming)
  
  References:
  • Shacham, The Geometry of Innocent Flesh on the Bone: Return-into-libc without Function Calls (on the x86)
  
  
  
• [Oct 9: Lecture 12] Lecturer: Jeff Foster
  In-class lab: Shellcode/buffer overflow lab
  There is no homework associated with the lab, but we recommend you complete parts 1-4 of the lab on your own if you do not finish them in class.
  
  
• [Oct 11: Lecture 13] Lecturer: Jonathan Katz
  Password cracking, rainbow tables.
  In-class lab: Crack the passwords from here
  
  Homework: See here
  
  
• [Oct 16: Lecture 14] Lecturer: Jonathan Katz
  Host exploitation, nmap, metasploit
  
  References:
  • Offensive Security -- Metasploit Unleashed
  • Metasploit Megaprimer
  
  
• [Oct 18: Midterm]
  Sample solution
  
  
• [Oct 23: Lecture 15] Lecturer: Michel Cukier
  
  
• [Oct 25: Lecture 16] Lecturer: Michel Cukier
  
  
• [Oct 30: Lecture 17] Lecturer: Michel Cukier
  
  
• [Nov 1: Lecture 18] Lecturer: Michel Cukier
  
  
• [Nov 6: Lecture 19] Lecturer: Michel Cukier
  
  
• [Nov 8: Lecture 20] Lecturer: Michel Cukier
  
  
• [Nov 13: Lecture 21] Lecturer: Michel Cukier
  
  
• [Nov 15: Lecture 22] Lecturer: Gang Qu
  
  
• [Nov 20: Lecture 23] Lecturer: Gang Qu
  
  
• [Nov 27: Lecture 24] Lecturer: Gang Qu
  
  
• [Nov 29: Lecture 25] Lecturer: Gang Qu
  
  
• [Dec 4: Lecture 26] Lecturer: Gang Qu
  
  
• [Dec 6: Lecture 27] Lecturer: Gang Qu
  
  
• [Dec 11: Lecture 28] Lecturer: Gang Qu
  
  

Web Accessibility