Idea Transcript
Introduction to Computer Security Course Introduction Pavel Laskov Wilhelm Schickard Institute for Computer Science
Computer security in headlines
Motivation for security abuse
Motivation for security abuse
Intelligence and military use (5th century BC – 1980’s) “security by obscurity”, crypto-wars
Motivation for security abuse
Intelligence and military use (5th century BC – 1980’s) “security by obscurity”, crypto-wars
Hacker spirit, fun and fame (1980’s – 2000’s) Pwnie, CCC, CTF
Motivation for security abuse
Intelligence and military use (5th century BC – 1980’s) “security by obscurity”, crypto-wars
Hacker spirit, fun and fame (1980’s – 2000’s) Pwnie, CCC, CTF
Cybercrime, monetary gain (2000’s – currently) Phishing, botnets, spam
Motivation for security abuse
Intelligence and military use (5th century BC – 1980’s) “security by obscurity”, crypto-wars
Hacker spirit, fun and fame (1980’s – 2000’s) Pwnie, CCC, CTF
Cybercrime, monetary gain (2000’s – currently) Phishing, botnets, spam
Political goals, cyberconflict (2007 – currently) Attack on Estonia, Russian-Georgian conflict, Stuxnet
Motivation for security abuse
Intelligence and military use (5th century BC – 1980’s) “security by obscurity”, crypto-wars
Hacker spirit, fun and fame (1980’s – 2000’s) Pwnie, CCC, CTF
Cybercrime, monetary gain (2000’s – currently) Phishing, botnets, spam
Political goals, cyberconflict (2007 – currently) Attack on Estonia, Russian-Georgian conflict, Stuxnet
“Hacktivism” (2011 – currently)
» Internet als Risikofaktor? » Omnipräsenz von Computer in und numbers Angriffen,security Würmern Viren » Massive Schäden bei Unternehmen und Bürgern » Zunehmende Kriminalisierung von Schadsoftware Schadcode pro observed Jahr (Symantec) NewNeuer malicious code samples (Symantec) 3.000.000 2.250.000 1.500.000 750.000 0
2002 2003 2004 2005 2006 2007 2008 2009
Technische Universität B
Why are computer systems insecure?
Why are computer systems insecure?
Growing complexity of computer systems large number of components, complex interaction
Why are computer systems insecure?
Growing complexity of computer systems large number of components, complex interaction
High competition short “time-to-market”, high ROI
Why are computer systems insecure?
Growing complexity of computer systems large number of components, complex interaction
High competition short “time-to-market”, high ROI
Leveraging of risks through high connectivity worm outbreaks, botnets
Why are computer systems insecure?
Growing complexity of computer systems large number of components, complex interaction
High competition short “time-to-market”, high ROI
Leveraging of risks through high connectivity worm outbreaks, botnets
Slow incident response “incident hiding”, manual handling
Why are computer systems insecure?
Growing complexity of computer systems large number of components, complex interaction
High competition short “time-to-market”, high ROI
Leveraging of risks through high connectivity worm outbreaks, botnets
Slow incident response “incident hiding”, manual handling
Human error
Why are computer systems insecure?
Growing complexity of computer systems large number of components, complex interaction
High competition short “time-to-market”, high ROI
Leveraging of risks through high connectivity worm outbreaks, botnets
Slow incident response “incident hiding”, manual handling
Human error
What can go wrong will go wrong!
Human error: a case study
Human error: a case study
Human error: a case study
Human error: a case study
Human error: a case study
Human error: a case study
Human error: lessons learned
Users make errors elaborate social engineering design time pressure
Significant monetary motivation Business efficiency via Internet
Security instruments
Security instruments
Security instruments
Reaction
Detection
Prevention
Prevention instruments
Goal: enforce certain operational policies. Examples: Encrypt messages during transmission over public networks. Require user authentication for certain services. Control access to different resources.
Limitations: Not always applicable, e.g. in open systems such as web services. Strong assumptions, can be circumvented.
Detection instruments
Goal: detect violations of security policies. Examples: Antivirus scanners: detection of malicious code or behavior. Intrusion detection systems: detection of attacks in network traffic. Detection of malicious websites.
Limitations: Significant latency in decisions. Significant workload: a detection system without an operator is useless.
Reaction instruments
Goals: Understand the root causes of successful attacks. Update prevention mechanisms. Real-time response, autonomous decisions.
Examples: Computer forensics: investigation of infected systems. Malware collection and analysis. Intelligent firewalls.
Limitations: Even larger latency, “post-mortem” operation. Significant risk of real-time response.
What will you learn?
Findamental concepts of computer security dry, but important!
Basic security goals and mechanisms authentication, access control, encryption, etc.
Practical security instruments Windows and Linux security
Further selected topics network security software security web application and browser security
Coarse administration
Lectures: Wed, 14:00 (ct) – 16:00, A301
Formalities: Credit hours (diploma): 2 SWS (lectures) + 1 SWS (exercises) Credit points (B.Sc.): 3 LP (lectures) + 1 LP (exercises)
Exams and grading: diploma: oral exam by appointment, graded certificate for exercises B.Sc: written exam at the end of semester, 30% of the final grade from exercises
Office hours: by appointment Course web page: http://www.cogsys.cs.uni-tuebingen.de/lehre/ws12/it sicherheit.html
Homework assignments
Meetings: Thu, 14:00 (ct) – 16:00, F122, on selected dates First meeting: 08.11
3 written homework assignments 2 lab meetings and practical assignments ˇ Teaching assistant: Nedim Srndi´ c Evaluation and grade: diploma: a grade reflects the percentage of points acquired. B.Sc.: a grade contributes 30% to the final grade.
Bibliography
Dieter Gollmann. Computer Security. 3rd edition. Wiley & Sons, 2010. Ross Anderson. Security Engineering. Wiley & Sons, 2001. Bruce Schneier. Secrets and Lies: Digital Security in a Networked World. Wiley & Sons, 2004. (fun to read)
A typical web application
Legitimate user
Internet Browser Webserver
Attack: interception of communication
Legitimate user
Internet Browser Webserver Credit card number
Attacker
Security goal: confidentiality
Prevention of unauthorized reading of data Prevention of unauthorized learning of information Potential abuse scenarios: Discovery of confidential information (e.g. details of a business contract) Discovery of authentication credentials (e.g. password sniffing)
Enforcement intruments: Symmetric or asymmetric cryptography Anonymization techniques
A typical web application
Legitimate user
Internet Browser Webserver
Attack: identity spoofing
Legitimate user
Internet Browser Webserver
Attacker Fake webserver
Security goal: authentication
Verification of an identity of a person or a computer Prerequisite for access control Authentication methods: Shared secrets (e.g. password or PIN) Ticket systems (identity cards, digital certificates) Challenge-response techniques Biometric techniques Human authentication: CAPTCHA’s, Turing test, etc.
Authentication risks: identity theft
A typical web application
Legitimate user
Internet Browser Webserver
Attack: injection of malicious code
Legitimate user
Internet Malware Browser Webserver
XSS attack
Attacker
Security goal: integrity
Prevention of malicious tampering of data Potential abuse scenarios: Fraudulent modification of data (e.g. 100,000¤ instead of 100¤ in an online transaction request) Injection of malicious code in downloaded software Evading detection by modification of a compromized operating system
Enforcement instruments: Integrity checking using cryptographic hash functions
A typical web application
Legitimate user
Internet Browser Webserver
Attack: server overload
Server overloaded Legitimate user
Internet Browser
Security goal: availability
A joint objective of security and dependability May be caused by attacks as well as natural phenomena, such as design errors or flash crouds. Enforcement instruments: Detection of DoS-attacks Router and firewall reconfiguration Service redundancy Virtualization
A typical web application
Legitimate user
Internet Browser Webserver
Attack: transaction denial by a user
Legitimate user
Internet Browser Webserver Transaction cancelled Credit card charded
Security goal: accountability
An audit trail of security-related events A key instrument of detection/response A stronger form of accountability is non-repudiation: unforgeable evidence that a certain action occurred. Similar attacks as integrity Enforcement instruments: Integrity checks Read-only audit Digital certificates and trusted third parties
Summary
Security is not a solution but an ongoing process. Security can only be achieved by a combination of technical and organizational measures. One of the biggest security risks is a user. Security is a big challenge but lots of fun as well: a great field of study and research.
Next lecture
The economics of computer security Security threats Security design principles