|Elk Cloner||First virus on a microcomputer (Apple II)||1982||Rich Skrenta|
|Morris Worm||November 2, 1988||Bob Morris, Jr.|
The Art of Computer Virus Research and Defense by Peter Szor
Von Neumann machine - no difference between code + data
- A Universal Machine
- A Universal Constructor
- Information on tape
Instead of writing computer viruses, I strongly recommend playing this harmless and interesting game. In fact, if worms fascinate you, a new version of Corw Wars can be created to link battles in different networks and allow warrior programs to jump from one battle to another to fight new enemies on those machines. Evolving the game to be more networked allows for simulating worm-like warrior programs.
Created by Robert Morris Sr. (NSA Chief Scientist), Victor Vyssotsky and Dennis Ritchie (Bell Labs)
Originally called Darwin, ran on a PDP-1 in Bell Labs
Redcode Assembly Language
10 instructions in original set, 14 in 1999
Memory Array Redcode Simulator (MARS)
Types of programs
Moves 1 address forward each cycle
Learning how to break software helps you build more robust software.
Confidentiality, Integrity, Availability
Understand the issues, risks
Assess, plan, design/architect
Principle of Least Privilege
When designing a security policy, be it a firewall rule, or filesystem permissions, never give more than the necessary permissions to get the job done. Doing so reduces the attack surface, and weakens (though does not eliminate) the potency of compromise. It’s easier to loosen rules than to tighten them later.
The appropriateness of a security architecture is that it meeds the confidentiality/integrity/availability needs of an organization. It balances security, risk mitigation, usability, and costs.
Where an action cannot be denied, proof of data integrity.
One of the chief goals of security is that business continuity is ensured. Beyond simple security practices, this is having systems in place that can tolerate failure so that business continues with little/no affect.
A hardened system has these characteristics:
- Minimal amount of software (and hardware) installed and running
- only what is needed
- Regular updates
- Privileges only for what is needed
Computer Security Incident Response Teams (CSIRT)
Team responsible for receiving, reviewing, and responding to computer security incident reports and activity
Security incident and event management
Monitors security-related events from network devices, servers, etc
Logs and alerts on anomalies, malicious activity, puts it into “single pane of glass”
References / citations
- SP800-12 “An Introduction to Computer Security: The NIST Handbook
- SP800-14 “Generally Accepted Principles and Practices for Securing Information Technology Systems”
- OECD Digital Security Risk Management
- Generally Accepted System Security Principles by International Information Security Foundation
- S. Bhatt, P. K. Manadhata and L. Zomlot, “The Operational Role of Security Information and Event Management Systems,” in IEEE Security & Privacy, vol. 12, no. 5, pp. 35-41, Sept.-Oct. 2014, doi: 10.1109/MSP.2014.103.
Looks for ‘gadgets’ in code, snippets of code in libc and elsewhere in lieu of providing the code itself.
BeyondCorp is a zero-trust approach to building infrastructure, where instead of an internal network accessible by VPN, each service is publicly available and limited to known users.
- At a certain point, you’ll look into the void, and realize that there is no way to secure software, only raise the cost of exploitation, and that there’s always someone willing to pay that cost. Learn to be fine with that.
Reflections on Trusting Trust by Ken Thompson
Ken Thompson and Dennis Ritchie both wrote a 20 line assembly program that matched character-for-character :-P
In college, before video games, we would amuse ourselves by posing programming exercises. One of the favorites was to write the shortest self-reproducing program. Since this is an exercise divorced from reality, the usual vehicle was FORTRAN. Actually, FORTRAN was the language of choice for the same reason that three-legged races are popular.
Since you have to “teach” a bootstrapping compiler new tricks, it’s possible to put a deliberate bug in the compiler. By using a quine, you can perpetuate the bug. The lower-level the code, the harder it’ll be to find.
The moral is obvious. You can’t trust code that you did not totally create yourself. (Especially code from companies that employ people like me.) No amount of source-level verification or scrutiny will protect you from using untrusted code.
Communication of the ACM, Vol. 27, No. 8, August 1984, pp. 761-763.