Security Pre-Cognition
summary
Argues for flipping postmortems into proactive experiments using chaos engineering, with an AWS-native open source framework Chaos Slinger that runs controlled experiments by siphoning ~0.01–0.05% real traffic and comparing control vs experiment. Describes a misconfigured-port scenario that shows cloud-native config blocks changes repeatedly while a Palo Alto firewall misses many, and cites Netflix’s Chap running 10–20k experiments to expose cascading failures. Focuses on finding human error and system glitches (roughly 52% of breaches) and validating runbooks and controls.
AI-written summary · about
chapters (12)
- 0:00Intro
- 1:16Security Precognition
- 4:33Our systems have evolved beyond human ability to mentally model their behavior.
- 6:24Microservice
- 9:32Software Complexity
- 11:30How well do you really understand how your system works?
- 13:29Failure Happens.
- 22:37Who is doing Chaos? NETFLIX
- 26:00Lets Flip the Model
- 29:44Testing vs. Experimentation
- 30:33Security Crayon Differences
- 36:01How does My Security Really Work?
related talks
40mNavigating the Unknowable: Resilience through Security Chaos Engineering
43mCyber Chaos Engineering
31mNavigating the Unknowable: Resilience through Security Chaos Engineering
28mBuilding Safety, Security and Resilience through Chaos
40mSecurity & Chaos Engineering: Resilient Distributed Systems
28mUnknown Unknowns + Chaos = Security Chaos Engineering
24mSecurity Chaos Engineering for Fun and Profit
34mDefining Steady States and Developing Hypotheses