How to Lose a Container in 10 Minutes
summary
Claims common cloud/container mistakes: default Kubernetes configs expose APIs and dashboards, secrets sit in etcd base64 and aren’t encrypted, and public images often contain malware. Recommends controlling base images, not running containers as root, rotating keys, using managed clusters or the CIS Kubernetes benchmark, and validating CI/CD and runtime tools. Describes an experiment leaving containers Internet‑exposed for months using psad to monitor traffic, expecting to be owned in 10 minutes but finding almost no compromises.
AI-written summary · about
chapters (14)
- 0:00Introduction and speaker intro
- 3:52Container security overview
- 7:47Good data protection practices
- 9:16Zero trust and isolation
- 10:39Shared responsibility model
- 12:05Controlling container images
- 14:12Kubernetes default flaws
- 15:56Secrets management mistakes
- 18:05Checking container privileges
- 19:10Orchestrator privilege pitfalls
- 21:14Shifting left with security tools
- 23:21Experiment losing a container
- 26:36Conclusion and key takeaways
- 28:22Q&A session
related talks
32mHow to lose a container in 10 minutes
32mHacking the Cloud Native Renaissance: Lowering Risk of Kubernetes Adoption
26mThe 10 Container Security Tricks That (...)
32mContainer Security: Exploits, Vulnerabilities and Defense
58mTop Secret Cloud Native Security Lessons
35mInsights into Unsecured Kubernetes in the Wild
40mContainers, Kubernetes and Practical DevSecOps
28mContainer Security: A Horror Story