What We Learned from Reading 100+ Kubernetes Post-Mortems

Explore insights from over 100 Kubernetes post-mortems in this conference talk. Discover recurring patterns, anti-patterns, and root causes of typical outages in Kubernetes-based systems. Learn from real-world experiences to prevent production outages, covering topics such as concurrency policies, YAML structure, ingress resources, and pod configurations. Gain valuable knowledge on integration, control, review, and monitoring practices. Understand the importance of not trusting default configurations and delegating knowledge effectively. Benefit from the speakers' analysis of common mistakes and their prevention strategies, including insights on tools like Gatekeeper and Datree.