Mastering Kubernetes: Challenges, Strategies, and Best Practices
summary
Panel covers concrete wins and disasters: CD corruption and relentless rollbacks that backups would have prevented, a production bug that created “walking dead” pods, and a platform that scales from 500 to 15,000 developers without growing the ops team. Tools and patterns named include Helm, GitOps, Carpenter, Cloud Custodian, Fluent Bit/Fluentd, OpenTelemetry, UI tools like Lens, and APIs like Gateway API and runtimes like WebAssembly with practical pros and cons.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 2:34Kubernetes industry oversells
- 4:26Kubernetes transforms work
- 6:56Kubernetes oh no moments
- 9:04Selling Kubernetes to management
- 10:27Kubernetes complexity benefits
- 12:55Audience pain points security
- 16:26Troubleshooting production updates
- 19:17Dashboards versus kubectl
- 21:03Navigating CNCF landscape
- 25:21Quick solutions to Kubernetes challenges
- 27:08Kubernetes best practices
- 31:00Kubernetes future trends
- 33:04Q&A on Kubernetes gaps
- 37:44Network segregation enterprise
- 40:13Should students learn Kubernetes
- 40:54Closing remarks
related talks
46mPanel Discussion: The number of orche
40mThe Good, the Bad, and the Ugly of Managing Production Systems with Kubernetes
34mReading 100+ Kubernetes Post-Mortems - lessons learned
20mKeynote: Kubernetes - Don't Stop Believin'
29m11 5 years of kubernetes and looking to the future
36mKubernetes Confessions: Real Mistakes We’ve All Made (and What They Teach Us)
35mThriving With Kubernetes On-Call: Best Practices & Lessons Learned
33mKubernetes: Crossing the Chasm