No Requests, No Limits, No Service - How to Bring Kubernetes to a Grinding Halt

Learn from a real-world incident analysis where thousands of ephemeral test environment pods caused a complete cluster jam in Octopus Deploy's internal build platform during this 17-minute CNCF talk. Explore the systematic investigation process, discover the key contributing factors that led to the Kubernetes cluster failure, and understand the continuous challenges faced when managing internal development tooling. Gain valuable insights into cluster resource management and learn how to prevent similar incidents in your own Kubernetes environments.