Teaching OpenSearch To Read YAML: Finding Broken Infrastructure Before It Breaks
summary
A pipeline converts committed Kubernetes YAML into structured JSON via a custom parser that mimics the API server, resolves label/selector joins into arrays, and indexes relationship-rich documents into OpenSearch so queries run in milliseconds. The demo finds an over‑privileged test user with cluster‑admin rights and orphan network policies across thousands of clusters in milliseconds; configured OpenSearch monitors block PRs on violations. The project is open source, includes a Dockerized OpenSearch, Python enrichment scripts, and a reproducible demo.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:06YAML tax impact
- 3:33Why tools fail on YAML
- 4:39Kubernetes YAML as graph
- 6:03OpenSearch for Kubernetes data
- 7:07YAML parsing pipeline
- 8:12Semantic enrichment importance
- 9:16RBAC blast radius use case
- 9:59Orphan network policies use case
- 10:43Live demo: OpenSearch graph queries
- 13:34Infrastructure immune system
- 14:38Vector search and LLMs future
- 15:41Drift detection in clusters
- 16:04Key takeaways
related talks
21mKubernetes Under the Magnifying Glass: Observability With OpenSearch
06mSponsored Keynote: OpenSearch - See Everything: Open Observability for Agentic AI
22mFrom Telemetry to Insight: Building AI-Powered Observability Pipelines with OpenSearch
23mCreating the OpenSearch Kubernetes Operator
52mOperating OpenSearch at Scale: Lessons From Managing 10,000+ Clusters Across...
21mNoOps for OpenSearch: The Future of Self-Healing Search Infras...
36mAutoscaling Search and Analytics in Kubernetes: The OpenSearch Operator Advantage
42mOpenSearch in Kubernetes: Helm Charts and Operator