A Simple and Practical Guide to Obser...
summary
Explains a minimal Kubernetes observability baseline: a short list of meaningful metrics (API server latencies, request rates, node CPU/memory, pod restarts, scheduling latency), consistent labels to avoid high cardinality and duplication, and a clear pipeline to storage and dashboards. Shows a working audit-log parser that sends Kubernetes audit JSON to VictoriaLogs and visualizes verbs, namespaces and user ops on a Grafana dashboard. Recommends structured logs, instrumentation from day zero, focused alerts, and using OpenTelemetry for tracing.
AI-written summary · about
chapters (14)
- 0:00Observability in Kubernetes
- 0:41Kubernetes metrics sources
- 1:21Stability metrics
- 2:26Kubernetes metrics challenges
- 2:49Kubernetes logs sources
- 3:32Audit logs value
- 4:13Audit log parsing demo
- 5:27Visualizing audit logs
- 6:09Structured logs advice
- 6:50Tracing benefits
- 7:32SIG Instrumentation role
- 8:15Improving instrumentation
- 9:20Community collaboration
- 10:04Q&A
related talks
32mUnderstanding Observability in Kubernetes
11mKubernetes Observability
16mShrinking the Observability Bill
38mPractical Guide to Application Observability in 10 Steps
55mBeginner's guide to observability
36mMetadata: The Secret Sauce for Full Observability
45mFrom k9s to OpenTelemetry: A guide to observability for your apps in K8s
43mA Guide to Observability for Your (Java) Apps in K8s