Understanding Observability in Kubernetes
Talkat Open Source Summit Europe 2023 · on Tuesday, 2023-09-19· at 14:30· Room 0E-1 (Floor 0), Abandoibarra Etorb., 4, 48011 Bilbao, Bizkaia, Spain·
summary
Explains observability as four concrete signals — metrics logging tracing profiling — and shows tooling and steps: enable metric-server then run kubectl top after tweaking metric-server args in the playground, add Prometheus, Grafana and Jaeger for collection and dashboards, use OpenTelemetry collector and sidecar log shippers, and generate flame graphs with a kubectl plugin or Pyroscope for CPU hotspots. Emphasizes probes (liveness/readiness/startup), security for metrics access, and starting small and validating in a sandbox.
AI-written summary · about
chapters (12)
- 0:00Kubernetes observability intro
- 3:15Observability vs monitoring
- 5:03Four pillars of observability
- 11:07Kubernetes built-in tools
- 14:18Demo: kubectl monitoring
- 16:24Metrics server install & troubleshoot
- 19:41Prometheus Grafana on Kubernetes
- 21:06External logging & tracing
- 22:54Best practices for observability
- 25:42Cloud-native teams strategies
- 27:26Conclusion scaling observability
- 29:15Q&A: New pillars & events
related talks
11mKubernetes Observability
28mImplementing Observability in Kubernetes
36mMetadata: The Secret Sauce for Full Observability
09mObservability in Kubernetes: A Signal- and SLI-Centered Approach to Reliability
10mA Simple and Practical Guide to Obser...
36mObservability? It's a Data Analytics Problem, You Fool!
46mObservability Ecosystem in Kubernetes
43mA Guide to Observability for Your (Java) Apps in K8s