Implementing Observability in Kubernetes
summary
Demonstrates deploying an OpenTelemetry demo app to Kubernetes and using an OpenTelemetry Collector pod to route telemetry: metrics to Prometheus, dashboards in Grafana, and traces to Jaeger (Zipkin-compatible). Shows both automatic and manual instrumentation, explains how async call fragmentation splits traces and how to fix gaps with selective manual spans, and recommends running replicated collectors for high availability while showing saved metrics and live traces in the UIs.
AI-written summary · about
chapters (13)
- 0:00Observability in K8s
- 1:48Why observability matters
- 3:13Prometheus metrics in K8s
- 4:37OpenTelemetry instrumentation
- 7:30Prometheus & Grafana benefits
- 7:50OpenTelemetry standard
- 9:33Instrumentation methods
- 12:43Distributed tracing explained
- 15:30OpenTelemetry collector role
- 17:36OTLP export protocols
- 19:05OpenTelemetry demo in K8s
- 21:56Metrics and traces
- 24:04Scalable architecture
related talks
46mObservability Ecosystem in Kubernetes
34mKubernetes Observability with OpenTelemetry
32mUnderstanding Observability in Kubernetes
45mFrom k9s to OpenTelemetry: A guide to observability for your apps in K8s
02h 02mHow To Troubleshoot and Monitor Applications using OpenTelemetry
35mOpenTelemetry for Busy People
50mGuide to Observability with OpenTelemetry
23mModern Application Debugging: An Intro to OpenTelemetry