Efficient Debugging and Logging with OpenTelemetry in Go
summary
Company uses tracing 100% of traffic with OpenTelemetry in Go, running traces across a microservice demo (controller, pubsub, Cloud Functions) to find latency and errors. Shows concrete middleware added for orm and pubsub clients, how spans carry attributes/events, sampling choices and exporters, and integration with Google Cloud Trace and Query Insights to link DB queries to traces. Describes structured logging with Trace ID, using traces in unit and e2e tests and a Trace Test tool for assertions.
AI-written summary · about
chapters (17)
- 0:00Introduction
- 2:58Debugging with logs and observability
- 7:17Demo app architecture
- 8:23Traces and flame graphs
- 10:56OpenTelemetry benefits
- 13:07Spans, attributes, and events
- 17:56Tracing vs traditional logging
- 18:17OpenTelemetry in Go setup
- 20:06Span status and context propagation
- 21:09Trace visualization and debugging
- 23:16Traces with logs and DB queries
- 25:21Adopting OpenTelemetry
- 28:35Spans in business logic
- 30:24Middleware and SDK
- 31:09Traces for testing
- 32:57Live demo debugging
- 37:56Summary
related talks
30mMastering Distributed Tracing with Go
34mOpenTelemetry for Dummies: Instrumenting Go Apps
19mInstrumenting Go Apps With OpenTelemetry
40mInstrumenting GO Applications
01h 28mUnderstanding Distributed Systems with OpenTelemetry
55mCloud-Native Observability with OpenTelemetry
29mThe Fundamentals of OpenTelemetry
42mObservable Applications Using OpenTelemetry