Are we observable yet?
summary
Builds a production Rust backend (Donate Direct at TrueLayer) in a few weeks and shows the concrete telemetry stack used and why it works. Uses tracing crate / tracing::instrument for spans and structured logs, exposes Prometheus → Grafana metrics, ships logs via Vector → Elasticsearch → Kibana, and pushes distributed traces to Jaeger (via OpenTelemetry). Demonstrates what was implemented, how traces correlate across services, and that Rust tooling for metrics logging and tracing is sufficient for real production observability.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 1:06Agenda
- 1:48The Project
- 2:39Production Challenges
- 3:45Production Systems
- 5:52Operational Standards
- 7:11The Dilemma
- 8:29Observability Basics
- 10:33Metrics
- 14:02Metrics in Rust
- 15:24Logging
- 19:12Structured Logging
- 23:23Distributed Tracing
- 25:58Processing Traces
- 28:13Microservices Tracing
- 30:05Conclusion
- 32:08Future Projects
- 33:36Q&A
related talks
13mArchitecting an open observability stack
33mThe Observant Developer - OpenTelemetry from code to prod
38mPractical introduction to OpenTelemetry tracing
43mPractical Introduction to OpenTelemetry Tracing
08mGraphQL Observability
43mObservability for React applications
35mOpenTelemetry for Busy People
28mImplementing Observability in Kubernetes