Unlocking Observability with React & Node.js
summary
Claims that unified fullstack observability for React and Node.js cuts MTTR by linking user actions to backend spans, demonstrates trace propagation using traceparent/trace-state headers and correlation ids, and relies on OpenTelemetry for vendor neutral context. Shows dashboards combining user perceived latency end to end request latency error rates core web vitals and concrete alerts that trigger only when front end error spikes coincide with backend latency. Mentions tooling such as Datadog New Relic Honeycomb Dynatrace Elastic Grafana Cloud Observe OpenObserve and suggests POCs.
AI-written summary · about
chapters (12)
- 0:00Introduction and Speaker Background
- 1:13Agenda Overview
- 1:43The Need for Unified Observability
- 2:40What is Unified Full Stack Observability?
- 3:58Benefits of Unified Observability
- 5:40Correlating Frontend and Backend Data
- 7:22Key Performance Indicators (KPIs)
- 9:19Distributed Tracing Explained
- 10:38Real-World Scenarios
- 12:15Tools and Products for Observability
- 14:01Best Practices for Unified Observability
- 15:36Conclusion and Final Thoughts
related talks
19mUnlocking Observability with React & Node.js
17mObservability in frontend: React Open Telemetry
43mObservability for React applications
30m"Observability for the Front End"
45mObservability is for User Happiness
21mtalks about mastering Observability for DevOps
44mObservability: Superpowers for Developers
38mTwo Roads Diverged in the Observability Woods: I Took the E2E One