Using OpenTelemetry to Build More Human-Friendly Systems
summary
Clear, actionable steps: install open_telemetry-sdk, configure an OTLP exporter with vendor OTLP headers (example New Relic), and add open_telemetry-instrumentation or pick gems manually; traces are stable now, metrics and logs remain experimental and require the metric SDK/ exporter and the instrumentation-logger gem. Add a tracer for custom spans to find a slow dashboard, create an up/down counter meter for worker capacity, and emit correlated logs with trace/span IDs so failures replay with minimal guesswork.
AI-written summary · about
chapters (10)
- 0:00OpenTelemetry intro
- 1:073 a.m. problem basics
- 3:55OpenTelemetry vs vendor tools
- 4:59Core telemetry types
- 6:43OpenTelemetry in Ruby
- 8:49Slow dashboards empathy
- 11:20Teammate capacity empathy
- 14:06Future self logs empathy
- 16:31Summary and call to action
- 17:54Q&A stability support
related talks
20mObservability and Open Telemetry
32mOpenTelemetry feat. logs, metrics and traces
01h 39mHow to instrument your Rails app with OpenTelemetry
36mObservability For You and Me with OpenTelemetry
15mOpenTelemetry: The Vision, the Reality and How To Get Started
34mOpenTelemetry: The Vision, Reality, and How to Get Started
30mOpenTelemetry 101
01h 00mPractical OpenTelemetry for .NET