OpenTelemetry feat. logs, metrics and traces
summary
OpenTelemetry delivers a single SDK plus the OpenTelemetry Collector with receivers, processors and exporters so logs, metrics and traces share a schema and a protocol (OTLP), enabling trace-to-log correlation and cross-vendor export to Prometheus, Jaeger, Datadog or Splunk. Auto-instrumentation for Java works as a drop‑in agent, tracing support is mature, metrics are improving and logs lack full semantic conventions; example pipeline adds host metadata via a resource detector and exports over port 4317.
AI-written summary · about
chapters (14)
- 0:00Introduction and welcome
- 1:07What is observability
- 3:38Logs metrics and traces explained
- 5:46How traces work with spans
- 8:58Why consistent metadata matters
- 10:43Problems with current observability tools
- 12:52OpenTelemetry to the rescue
- 14:38How OpenTelemetry differs
- 17:51OpenTelemetry specification
- 20:37Language support in OpenTelemetry
- 22:45OpenTelemetry collector explained
- 26:42OpenTelemetry distributions
- 29:07OpenTelemetry community
- 31:11Closing and resources
related talks
30mOpenTelemetry 101
22mOpenTelemetry – An Observability Framework for Cloud-Native Software
46mThe State of OpenTelemetry
43mWhat is this OpenTelemetry thing?
36mObservability For You and Me with OpenTelemetry
34mOpenTelemetry: The Vision, Reality, and How to Get Started
20mObservability and Open Telemetry
50mGuide to Observability with OpenTelemetry