What is your app really doing? Make it tell you with OpenTelemetry.
summary
Claims that OpenTelemetry lets production systems tell you what happened, demonstrates a Java setup using the OpenTelemetry Java agent plus Docker Compose, Jaeger and Honeycomb, and shows spans created by Spring annotations and automatic library instrumentation. Shows traces with end-to-end durations (1.1s, 3.74s, 2.43s and a 7s user-facing request where backend is 2.3s), uses the Traceparent header for propagation and recommends dynamic tail sampling to control storage cost.
AI-written summary · about
chapters (13)
- 0:00Observability with OpenTelemetry
- 2:16Why software needs to tell us
- 9:04From logs to observability
- 11:10Live demo: phrase picker service
- 15:46Logs vs distributed traces
- 24:39How OpenTelemetry enables tracing
- 32:12Adding OpenTelemetry to Java
- 36:02Production observing with Jaeger & Honeycomb
- 44:14Why traces scale better
- 48:15OpenTelemetry production best practices
- 51:25Q&A on propagation & reactive apps
- 56:17Cost vs data retention balance
- 1:00:40OpenTelemetry adoption closing
related talks
37mMake your Java application observable with no code changes using OpenTelemetry
38mPractical introduction to OpenTelemetry tracing
31mTroubleshooting and Monitoring Production Applications using OpenTelemetry
28mPractical Introduction to OpenTelemetry Tracing
47mA Practical Introduction to Open Telemetry
02h 02mHow To Troubleshoot and Monitor Applications using OpenTelemetry
47mUsing metric and traces to diagnose our cloud-native applications
30mInstrument Your Java Applications with OpenTelemetry!