Dude, Where’s My Error?: How OpenTelemetry Records Errors, and Why...
summary
OpenTelemetry records errors as either enriched spans or correlated logs, using methods like record_exception (or record_error in Go) and automatically adding a span event with stack trace while requiring an explicit call to set span status to error. A Python Flask demo (Docker Compose, OTel Collector) shows that Jaeger displays span events as logs, other backends surface a distinct span events signal, and span kind rules can change which errors count toward error rates.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:50What this talk will cover
- 2:34Error handling background
- 3:38OpenTelemetry refresher and its role
- 4:00Errors and exceptions
- 5:25OpenTelemetry error handling
- 7:15Recording errors with spans/logs
- 14:17Vendor differences
- 14:38Demo setup
- 18:09Running demo
- 19:32Jaeger error views
- 20:56Backend error views
- 22:42Third backend errors
- 24:09Key takeaways and wrap-up
related talks
33mDude, Where’s My Error?: How OpenTelemetry Records Errors, and Why It...
24mObservability and instrumentation via OpenTelemetry
27mDistributed Tracing with OpenTelemetry
02h 02mHow To Troubleshoot and Monitor Applications using OpenTelemetry
37mBeyond Getting Started: Using OpenTelemetry to Its Full Potential
10mopentelemetry_testing
38mPractical introduction to OpenTelemetry tracing
46mThe State of OpenTelemetry