Dude, Where’s My Error?: How OpenTelemetry Records Errors, and Why It...
summary
A small Python Flask app is instrumented both manually and with auto-instrumentation, forcing an exception whenever the die roll is even so errors reproducibly occur; telemetry flows through an OpenTelemetry Collector to Jaeger, ServiceNow Cloud Observability, and New Relic where the same span and log data render differently. The demo contrasts record_exception vs record_error, span events vs logs, how span kind and status affect whether an error counts toward application error rate, and it works end to end.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 2:28OpenTelemetry error recording
- 3:52Errors vs exceptions
- 5:59OpenTelemetry error handling
- 8:08Errors in spans and logs
- 16:40Demo overview
- 20:36Live error recording demo
- 25:37Error visualization in New Relic
- 28:53Summary
- 31:01Resources and closing
related talks
17mOpenTelemetry Looks Good To Me: Loki, Grafana, Tempo, Mimir
32mAntipatterns in Observability: Lessons Learned and How OpenTelemetry Solves Them
38mReaching Operational Excellence using OpenTelemetry
36mObservability For You and Me with OpenTelemetry
06mIgnite | Discovering the Magic Behind OpenTelemetry Instrumentation
37mOpenTelemetry & Distributed Tracing: A Jaeger v2 Deep Dive
30mOpenTelemetry 101
16mHacking OpenTelemetry