Lightning Talk: Debugging OpenTelemetry
Lightningat OTel Community Day North America 2022 · on Monday, 2022-06-20· at 13:35· 303+304, JW Marriott Austin | 110 E 2nd St·
summary
Common install failures are listed with fixes: start by checking exporter connection errors and set log level to debug, then account for the buffering default for production that delays data visibility, and finally diagnose silent problems like missing instrumentation or broken context propagation which produce no errors. Proposes a simple testing language for expectations (YAML + collector checks or a web UI) that reads the package manifest, declares expected transactions and spans, and validates incoming telemetry.
AI-written summary · about
chapters (9)
- 0:00Lightning Talk: Debugging Open Telemetry Ted Young, Lightstep
- 0:36But what happens when the installation don't work?
- 0:52Once metrics and logs are stable...
- 1:19so check for connection errors.
- 2:17Because debugging data issues is the worst.
- 2:27With a connection failure, at least there's an error.
- 3:33Users test their instrumentation...
- 4:01What would this testing look like?
- 5:29Maybe even use the package manifest for the app as input?
related talks
10mLightning Talk: Is OpenTelemetry Too Complicated to Get Started?
05mProject Lightning Talk: New Frontiers for OpenTelemetry - A Roadmap for the Future
23mModern Application Debugging: An Intro to OpenTelemetry
11mLightning Talk: OTel Me How to Build a Data Pipeline for Observability
41mModern Application Debugging: An Introduction To OpenTelemetry
09mOpenTelemetry or OpenTeleMessy?
21mUsing OpenTelemetry to Build More Human-Friendly Systems
50mGuide to Observability with OpenTelemetry