Coding with the lights ON - OpenTelemetry from code to prod
summary
Shows that flipping on OpenTelemetry yields immediate, actionable traces with almost no code change using the Java agent and Jaeger and surfaces real issues in the Pet Clinic demo: a request producing 256 spans, an N+1 select pattern, and many hidden HTTP calls causing ~7s async delays. Adds a local IDE plugin called digma that analyzes traces and CI runs to flag an anti pattern and a scaling regression of ~3s per execution at ~31 concurrent calls. The demo code and tools work end to end.
AI-written summary · about
chapters (21)
- 0:00intro
- 1:43preamble
- 2:10about roni
- 2:54coding practices are changing
- 3:38how i wrote code yesterday
- 5:37technology is allowing us to leverage more data
- 6:40what you know about your code?
- 8:33here be dragons
- 9:20instead...
- 10:28those who don't know, troubleshoot
- 11:35fact or fiction?
- 14:25the would be 10x developer
- 14:52where feedback is missing
- 17:33back to the source
- 20:21why is otel so important?
- 23:35what is tracing?
- 25:32what a trace looks like?
- 26:03automatic vs. manual instrumentation
- 26:45how to use otel when developing
- 27:56demo
- 54:08reach roni
related talks
51mCoding with the lights ON - OpenTelemetry from code to prod
21mBetter Observability with No Code Changes
38mPractical introduction to OpenTelemetry tracing
31mTroubleshooting and Monitoring Production Applications using OpenTelemetry
41mPractical introduction to OpenTelemetry tracing
01h 02mWhat is your app really doing? Make it tell you with OpenTelemetry.
41mModern Application Debugging: An Introduction To OpenTelemetry
40mPractical introduction to OpenTelemetry tracing