An introduction to distributed tracing
summary
Claims that distributed tracing is practical now with vendor-neutral hooks and concrete bits to pass between services, shows actual formats and numbers not just buzzwords: a 128-bit trace ID and 64‑bit span IDs carried in a traceparent header, sampling rates like 1–5% to control cost, browser bootstrap using a page_load trace/span, OpenTracing APIs for start/finish/inject/extract, an agent that auto-instruments Express and sends data to Elastic APM backed by Elasticsearch, demoing cross-service traces with ~940 requests and a working UI despite demo hiccups.
AI-written summary · about
chapters (4)
- 0:00Start
- 8:10Agents, Clients, and Tracers
- 23:29Navigation Timing API
- 38:25Don't try to instrument everything in the beginning •Instrument important services first
related talks
57mDistributed Tracing: How the Pros Debug Concurrent and Distributed Systems
37mObservability, Distributed Tracing & the Complex World
49mDistributed Tracing: How the Pros Debug Concurrent and Distributed Systems
42mDistributed Tracing Opening the Door to Visualisation and Correlation
24mDistributed Tracing : New DevOps Foundation.
31mUnderstanding Microservices with Distributed Tracing
42mAn introduction to distributed tracing and Zipkin
23mIntroduction to OpenTracing