Intro + Deep Dive: Jaeger
summary
Deploys Jaeger with an all-in-one docker image and instruments three Quarkus microservices to show trace timelines, service graphs and trace comparison working end to end. Collection uses an agent sidecar via UDP to a collector, optional Kafka ingestion and storage on Elasticsearch and Cassandra (Badger/in-memory or custom plugins supported). Default sampling is 1 in 1,000, adaptive sampling is planned, and end-to-end tracing adds roughly a 15% throughput overhead while dropped spans appear when queues are underprovisioned.
AI-written summary · about
chapters (18)
- 0:00Intro
- 1:38Tracing
- 5:28Demo
- 12:29Developers
- 15:15Components
- 19:52Contribution
- 23:33Logging vs Trace
- 34:11Deep Dive
- 39:07Tracer
- 40:09Agent
- 41:27Collector
- 42:54Adjuster
- 43:16Query
- 44:11Why
- 46:10Installation
- 47:34User Configuration
- 51:01Performance
- 55:33Drop Spans
