Intro: Jaeger
summary
Live demo shows Jaeger running as an all-in-one Docker plus a Hot Rod microservice app and visualizes full traces across frontend, customer, driver, route, MySQL and Redis; one request surfaces a Redis timeout and a 1.39s blocked MySQL query. Backend supports Cassandra/Elasticsearch or in-memory storage, uses Kafka + Flink-style pipeline for feature extraction, and plans post-trace sampling and incremental (incomplete) spans. Claims: end-to-end context propagation via OpenTracing works, adaptive sampling reduces missed anomalies, and aggregation enables automated anomaly detection.
AI-written summary · about
chapters (13)
- 0:00Start
- 1:09UBER Service Dependency Graph
- 1:58Transaction Monitoring for Microservices
- 3:12Context Propagation & Distributed Tracing
- 4:46OpenTracing
- 18:46Jaeger, a Distributed Tracing System
- 19:36Technology Stack
- 20:11Community
- 20:42Distributed Tracing Systems
- 21:44Recent Highlights
- 22:26Post Trace Sampling
- 23:16Data Pipeline
- 24:14Incomplete Span Support
