Deep Dive: Jaeger – Steve Flanders
summary
Jaeger is a distributed-tracing backend with agent/collector/query/UI components, a pluggable storage layer supporting Cassandra Elasticsearch Badger plus community plugins, and two ingest architectures: batch analytics via Spark (v1) or real‑time processing with Kafka + Flink (v2). It demos end‑to‑end tracing using the hotrod demo, finds a Redis timeout causing 1.6–1.8s requests, and shows trace comparisons. Features include adaptive sampling, Zipkin/OpenTracing/OpenTelemetry compatibility, a Kubernetes operator, all‑in‑one Docker deploy, and roadmap items for duration‑based trace comparisons and harder enterprise hardening.
AI-written summary · about
chapters (12)
- 0:00Introduction to Jaeger
- 0:56Distributed tracing basics
- 5:31What is Jaeger
- 7:27Jaeger v1 and v2 comparison
- 9:41Architecture flexibility
- 11:17Open standards and future
- 12:14Technology stack
- 13:34Integration strategies
- 14:11New features and maturity
- 16:45Live demo and UI walkthrough
- 21:57Future roadmap and roadmap
- 26:39Q&A session
