One Log
summary
Argues for treating every application output—logs, metrics, traces, domain events—as a single, append-only one log stream (JSON-structured events are used in the demo). Shows a petstore in Haskell + Java payment service, a node-exporter query feeding the same stream, traffic replay and post-processing via syslog/Kafka. Demonstrates a stream-driven circuit-breaker via stream controller that counts error tokens and issues control commands, and reports the small demo implementation and controller logic working end-to-end.
AI-written summary · about
chapters (9)
- 0:00Start
- 1:49Functional versus Operational Plane
- 3:17What We Believe
- 14:56The way to One Log
- 21:14Event Sourcing
- 23:54Log events as Words of our Language
- 27:04Relation between Words
- 29:03Circuit Breaker
- 34:41Log streams as sentences
related talks
01h 00mApplication Logging & Logstash
30mLogging Rethought 2: The Actions of Frank Taylor Jr.
30mBuilding a Large-Scale, Streaming-Based Logging and Monitoring Solution
27mLogs Are Good, Actually
31mThe Long and Winding Road to Application Logging [A]
39mMaking Logs Work for you with Fluentd
39mFast log management for your infrastructure
55mThe Incredible Efficiency of Log Unification!