Observability with Spring-based distributed systems
summary
Spring Boot Actuator gives local endpoints but not system-wide insight so add centralized logging (ELK or Splunk) and stream logs for searchable, request-correlated logs. Use Spring Cloud Sleuth to propagate a trace ID in logs across services and report spans to Zipkin or another backend. Use Micrometer as a metrics façade with exporters like Prometheus or Datadog and configure common tags via properties. Tracing samples by default at 10% and transports to Zipkin can use HTTP, Kafka or RabbitMQ for async reporting.
AI-written summary · about
chapters (24)
- 0:00Intro
- 2:01What is observability?
- 3:53Why care about observability?
- 7:01Spring Boot Actuator
- 8:30Non-distributed system
- 9:19Distributed systems are hard
- 10:41Logging and metrics and tracing... oh my!
- 11:57Effort to reward
- 13:13Logging In general
- 15:27Problems with basic logging
- 16:42Centralized logging
- 18:30Logging and Spring
- 21:35Metrics in general
- 24:03Metrics examples
- 25:15Basic metrics
- 26:25Metrics for observability
- 27:34Metrics and Spring
- 31:58Distributed tracing for observability
- 33:54Zipkin UI
- 34:19Zipkin architecture
- 34:54Tracing and Spring
- 37:12Correlation everywhere
- 38:33Observability cycle
- 40:06Key takeaways
related talks
58mObservability Beyond the three pillars with Spring
47mSpring Boot Observability in Practice: Actuator, Micrometer & OpenTelemetry
26mSpring Boot Observability
46mImplementing Observability in Java-Based Projects
55mDemystifying Observability for Spring Applications
49mMonitor Your Microservices with Logs, Metrics, Pings and Traces
38mObservability 101 with Spring and Micrometer
44mThe Holy Trinity of Observability: Logs, Metrics, and Traces