You cannot operate what you cannot see. Observability is how you know whether a system is healthy, and how you diagnose it when it isn’t — through metrics, logs, and traces. Part of the DevOps curriculum.
Partly grounded: the Java logging note already covers SLF4J/Logback and a Micrometer→Prometheus path from a real pipeline; this section is the infrastructure-side, tool-and-concept view. The specific stacks are otherwise [reference].
Reading order
- Observability Fundamentals — [Advanced] — the three pillars (metrics, logs, traces), monitoring vs observability, SLIs/SLOs/SLAs and error budgets, alerting, and the four golden signals
- The Observability Stack — [Advanced] — Prometheus + Grafana (metrics), the ELK stack / Loki / Graylog (logs), OpenTelemetry + Jaeger (tracing), and the APM landscape (Datadog, New Relic, Dynatrace)
Related
- Logging & Observability (Java) — the application-side view, grounded in a real pipeline
- Orchestration — Prometheus is the de-facto k8s monitoring stack
- DevOps curriculum map
Jump to a subject
Software engineering
Infrastructure & security
Computer science
Hardware & engineering