Theory · Monitoring & logging

17 min · Module 10 · Theory

How to study this topic

Read the theory below (concepts, bullets, and shell/YAML examples).
Try the Exam pool and Book quizzes in this module.
After each quiz, use Show answer on every question to review.

Monitoring & logging

Observability = metrics, logs, traces. Prometheus scrapes metrics; Grafana dashboards alert via Alertmanager. Structured JSON logs enable fast search in ELK/Loki.

SLI/SLO: define reliability targets (99.9% availability).
Alerting: page on symptoms users feel, not every CPU spike.
RED/USE: rate, errors, duration for services; utilization, saturation, errors for resources.

Shell, YAML & config examples

# Prometheus metric types: counter, gauge, histogram
# Example alert: high error rate
# rate(http_requests_total{status=~"5.."}[5m]) > 0.05

Hands-on practice: /lab/