o11y-metrics-prometheus-ops

Prometheus monitoring requires configuring scrape targets in prometheus.yml to define which endpoints collect metrics. The Skill provides a template with global intervals, scrape configs, and service discovery patterns for Kubernetes, Docker Compose, and cloud environments. Copy the template, adjust external labels and target endpoints, then start Prometheus to begin metric collection.

Yes, Prometheus supports Kubernetes service discovery through SD configurations. This Skill covers Kubernetes SD setup alongside static targets and file-based discovery, enabling automatic target detection and metric scraping across containerized deployments without manual endpoint management.

Alerting rules in Prometheus define conditions that trigger notifications when metrics cross thresholds. This Skill covers alert rule design, rule loading configuration, and alert integration patterns. Define rules in YAML, load them via prometheus.yml, and connect to alerting backends for notification delivery.

Scrape configs define which targets Prometheus pulls metrics from and how often; recording rules pre-compute frequently queried metric expressions and store results as new time series. This Skill covers both: scraping handles collection and storage, while recording rules optimize query performance for complex calculations.

Prometheus stores metrics that Grafana queries and visualizes through dashboards. This Skill configures the Prometheus backend—metric collection, alerting, and storage—which Grafana uses as a data source. The setup enables end-to-end monitoring pipelines combining Prometheus and Grafana.

Use promtool to validate prometheus.yml syntax and rule files for errors before starting Prometheus. The Skill recommends verifying targets and rules in the Prometheus UI after deployment. This prevents misconfiguration and ensures scrape intervals, service discovery, and alert rules load correctly.