monitoring-observability

SLO-based alerting prevents alert fatigue by triggering notifications only when service level objectives are genuinely threatened. This Skill guides you through defining error budgets and configuring Prometheus alerts in Grafana to focus on real operational issues.

Structured logging formats log data into consistent, searchable JSON fields. This Skill helps you configure Pino or Winston to emit structured logs with correlation IDs, enabling rapid tracing across distributed microservice requests.

OpenTelemetry provides end-to-end distributed tracing by propagating trace contexts across service boundaries. This Skill offers implementation guidance to integrate OpenTelemetry, allowing you to diagnose request latency and failures.

RED metrics measure service performance (Rate, Errors, Duration), while USE metrics track resource utilization (Utilization, Saturation, Errors). You need them to establish comprehensive observability and detect bottlenecks in your software services.

Yes, Grafana dashboard design patterns help visualize Prometheus metrics effectively. This Skill provides guidance on structuring your dashboards to display RED/USE metrics and SLO status, ensuring clear visibility into your system's operational health.

The three pillars of observability are metrics, logs, and traces. They work together to detect, diagnose, and resolve issues, combining resource monitoring, structured event logging, and request tracing for comprehensive system health analysis.