Prometheus: Metrics Collection 📊

Metrics collection system for early warning about problems.

Functions:

  • 🖥️ Server metrics: CPU, RAM, disk, network via node_exporter
  • 🌐 Service availability: blackbox HTTP/TCP/ICMP checks
  • 📈 Collecting metrics from applications: Nextcloud, Home Assistant, and others
  • 🚨 Alertmanager: notifications in Telegram/Discord when thresholds are exceeded
  • 🔍 Queries via PromQL for deep analysis

How it works:

  1. Prometheus collects metrics on a schedule (scrape)
  2. Charts are displayed in Grafana
  3. On anomaly, an alert is triggered - the administrator receives a notification

For administrators: Alert rules, recording rules, federation of metrics, long-term storage via Thanos.

Access: via Grafana (grafana.potatoenergy.ru) • according to Potato Energy credentials (management is only based on the rights of the admin group)