mydocker/grafana

Grafana - Monitoring Dashboards

Visualization platform for metrics (VictoriaMetrics) and logs (Loki). Sends alerts via Telegram.

Access

Endpoint	Details
Public URL	https://grafana.sebson.space
Internal	http://localhost:3001
Login	seba (password in .env: GRAFANA_ADMIN_PASSWORD)

Configuration

Automatically provisioned from ./etc/grafana/provisioning/:

Data Sources

Source	URL	UID	Notes
VictoriaMetrics (default)	http://victoriametrics:8428	prometheus	Long-term metrics
Prometheus-Raw	http://prometheus:9090	-	With basic auth
Loki	http://loki:3100	loki	Max 1000 lines per query

Pre-installed Dashboards

• home-assistant.json - HA entity metrics
• ping-monitoring.json - Network availability
• node-exporter-*.json, node.exporter.server.metrics.json - Host metrics
• traefik2.json - Reverse proxy metrics
• system-monitoring-complete.json - Overall system monitoring
• mazda.json, windows-wmi.json - Additional sources

Plugin: grafana-piechart-panel

Alerts: Telegram via Grafana Unified Alerting (GF_UNIFIED_ALERTING_ENABLED=true). Bot token and chat ID from .env.

Monitored Metrics

Grafana visualizes data from other services:

• Host system metrics (CPU, RAM, disks, network)
• Docker container metrics (cAdvisor)
• Home Assistant metrics (entities, states)
• Traefik metrics (requests, latency)
• Container logs (via Loki)
• Ping / device availability

Dependencies

Service	Role
VictoriaMetrics	Primary data source (long-term metrics)
Prometheus	Raw metrics source (short-term)
Loki	Container log source
Alertmanager	Optional (alerts managed by Grafana Unified Alerting)

Useful Commands

# Restart
docker compose restart grafana

# Follow logs
docker compose logs -f grafana

# Upgrade
./upgrade-container-by-name.sh grafana