Get Ampel's observability stack running in 5 minutes.

• Docker and Docker Compose installed
• Ampel services running (API, Worker, PostgreSQL, Redis)

# Start monitoring services
make monitoring-up

# Or manually
docker-compose -f docker/docker-compose.monitoring.yml up -d

This starts:

• Prometheus (http://localhost:9090) - Metrics storage
• Grafana (http://localhost:3000) - Dashboards (admin/admin)
• Postgres Exporter (http://localhost:9187/metrics)
• Redis Exporter (http://localhost:9121/metrics)
• Loki (http://localhost:3100) - Log aggregation

# API health check
curl http://localhost:8080/health

# API readiness check
curl http://localhost:8080/ready

# View Prometheus metrics
curl http://localhost:8080/metrics

1. Open http://localhost:3000
2. Login with admin/admin
3. Navigate to Dashboards > Ampel Overview

1. Open http://localhost:9090
2. Try these queries:

# Request rate
rate(http_requests_total[5m])

# Error rate
sum(rate(http_requests_total{status=~"5.."}[5m])) / sum(rate(http_requests_total[5m]))

# Database connections
pg_stat_database_numbackends

Pre-configured dashboard shows:

• HTTP request rate by endpoint
• Request duration (p95)
• Status code distribution
• Database connection count
• Active pull requests

# View logs
make monitoring-logs

# Restart services
make monitoring-restart

# Check health
make monitoring-health

# Stop monitoring
make monitoring-down

# Clean all data
make monitoring-clean

Alerts are configured in /monitoring/alerts/ampel.yml:

• High error rate (>5% for 5min)
• High latency (p95 >1s for 10min)
• Database down
• Service unavailable

• Full Observability Guide
• Custom Metrics
• Monitoring Setup
• Distributed Tracing

Check prometheus.yml targets match your service ports:

- targets: ['api:8080'] # Should match your API port

1. Verify Prometheus datasource: Configuration > Data Sources
2. Check Prometheus is scraping: http://localhost:9090/targets
3. Verify metrics endpoint returns data: curl http://localhost:8080/metrics

Reduce retention period in prometheus.yml:

global:
  scrape_interval: 30s # Increase from 15s

Or add storage retention flags:

--storage.tsdb.retention.time=15d
--storage.tsdb.retention.size=10GB