Overview
- Containerized monitoring stack
- Real-time and historical metrics
- Alerting foundation
- Portfolio-quality documentation
Monitoring Stack
Building a production-style monitoring platform using Prometheus, Grafana, Node Exporter, Alertmanager, and cAdvisor.
Technology Stack
- Prometheus (metrics)
- Grafana (visualization)
- Node Exporter (host stats)
- cAdvisor (container stats)
- Alertmanager (alerts)
- Docker Compose (orchestration)
System Architecture
- Exporters → Prometheus
- Prometheus → Grafana
- Prometheus → Alertmanager
- Docker internal networking
Key Challenges
- Port conflicts with Cockpit
- YAML syntax failures
- Grafana datasource routing
- Dashboard variable resets
- Container restart loops
Solutions Implemented
- Custom port remapping
- Prometheus config validation
- Docker DNS-based routing
- Variable refresh automation
- Centralized health checks
Operations Checklist
docker compose pscurl /-/ready- Target health validation
- Dashboard review
- Log inspection
Results
- 24/7 visibility
- Early issue detection
- Improved system reliability
- Enterprise-style monitoring