Docker Monitoring System with Prometheus and Grafana

This project provides a complete monitoring stack using Prometheus and Grafana, including exporters for host metrics, Docker containers, and system processes.

📋 Components

The project includes the following services:

• Prometheus - Monitoring system and time series database
• Grafana - Metrics visualization and analytics platform
• Node Exporter - Hardware and OS metrics exporter
• cAdvisor - Container resource usage and performance analyzer
• Process Exporter - System process metrics exporter

🚀 Quick Start

Prerequisites

• Docker
• Docker Compose

Installation

1. Clone the repository (or navigate to the project directory)

2. Start all services:

docker-compose up -d

3. Verify that containers are running:

docker-compose ps

📊 Accessing Services

Grafana

• URL: http://172.25.0.13:3000
• Username: admin
• Password: MYPASSWORD

Prometheus

• URL: http://172.25.0.10:9090

Exporters

• Node Exporter: http://172.25.0.11:9100/metrics
• cAdvisor: http://172.25.0.12:8080
• Process Exporter: http://172.25.0.14:9256/metrics

📁 Project Structure

.
├── docker-compose.yml                          # Services configuration
├── dashboards/
│   ├── Docker Monitoring-1764091626474.json    # Docker monitoring dashboard
│   └── Host Monitoring-1764091636759.json      # Host monitoring dashboard
└── data/
    ├── prometheus/
    │   └── config/
    │       └── prometheus.yml                  # Prometheus configuration
    └── process-exporter/
        └── config/
            └── process-exporter.yml            # Process Exporter configuration

⚙️ Configuration

Prometheus

Prometheus is configured to collect metrics every 120 seconds from the following targets:

• Prometheus (localhost:9090)
• cAdvisor (cadvisor:8080)
• Node Exporter (node-exporter:9100)
• Process Exporter (process-exporter:9256)

To modify the configuration, edit the data/prometheus/config/prometheus.yml file.

Grafana

The default Grafana password is defined in docker-compose.yml through the GF_SECURITY_ADMIN_PASSWORD environment variable.

⚠️ IMPORTANT: Change the default password for production environments!

Dashboards

The project includes two pre-configured dashboards in the dashboards/ folder:

Docker Monitoring Dashboard

Available metrics:

• Running Containers - Total number of active containers
• Total Docker Containers Memory Usage - Aggregated memory consumption across all containers
• Total Docker Containers CPU Usage - Aggregated CPU usage across all containers
• Docker Containers Memory Usage - Time series graph showing memory usage per container
• Docker Containers CPU Usage - Time series graph showing CPU usage per container
• Docker Containers Disk I/O - Read/write operations per container
• Docker Containers Network Throughput - Network receive/transmit bandwidth per container
• Docker Container Memory Reservation Usage - Table showing memory usage against limits
• Docker Container Memory Usage - Table with current memory consumption per container

Host Monitoring Dashboard

Available metrics:

• Host Uptime - System uptime in seconds
• Memory Usage - Current percentage of memory in use
• CPU Usage - Current percentage of CPU utilization
• Disk Space - Percentage of disk space used on root filesystem
• CPU Usage History - Time series graph of CPU utilization over time
• Top Processes by CPU Usage - Configurable top-k processes consuming CPU
• Memory Usage History - Time series graph of memory consumption over time
• Top Processes by Memory Usage - Configurable top-k processes consuming memory
• Disk IOPS - Input/output operations per second (read/write)
• Network Throughput - Network bandwidth usage (download/upload)

To import dashboards into Grafana:

1. Access Grafana
2. Go to Dashboards → Import
3. Upload the JSON files from the dashboards/ folder

🔧 Useful Commands

Start services

docker-compose up -d

Stop services

docker-compose down

View logs for a specific service

docker-compose logs -f <service-name>
# Example: docker-compose logs -f prometheus

Restart a service

docker-compose restart <service-name>

Remove volumes (⚠️ deletes data)

docker-compose down -v

🌐 Network

All services are on the same Docker network (monitoring) with the following IPs:

• Gateway: 172.25.0.1
• Prometheus: 172.25.0.10
• Node Exporter: 172.25.0.11
• cAdvisor: 172.25.0.12
• Grafana: 172.25.0.13
• Process Exporter: 172.25.0.14

📦 Volumes

The project uses Docker volumes for data persistence:

• prometheus_data: Stores Prometheus time series data
• grafana_data: Stores Grafana configurations and dashboards

🛡️ Security

• Change the default Grafana password in production
• Services are exposed only on the internal network (using expose instead of ports)
• For external access, configure a reverse proxy (nginx, traefik, etc.)

📝 Notes

• Collection interval configured for 120 seconds (can be adjusted as needed)
• cAdvisor requires privileged mode to access container metrics
• Process Exporter uses pid: host to monitor host system processes

🤝 Contributing

Feel free to make improvements and adjustments to the project!

📄 License

This is a personal study project.