📊 Monitoring Stack with Coolify, Grafana & Elasticsearch

This project provides a fully automated, containerized monitoring solution designed specifically for Coolify. It features log aggregation via Elasticsearch, metric collection via Prometheus, and a pre-configured Grafana dashboard suite.

🚀 Features

• Log Management: Centralized logs for all your containers.
• Metric Visualization: Real-time system and container metrics.
• Auto-Provisioning: Dashboards for Elasticsearch, Prometheus, and Node Exporter are automatically downloaded and configured on first boot.
• Coolify Native: Integrated health checks and automatic Traefik routing.

⚠️ Important Note

Use at your own risk.

This configuration is provided "as-is" without warranty of any kind. By deploying this stack, you acknowledge that I am not responsible for lost data, security breaches, server meltdowns, or any other digital catastrophes that may occur. A few friendly reminders:

• Passwords: If you lose your ELASTIC_PASSWORD, don't come knocking—you’ll have to wipe the volumes and start over.
• Resources: Elasticsearch is a memory-hungry beast. If your VPS starts sweating, give it more RAM or adjust the ES_HEAP_SIZE.
• Security: This stack is configured for ease of use. If you expose Elasticsearch's port 9200 to the public internet without extra hardening, you're essentially leaving your front door open in a bad neighborhood.

---

🛠️ Setup Instructions

1. Environment Variables

Before deploying, add the following variables to your Coolify Service Stack:

Variable	Description
ELASTIC_PASSWORD	The master password for Elasticsearch.
GRAFANA_ADMIN_USER	Admin username (default: admin).
GRAFANA_ADMIN_PASSWORD	Admin password for Grafana UI.
SERVICE_FQDN_GRAFANA	Your full URL (e.g., https://stats.example.com).
ES_HEAP_SIZE	Memory for Elasticsearch (e.g., 2g).

2. Deployment

1. Paste the docker-compose.yaml into your Coolify stack.
2. Ensure you have deleted any old volumes (elasticsearch-data, grafana-data) if you are resetting passwords.
3. Click Deploy.

---

📁 Included Dashboards

The stack includes a dashboard-loader sidecar that automatically provisions:

• Node Exporter Full: Detailed OS metrics (CPU, Memory, Disk, Network).
• Elasticsearch Stats: Cluster health, shard status, and indexing rates.
• Prometheus 2.0: Scraper health and performance metrics.

---

🔍 Troubleshooting

"No date field named @timestamp found"

This error appears in Grafana when Elasticsearch is empty.

• Fix: This will disappear automatically once your containers begin sending logs to Elasticsearch. Ensure your other applications are configured to log to this stack.

"401 Unauthorized" (Elasticsearch)

This happens if the ELASTIC_PASSWORD was changed after the first deployment.

• Fix: Elasticsearch only sets the password on the first boot. To update it, you must go to the Storage tab in Coolify, delete the elasticsearch-data volume, and redeploy.

Dashboards Missing

The dashboard-loader container runs once to download JSON files.

• Fix: Check the logs of the dashboard-loader container. If it failed to download, ensure your server has outbound internet access to grafana.com.

---

🛡️ Security Note

Elasticsearch is configured with xpack.security.enabled=true but http.ssl.enabled=false. It is intended to run inside a private Docker network. Do not expose Port 9200 to the public internet without setting up SSL/TLS.