Skip to main content
The System page gives you a real-time view of your server’s resource usage and current status. Use it to spot performance issues, verify that services are running, and plan storage before large downloads. Open it by going to Settings → System.

​
What you can monitor

CPU usage

Current CPU utilization as a percentage. High sustained usage may indicate a service is under heavy load or stuck.

Memory usage

Current RAM consumption. AI inference is the most memory-intensive workload — large models require 16–32 GB or more.

Storage usage

Disk space used and available. N.O.M.A.D. prefers to display real block devices (e.g. /dev/sda) rather than temporary filesystems like tmpfs, so the figure reflects your actual usable storage.

System uptime

How long the server has been running since its last restart.
Check storage here before downloading large ZIM files or map regions. Full Wikipedia is approximately 95 GB; US state maps are 2–3 GB each.

​
Service logs and metrics

For deeper visibility into individual services, N.O.M.A.D. can run Dozzle — a lightweight container log viewer that streams live logs from all running Docker containers. If Dozzle is installed, access it at: 
http://YOUR_SERVER_IP:9999
Dozzle shows real-time log output and basic container metrics for every nomad_* container. It is useful for diagnosing errors that don’t surface in the Command Center UI.

​
Health check endpoint

You can programmatically verify that the N.O.M.A.D. Command Center is running by querying the health endpoint: 
GET /api/health
A running server returns HTTP 200. Use this in monitoring scripts or uptime checks to confirm the server is responsive. Example: 
curl -f http://YOUR_SERVER_IP:8080/api/health
A successful response confirms the Command Center is up. If the request fails or returns an error code, the server may still be starting up — wait 1–2 minutes and try again.

​
Troubleshooting with system health

  1. Check Settings → System to see if the service shows as running.
  2. Go to Settings → Apps and try Restart on the affected tool.
  3. Wait 30 seconds after restarting, then try accessing the feature again.
  4. If the issue persists, check container logs in Dozzle (port 9999) for error messages.
High CPU usage during AI inference is normal on CPU-only systems. To improve speed, add an NVIDIA GPU and Force Reinstall the AI Assistant from Settings → Apps. See the AI Assistant page for GPU setup instructions.
Go to Settings → Content Manager to delete ZIM files you no longer need. Go to Settings → Maps Manager to remove map regions. Unused content is the most common cause of full storage.

Apps Manager

Start, stop, restart, and manage installed tools.

Content Manager

Delete ZIM content to free up storage.

Maps Manager

View and delete downloaded map regions.

Helper Scripts

Start or stop all services from the command line.