Export Metrics to Prometheus

The console serves its metrics at /api/metrics in Prometheus format, so Prometheus, Grafana Agent or anything else that reads that format can scrape them into an existing monitoring stack. The metrics cover:

• hosts and VMs (CPU, memory, disk, network),
• datastores,
• alarms,
• tasks,
• notification deliveries.

A host's disk and network figures are the totals of that host's VMs.

Set it up

1. Settings → Metrics Export → Generate token (Super Admins only).
2. Copy the token now. It's stored only as a hash and isn't shown again.

3. The card shows a ready-made scrape job for your console's address. Add it to prometheus.yml:

   scrape_configs:
     - job_name: mfconsole
       scheme: https
       metrics_path: /api/metrics
       authorization:
         credentials: <scrape token>
       static_configs:
         - targets: ['console.example.com:443']
   

4. Reload Prometheus and check the target is UP.

The scrape token can read metrics and nothing else. Super Admins can also scrape with their own sign-in or API key, but a dedicated token is the better choice: it can't do anything else if it leaks.

Replace or revoke the token

• Generate token again replaces the current one. Prometheus gets a 401 until you give it the new token.
• Revoke removes it. Prometheus gets a 401 until you generate a new one.

The card shows who generated the current token and when. Generating and revoking are recorded in the audit log.

Certificates

If the console uses its own self-signed certificate, add the CA to Prometheus with tls_config: { ca_file: … }. Avoid insecure_skip_verify.

See also

• Notifications & Alarms
• Default Ports