Monitoring

Monitoring is pull-based probing from Mydle workers against targets you configure. Quiet charts mean no checks have run yet, not a fake green score.

• Rules — one monitor per target and kind (HTTP, SSL, or DNS).
• Checks — scheduled executions with pass/fail and timing from the probe.
• Health — project health contributors built only from observed results.
• Alerts — opened (and optionally notified) when evaluation rules match failures.

Create an HTTP monitor in a few minutes. Status and uptime appear only after the first real check — nothing is invented before that.

1. 1

   Step 1: Open Monitoring

   Choose Monitoring in the sidebar. Until you add a monitor the list stays empty. Use Add monitor or Create monitor.

   

2. 2

   Step 2: Configure the HTTP check

   Enter the public URL (for example https://example.com/health), an optional name, the expected status (default 200), and how often to check. Free runs at most every 10 minutes; Pro unlocks shorter intervals.

   

   The form also explains Email alerts (workspace members) and SSL and DNS — the hostname is checked once a day for certificate expiry and DNS resolution, separately from the HTTP probe. Click Create monitor to save; the first check starts right after.

3. 3

   Step 3: Verify the first check

   Open the monitor detail under /app/monitoring. After the probe runs you will see pass/fail and timing from the real response. Quiet charts mean checks have not run yet (or the scheduler is delayed) — not a fake green score.

4. 4

   Step 4: If checks do not run

   • Continuous scheduling needs production Trigger workers. Without them, monitors exist as configuration but may not run on a cadence.
   • Private or link-local URLs are rejected (SSRF-safe hostnames only).
   • Only GET and HEAD — no request bodies or custom auth headers.

• HTTP — availability and response expectations (Free and above).
• SSL — daily certificate validity and expiry checks for each active HTTP monitor’s hostname.
• DNS — daily DNS resolution checks for each active HTTP monitor’s hostname.

Setup catalog entries marked Coming soon (store listings, Firebase, privacy, app-ads) cannot be activated — they do not score health.

Continuous scheduling requires production Trigger workers. Without them, monitors exist as configuration but do not run on a cadence. Local or manual probe paths may exist for development; treat empty check history as honest, not estimated.

Check results feed project health, alerts, incidents, and status page components when those features are entitled and wired.

• Free — 3 HTTP monitors every 10 minutes.
• Pro — 10 HTTP monitors every 5 minutes, alert integrations, incidents, more than one status page, and a custom domain. Free includes one status page. Both plans include daily SSL and DNS checks and outage emails. See Limits & Quotas.

Continuous execution still needs Trigger credentials in production on every plan that schedules checks.

• HTTP Monitoring
• SSL Monitoring
• DNS Monitoring
• Health & Checks
• Alerts
• In the product: /app → project → Monitoring · API: OpenAPI (Enterprise API keys).