DADA Cloud
All guides

Monitoring: metrics, logs, and alerts

What it's for

Know whether your service is alive, see where a problem is, and get alerted before your users notice — for your own team, or as a reliability guarantee you can show clients.

How to set it up

  1. Go to Monitoring → Create Monitoring App, give it a name.
  2. You'll immediately see an onboarding card with 3 steps:
    • Step 1 (done automatically): the monitoring app is created.
    • Step 2: a one-shot API key is shown once — copy it now (there is no way to view it again later; losing it means creating a new key via the app's own page).
    • Step 3: instrumentation snippets in Node.js, Python, Env vars, or curl tabs — each pre-filled with your API key and the ingest endpoint. service.instance.id is optional and becomes a filterable "source" label.
  3. The card's status updates automatically: Checking… → Waiting for first telemetry → Receiving (polled every few seconds, no refresh needed).
  4. Once data arrives, open the app for Overview (health, last-seen, 15-minute error rate, firing alerts), Metrics, Logs, and Alerts tabs.

Alerts and notifications

  1. On an app's Alerts tab, add an Alert Rule: pick the metric, condition, threshold, duration, and a notification channel (or "— none —" to just record the firing state).
  2. Configure Notification Channels separately:
    • Telegram — bot token + chat ID.
    • Email — comma-separated addresses.
    • Webhook — a URL.
  3. Use Open in Grafana on the app for a deeper, ad-hoc view of the same data.

Gotchas

  • The API key is shown exactly once, at creation time. If you lose it, you can't recover it from the console — you'll need to generate a new one from the app's page (check its "manage keys" link) and re-instrument.
  • Monitoring is opt-in and manual today: deploying an app via GitHub/image does not automatically create a monitoring app for it. You have to create one yourself and wire the OTLP SDK/env vars in.
  • The curl test snippet uses the current clock second as the counter value specifically so the console renders a real rate instead of a flat line — re-run it a few times if you want to see a moving chart.

Not yet supported

  • Auto-attaching monitoring to a newly deployed app (still a manual "Create Monitoring App" step, separate from deploy).
  • Custom dashboards in the UI (the capability exists in the backend but is not exposed here — use Open in Grafana for anything beyond the built-in Overview/Metrics/Logs/Alerts tabs).
All guides