Connect your service

In this test, Sauron reads measurements from your service's /metrics address; you show the result in Grafana and send a test alert. Separately, you send two logs and find them in OpenSearch.

New to the tools? Explore your Sauron.

Before you start

Confirm your test Sauron in Console Endpoints. Get:

  • Service owner: a /metrics address the test Sauron can reach.
  • Sauron owner: API edit and OpenSearch write access, plus an approved test receiver whose route matches the alert rule.
  • Grafana owner: Editor access and a writable dashboard folder.

Ask the listed owner for anything missing. Confirm the alert recipient agrees to a test message before upload.

A simple path through the setup

  1. Connect a metrics endpoint. Follow the direct scrape test using the name sauron-first-metric-test. Check: Prometheus Targets shows UP and up{job="sauron-first-metric-test"} returns 1.
  2. Send two logs. Follow the first log test. Check: The upload response says "errors": false and a search finds both records in OpenSearch.
  3. Save a dashboard. Create a Grafana panel titled Metrics endpoint reachable with the same up query. Check: Reopen the saved dashboard; the panel shows 1 while the scrape succeeds.
  4. Test an alert. Have your Sauron owner confirm the alert will reach an approved test destination. Follow the first alert test. Check: The rule is Firing, appears in Alertmanager, and the recipient gets the message. Delete the temporary rule afterward.

up checks whether Sauron can read the configured metrics address; it does not prove service health. Next, set up ongoing logs or a service-health alert with your owners.

Which guide should I open?

For restricted networks, see PushProx or Shuttleproxy. For other starting points, see OCI metrics or automated client access.

For a missing result, use Solve a problem. If blocked, send the Sauron URL and checks completed to #sauron-support, without secrets.