Assign checks

Open Dashboard and click + Add check. The wizard has three steps. You can return to an earlier step without losing the values already entered.

1. Check: target and locations

1. Choose the check type: HTTP(s), Ping, DNS, TCP, SMTP or RabbitMQ.
2. Enter a recognizable name and the target URL or host. Complete the fields shown for that check type.
3. In Run checks from, choose Agents, Countries, Regions or All regions.
4. For a selected category, move locations from Available to Selected with +. Use − to remove a selection.
5. Click Next.

If no suitable location is available, first add an agent or check its group, region, sharing and enabled state. See location selection for the difference between a region and an individual agent.

2. Settings: frequency and success conditions

Check interval and Timeout are in seconds. Retries controls how many failed attempts precede a down alert. Threshold timeout is in milliseconds; 0 disables the slow-response warning.

For HTTP(s), set Accepted Status Codes. Expand Request and connection, Response validation, Authentication or Proxy only when needed. Additional settings contains the optional expiration date. Set the check group, description and enabled state on the Notifications step. Review the fields shown for your selected check type before continuing.

3. Notifications: choose who responds

Select saved notification destinations, choose the incident priority, and optionally add a runbook link. A destination set to Disabled receives no alerts from this check. Create and test notification channels before assigning them.

Click Create check, then open the check on the Dashboard and wait for a new result from each selected location. A saved check or an initial UP label alone does not confirm that an agent has performed it.

Use the optional Group field to organize the check under a Dashboard check group. User groups and roles determine access separately.

Change an existing check

Use the card's edit action to change its target, locations or conditions. Clone creates a separate check that you can adjust before saving. The history action opens its alert history. Deleting a check removes it from monitoring; disable a temporary check when you intend to use it again.

Continue with the HTTP, DNS, Ping, TCP, SMTP or RabbitMQ guide, or learn to read check results.

Worked example: monitor a website from one agent

Use a website you control and an agent whose installation has completed. The values below are an initial test configuration; adjust them to the service's normal response time and your monitoring policy.

Scroll horizontally to view the full table.

Click Create check, open its name from the Dashboard and wait for the selected location's next result. Confirm that the timestamp is later than creation and that the recorded response matches the expected condition. Observe a further result at the configured interval to check that monitoring continues.

If the result is DOWN, read the error before changing success conditions. A redirect may need a different final URL; a certificate error needs trust or certificate correction; a timeout needs a connectivity or response-time check. Use the HTTP result guide for the reported condition.

Verify an edit or clone

1. After saving a changed target, location or condition, reopen the check and confirm the values were retained.
2. Wait for a new result from each intended location; old history belongs to earlier runs.
3. For a clone, give it a distinct name and review copied notification destinations before enabling it.
4. For a temporary test, disable or delete it when verification is complete.

If a field prevents saving, return to the highlighted wizard step and correct it. At least one valid monitoring location and the required type-specific fields must be present. A selected region chooses a location within that region; select individual agents when every specific host must run the check.