← Back to help centre

Notifications

How WebCheck notification retries work

See what WebCheck retries after an alert delivery fails and what to check when a notification does not arrive.

Check retries and notification retries are separate. An HTTP connection failure can be retried during a check run; sending an alert happens later through a background queue.

What is retried

Incident and recovery email alerts use a queued job configured for up to three attempts. When an email send fails and the job reports that failure, the queue uses back-offs of 30 seconds and then 5 minutes before its next attempts.

This does not mean every destination is automatically sent again. A failed incident webhook is recorded as failed, but its delivery error does not currently trigger another queue attempt. Certificate and domain-expiry event senders also record destination errors without automatically resending that destination.

What you can check

The site detail page does not show notification delivery records or provide a manual retry control. Confirm the Account email address, destination settings and receiving service logs. For a webhook, check that the endpoint is reachable and accepts the request; a non-success response is recorded as a failed delivery.

For the separate connection retries that happen before an incident is confirmed, see How WebCheck confirms an incident. For destination setup, see Setting up WebCheck notifications.