Customer communication
Publish monitor-backed status without hiding what the evidence says.
A status page turns selected monitor health into a customer-facing board, with explicit incident updates, maintenance windows, and confirmed email subscriptions.
Configure the page
- Open Status pages. Set the public name, headline, maintenance-window note, and update policy.
- Choose access. Keep the page public or enable password protection and set a shared password. Leaving the password blank on a later save preserves the existing password.
- Choose subscriptions. Enable email subscriptions when visitors should receive incident and maintenance updates through confirmed double opt-in.
- Attach monitors. Add the uptime and heartbeat monitors customers should see. Use customer-safe component names that describe the service outcome, not command names or infrastructure details.
- Open the public page. Verify access, component order, current state, recent updates, and the subscription form in a private browser session.
How component and page state is derived
Monitor-backed components
Each attached component follows the linked monitor. Configuration or probe-infrastructure failures remain operator-visible but do not present the target as down.
Evidence before recovery
A successful production run can restore operational state. An operator override leaves the component and page at monitoring until fresh evidence arrives.
Independent services
One healthy component can remain operational while another reports degraded service or an outage. The page aggregate reflects the most serious visible component state.
Maintenance truth
A maintenance window can hold selected outbound alerts, but monitoring, incidents, component health, and the public board continue to show the observed state.
Publish an incident update
- From the status-page editor, write a short title and customer-facing body; or publish from an incident's Status-page update panel.
- State the observed customer impact, what is being investigated, and when the next update is expected. Do not expose tokens, payloads, customer records, raw errors, or infrastructure identifiers.
- Review the exact status page and content before publishing. Publication is immediate and creates a durable update in page history.
- After publishing, open the public page and verify the message, timestamp, and component state independently.
Schedule or cancel maintenance
Describe the window
Enter a customer-facing title and summary, start and end time, and expected impact.
Choose affected monitors
Select the status components covered by the maintenance event.
Choose alert behavior
Optionally hold selected incident or escalation alerts during the window. Still-open alerts are reconsidered after the window ends.
Verify publication
Confirm the maintenance announcement appears on the intended public page with the correct times and impact.
Cancel with context
Provide a customer-facing cancellation reason. The history remains available instead of silently removing the event.
Verify subscriber delivery
- A visitor submits an email address and must use the signed confirmation link before becoming an active subscriber.
- Signed unsubscribe links remove future delivery without requiring a product account.
- The editor shows subscriber delivery evidence separately from operator alert delivery.
- Provider acceptance proves that the mail transport accepted the message; it does not prove inbox placement, reading, or human response.
- Evidence exports omit subscriber email addresses while retaining delivery state and timing.