Which check catches an unavailable page or a missed job?
An uptime check asks a web address for a response. A heartbeat waits for a scheduled job to send a short message. Use separate checks for a report download page and the job that writes the report. A successful page request alone cannot tell you whether the latest file exists.
Send a success signal only after the job saves and checks its result. Keep the heartbeat address out of public material.
When does an internal alert become a customer update?
OnlineOrNot can create and resolve an incident when you link a check to a status page component, a named service visitors can recognize. You can also manage incidents yourself. Before connecting a check, decide whether every failure on it deserves a customer-facing update.
The pricing table lists private status pages on paid plans, with availability and charges by plan. Check audience and plan separately from where team alerts are delivered.
What does Hobby include and what should you keep yourself?
Hobby lists up to three uptime monitors, a three-minute check interval and 14 days of data retention. Pro counts browser check runs separately from uptime monitors. Keep check targets, alert contacts and important incident records outside the dashboard if you may need them later.
A developer can read checks and results through the API. That is different from a complete archive. Check which dates and fields are available before relying on a move.
What would you publish when a small part of the app fails?
Draw a fictional download page, report job and status page. Use no account or actual notification.
- Keep the incident smallMark the site as reachable and the sample report as late. Write a short customer update only if the delay affects what they can do.
- Draw the automatic pathIf a linked check automatically opens and closes an incident, show who would see it. Keep internal details and recipients off the public drawing.