Getting started
Concepts
Regions, confirmation, intervals, states, and the vocabulary StatusCheck uses everywhere.
A few ideas show up on every screen in StatusCheck. Understanding them makes the rest of the product predictable.
Projects
Section titled “Projects”A project is the tenant. It owns monitors, channels, status pages, members, and a subscription. Most companies use one project; agencies and platform teams often use one per client or per product.
Monitors and checks
Section titled “Monitors and checks”A monitor is a thing you want to watch, configured once: a target, a check type, an interval, and a set of regions. A check is a single run of that monitor from a single region. A monitor that runs every 60 seconds from five regions produces five checks a minute.
Every check records timing detail, not just up or down. For HTTP that is the full waterfall of DNS lookup, TCP connect, TLS handshake, time to first byte and download, plus the final status code, redirects followed, and whether any content assertion passed.
Regions
Section titled “Regions”StatusCheck runs probes in 18 regions across the Americas, Europe, Asia-Pacific, and the Middle East & Africa. Each region is an independent machine with its own network path to your target. You choose which regions each monitor uses. The regions page lists them all.
Confirmation
Section titled “Confirmation”A single region reporting a failure does not make a monitor down. StatusCheck waits for a second region to see the same failure before it changes the monitor’s state and sends alerts. This is what stops one provider’s network incident, or a transient DNS hiccup, from waking you up.
Confirmation happens at the next scheduled check from another region, so the time from first failure to alert is typically one interval or less. Monitors with a single region skip confirmation, and the dashboard says so.
States
Section titled “States”Every monitor is in exactly one state, and the same vocabulary is used on the dashboard, in alerts, and on status pages.
| State | Meaning |
|---|---|
| Up | The most recent confirmed checks succeeded. |
| Degraded | Checks succeed but exceed the response-time threshold you set, or an SSL certificate is inside its expiry warning window. |
| Down | A failure has been confirmed from more than one region (or from the monitor’s only region). |
| Maintenance | A scheduled maintenance window covers this monitor; failures are recorded but do not alert. |
| Paused | You paused the monitor. No checks run. |
Status is never communicated by color alone. Each state has a label and an icon so it survives color blindness and monochrome displays.
Incidents
Section titled “Incidents”An incident is a period during which one or more monitors were not up. StatusCheck opens one automatically when a failure is confirmed and closes it when the recovery is confirmed. You can also open incidents manually, for example for a problem your monitors can’t see, and attach them to status-page components. Incidents carry a timeline of updates that is posted to the status page and to subscribers.
Maintenance windows
Section titled “Maintenance windows”A maintenance window is a scheduled period during which affected monitors do not alert and the status page shows “Under maintenance” instead of “Down”. Schedule them ahead of time so your customers hear from you before their monitoring does.
Status pages
Section titled “Status pages”A status page is a public (or private) view of your project’s health, organized into components your customers recognize, such as “API”, “Dashboard” and “Payments”, rather than internal monitor names. Each component can be backed by one or more monitors, or updated manually. Subscribers can follow the page by email, RSS, or Slack.