Cloud Status Page
What customers see
The cloud status page is a public page for customers, with no sign-in. It opens with a single overall banner that states the estate's worst current state in plain terms: all systems operational, degraded, or an outage in progress. Below it sits a short list of named components, grouped by stack, each showing one coarse state. A customer reads their service's health at a glance without needing to know what a sensor is.
Components
A component is a customer-facing name - "Cloud Receiver Connections", "Central Station Automation" - that stands in front of a set of sensors the operators watch. The mapping from a component to the sensors behind it lives only in SBN Monitor's central configuration. The page deliberately shows component names, their state, and timestamps. A customer learns that a service is degraded.
Automatic incident history
Underneath the components is a history of incidents. Each entry is derived automatically from the components' own state changes: when a component leaves operational the page opens an incident, and when it returns the page closes it and records how long it lasted. No one authors these by hand - the history is a faithful, timestamped account of what customers could actually observe. Operator-written commentary on an incident is a possible later addition; today's history is automatic.
Planned maintenance
Scheduled work is shown ahead of time. A maintenance window that has been marked for public display appears on the status page as an upcoming or in-progress maintenance notice, so a customer who sees a component dip during planned work understands it was expected rather than a fault. Windows not marked for public display never appear here.
How it differs from the operator dashboard
The status page and the operator dashboard are built from the same live state but serve opposite audiences. The dashboard is a signed-in operator tool: every sensor, every host, full messages and readings, snooze and pause controls. The status page is public and deliberately thin - components, coarse states, timestamps, and incident and maintenance summaries, with no control of any kind. The thinness of the status page is a designed boundary, enforced where the page is built.
Availability
Because it is what customers reach during an incident, the status page is served by both cores - the active one and the standby - so it stays up even if the active core is having the very trouble customers are checking on. It is a lightweight page that holds up under a rush of visitors during an outage.