Operations

Component health, service objectives and incidents.

Operations is about the state of the running system: are the components healthy, are the service objectives being met, is there an open incident, where has an install in a tenant stopped.

This section covers the platform's Operations screen and the tenant operations screen, and how an incident is opened and closed. A component with no measurement is marked as a “blind spot”: it isn't healthy, its state is unknown. This distinction is made on purpose.

Platform health is monitored in the platform scope. Opening, taking on and closing an incident requires operations.write; operations.read is enough to only monitor.

Articles in Operations

Screen guideOperationsShows the platform's component health, service objectives and open incidents. The platform operator runs checks and manages incidents here.
Screen guideOperationsShows which capabilities are installed in this Tenant, which are installing or stuck, and the status of platform components that also affect this Tenant.
How-toOpen, claim and close an incidentOn the Operations screen you open an incident with a title, a severity and the affected components; someone claims it, and it is closed once resolved.