Workspace lifecycle

The Lifecycle

The diagram summarizes the observed transition rows; declared edges without an observed writer remain outside the operational graph.

Transitions Explained

The structured transitions describe ordinary readiness and destruction changes and the explicit restore exception. Business status is a separate model; activation after a load preserves it and the stored first-activation time.

The loading-to-retirement edge is declared but unreachable through the request operation. It remains an unknown implementation path with a finding. Failure handling preserves partial work and permits forward retry; notification claims prevent repeated dispatch claims but do not guarantee delivery.

What Cannot Happen

Destruction may already have removed data, so purge failure never makes the workspace operational again.

Initialization can supply later participant defaults without re-entering provisioning.

A running provisioning attempt must finish or fail before retirement can be requested.

An outstanding retirement request cannot open an initial load.

A late provisioning failure consequence cannot undo committed activation.

Concurrency and Timing

Lifecycle services use locked or conditional updates to protect state changes. Cancellation and the destructive boundary compete for the same request/workspace locks. Repeating a lifecycle target preserves state but does not necessarily preserve version or modification time. The operator restore path bypasses the ordinary graph. Availability in other warm processes can remain cached for thirty seconds.