Manage environments
Containers, governance and lifecycle
Environments hold Dataverse, apps, flows and connections. Know the types: Default (one per tenant, cannot be deleted), Production, Sandbox (supports copy and reset for test/UAT), Developer (personal), and Trial, plus Managed Environments with extra governance features. Environment roles are separate from Dataverse security roles: Environment Admin manages the environment and Environment Maker can create resources, but you still need System Administrator or System Customizer inside Dataverse to change the data model.
Data Loss Prevention policies are tenant-level connector rules with three groups - Business, Non-business and Blocked. A Business connector cannot be used in the same app or flow as a Non-business one, and Blocked connectors cannot be used at all. Policies can be scoped to specific environments, so a question asking whether DLP can restrict one environment while leaving others open is answered yes.
Solution lifecycle: build unmanaged in development, export as managed, and import into test and production. Managed solutions cannot be edited in the target and uninstalling removes their components. Always use a publisher prefix and never build production assets in the Default environment.
Capacity is licensed across Dataverse database, file and log storage plus request limits; the admin centre shows consumption and supports backup/restore for sandbox environments and environment copy.
Rule of thumb: development holds unmanaged solutions, test and production hold managed ones - and configure DLP before you hand makers access, because retrofitting policies across live apps is far more painful than designing them up front.