Workspace configuration¶

Each workspaces/*.yml file must contain a top-level workspace mapping.

Workspace fields¶

Field	Type	Required	Description
`name`	string	yes	Logical workspace name
`cloud`	enum	yes	`azure`, `aws`, or `gcp`
`region`	string	yes	Provider region label
`tier`	string	no	Informational workspace tier
`catalogs`	list	no	Catalog declarations
`principals`	mapping	no	User, group, and service-principal declarations

Catalogs¶

A catalog may be a string in internal normalization or an object:

- name: analytics
  managed_location: abfss://container@account.dfs.core.windows.net/analytics
  use_default_storage: false
  schemas:
    - raw
    - name: curated
      managed_location: abfss://container@account.dfs.core.windows.net/analytics/curated

Use either managed_location or use_default_storage according to your Databricks setup.

Principals¶

Principals must be nested under workspace.

String entries are existing references:

principals:
  groups:
    - existing-data-consumers

Object entries support lifecycle metadata:

principals:
  users:
    - email: analyst@company.com
      display_name: Data Analyst
      lifecycle: managed
      allow_delete: false

Lifecycle values are existing, managed, and external. Managed lifecycle is an Enterprise feature and requires account SCIM capability.