Import an existing workspace¶

This tutorial discovers supported Databricks resources and generates a starter workspace YAML file. Import is a review workflow, not automatic ownership transfer.

Prerequisites¶

live-readonly or live-apply execution mode;
successful workspace connectivity;
exactly one workspace declaration in scope.

Discover resources¶

datamuru import discover --config datamuru.yml

By default, DataMuru filters common system catalogs, schemas, and groups.

For machine-readable output:

datamuru import discover --config datamuru.yml --output json

Generate a starter file¶

Select catalogs explicitly:

datamuru import generate `
  --config datamuru.yml `
  --catalog existing_sales `
  --catalog existing_marketing `
  --out .\workspaces\imported-review.yml

Add --include-groups only if you intend to review group references.

Generate an Enterprise review suite¶

For enterprise brownfield onboarding, generate all review files together:

datamuru import generate `
  --config datamuru.yml `
  --catalog existing_sales `
  --suite-out .\import-review

This writes:

import-review\workspaces\imported-dev.yml
import-review\governance\rbac.imported.yml
import-review\governance\taxonomy.imported.yml
import-review\governance\masking.imported.yml

The suite attempts to include identity context and grant-derived RBAC when the workspace supports account SCIM and a SQL warehouse is configured. If those capabilities are unavailable, DataMuru still writes the catalog/schema review file and leaves the missing enterprise sections empty for manual curation.

Review before adoption¶

Inspect the generated file for:

system or vendor-managed objects;
resources owned by another team;
names that should remain external references;
missing managed locations or governance intent;
groups that DataMuru should not manage.

Generated YAML describes discovered shape. It does not establish that DataMuru has permission or authority to mutate every object.

Preview state adoption¶

Back up the existing local state file.
Keep execution_mode: live-readonly.
Move the reviewed YAML into the intended workspace scope.
Run validation, a targeted plan, and an adoption preview.

datamuru validate --config datamuru.yml
datamuru plan --config datamuru.yml --target catalog:existing_sales
datamuru import adopt --config datamuru.yml --target catalog:existing_sales

The preview must show only resources you intend DataMuru to manage. Adoption is blocked if a selected resource is missing live, differs from the declaration, or already has conflicting local state.

Commit ownership to state¶

datamuru import adopt `
  --config datamuru.yml `
  --target catalog:existing_sales `
  --auto-approve

Expected result:

Adopted 3 resources into state.

Re-run the targeted plan. Adopted resources should be no-op:

datamuru plan --config datamuru.yml --target catalog:existing_sales

Adoption writes local state only. It does not change Databricks resources. Do not switch to live-apply until the resulting plan contains only understood actions.

Include system objects only for diagnosis¶

--include-system can expose system, samples, workspace, information_schema, admins, and users. Do not adopt or delete these objects merely because they appear in discovery.