Home / DE Associate practice test / Implementing CI/CD / Question

Databricks Data Engineer Associate · free practice question

In a CI pipeline running on a pull request, the team wants to confirm an Automation Bundle's configuration is syntactically valid and all referenced files…

Exam-style question on Implementing CI/CD from the free DE Associate practice test, with the answer and a full explanation below.

Updated November 10, 2026

Implementing CI/CD

In a CI pipeline running on a pull request, the team wants to confirm an Automation Bundle's configuration is syntactically valid and all referenced files exist, WITHOUT connecting to or changing any Databricks workspace. Which Databricks CLI command is appropriate for this CI step?

  1. Adatabricks bundle validate, which checks the bundle's configuration syntax, schema compliance, and all file references entirely on the local machine without connecting to or making any change to a workspace.
  2. Bdatabricks bundle deploy -t prod, which is safe to run inside CI because in a pull-request context it only simulates the deployment and never actually creates, updates, or mutates any workspace resources.
  3. Cdatabricks bundle plan, which previews the exact set of changes the bundle would make as a Terraform-style diff and is the officially recommended command for validating bundle configuration inside a CI pipeline.
  4. Ddatabricks bundle apply, which reconciles the bundle's declared state and validates the configuration thoroughly while guaranteeing that no changes are ever pushed out to the configured target workspace.
  5. Edatabricks bundle run, which executes the bundle's defined job in a special dry-run mode so that the configuration is exercised end-to-end and validated without producing any real side effects.
Answer & explanation

Correct answer: A

WHY A: databricks bundle validate performs local checks — syntax, schema compliance, variable and file references — without contacting a workspace, exactly what a PR-triggered CI check needs. WHY NOT B: deploy actually creates or updates workspace resources; it does not merely simulate, even on a pull request. WHY NOT C: bundle plan is a real command, but it contacts the target workspace to compute a Terraform-style diff of pending changes, so it violates the 'without connecting to a workspace' requirement that validate meets locally. WHY NOT D: bundle apply is not a bundle subcommand — the apply step is performed by deploy, not by a local validator. WHY NOT E: bundle run triggers an already-deployed job; it is not a configuration validator and has no dry-run validation mode.

Take the full DE Associate practice test free →

Would you pass today?

This question is one of 45 in the free, timed Databricks Data Engineer Associate practice test. Take it and get a section-by-section score.

Start the free test →