Question 330 A feature branch has new commits locally in the Databricks Git folder, but the remote repository does not show them. Which action is missing? Increase Spark executor count Push the branch commits to the remote repository Run VACUUM on a Delta table Rename the Git folder without pushing Explanation Correct Answer B:Local commits are not automatically visible in the remote until pushed. Incorrect Answers A:Compute scaling cannot transfer source-control commits. C:Delta cleanup does not synchronize Git commits. D:A folder rename does not upload the missing commits.
Question 331 A developer is about to switch from one Git branch to another and has uncommitted notebook changes. What should be considered first? Commit, stash, or otherwise preserve the changes safely Grant the developer metastore admin Drop every production Delta table Restart all production jobs Explanation Correct Answer A:Uncommitted edits can conflict with branch switching and should be handled deliberately. Incorrect Answers B:Broad catalog rights do not preserve uncommitted source edits. C:Data deletion is unrelated to Git working tree changes. D:Job restarts are not a version-control preservation strategy.
Question 332 A team deploys the same Lakeflow Job definition into dev, test and prod, but requires different workspace URLs and compute sizes. What design is suitable? Keep separate manually edited job definitions for each environment Use Declarative Automation Bundle targets with environment-specific variables and overrides Hardcode production credentials in the dev notebook Create a different manually edited codebase for each environment Explanation Correct Answer B:One codebase can be promoted while target configuration supplies the intended differences. Incorrect Answers A:Multiple unlinked copies drift over time instead of sharing a declarative definition with target-specific overrides. C:Embedded secrets are unsafe and couple environments. D:Multiple divergent copies increase drift and make promotions harder to audit.
Question 333 A CI job must catch invalid Declarative Automation Bundle configuration before creating workspace resources. Which CLI action is appropriate? DROP TABLE in Unity Catalog databricks jobs run-now on an arbitrary ID databricks bundle validate databricks bundle destroy Explanation Correct Answer C:Validation checks the bundle configuration before deployment. Incorrect Answers A:Deleting data objects does not check bundle YAML syntax or resource references. B:Running a job does not validate the whole bundle definition. D:Destroy removes deployed resources rather than validating a new definition.
Question 334 A validated Declarative Automation Bundle should create or update its defined job resources in the test environment. What CLI operation should the pipeline run? databricks bundle destroy -t test databricks bundle validate -t test only databricks jobs list only databricks bundle deploy -t test Explanation Correct Answer D:Deploy applies the bundle resource definitions to the selected test target. Incorrect Answers A:Destroy removes deployed resources, the opposite of creating them. B:Validation checks configuration but does not by itself deploy resources. C:Listing existing jobs does not apply bundle definitions.
Question 335 A bundle defines a job but CI executes only `bundle validate`, then asserts the new job exists in the workspace. What is missing? A bundle deploy step against the intended target An extra Git commit with no deployment action Run `bundle summary` and assume it provisions the missing job A new data table owner grant Explanation Correct Answer A:Validation does not create or update the bundle-defined workspace resources. Incorrect Answers B:Committing source alone does not apply resources to the workspace. C:Summary reports deployment resource details but is not the command that deploys the bundle definition. D:Data permissions are not equivalent to deploying job configuration.
Question 336 An engineer wants the CI/CD system to launch a job already defined and deployed in a bundle. What must happen before a bundle run action can succeed? The table schema must be dropped and recreated The bundle must be destroyed after validation The bundle resources must be deployed to the selected target Every notebook must be manually opened by an analyst Explanation Correct Answer C:The intended job must exist in that environment before orchestration can run it. Incorrect Answers A:Data schema deletion is unrelated to launching a bundle-defined job. B:Destroying removes resources needed for execution. D:A CI/CD run need not depend on user clicking notebooks.
Question 337 A team wants a declarative source-controlled definition of jobs and pipelines rather than manual click-by-click creation. Which capability fits? Only a local Python variable named jobs Ad hoc workspace clicks with no version-controlled record Declarative Automation Bundles resources configuration Keep a manually edited job in production and export its definition only after every change Explanation Correct Answer C:Bundles package workspace resources as deployable configuration alongside code. Incorrect Answers A:An in-memory variable is not an environment-targeted deployment artifact. B:Manual changes are harder to reproduce and compare across environments. D:Post-change exports do not make a deployment reproducible from source-controlled declarative resources; they preserve manual drift.
Question 338 A team uses a single bundle project for dev and prod. Production requires a distinct job schedule, while the Python ETL code must remain identical. What should be varied? The main Python transformation algorithm for production only The source table business keys The production target resource override for the schedule The Git history to omit dev commits Explanation Correct Answer C:Target overrides can supply environment-specific schedule behavior without forking ETL source. Incorrect Answers A:Forking business code introduces avoidable environment drift. B:Changing key definitions is not a deployment schedule configuration. D:Rewriting history is not necessary to set target-specific schedule parameters.
Question 339 A production bundle accidentally points at a development workspace URL. Which configuration should the CI pipeline check? Validate the bundle resource schema without checking the resolved deployment target or workspace URL The selected target and its workspace host mapping Compare development and production job resource names but skip checking the workspace hosts Execute bundle validate only for the development target and assume production points at a different host Explanation Correct Answer B:Target selection must resolve to the intended environment to prevent misdeployment. Incorrect Answers A:A schema-valid bundle can still select the wrong environment; the actual target-to-host mapping is the relevant safety check. C:Similar resource names across environments do not prove that the selected target resolves to the production workspace. D:Validation of one environment does not prove the selected production target is mapped to the intended workspace host.
Question 340 A CI runner authenticates to Databricks using a personal engineer token that will expire when the engineer leaves. What is a safer production pattern? Commit the personal token into the repository for permanence Use an approved automation identity or service principal with scoped permissions Embed the engineer's personal access token in a protected branch of the repository Share the token with all notebook users Explanation Correct Answer B:A dedicated identity gives CI/CD durable accountable access without depending on a person's credentials. Incorrect Answers A:Hardcoding a token exposes a secret and does not solve personnel ownership safely. C:Branch protection does not address the personal token lifecycle or keep plaintext deployment credentials out of tracked files. D:Broad secret sharing increases security risk.