Prepstellar
Sign in Get it on Google Play Download on the App Store

DP-700 · Practice set 11 of 17

Warehouse Database Projects: 10 practice questions

10 questions · Untimed · Free

10 free DP-700 practice questions on Warehouse Database Projects, with an explanation for every answer. Untimed. The full mock exam and the timed version are in the app.

Set 11 · Warehouse Database Projects · 10 questions Read the lesson

  1. Question 1 of 10

    A developer creates a new database project for a Fabric warehouse in Visual Studio Code. Which settings does the project need?

    1. ATarget platform Synapse Data Warehouse in Microsoft Fabric, as a project that is not SDK-style
    2. BAny target platform, as long as the project is created from an existing warehouse
    3. CTarget platform Azure SQL Database, as an SDK-style project that uses the warehouse connection
    4. DTarget platform Synapse Data Warehouse in Microsoft Fabric, as an SDK-style project
    Show the answer

    Two answers in the wizard are fixed for a warehouse. The target platform names the engine the project is validated for, and the project format must be SDK-style, because Fabric Data Warehouse supports no other format.

    Next → 1 / 10

  2. Question 2 of 10

    A developer wants to build a database project in Visual Studio Code and publish it to an existing Warehouse item. What is the lowest workspace role that allows this?

    1. AContributor
    2. BAdmin
    3. CMember
    4. DViewer
    Show the answer

    The requirement is Contributor or higher permissions in the workspace of the warehouse. Start from the idea that publishing changes the schema, so a read-only role is not enough, and then remember that the bar is the first role above it.

    Next → 2 / 10

  3. Question 3 of 10

    A developer is about to deploy a warehouse project to a test warehouse for the first time and wants to read the T-SQL that would be applied, without changing the warehouse. What should the developer do?

    1. AChoose the Generate Script action.
    2. BChoose the Publish action with a saved profile.
    3. CBuild the project and open the dacpac file.
    4. DChoose Publish with `BlockOnPossibleDataLoss` True.
    Show the answer

    The publish steps end with a choice between two actions. One writes to the warehouse and the other only produces the T-SQL for review. For a first deployment, review comes first.

    Next → 3 / 10

  4. Question 4 of 10

    A warehouse project builds without errors. Publish then fails with error SQL72033: a procedure depends on `[sys].[views]`, which does not exist in the target database. What should the developer fix in the `.sqlproj` file?

    1. AAdd `DatabaseVariableLiteralValue` set to `master` to the package reference.
    2. BAdd a `PreDeploy` script that creates the missing `sys.views` object.
    3. CRemove the reference to the `Microsoft.SqlServer.Dacpacs.FabricDw` package.
    4. DChange the target platform to SQL Server so that system views resolve.
    Show the answer

    The pattern is build fine, publish fails, and a `sys` object that supposedly does not exist. It points to the reference to the system objects package. That reference needs the `DatabaseVariableLiteralValue` metadata with the value `master`.

    Next → 4 / 10

  5. Question 5 of 10

    A team versions a warehouse with Fabric Git integration. Table permissions that were granted in the development warehouse must also exist in the test warehouse. What does the team need?

    1. AA commit of the warehouse, which exports its permissions
    2. BA post-deployment script in a SQL database project
    3. CA selective commit that holds only the permission objects
    4. DThe `XMLA.json` file, which Git stores with the warehouse
    Show the answer

    Warehouse Git integration carries the schema, and SQL security features such as permissions are outside it. They need a script-based approach, and the usual place is a post-deployment script in a database project.

    Next → 5 / 10

  6. Keep the ones you got wrong

    In the app, every question you miss comes back exactly when you’re about to forget it.

    Opens in your browser. Nothing to install.

    Start studying

    On your phone, with the same account

    Get it on Google Play Download on the App Store

  7. Question 6 of 10

    In a Git-connected workspace, a developer changed three views and one table in the same warehouse. Only the views are ready for review. What can the developer commit from the Source control panel?

    1. AThe three views, selected one at a time
    2. BThe warehouse item, with all four changes
    3. CThe folder of views, as a single unit
    4. DNothing, until the table change is undone
    Show the answer

    In Fabric Git integration the unit of a commit is the item. A warehouse is one item, so its changes are committed together and not object by object. The `.sql` files in the repository do not change that.

    Next → 6 / 10

  8. Question 7 of 10

    Five developers build one warehouse. Each developer needs an isolated copy to work in, every schema change must have version history, and promotion from dev to test to prod must be tracked inside Fabric without external build agents. Which combination fits?

    1. AVisual Studio Code with DacFx, plus manual publish to each stage
    2. BFabric web with Git integration, plus Fabric deployment pipelines
    3. CFabric web without Git integration, plus Fabric deployment pipelines
    4. DFabric web with Git integration, plus the SqlPackage CLI in Azure DevOps
    Show the answer

    Match each requirement to a mode. Isolation and history come from Git integration, which adds version control and branching out to feature workspaces. Tracked stage promotion inside Fabric is what deployment pipelines give.

    Next → 7 / 10

  9. Question 8 of 10

    A team publishes one warehouse project to a dev warehouse and to a production warehouse. In dev, tables left over from experiments should disappear on every publish. In production, a publish must never remove data, and objects created outside the project must stay. Which publish profiles fit?

    1. ADev: `DropObjectsNotInSource` True. Prod: `BlockOnPossibleDataLoss` True, `DropObjectsNotInSource` False.
    2. BDev: `BlockOnPossibleDataLoss` True. Prod: `DropObjectsNotInSource` True, `BlockOnPossibleDataLoss` False.
    3. COne profile for both: `DropObjectsNotInSource` False and `BlockOnPossibleDataLoss` False.
    4. DOne profile for both: `DropObjectsNotInSource` True and `BlockOnPossibleDataLoss` True.
    Show the answer

    The two environments want opposite things, so they need two profiles. Cleaning up leftovers is `DropObjectsNotInSource` True. Never losing data is `BlockOnPossibleDataLoss` True, and keeping outside objects is `DropObjectsNotInSource` False.

    Next → 8 / 10

  10. Question 9 of 10

    A warehouse release drops the table `dbo.SalesArchive`. Its rows must first be copied to a backup table, and a small lookup table must be refilled once the new schema is in place. If the release fails halfway, the team runs it again from the start. Which design fits?

    1. ARun-once scripts: the copy in `PreDeploy` and the refill in `PostDeploy`
    2. BIdempotent scripts: the refill in `PreDeploy`, the copy in `PostDeploy`
    3. CIdempotent scripts: the copy in `PreDeploy`, the refill in `PostDeploy`
    4. DIdempotent scripts: the copy and the refill together in `PostDeploy`
    Show the answer

    Two decisions. Timing: `PreDeploy` runs before the schema deployment, while the old table still exists, and `PostDeploy` runs after it. Repetition: the release can run twice, so each script must give the same result every time.

    Next → 9 / 10

  11. Question 10 of 10

    A Git-connected workspace holds two warehouses, Sales and Finance. A view in Sales reads a table in Finance, and a view in Finance reads a table in Sales. Branch out to a new workspace fails. What should the team change?

    1. ARemove the cyclic dependency between the two warehouses.
    2. BCommit the two views one at a time, so that each one is synced alone.
    3. CTurn on version control for the SQL analytics endpoints first.
    4. DCreate one warehouse with a different collation, so they sync apart.
    Show the answer

    Look at the direction of the references. When warehouse items reference each other they form a loop, and Fabric detects that loop during branch-out or a sync from Git and fails the operation. The design fix is to make the dependency go one way only.

    Next → 10 / 10

  12. You’ve finished this set

    That’s 10 questions on Warehouse Database Projects. In the app the ones you miss come back exactly when you’re about to forget them.

    Back to the course Next lesson: Create Deployment Pipelines

    Opens in your browser. Nothing to install.

    Start studying

    On your phone, with the same account

    Get it on Google Play Download on the App Store
Start studying Get it on Google Play Download on the App Store

The whole course, on your computer and your phone

Lessons you can read, audio you can listen to on the way to work, and practice that remembers what you got wrong.

Opens in your browser. Nothing to install.

Start studying

On your phone, with the same account

Get it on Google Play Download on the App Store

“This app is the ‘One Lake’ for preparing the certification exams. Really worth it.”

Abdelhak Y. Google Play review