Menu
CoddyTech

Power BI Service, Sharing and Security Interview Questions

Workspaces and roles, apps and sharing, gateways and refresh, row-level security and deployment pipelines.

11 questions
  1. What are the workspace roles in the Power BI Service?

    Admin, Member, Contributor and Viewer. Admins manage the workspace and its access. Members can also add people and publish or update the workspace's app. Contributors create, edit and delete content, but cannot add people or publish the app (an Admin can let them update it). Viewers can only view and interact with content, and they are the only role row-level security applies to.

  2. What are the ways to share a Power BI report?

    Publish an app from the workspace (the usual way for many readers), share a single report or dashboard by link or with named people, give access to the workspace (for people who build), embed it in Teams or SharePoint, set up email subscriptions, or export to PDF or PowerPoint. Publish to web makes a report public to anyone on the internet with no sign in, so it is never for internal data.

  3. What is the difference between a workspace and an app?

    A workspace is where people build and maintain content: reports, semantic models, dataflows, all editable by its contributors. An app is a read only package of chosen content from one workspace for readers, with its own navigation and audiences that see different pages. Changes in the workspace reach readers only when the app is updated, so work in progress never shows up in front of them.

  4. What is an on-premises data gateway, and when do you need one?

    A gateway is software installed on a machine inside your network that lets the Power BI Service reach sources it cannot reach from the cloud: a SQL Server in your data center, files on a network share. You need one to refresh or DirectQuery those sources. Standard mode is shared by many users and supports DirectQuery; personal mode serves one user and only Import. Cloud sources such as Azure SQL need no gateway.

  5. How does scheduled refresh work, and what are its limits?

    In the semantic model's settings in the Service you set credentials and up to 8 refreshes a day with Pro, or up to 48 on a Premium or Fabric capacity or with Premium Per User. A refresh on shared capacity times out after 2 hours, on a capacity after 5. Common failures: expired credentials, the gateway offline, a source schema change (a renamed column), and timeouts on large imports, which incremental refresh fixes.

  6. How do you set up row-level security?

    In Desktop, open Modeling, Manage roles, create a role and give it a DAX filter on a table, such as Region[Region] = "Europe". The filter flows through the relationships to the facts. Test it with View as. After publishing, add users or security groups to the role in the semantic model's Security settings. Once roles exist, a reader who is in no role sees no data.

  7. How do you write dynamic row-level security?

    One role whose filter compares the signed in user with a mapping table, instead of one role per region:

    -- role filter on the Region table
    Region[Region]
        IN CALCULATETABLE (
            VALUES ( Security[Region] ),
            Security[Email] = USERPRINCIPALNAME ()
        )
    

    Security has one row per user and region. USERPRINCIPALNAME returns the user's sign in email in the Service. Adding a person or a region is a data change, not a new role.

  8. Who does row-level security not apply to?

    Workspace Admins, Members and Contributors: anyone who can edit the semantic model sees all its data, so give readers the Viewer role or an app instead. RLS also does not apply to the model's author in Desktop except through View as. It filters rows only; to hide whole tables or columns from a role you need object-level security, set with an external tool such as Tabular Editor.

  9. What are deployment pipelines?

    A deployment pipeline links workspaces as stages, usually development, test and production, and copies content from one stage to the next. Deployment rules change data source connections and parameter values per stage, and a comparison view shows what differs between stages. Pipelines need Premium Per User or a capacity. For version history, add Git integration with the project (.pbip) format.

  10. What is the difference between a subscription and a data alert?

    A subscription emails a snapshot of a report page or dashboard on a schedule (daily, weekly, or after a refresh), with a link back to it. A data alert fires when a value on a dashboard tile (a card, KPI or gauge) crosses a threshold you set, and notifies you by email and in the Service. Alerts work on dashboard tiles only, not on report visuals.

  11. How do several reports share one semantic model?

    Publish the model once, then build reports that connect to it live (Get data, Power BI semantic models) instead of importing the data again. Those thin reports hold only pages and report level measures, so the measures and security are defined in one place. Readers need Build permission on the model to create reports on it. Mark the model Promoted or Certified so others know which one to use.

Coddy programming languages illustration

Learn to code with Coddy

Get started