Skip to content

Layering row, column, object, and file-level security rules

Microsoft Fabric provides multiple layers of granular access control across its analytics items, allowing administrators to restrict data visibility by row, column, table/object, or folder/file path depending on the underlying item type (Warehouse, Lakehouse, Power BI semantic model, or OneLake). Each control type uses a different implementation mechanism: T-SQL security policies, OneLake data access roles, or Power BI RLS/OLS, so engineers must choose the right tool for the item and storage layer involved.

Must-know

  • Row-Level Security (RLS) in Fabric Warehouses and SQL analytics endpoints is implemented using CREATE SECURITY POLICY with inline table-valued functions (Predicate-based, like SQL Server RLS), filtering rows returned to specific users or roles.
  • Column-Level Security (CLS) in Warehouses/SQL endpoints is implemented via GRANT/DENY SELECT on specific columns or by using views that expose only permitted columns to certain roles.
  • Object-Level Security (OLS) restricts access to entire tables, views, or stored procedures using standard SQL GRANT/DENY/REVOKE permissions, and it takes precedence over RLS/CLS since users need object access before row/column filters apply.
  • OneLake data access roles (a Fabric Lakehouse feature) allow folder- and file-level access control by defining roles scoped to specific folders within a Lakehouse, using role assignments to Microsoft Entra ID users, groups, or Fabric workspace roles, separate from SQL permissions.
  • Power BI semantic models support RLS and Object-Level Security (OLS) via roles defined in the model (using DAX filter expressions for RLS) and OLS configured through external tools like Tabular Editor, since OLS is not natively supported in the standard Power BI Desktop UI.
  • Security policies and role assignments defined at the Warehouse/SQL endpoint are separate from OneLake folder-level roles, so implementing consistent security across both a Lakehouse's files and its SQL analytics endpoint requires configuring both layers independently.
Check this objectiveFree · always available

A data engineer needs regional sales managers who query a Fabric Warehouse table to see only the sales rows belonging to their own region, without changing how the application connects or maintaining separate views per region. Which T-SQL based approach satisfies this requirement?

Your objective map0 tried · 0 right · 54 untouched

What you have tried across DP-700's objectives, not a readiness score.

Coverage checked against the published exam guide on Aug 11, 2026.

These are independent practice questions, written against this certification's published exam guide. They are not the certification vendor's own questions, and not the real exam.