Question

A data engineer manages a Direct Lake semantic model in Fabric that is scheduled to refresh every morning. Business users have complained that on several occasions they viewed stale data without knowing the scheduled refresh had failed silently overnight. The engineer wants to be automatically notified by email whenever a scheduled refresh of this semantic model fails, without writing custom code. What is the most appropriate approach?

Accepted Answer

Enable the built-in refresh failure notifications in the semantic model's scheduled refresh settings and specify the recipients.

Answer

Create a Data Activator (Activator) reflex triggered on the semantic model's dataset query results to detect stale rows.

Answer

Write a notebook that queries the model daily and sends an email using a Python SMTP library.

Answer

Configure a Fabric pipeline with a Web activity that polls the refresh history REST API and emails on failure.