Edit

Share via


Understand callback registration for Dataverse triggers

Power Automate flows that use the When a row is added, modified or deleted trigger rely on the Callback Registration (CallbackRegistration) table in Microsoft Dataverse. This article explains:

  • How Dataverse triggers work internally
  • The relationship between flows and callback registration records
  • How to monitor flow execution using system jobs in the Power Platform admin center.

How Dataverse triggers work

When a flow uses the Dataverse When a row is added, modified or deleted trigger, Dataverse must notify Power Automate when a data event occurs. To enable this communication, Dataverse creates a callback registration record that acts as a webhook subscription stored in Dataverse.

Flow and callback registration relationship

Each Dataverse When a row is added, modified or deleted trigger in a flow creates exactly one callback registration record. This record acts as the subscription that tells Dataverse when and how to call Power Automate. It contains metadata about the trigger configuration. This metadata allows Dataverse to subscribe to table events and call the correct flow with the appropriate context.

Power Automate flow and callback registration relationship diagram.

Callback registration lifecycle

  • Saving or turning on a flow creates the callback registration record.
  • Turning off or deleting a flow removes the callback registration record.
  • If the callback registration is missing or invalid, the flow doesn't trigger.

Asynchronous execution model

Dataverse triggers run in the background (asynchronously) through the Dataverse asynchronous service, not inside the database transaction. This model has important implications:

  • Flow failures don't roll back data changes in Dataverse.
  • Trigger execution depends on the asynchronous service health and availability.
  • The user's operation finishes before the flow executes.
  • Flow execution happens separately in Power Automate's infrastructure.

Asynchronous execution model diagram.

Monitoring flow execution by using system jobs

Because Dataverse triggers run through the asynchronous service, each trigger execution creates a system job that you can monitor in the Power Platform admin center.

How to check system jobs

  1. Open Power Platform admin center.
  2. Select your Environment.
  3. Go to Settings > System Jobs.
  4. Filter by:
    • Type: Callback Registration
    • Status: Succeeded or Failed

Why system jobs matter

System jobs provide the most reliable way to confirm:

  • Whether Dataverse attempted to trigger the flow
  • Whether the trigger failed before reaching Power Automate
  • Details about validation failures or errors

If no system job exists for an expected trigger, the callback registration might be missing or invalid, or the event might not match the trigger criteria.

System job monitoring diagram

System job monitoring diagram.

Summary

The following table summarizes the key concepts of callback registration in Power Automate:

Concept Description
Dataverse trigger Implemented using callback registration webhook subscription
Flow ↔ callback registration relationship One flow trigger = One callback registration record
Execution model In the background via Dataverse asynchronous service
Transaction isolation Flow failures don't roll back data changes
Monitoring Use system jobs in Power Platform admin center to verify trigger execution