What causes Fabric Pipeline Error 4123?

The 'pipelineParameters' defined in the Fabric pipeline activity do not match the parameters expected by the published AzureML pipeline. The published AzureML pipeline was updated (parameters added, removed, or renamed) without updating the corresponding Fabric pipeline activity. Parameter values are in the wrong data type or format (e.g., passing a string where a numeric value is expected). A required parameter is missing from the activity definition while the AzureML pipeline marks it as mandatory

How do I fix Fabric Pipeline Error 4123?

Step 1: Open the AzureML pipeline in Azure Machine Learning Studio and review the published pipeline's expected parameters, their names, types, and whether they are required or optional.. Step 2: In Fabric Data Factory, open the AzureMLExecutePipeline activity and inspect the 'pipelineParameters' property to identify any mismatches with the AzureML pipeline's parameter schema.. Step 3: Update the activity's pipelineParameters JSON to exactly match the parameter names, types, and structure expected by the published AzureML pipeline.. Step 4: If the AzureML pipeline was recently republished with changes, ensure the Fabric pipeline activity references the correct published pipeline version.. Step 5: Save the updated activity definition and run the pipeline in debug mode to validate the parameters are accepted before deploying to production.

High severityconfigurationMicrosoft Fabric →

Power BI Refresh Error:
Fabric Pipeline Error 4123

What does this error mean?

A request to Azure Machine Learning failed because one or more activity properties, such as pipeline parameters, do not match the expected payload of the published AzureML pipeline. The parameter structure or values are incompatible with the target ML pipeline.

Common causes

1The 'pipelineParameters' defined in the Fabric pipeline activity do not match the parameters expected by the published AzureML pipeline
2The published AzureML pipeline was updated (parameters added, removed, or renamed) without updating the corresponding Fabric pipeline activity
3Parameter values are in the wrong data type or format (e.g., passing a string where a numeric value is expected)
4A required parameter is missing from the activity definition while the AzureML pipeline marks it as mandatory

How to fix it

1Step 1: Open the AzureML pipeline in Azure Machine Learning Studio and review the published pipeline's expected parameters, their names, types, and whether they are required or optional.
2Step 2: In Fabric Data Factory, open the AzureMLExecutePipeline activity and inspect the 'pipelineParameters' property to identify any mismatches with the AzureML pipeline's parameter schema.
3Step 3: Update the activity's pipelineParameters JSON to exactly match the parameter names, types, and structure expected by the published AzureML pipeline.
4Step 4: If the AzureML pipeline was recently republished with changes, ensure the Fabric pipeline activity references the correct published pipeline version.
5Step 5: Save the updated activity definition and run the pipeline in debug mode to validate the parameters are accepted before deploying to production.

Frequently asked questions

How do I find the exact parameter schema expected by a published AzureML pipeline?

In Azure Machine Learning Studio, navigate to Pipelines > Pipeline Endpoints, select your published pipeline, and view the 'Parameters' tab. You can also retrieve the schema programmatically using the AzureML Python SDK or REST API.

Can dynamic expressions in pipelineParameters cause this error?

Yes — if a dynamic expression evaluates to null, an unexpected type, or a malformed string at runtime, the resulting value may not match what the AzureML pipeline expects. Test dynamic parameter expressions using Fabric pipeline debug runs with sample input data.