How to Handle Dynamic Environment Configuration in Microsoft Fabric Deployment Pipeline?

[bhavya5903]

Hi Fabric Community,

I'm deploying my Microsoft Fabric project using Deployment Pipelines, moving from Dev to Prod workspaces.

However, I'm struggling to dynamically configure environment-specific resources like Lakehouses, Warehouses, Pipelines, and more during deployment.

Example Scenario:

I have a pipeline with a Copy Data activity where the destination is set to Dev_LH (Lakehouse) or Dev_WH (Warehouse).
But this is just one example - I want the entire project (including datasets, notebooks, pipelines, and connections) to automatically point to the corresponding Prod resources (e.g., Prod_LH, Prod_WH) after deployment.

My Question:

What is the best way to handle environment-specific configurations in Microsoft Fabric Deployment Pipelines?
Are there any known workarounds, best practices, parameterization options, or external config management strategies that can help switch resources automatically from Dev to Prod?

Looking for a scalable and maintainable solution for enterprise-level projects.

[v-agajavelly]

Hi @bhavya5903 ,

Thanks for raising this important question.

As @AntoineW  correctly pointed out, the Variable Library currently in preview is the right and recommended solution to handle dynamic, environment-specific configurations during deployment in Microsoft Fabric.

This capability was introduced to exactly solve the issue you're facing allowing you to define variables once and assign different values for Dev, Test, and Prod stages. It's scalable, clean, and works well across pipelines, notebooks, and more. I suggest following the approach detailed by super user, It’s currently the most maintainable and enterprise-aligned method to achieve automated configuration switching in Fabric Deployment Pipelines.

Regards,
Akhil.

[AntoineW · Accepted solution]

Hi @bhavya5903 ,

Thanks again for raising this — I wanted to provide an important update based on recent capabilities introduced in Microsoft Fabric (currently in preview):

The Variable Library is now available and designed exactly to address environment-specific configurations within Deployment Pipelines.

What is the Variable Library?

The Variable Library acts as a centralized store of configuration variables (e.g., Lakehouse names, Warehouse references, wait durations, endpoints), with different values per stage of the deployment pipeline (e.g., Dev, Test, Prod).

This enables dynamic resource switching across environments without modifying your pipeline or notebook logic.

Key Capabilities:

• Define variables once and reuse across pipelines, shortcuts, and notebooks
• Assign different values per environment/stage (e.g., Dev_Lakehouse vs. Prod_Lakehouse)
• Compatible with CI/CD, Git integration, and Fabric REST APIs
• Automatically resolves the correct value based on the active deployment stage

Current Limitations:

• Still in Preview – not yet GA
• Not supported on all Fabric items yet (e.g., semantic models or dataflows)
• Variable resolution is scoped per workspace, not shared across workspaces

Conclusion:

If you're deploying Fabric assets across environments, the Variable Library is the most promising approach to ensure scalability, consistency, and automation for configuration management.

Hope this helps others looking for the same flexibility — happy to collaborate and share examples if needed!

Best regards,

Antoine