How to change the Power Apps Databricks catalog name based upon the environment I deploy to

(0) Share

Report

Posted on by DAnderson

28 Season of Giving Solutions 2025

I have a Power App that references several tables in Azure Databricks. The catalog name is "dev" for my dev and qa environments. I just deployed it to my STG environment and need a way to switch the catalog name from "dev" to "test".

Has anyone run into this issue and might you have a resolution.

I have even exported the solution, updated the connection references from "dev.schema_name" to "test.schema_name" and when running with the live monitor it receives a 403 response and is still referencing "dev.schema_name".

Thank you!

Categories:

Building Power Apps

Power Apps Pro Dev & ISV

General topics

I have the same question (0)

All responses (4)

Answers (0)

Sort by

Soufyane 14 on at

Like (0)

Report
Copy link

Link copied!

Was this reply helpful? Yes No
Suggested answer

Soufyane 14 on at

Like (0)

Report
Copy link

Link copied!

The Problem
The catalog name is hardcoded inside the app’s data source. Editing the ZIP doesn’t work.

Try this instead :
   1.   Create an Environment Variable (type: Text) in your solution, and set its value to match the catalog name of that environment.
   2.   In your Canvas App, remove the existing Databricks data source, re-add it, and when the catalog picker appears -> hit Advanced -> bind it to your environment variable.
   3.   On import to each environment, the wizard will prompt you to set the environment variable value, just enter the right catalog name for that environment.

That’s it. The app will dynamically pick the correct catalog at runtime based on where it’s deployed.
Also make sure your service principal has the right permissions on the target catalog in Unity Catalog — that’s likely why you’re seeing the 403.

Was this reply helpful? Yes No
Suggested answer

11manish 871 on at

Like (0)

Report
Copy link

Link copied!
Best and simplest approach:

Use environment variables for catalog name

Or create identical views per environment and avoid catalog dependency

You cannot reliably change catalog via solution export
Use environment-aware configuration (variables or views)
Ensure connection references are re-bound per environment

Was this reply helpful? Yes No
DAnderson 28 Season of Giving Solutions 2025 on at

Like (0)

Report
Copy link

Link copied!

I truly appreciate the feedback thus far! Unfortunately, the advanced option is not available with the Azure Databricks connector. I am also using a service principal to authenticate with Azure Databricks, which uses client id and secret. The standard Databricks connector allows only a bearer token and doesn't support service principal authentication.

Was this reply helpful? Yes No