Update DataLake AccessKey when deploying

[shoefried]

We are currently trying to automate the deployment of our PowerBI reports. Therefore we have 3 workspaces within PowerBI (one for each environment) and want to deploy from one to another at each stage. Our report queries from a ADLS Gen 2, but be have a different DataLake per environment. Now i want to change the DataLakeConnection while deploying and I thought it would be easiest to parametrize the ConnectionString and AccessKey and set it accordingly via M-Skript within PowerBI. To set the parameters externally when deploying with an Azure DevOps pipeline is no issue but I dont find a way to set the AccessKey that way. Setting the new ConnectionString is no issue but the access credenatials should be set appropriately too.

Therefore my questions:

-  Is that the appropriate way to change the DataLake connection when deploying? If not what is the common way to go? I can only find tutorials and forumquestions for SQL-Dbs unfortunately and also the docu didnt help me too much so far

- If thats how to do it, how to dynamically set the AccessKey for a ADLS Gen 2 connnection via parameters in PowerBI?

Thanks in advance:)

[v-rzhou-msft]

Hi @shoefried ,

I think you can try to create a Microsoft Entra ID service principal for your Azure storage.

For reference:

Tutorial: Connect to Azure Data Lake Storage Gen2 - Azure Databricks | Microsoft Learn

Then you can securely access data in the Azure storage account using OAuth 2.0 with your Microsoft Entra ID application service principal for authentication.

Best Regards,
Rico Zhou

If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.