Discussion on "Automating Feature Workspace maintainance in Microsoft Fabric"

Peer Grønnerup · 2025-04-21T08:53:40.187Z

📣 Update – July 2025: This post has been updated to reflect new support for connecting and synchronizing Microsoft Fabric workspaces with Azure DevOps Repos using a Service Principal. The update includes a new section on Azure DevOps setup, covering...

Hi Peer,

I'm encountering an error while using Azure DevOps with Microsoft Fabric:

{ "status_code": 400, "text": { "requestId": "4ec14443-fe2b-472a-b2d4-5e0d6edc5f27", "errorCode": "GitCredentialsConfigurationNotSupported", "message": "Credentials source ConfiguredConnection is not supported for AzureDevOps." } }

I'm calling the following API: POST workspaces/*/git/connect -i

POST workspaces/*/git/connect -i When I retrieve my Git credentials, the response is:

{ "status_code": 200, "text": { "source": "Automatic" } }

However, when I try to set the credentials source to ConfiguredConnection, I get this error:

So in my case, it seems I'm unable to make it work. Do you have any idea what could be causing this or any suggestions on how to proceed?

Thanks in advance!

Best regards, Elbert

Really great post Peer. Looking forward for a version with azure repos! One question with regards to your architecture: You’re placing the lakehouse in the same workspace, I assume, partly, to avoid issues with moving data across workspaces. But which workspaces do you associate with which lakehouse? Ingest i presume is related to bronze, but which workspace is related to silver and gold?

Hope my question makes sense

Hi Raeed, thanks for reading and liking the post. Silver and Gold lakehouses are related to data engineering workspaces. I also refer to these as Prepare or Transform workspaces. The Prepare/Transform stage handles the movement of data from Bronze to Silver and from Silver to Gold. The data transformantion is imlemented using Notebooks. Data in the Gold layer is then further on consumed by Serve or Insights workspaces containing semantic models.

I have been a bit busy lately but I will prioritize enhancing my repo to also include metadata store, dynamic ingestion pipelines and notebooks etc.

Also stay tune for an update to include automatically connecting and syncronizing with Azure DevOps Repos. I will write an additional post for this enhancement within the next couple of days.

This works great for me, until I add a pipeline. Then the update from git always fails saying missing dependencies. This is with a brand new pipeline with just one object of set variable. Not sure why..

How are you refering to your child activities, i.e. when using the Invoke Pipeline og Invoke Notebook activities?

Be sure to design your pipelines with automation in mind and use dynamic references when possible.

Also read my blog post on dynamically configuring pipelines here: peerinsights.hashnode.dev/automating-fabric-dynam… and the post on designing for automation here: peerinsights.hashnode.dev/designing-for-automatio…

Discussion

Automating Feature Workspace maintainance in Microsoft Fabric

Responses(3)

Recent in Forum

Search Hashnode

Automating Feature Workspace maintainance in Microsoft Fabric

Responses(3)

Recent in Forum