Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowData Days is here! Join us now for 60+ days of learning, challenges, and connection. Learn more
Hi,
I was wondering if anyone knew of a work around to let existing resources in ADF or Azure Synapse write to One Lake.
I was hoping I could configure a service principal and then setup a linked service in ADF. However it enforces a name convention of the URL:
For reference I took the URL from: OneLake access and APIs - Microsoft Fabric | Microsoft Learn
Thanks
Ben
Ben
@bcdobbs It depends on what you want to do with the data, if your source data is already structured/curated then you don't generally need to copy it again in the lakehouse/warehouse, shortcuts will work.
@GeethaT-MSFT Will the connectors for the Fabric copy data activity be expanded to match what is available in Azure Data Factory? For instance, I have on premises Oracle data that I would like to land in OneLake.
Thank you.
@kdoherty Yes, on-prem connectivity is tracked for GA, Until then a workaround is to use on-prem gateways in pbi to stage the data in a cloud location and then use copy activity.
That's good news. Thank you for replying.
I fully intend to once it supports what I need! Currently Fabric copy data activity in pipeline only has very basic API support; I need to be able to pass a bearer key in the auth header (which I've ideally retrieved from key vault or new fabric equivalent), or for internal Microsoft stuff use a managed identity/service principal.
I was just thinking until that was available (I understand it's coming) I could use ADF in azure to land the data while I experiment with fabric.
That said having now played with it more I think the way to go is simply create a shortcut to my exisiting ADLS storage and get the data in that way.