Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

Don't miss out! 2025 Microsoft Fabric Community Conference, March 31 - April 2, Las Vegas, Nevada. Use code MSCUST for a $150 discount. Prices go up February 11th. Register now.

Reply
IntegrateGuru
Advocate I
Advocate I

Dynamic/Widespread SQL Ingestion With On-Premises Gateway

Use Case/Story:

I have many SQL Server datasouces configured with my on-premises gateway (Lets say 20-50 datasources). I need to copy a sizable amount of data from each database (lets say 50-200 tables) from each datasource. I need to put that data into a corresponding Lakehouse (20-50 lakehouses, one for each datasource). This is a prequisite to creating my semantic models, reports, etc.

Method:

To copy that much data, it is my understanding that Microsoft would reccomend a data pipeline that utilizes the copy data activity to perform the data transfer. Ideal solution is to loop over all my datasources and invoke a copy data activity for each datasource/lakehouse pairing. 

Problem:

 

The copy data pipeline activity does not support dynamic content for SQL Server connection types. I would need to create, deploy, and maintain 20-50 pipelines, all almost exactly identical, with the only distinguishing feature being the selected SQL server connection on the source configuration. This solution is untenable due to the maintenance requirement of having that many pipelines, along with the added requirement of having to have a person designate the connection via the UI, or implment a git script to identify the datasource and apply it to the pipeline (or otherwise store 50 copies of basically the same pipeline in my repository?).

(Im)possible Workarounds/Non-Solutions:

  • Notebook/Spark: Create one notebook that uses Spark to connect to the databases and copy the data over to the warehouse. But there is no Spark connector for gateway datasources, you would need to provide the full JDBC connection information to spark, therefore bypassing the intended purpose of having created a gateway/datasource to begin with. I imagine this is also less performant than copy data.
  • Dataflow Gen2: I don't know if dataflows support dynamic connection settings, but they have no git support, so this is completely unrealistic from a maintainability/deployability standpoint and not worth considering.
  • 20-50 copy data activites in a single pipeline. I would laugh, if it didn't sound like it just might work. 


Questions:

  1. Is there any effort underway to add support for dynamic content for SQL Server Connection types in the copy data activity?
  2. What is the suggested path forward to support my use case in the meantime?
3 REPLIES 3
jpelham
Helper I
Helper I

I don't have a solution but I can see where you would find that having this feature could be beneficial. While having that many similar pipelines will require a bit of initial setup, I have found that there are benefits to having multiple pipelines.

 

My organization has a simliar setup with many servers, usually one for each vendor. Sometimes we experience an issue with a load failing from the vendor to our sql server, which will then cause our pipeline to fail. With having individual pipelines for each server, the impact of a load failing is limited to just the tables loaded in that pipeline, rather than our entire data library.  

v-yilong-msft
Community Support
Community Support

Hi  @IntegrateGuru ,

Have you solved your problem? If so, can you share your solution here and mark the correct answer as a standard answer to help other members find it faster? Thank you very much for your kind cooperation!

 

 

Best Regards

Yilong Zhou

v-yilong-msft
Community Support
Community Support

Hi @IntegrateGuru ,

Here are some of my thoughts and suggestions that I hope will help you.

 

Regarding dynamic content support: Currently, the Copy Data Activity (CDA) really does not support dynamic content for SQL Server connection types. This means that you need to create, deploy and maintain multiple nearly identical pipelines for each different SQL Server connection, which does create a significant maintenance effort. You can also check out this topic for more information: Solved: Re: Dynamic Connection To Sources - Microsoft Fabric Community


Regarding whether there are plans to add support: Based on current information, there are no definite plans to add dynamic content support for SQL Server connection types. You can also raise your expectations in the Idea, here is the website: Home (microsoft.com)

 

 

Best Regards

Yilong Zhou

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

Helpful resources

Announcements
Las Vegas 2025

Join us at the Microsoft Fabric Community Conference

March 31 - April 2, 2025, in Las Vegas, Nevada. Use code MSCUST for a $150 discount!

December 2024

A Year in Review - December 2024

Find out what content was popular in the Fabric community during 2024.