Forum Discussion
DataFlow Choose destination settings "Use automatic settings" disabled
Please check the documentation available for this:
answering some of the questions that I've seen on this thread:
- Automatic settings are only supported for Lakehouse and Azure SQL database
- Automatic settings can only be applied to new tables
- Dynamic table schema options on publish can only be selected when using the "Replace" update method
As mentioned in some previous replies, there are some experiences that perhaps don't exist today in Dataflow Gen2. For any of those please feel free to post an idea in the Fabric ideas portal (aka.ms/FabricIdeas)
if you can't use the "automatic settings" when trying to create a new table to a Lakehouse or an Azure SQL DB, please feel free to open a support ticket and reply to this message with the case number. I'll elevate it and have the engineering team take a closer look as to what might be causing such bug.
Please do refer to the documentation available that covers the current supported scenarios and limitations.
On your last reply you stated that you did tested the functionality successfully layout the steps on how you did so.
From your narrative, it is suggested that "Workflow Gen2" and while choosing the destination target the table should be an "Existing table" and also stated that you could see the database schemas (such as "dbo", and others you may have created).
I have done this exersice multiple times and in the "Choose destination target" no database schemas are shown and therefore none could be selected as a destination.
Furtheremore, I also tried to prefix the schema to the table name as you will normally do such as "[Accounting].[Ledger]" in were [Accounting] is the schema and "Ledger" is the table. But it does not works since it creates a table named "[[Accounting].[Ledger]]" and that is not what is intended.
Maybe the option is to find a way to see the underlying script and with kind of an advance option edit it to specify "Accounting" instead of "dbo", as I could do in the source script.
Anyways, this is not solved yet... as possible scale-this-up to find a way to write to a database schema different from "dbo".
While waiting for the above, I may create the table in a target schema before creating the workflow and then append the data to it which is not the ideal choice since is more work (an additional step), I rather be allowed to identify the schema on the copy.
I can't accept your suggestion as a solution since I can't replicate it and it does behave different from your narrative. Topic is still open...
First, I created the schema "testSchema" and the table "table1" in my Data Warehouse:
In my Dataflow Gen2, I can find the table:
frithjof,
Presicely the point! The table needs to be first created in the schema, and this is not the scenario that I am describing or described... There should be a way to assigned the target schema not forcing the creation of the table before hand.
My vote is to have the as described above and not force the pre-creation of the tables to be targets.
I see, now I understand,
Okay I have created an idea for this, please vote:
https://ideas.fabric.microsoft.com/ideas/idea/?ideaid=452b78e0-0300-ef11-a73c-6045bdbe85bd
Here is also another thread about the same topic: