Forum Discussion
Dataflow Gen 2 fail error message: WriteToDatabaseTableFrom_TransformForOutputToDatabaseTableFrom_20
Hey Miguel,
Yeah it definitely had something to do with the staging artifacts.
While troubleshooting / learning how it all works, I was deleting the staging lakehouse and any dataflows (gen 2) and then recreating the dataflows which recreated staging lakehouse.
However, now I am pretty sure that under no circumstances should staging lakehouse (or other staging artifacts) be deleted. I get the idea that they really should be hidden or otherwise considered "system" artifacts and to just let them do their thing.
I had to delete my Workspace and create a new one and start all over again to make it work.
Can you confirm the above is true?
Thanks!
You are correct, the automatically created LH / DW artifacts should not be deleted. Even if you create a new dataflow (which will provision new artifacts) it will leave existing dataflows in a broken state right now.
We have several improvements planned for this experience.
Hi!
When it comes to your workspace, do you see a Lakehouse with the word "Staging" available in it?
You can read more about a similar situation from the thread below:
Solved: Re: Dataflow Gen2 Web.Contents : Error : PowerBIEn... - Microsoft Fabric Community
Hey Miguel,
Yeah it definitely had something to do with the staging artifacts.
While troubleshooting / learning how it all works, I was deleting the staging lakehouse and any dataflows (gen 2) and then recreating the dataflows which recreated staging lakehouse.
However, now I am pretty sure that under no circumstances should staging lakehouse (or other staging artifacts) be deleted. I get the idea that they really should be hidden or otherwise considered "system" artifacts and to just let them do their thing.
I had to delete my Workspace and create a new one and start all over again to make it work.
Can you confirm the above is true?
Thanks!
You are correct, the automatically created LH / DW artifacts should not be deleted. Even if you create a new dataflow (which will provision new artifacts) it will leave existing dataflows in a broken state right now.
We have several improvements planned for this experience.