Forum Discussion
Modification of Warehouse & Deployment Pipeline structure: Drop & Create table ????
Hello everyone,
I made a change in a DEV Warehouse to rename a column.
Then I deployed this change to the UAT environment via the deployment pipeline.
To my surprise, the table was then empty! But it was a table with historical data.
Does this mean that the deployment pipeline performs a DROP and CREATE to promote this type of change?
If so, this is unacceptable because it could lead to data loss.
Have you had a similar experience?
Thank you in advance for sharing your experience.
Have a nice day,
Vivien
Hi vivien57,
This is a known limitation.
Source Control with Fabric Data Warehouse (Preview) - Microsoft Fabric | Microsoft Learn- Currently, if you use
ALTER TABLEto add a constraint or column in the database project, the table will be dropped and recreated when deploying, resulting in data loss.
Warehouse support in deployment pipelines is still in preview, so hopefully this is fixed before it goes GA.
- Currently, if you use
2 Replies
- tayloramy
Super User
Hi vivien57,
This is a known limitation.
Source Control with Fabric Data Warehouse (Preview) - Microsoft Fabric | Microsoft Learn- Currently, if you use
ALTER TABLEto add a constraint or column in the database project, the table will be dropped and recreated when deploying, resulting in data loss.
Warehouse support in deployment pipelines is still in preview, so hopefully this is fixed before it goes GA.
- Currently, if you use