I am getting the following error.
Missing or wrong flow
(ICM18954) Missing transforms for required (NOT NULL) columns in target.
I am loading a star schema. I created a flow and it created a table with no nulls for the surogate keys.
The only coluns without nulls are the surogate keys.
Then I created a flow to add daily sales to the fact table and I get this error. I don't understand why it can't have no nulls on the surrogate key?This message has been edited. Last edited by: Kathleen Butler,
A surrogate key in a dimension table uniquely identifies the row and to do so it must have a value and so should be described as NOT NULL. That said, if your are using the "Slowly Changing Dimension" load type then DataMigrator assigns a value for the surrogate key so you never need to so.
On the other hand in a fact table, it's the combination of all the surrogate keys that must be unique. So you could have one or more surrogate key as nullable although I would say a better approach is a key value with a corresponding row in the dimension table that indicates the value was not available. The measures in the fact table can be nullable or not as needed.
Please check the target transformations to ensure that you do indeed have values assigned for all the columns that require values. If you right click on the header line you can add Nulls to the information that is displayed so you can see which columns allow nulls.
|Powered by Social Strata|