A connector is changing the data-type of a field.
We change the types of columns in various cases depending on the source and the destination types. You have more detailed information regarding this in our documentation section Editing Fivetran-delivered tables.
We recommend that you use transformations to adjust the tables to your own needs. As we normalize the data in the process of ELT (Extraction - Load - Transform), types are inferred by our code base. You will need to use transformations to ensure that the data suits your needs.
When you change the data type of a column in your source, we look at whether it is a widening or a narrowing change.
Widening change: (e.g.: from INTEGER to BIGINT), we update the column type in your destination.
Narrowing change: (e.g.: from BIGINT to INTEGER), we do not update the column type in your destination.
You can read more about this in our section Column type changes.