1) When we run a DI job it fetches data from Oracle table and loads it into Hadoop table using Table loader transformation.
The data gets loaded into Hadoop Table and then when you open the table in SAS environment you are getting some blank rows.
2) As you suggested, there is no such blank values in Oracle table but while data loading into Hadoop dataset,some extra missing rows were generated in hadoop table.
DDL is created in Hadoop using CREATE TABLE node. and then using DATA LOADER node data was loaded into Hadoop tables.
Should I share backend code for these transformations?
I have attached backend coed for the same.
Please find attached log.