I'm doing an ETL for a job. For the data source stage, I input custom select statement. In the output tab of data source stage, I defined the INCEPTION column data type is Timestamp. The right data type for INCEPTION is date. I check it via DBEAVER. But somehow the IBM Datastage assumes that it is WVARCHAR. It says ODBC_Connector_0: Schema reconciliation detected a type mismatch for field INCEPTION. When moving data from field type WVARCHAR(min=0,max=10) into DATETIME(fraction=6), data corruption can occur (CC_DBSchemaRules::reportTypeMismatch, file CC_DBSchemaRules.cpp, line 1,838). I don't know why it is, since from the database shows that INCEPTION is definitely a Date column. I don't know how to fix this since I don't think I'm making mistake. What did I do wrong and how to fix this?
IBM Datastage assumes a column is WVARCHAR while it's date
1.5k views Asked by ImBaldingPleaseHelp At
1
Where did DataStage get its table definition? DataStage is a computer program; it can't "decide" anything. If you import the table definition from the source, what data type is INCEPTION ? If it is either Date or Timestamp, load that table definition into your DataStage job. Otherwise explicitly convert the string using StringToTimestamp() function in a Transformer stage.