My source is a table on Oracle 9i. In the Oracle 9i stage I have set it to work in parallel. When I run the job I am getting duplicate records from Oracle.
How can I set up the stage so that each partition gets a unique set of rows to process?
duplicate rows from oracle when running in parallel
Moderators: chulett, rschirm, roy
I have an Oracle stage going into a transformer stage which writes out to a flat file.
In the oracle stage I have changed from the default of sequential to parallel partitioning. It does not give me any options to specify entire , round robin etc...
In the transformer stage partitioning is set to the defaults which are parallel and auto.
The sequential file is set to the default which is auto collector.
I am seeing rows twice in the flat file.
I don't have a lot of experience with DataStage but I have used Informatica and in Informatica the query for each partition would be set up with a where clause to ensure that each partition got a unique set of rows from the source. But I don't see that functionality in DataStage.
What am I missing?
In the oracle stage I have changed from the default of sequential to parallel partitioning. It does not give me any options to specify entire , round robin etc...
In the transformer stage partitioning is set to the defaults which are parallel and auto.
The sequential file is set to the default which is auto collector.
I am seeing rows twice in the flat file.
I don't have a lot of experience with DataStage but I have used Informatica and in Informatica the query for each partition would be set up with a where clause to ensure that each partition got a unique set of rows from the source. But I don't see that functionality in DataStage.
What am I missing?