Hi All
I have a DS Parallel job in (DS 11.7.1) with 3 Sequential file stage,3 Transformation Stage,1 Funnel and Output to a Dataset.
3 Sequential file has different layout. But using Transformation stage we got all Fields into one layout. Then sending those 3 outputs to Funnel. But the Funnel output is giving incorrect output for few Records but few records output looks correct.
Output from Transformer stage :
Key : 101 Date1:2018-10-31 Date2:2018-10-31 Date3:2018-10-31
Key : 102 Date1:2018-10-31 Date2:2018-10-31 Date3:2018-10-31
Output from Funnel stage :
Key : 101 Date1:2018-10-31 Date2:2018-10-31 Date3:2018-10-31
Key : 102 Date1:NULL Date2:NULL Date3:NULL
Can you please let me know where could be the issue.
Funnel Stage Output is incorrect
Moderators: chulett, rschirm, roy
-
- Participant
- Posts: 84
- Joined: Sat Jul 19, 2008 8:50 am
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
Please let us know the settings in your Funnel stage and the per-node row counts coming out of the Transformer stage. (You can determine these using Monitor view in the Director client.)
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
-
- Participant
- Posts: 54607
- Joined: Wed Oct 23, 2002 10:52 pm
- Location: Sydney, Australia
- Contact:
I guess I should also have made the point that the Funnel stage expects to have TWO input links, and returns the UNION ALL of the input rows. The Funnel stage settings only determine the order in which input rows are written to the output.
Both input links and the output link require identical metadata.
Both input links and the output link require identical metadata.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.