How to avoid this warning

Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.

Moderators: chulett, rschirm, roy

Post Reply
bollinenik
Participant
Posts: 111
Joined: Thu Jun 01, 2006 5:12 am
Location: Detroit

How to avoid this warning

Post by bollinenik »

Hi,

I am reading data from seq.file and next i am using Sort Stage and next using Transformer for avoid duplicates and next lookup i am getting this warning for these 3 stages like this
where ever i am using seq.file as target in this job ia m getting warning like " Trgt_Rej_data_FF: When checking operator: A sequential operator cannot preserve the partitioning
of the parallel data set on input port 0." any can help me
thanks in advance
kOTI,
1586-738-7785
kris007
Charter Member
Charter Member
Posts: 1102
Joined: Tue Jan 24, 2006 5:38 pm
Location: Riverside, RI

Post by kris007 »

I didn't quite get the clear picture of what you are doing in your job. The order of stages are confusing.But what you will need to do get rid of the warning is set your PreservePartioning flag to "Clear" in one of the stages downstream of the Sort stage because sequential file writes records only sequentially. How exactly are you removing duplicates from Transformer. Using Constraints? Wouldn't Remove Duplicates stage help you in this case? Just wondering.

P.S. Please delete the other duplicate post you have in place before anyone starts replying to it.
Kris

Where's the "Any" key?-Homer Simpson
vmcburney
Participant
Posts: 3593
Joined: Thu Jan 23, 2003 5:25 pm
Location: Australia, Melbourne
Contact:

Post by vmcburney »

Assuming you are sending reject rows from the transformer (via Trgt_Rej_data_FF) to a sequential file the job is warning you that you are switching from partitioned data to sequential data. Each transformer across each node will have to send rejects to a link that repartitions (or unpartitions?) the data into a sequential stream to write it to a single file.

This is not a big deal since you would hope to get a very low number of rejects.

You could switch this message across to an information message via the message handler. Would be interested to hear from other programmers as to whether there is an easy way to avoid the warning altogether. Maybe if you switch the partitioning output link from SAME to sequential.
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
Post Reply