Error while running with 4 nodes

Madhu1981 · Post by **Madhu1981** » Mon Apr 24, 2006 1:54 am

Hi,

I have two jobs. First job will create a dataset, which is the source of the second job.

My sencond job looks like

Dataset --> Copy --> DB2. I am unsing Upsert menthod in DB2 stage.

When i ran the job with 1 node it is runnig succewsfully. When i ran with 4 nodes it is throwing the following error

main_program: Fatal Error: There are irreconcilable constraints on the number of
partitions of an operator: parallel Dset_TEDSZ06002JB_00001.
The number of partitions is constrained to 1,
but the eSame input file, /work/EDWCST/WHSE/working/ds_TEDSZ06002JB_00001.ds, has 4.
An eSame input file cannot be repartitioned to match
the number of partitions of the consuming operator.

ArndW · Post by **ArndW** » Mon Apr 24, 2006 2:33 am

Madhu - there is no need to continually post the same question; it has the effect that some people (myself included) just ignore the question and won't post a response.

Madhu1981 · Post by **Madhu1981** » Mon Apr 24, 2006 3:06 am

ArndW wrote:Madhu - there is no need to continually post the same question; it has the effect that some people (myself included) just ignore the question and won't post a response. ...

HI Arnd.

I am sorry, Could you please tell me how to delete the post here

ArndW · Post by **ArndW** » Mon Apr 24, 2006 3:16 am

Once a post has been replied to it cannot be deleted. But I think that our moderator will see these posts and take care of deletions.

madhu851 · Post by **madhu851** » Tue Apr 25, 2006 8:53 pm

anyone please answer this question.

ray.wurlod · Post by **ray.wurlod** » Tue Apr 25, 2006 9:08 pm

Everything you need to know is here.
main_program: Fatal Error: There are irreconcilable constraints on the number of
partitions of an operator: parallel Dset_TEDSZ06002JB_00001.
The number of partitions is constrained to 1,
but the eSame input file, /work/EDWCST/WHSE/working/ds_TEDSZ06002JB_00001.ds, has 4.
An eSame input file cannot be repartitioned to match
the number of partitions of the consuming operator.

Somewhere the number of partitions is being constrained to 1. Are you running a stage in sequential mode? In a node pool that contains only one node?

Whatever you've set up, you can't use Same as the re-partitioning algorithm if the upstream stage is running on four nodes.

praneeth · Post by **praneeth** » Wed Apr 26, 2006 9:18 am

hi could you please elaborate.

ray.wurlod · Post by **ray.wurlod** » Wed Apr 26, 2006 3:57 pm

Not without seeing the actual job design.

DSXchange