Null display in datastage

Jayb1 · Post by **Jayb1** » Tue Jun 14, 2011 12:18 am

Hi every one

I am facing the problem in datastage with null population. We have 2 environments, Dev and Integration testing env. In the dev environment We have dev a parallel job. When we run this job the job is sucessfull and all the columns are populating correctly i.e all the joins are working correctly.

No when we run the same job in the Integration testing environment

Some of the rows get populated with all the column and for some of the row only some of the colums are populated. Expected results are the non populated column should get populate.

So what we did we took the dump of the data from Integration testing environment and imported into the dev environment.

All the columns are all the rows were populated.

Has any one faced this issue.

Need help and way out for the problem.

Regards

Jay

Sreenivasulu · Post by **Sreenivasulu** » Tue Jun 14, 2011 1:32 am

Datastage in test and dev are not in sync

Regards
Sreeni

ray.wurlod · Post by **ray.wurlod** » Tue Jun 14, 2011 1:50 am

Since you're running version 8.1 or higher, you can compare the jobs between projects. In DEV right click on the job in the Repository toolbar and choose, from the context menu, Cross Project Compare...

From the dialog that opens connect to the TEST project and identify the job that you want to compare.

Determine, from the comparison report, why the symptoms you are noticing occur.

Jayb1 · Post by **Jayb1** » Tue Jun 14, 2011 1:52 am

Your are right , the environments are not in sink.

But is this a bug in datastage or some patch needs to be installed ?

Regards

Jay

jwiles · Post by **jwiles** » Tue Jun 14, 2011 5:24 am

Before asking whether or not there is a bug causing your problem, you need to eliminate the possibility that the problem is not in your job design, environment and/or how you're running the job.

1) Is the job identical in both environments?
2) Are you running the job identically in both environments? (same config file settings, parameters)
3) Are the projects configured identically in both environments? (same environment variable values)

A wild guess: Your job is performing one or more joins or merges? Look at your partitioning and sorting, especially the values of $APT_NO_PART_INSERTION and $APT_NO_SORT_INSERTION for the two projects if you're using automatic partitioning.

Regards,

Hema.k · Post by **Hema.k** » Thu Jun 16, 2011 10:41 pm

Has the issue solved?
Datastage 8.1 has some issue with the odbc stage it seems.
If your job design has ODBC stage ,then null will be populated in the columns whose datatype is NVARCHAR.Use VARCHAR instead and try...

ray.wurlod · Post by **ray.wurlod** » Thu Jun 16, 2011 11:46 pm

Are you using a single-node configuration in DEV but not in TEST?