Interpreting The Job Score

[jonesv]

Hi,

I have a job that APPEARS to be the same between our DEV and QA environments. I took a look at the generated OSH code and it is identical in both environments. I then took a look at the job score and noticed some differences. One that I would like to get to the bottom of is that the job score for the DEV job reports that it has 6 datasets and the QA job has 7 datasets. Here are the diffs:

DEV: (datasets 0 through 4 are identical in both envs)

Code: Select all

ds5: {op1[2p] (parallel APT_CombinedOperatorController:ds_RegData)
      =>
      /DataStageProjects/Multi_Channel_Rtl/ILOS_Online_Shopping/CheetahMailRegistrationFeed/datasets/RegData.ds}

QA:

Code: Select all

ds5: {op5[2p] (parallel delete data files in delete /DataStageProjects/Multi_Channel_Rtl/ILOS_Online_Shopping/CheetahMailRegistrationFeed/datasets/RegData.ds)
      >>eCollectAny
      op6[1p] (sequential delete descriptor file in delete /DataStageProjects/Multi_Channel_Rtl/ILOS_Online_Shopping/CheetahMailRegistrationFeed/datasets/RegData.ds)}
ds6: {op1[2p] (parallel APT_CombinedOperatorController:ds_RegData)
      =>
      /DataStageProjects/Multi_Channel_Rtl/ILOS_Online_Shopping/CheetahMailRegistrationFeed/datasets/RegData.ds

My question is why does the QA job have the delete data files and delete descriptor files part? I looked at the properties of the only dataset in my job, and didn't see any differences between the two environments. Both jobs ran with a 2 node config.

Thanks!

[ray.wurlod]

You should have delete descriptor file whenever the mode in the Data Set stage is Overwrite, but not when it's Append. Can you please check this property?

[jonesv]

Thanks Ray. That is what I thought, but just wanted to get confirmation from the experts. :D

They are both set to Overwrite. But I am suspecting that something is amiss in our dev environment. They are small jobs, so I am going to rewrite the dev job from scratch and re-export to qa.

I am going to mark this thread as resolved.