CDC Issue

[Vishvas]

Hi,

We are using CDC in our job. For that we use current dataset and history dataset as inputs. At the end of the job we create one more dataset which will act as hist dataset in the next run. We data is hash partitioned on the KEYS in both datasets. The job works fine if the history dataset is created in the same job.

The problem happened when we use another job to create a history dataset. That job just reads the data from teradata using ODBC connector and puts it in history dataset (this is again hash partitioned). When this history dataset is used in CDC job I didn't get the match for some KEY values although the values are there in the history dataset.

Did anybody face similar issues.

Vishvas

[sankar.td]

May be it is due to configaration file and partition technique you are using for creating dataset.
make sure same config file and partion technique should be there while making dataset and reading datset as well.