DSXchange

Oracle is ... different. Storage is reported as NUMERIC and, by default, precision 38. That's too big for a BigInt, so DataStage raises an alert. If you can be 100% confident that the numbers will always be small enough to be represented as BigInt, then you can ignore, even demote, this warning. But...

Try setting the environment variable that prints the record schema used by each operator into the log. From memory it's OSH_PRINT_SCHEMAS (but please do check).

One quote only? Did you specify an environment variable called APT_PERFORMANCE_DATA" (sic) inadvertently?

The essential difference is based on the eventual use to which they are to be put. The Change Capture stage generates a structure that can be used by a subsequent Change Apply stage to reconstruct the After input from the Before input. A Compare stage yields not just the fact that there is a differe...

Tell us what happens when you try.

Use pairs of stage variables (the technique has been discussed here often) to detect change in the sorted keys - here "keys" has the same meaning as in the Remove Duplicates stage - and set a flag that can be used to constrain the output. :idea: Get yourself a premium membership so you can...

Those default values have to be coming from somewhere! Check all upstream stages.

Remove the default values from the record schema. I assume you've used different names to receive the key values from left and right inputs.

... or post in the correct forum.

Yes there is. This manual does not address the DB2 Connector. Nor is it in the on-line help until you apply the patch that adds the omitted documents relating to DB2 and ODBC Connector stages.

An INI file is not code. An INI file contains data read by code during its initialization phase.

Specify a conversion function date_from_timestamp()

... and bring them back together with a Funnel stage.

Enable RCP and let DataStage decide that, then inspect the output and see how you want to label things.

Two subrecords ---> two Promote/Split Subrecord stages.