DSXchange

If you reset the job (not recompile, reset), any diagnostic information in &PH& should be transferred to the job log, in an event called "From previous run...". Otherwise, &PH& is a directory within the project on the server. The file will have a name made up of the program name (DSD.RUN or DSD.Stag...

Given that Ascential do not supply the compiler, they probably won't be able to help if it's a compiler problem.
But they may know whether you need a patch from the compiler vendor. Have you sought support?

Why not just use the inbuilt link row counter (@OUTROWNUM)? Run this through an Aggregator with the Last set as its aggregate function and thence to your file.

Does your design also include an Exception handler? If so, what happens when it fires?

Welcome aboard! :D When QualityStage is invoked from DataStage, the concept of "file" can mean "data stream". It still needs "file", so that it can find the file definition in the QS project directory, but it does not need a physical file; the data stream can be managed in memory. This becomes even ...

No rows will be rejected if toFilter.UPDT_TXN_TYPE_CD is "D" for every row.

If it's accepting Date() then it appears that the stage is expecting a date in DataStage internal format.

Try using just the Iconv(), omitting the Oconv() from your derivation.

I've noticed (7.1 on AIX) that the job entries are to be found in atjobs rather than in cronjobs. But this really addresses the "can't schedule" problem rather than the "won't start" problem.

The database ends up doing the same amount of work - maybe more - than if you'd done the original joins!

You will find comments at the top of the routine in RT_BPnnn that show you the mapping from your column names to the internal ones used in the routine. You could read that code in a routine and parse it.

It's not something I've ever wanted to try.

Or she.

Of course, you may be able to discern what's in what column by inspection but you would need to be lucky (for example string, number, string, number kind of layout). For all numeric data it's far more difficult.

Your assumption that UniData is supported by Ascential DataStage does not mean that an ODBC driver is supplied. DataStage supports Red Brick Warehouse, too, but does not supply an ODBC driver. However, you will have, on your UniData client CD, an ODBC driver for UniData (called UniODBC). You can use...

In versions earlier than 7.5, your Exception Handler can trigger a Routine Activity that calls UtilityAbortToLog, which will abort the entire job sequence.
It's not a technique that I like (I prefer to create jobs that don't abort), but it works.

Find out what your default Oracle date picture is.
Then make sure that whatever you're passing to the OCI stage matches this format precisely using Iconv/Oconv functions or substring/concatenation techniques.

My guess is that it's the number on the sticker on your CD case.