DSXchange

Depending on exactly what you want the results to be, you might consider using the Rejects link in the Sequential File stage to capture the header lines, and process them after converting from raw format (if, indeed, you need to process them at all). This way only detail rows will appear on the main...

Curiously, this requirement CAN be fulfilled with a variable record layout. For this example I am assuming that there are no multi-valued fields in the hashed file, but it's an easy adjustment if there are. With hashed files the entire data record is accessible via the @RECORD system variable. There...

I don't understand your requirement. Is the job part of this sequence, or started from a different sequence? Where in your job sequence design do you propose to put the Routine activity? If downstream of the job, then the job status can never be "running", because it must finish before the Routine a...

You don't EVER need to do anything to the Repository at this level. If you either create or import the routine, indexes will be automatically maintained. DSU_BP.O is used - by DataStage - when the routine is compiled.

A similar approach for UNIX also involves the date command; the picture option is set according to your requirements. For example date +"%yyyy-%MM-%dd" > bippo.dat But, as Arnd noted, we are not sure whether you are asking about UNIX (your job type in post header) or DOS (your post body).

In server jobs there is one process running the job (DSD.RUN). Passive stages do NOT generate a process. There will be at most the same number of child processes as active stages (for example Transformer stages) but more likely directly-connected active stages will run in the same process. Any use o...

Open the row (Edit Row in the Columns grid) and set the Decimal Type sub-property. Click on Help to learn more about what this does. Beware that the data browser (View Data) may add leading zeroes to the display of decimal fields whether or not they are actually there in the data.

Yes it should be there. These are created when the job is created.

Perhaps you are looking for a solution involving two consecutive lookups; the first to determine the value associated with your "entered" value (a job parameter?), the second a reverse lookup to return all keys associated with that value. The second lookup would need "return multiple rows" enabled.

It's not clear WHERE you want the date, nor whether you want to generate dates. Certainly converting from 200603 to 20060301 is very simple; you make use of the fact that there are no data types in server jobs, and concatenate the "01". But I suspect there is more to your question than that. Do you,...

Why are you even bothering with a parallel job for one file to one file processing? How do you propose to "select" A etc.? Once we know this maybe we can provide sensible answer.

Why? Surely "in the middle of the job run" presupposes "for every row". The whole point of PX is to do as much as possible in memory; starting new shells will hammer your throughput.

Welcome aboard. :D This is as good a place to start as any. Based on the training you've just had, any questions should properly first be directed to the trainer(s), so that they can be aware of any areas where the training might be improved. There is a small FAQ forum here, a fairly recent addition...

Recycle your calendars! Monday comes around again every week, January comes around again every year, therefore 2004 will come around again!

It wasn't the Execute Command activity. Read the error message again. It was the Routine activity through which you invoked UtilityWarningToLog.