DSXchange

That crossed my mind briefly also, but I felt that INSTALLMENT_NO would be different in different rows, which makes that approach messy. Doable, but messy.

There's no CASE - you will need nested IF..THEN..ELSE

As to whether there's a simpler way, probably not. There does not appear to be any pattern to the test expression values or generated values.

I suspect that you are just using the default DataStage user in Information Server, and have not credentialled the individual users separately.

It doesn't matter whether it's two or 50 or millions, you're going to need some code. It might be a BuildOp, or you may prefer to script it and use an external filter stage. Or you may like to encapsulate your code in an Oracle stored procedure and call that. I believe there will be a looping constr...

There's a section in the Parallel Job Advanced Developer's Guide on writing custom stages. How good are your C++ coding skills?

Not "sequencer". "Sequence".

A sequencer is a component that you an use in a sequence.

The BAT script needs to begin with a CD to the project directory.

Then run the uv (or dssh) command with the query as its argument.

You are on version 8. Nothing is simple any more. There are two repositories - the local-to-project DataStage repository and the common, unified, metadata repository for Information Server. Can you delete the routine from the client? Otherwise have your support provider advise how to use the scripts...

There must be some other stages in your job - a Link Collector stage must have one or more input links. It is my understanding that the Link Collector stage creates a process for itself and that, as a result, you must have inter-process row buffering enabled if you are using this stage type. As far ...

As I understand it, it requires Oracle to use OS authentication rather than internal Oracle authentication. OS level authentication was used exclusively at a recent project (using Windows servers) where I worked, and totally successfully.

936 is a Windows code page.

Clearly not, because that is not "as per Ramesh's requirement".

Each pin record has, somewhere in its structure, the DSRID of the pin on the other end of its link. Its value has the form VnSnPn so it should be fairly easy to spot. For example, pin V0S1P1 (DS_JOBOBJECTS record J\n\V0S1P1) might have field #5 containing V0S2|V0S2P1 meaning that stage V0S2 is on th...

Nor can you have two adjacent Modify stages. You can have two (or more than two) Modify stages, but not adjacent. Place a Copy stage between them.

Parallel Job Developer's Guide, Chapter 2 (near the end). Why? Because the server shared container stage contents must execute in a server job context and that can only be guaranteed if you can guarantee that the server engine is accessible. You can not guarantee this in any multiple machine enviro...