DSXchange

Why not simply use your SQL as is?

Search here for "dscc" to do command line compiles. As noted, though, what exactly are you trying to accomplish here?

That wasn't what was being suggested, the assumption it was made under was that the ODBC stages were a target in your job. I assume.

People typically adopt that as a convention, but no it's not required.

Not in a Server job.

I would repeat #3 but this time do a complete uninstall as noted here.

Please start your own post with the relevant details of your particular version of this problem.

We know... and you've been given the solution for that.

You missed the "capture all duplicates" part of the requirement, hence the other suggestions for a fork join design.

Unless we are talking about before/after routines, those are still in BASIC even for Parallel jobs.

If this was Oracle, you'd have to preface it with "alter session ", is that the case here as well?

You should be scheduling a wrapper script rather than dsjob directly, that way you can prepare the way and handle any 'before' or 'after' activities that may be needed. In order to create the same environment from the scheduler that the jobs would have when run from the client, you need to "sou...

Did you or did you not successfully create a VOC record for this hashed file?

I'm wondering if there is a new guiding spirit behind the homunculus.

Seriously? That's what you decide to put up as your 61st post?

If you are at all serious, start by googling for 'DataStage' or by visiting IBM's website dedicated to the product.