DSXchange

Not impossible at all, just not generally practical. As I noted, I've created a 'keyless lookup' only for doing a 'select xxxxx.nextval() from dual' statement. Each row gets one with no key expression needed or desired. Other than something like that, however, I agree with you...

Not in a Parallel job.

A question for Support, I would think, as that's definitely not polite behaviour.

There really isn't anything like a 'lookup without keys'. If you are planning on joining tables then columns not in your where clause - i.e. used to join the tables - could be keys. The only time I've done something without keys was when the 'lookup' returned something regardless of the input row - ...

Please don't post the same question in multiple forums. I've already answered this over in the Server forum.

It pretty much said all there is to say. The 'branded' ODBC drivers that ship with the product are licensed for use by the stages the product provides. Any other use, including direct access in 'job control' like you are doing, violates the license agreement and throws this message. Build a job usin...

Ok, you gotta know what's coming next - what fixed it? Could you post that so that any poor slub who has a similar problem and comes searching will have an answer here, please?

Oh, and mark the thread as Resolved.

I'm here for ya.

Have you tried... oh, I don't know... reading the Parallel Job Developer's Guide pdf? Each stage has its own lovely little chapter in there.

Add a primary key to both tables. [/jk]

What are you trying to solve now? Are you saying it is working? According to Sunshine you're still generating warnings...

DSguru2B wrote:And if you feel like you want a fresh start all over, just reset the sequence, this way it will start from the first job.

Fixed that for ya.

Ah... then you needed to say so, because their goal as stated is not achievable.

And the same search that revealed that would have revealed the fact that all subsequent advise would be to delete it 'after job' if it was still empty then. So much for searching, eh?

Then why the heck not respond over there? Why start up the merry-go-round all over again?

Are you doing what was suggested there? The answer must be 'no' or it would be working. I'll wager you are using user-defined sql, yes?

Right now even if no records go in to the file,a 0 kb file is created,whereas i dont want a zero kb file to be created if there are no records..is it achievable? Certainly achievable. Search for specific methods. Someone forgot their 'not'. This is not achievable and what you would discover if you ...

I don't know about you, Ray, but I made the assumption that when someone complains about their end result being null that they know it shouldn't be. Call me crazy or a heretic, but... still, point taken and (hopefully) duly noted.