DSXchange

Not if you jump on the end of someone else's problem, I'm afraid.

I've split you out into your own topic but because this was your first post I had no idea what to put for your Version # or O/S flavor... can you clarify that for us please?

Once again, simply saying "it doesn't work" doesn't help us be able to help you. Doesn't work how - what happens, what errors do you get? :? You seem to be rapidly moving past the point where strangers on the Internet can help as we seem to be going around in circles... have you engaged yo...

Yes, expected. Any target file is created (if it doesn't already exist) when the job starts and the stage "opens"... regardless of it getting (or not getting) written to.

FYI, the use of Index() is unnecessary as Field() supports not only a start field but optionally how many to return... 1 is the default. Just make sure when you add it that the "how many" value is high enough. I tend to use 99 even if I know there could only be a handful. Makes what is goi...

"Preparing the table"?

If you're good, go ahead and mark this as resolved and I'll move along. I'm just morbidly curious at this point.

Can we revisit what exactly "not working" means at this point? What output do you see? What is displayed if you type "echo $?" immediately after dsjob runs?

Yup! Always best to become conversant with all of the functions at your disposal. Or at least know where to go look for something to do the trick.

For this, Field() would be my first go to function.

Well, that's how you do it in a Parallel job when you just need to trigger an action in the database. In a Server job you would start from a Transformer, they can be re-jiggered to serve as a "row generator". Are you doing something before/after SQL or some kind of "hard-coded" u...

Okay... at a high level if it only issued a commit after 2000 records or end-of-job then saying that when record 6 fails the job the previous 5 are still in the target didn't make sense to me, hence the question. Something else is going on, perhaps partitioning is affecting the behavior - one node c...

If you are letting the Sequence job run everything, then in your custom routine you would need to attach to the job(s) in question again - establish a new handle for the routine to use - and then call any other API functions like DSGetJobInfo. Don't forget to release it when you are done. You have m...

So... that addresses the subject of the question but not the substance of it. Hopefully more help will be coming.

We'll see what kind of answers you get but to me these sound like questions you should asking your sales person.

I'm also curious (based on your forum choice and job type noted) if you are really asking about a Server jobs only environment...

What's your commit interval set to in the job? Sounds like it is set to 1 to see that behavior.

Now hopefully someone can answer your question...

Seems to be Enabled for Information Services option in Job Properties will help on this No, it won't. If all you need to do is run a "normal" job from outside of DataStage proper, then you need to get familiar with the Command Line Interface and API. And everyone has access to the same on...