DSXchange

If your output link order is set properly, then it will process link 1 first, then link 2 and then link 3 - on a row by row basis. If you mean you want it to process everything through link 1 and then process everything through link 2 only after link 1 is complete - then, no, that's not the ways thi...

Read the original post. They are using a single link with one of the combo update actions so there's no way to get separate numbers for inserted versus updated records.

You'll get one number - the total number inserted and/or updated - but not the two distinct numbers the OP wants.

Check out the next to the last entry in this thread.

What exact 7.x version do you have, Peter? In the 7.5.x versions the parameter value in a Sequence supports expressions, so you wouldn't have to touch any jobs, just the calling sequence. Meaning, you should be able to strip the quotes the scheduler needs as it is passed to the job(s) that needs the...

rafidwh wrote:Now how can i get the count of number of records updated and number of records inserted? Note: Here I am using only one link not two links.

Can't be done.

Interesting... thanks for the clarification. Still seems to be something unique to PX jobs, but I'll keep my eye out for that behaviour over in my Server world.

Just curious - where did the [38,10] come from? A NUMBER in Oracle, one with no precision specified, is the same as saying NUMBER(38). There's no scale associated with it. I'd suggest trying [38] or maybe [38,0]. Not familiar enough with PX to get any more specific than that. Some stages over on the...

You need more than 700 lines of user defined sql?

I'm assuming Teradata has to concept of a 'view', yes? That's what I would do - have the query turned into a view so that all you need to do in your job is 'select * from' on it.

If you've solved your problem, you really should post the solution back here as well. That way when future peoples with the same problem come searching, they'll be able to benefit from your pain.

You mean to say you have no idea where DataStage is installed or where the Projects you have defined live? You will need access to the DataStage Server you are connecting to with your client. I'm not familiar with the directory structures on a Windows install but I'll assume it's close enough to a U...

Actually, TRANS1 is the compiled code from the first transformer stage in the job. Are you using stage variables there? Are you calling custom routines? Both of those can be sources of these kinds of phantom errors. It mentions a specific line of code, you could go look at or very near it to get an ...

Being new to DataStage, some of the advice you'll get will probably sound like Greek. If so, try taking the keywords and searching the forums here as pretty much any question you'd have has already been asked and answered. Two more ways: You can also create a new link in the job to an Aggregator sta...

hi thanks for reply. But... Just a thought, but if you are going to start off a new thread with 'thanks for the reply, but' you really should reply in the original thread - not start a new one. Then people know what started it all, what you are talking about and can bring it to its rightful conclus...

No. Remember the 'Server is typeless' mantra. And that the zeroes are not significant. So you need to explicitly 'format' them into a character field in order to preserve them.