DSXchange

Write it to something simple, a flat file for instance. Then an Execute Command stage can 'cat/read' the value and your triggers can decide to run Job2 or not based on the result.

You can't "group by item", you need to group across all items. Add a new field with a constant value and group on that. You'll get one row from the Aggregator with your sums that you'll need to assign 'JKL' to and then some kind of a fork-join design to put the two sets of data back togeth...

Need more information. Each xml file will be read by the stage as 'one record' so that in and of itself is not an issue. What is your expected output here? Are you generating a sequential file? Can you be more specific about your output, please? Examples are always good.

Server job.

Ah well, worth a shot... sometimes that 'invalid sql' message is simply the trailing semi-colon you've left in after pasting it in from Toad. :( Not sure what to try, you may need to take this to your official support provider or wait for someone who has done this before to chime in. Me, I've had ab...

Try it without the trailing semi-colon.

Didn't you already post this? Looks really familiar but I can't seem to find it, hmmmm... :? Regardless, a couple of things to check. Does the proc actually raise an exception when it has a problem? Also, have you looked at the tab in the stage where you define custom error levels? Don't have anythi...

You go out to your operating system, "source" your dsenv file to setup the environment and then run the commands. You'll see the output and you can then decide if that is appropriate for your needs. All of these will require some amount of "iteration", meaning you'll get a list o...

OK... gotta ask... why in the world would you *not* be "allowed" to use the stored procedure stage? What kind of "standard" is that?

When you said "after stage" in the OCI stage did you really mean "after sql"?

You need to "source" your dsenv file to setup the environment:

It returns TRUE if (StageVar = 'Y' and the lookup succeeds) or if the lookup fails.

Can you not load the results from Table A into a work table so that downstream from there you can join it to Table B to constrain the query?

How did you generate the metadata (XPath Expressions) that you used in the stage? From an xsd or something else?

Not really. Why not explain to us why you need to use 'less key columns'? Then perhaps people could provide more specific help.

Any chance you could let us know what 'system level tuning' IBM suggested?