DSXchange

Anything you can run from the command line you can call from the Execute Command stage.

There's no "View Data" from the Merge stage.

Would any 7.x installation have Information Server Manager installed?

So... do you understand what that does and why it fixed things for you?

That's not any kind of 'julian' format.

How many nodes are you running the job on?

The syntax *is* fully documented.

And here we go, moved to a more proper location.

No, not possible in just this job. You may want to start considering an alternate approach - one could load these files into work tables in your database of choice and let it then do the join there. That or put the 'right' file in a hashed file and stream the 'left' file in doing a normal reference ...

I'd also suggest a more descriptive routine name than simply "Count".

Technically, everything is just 'zeros and ones'... somedays I feel like I am, too. :wink: Have you looked at one yet, say via a hex editor or octal dump? I'm sure Ray will be along with a much more technical answer but at a high level it is the format that DataStage uses internally - meaning a stra...

Get your support provider to analyze the crash dump?

Defer the start of this job until after the three other jobs complete?

Post the complete, unedited error. And use an External Source stage, not a Sequential File stage to feed in the XML URL/Filepath, otherwise you are asking for trouble.

What's the failure message? Anything else useful in the log after you Reset the aborted job? How large are your files?

The Merge stage uses hashed files under the covers and I'll wager you're slamming into the dreaded 32bit addressing '2GB Barrier'.

In other words, you'll need to build your own 'job control' to be able to do anything like this, be it a simple 'pre' check in a Sequence as noted above or as complicated as you'd like.