DSXchange

That's already been explained here... and I'm sure it is documented as well.

No problem. Just as an FYI, importing the metadata from an xsd if you have one is preferable to importing from the xml itself.

If your issue has been Resolved please mark your post as such using the big button at the top of the page.

So... they don't match? Simply drag them into the proper order, that or Save and then Load the metadata directly in the stage. Depending on the location, you may need to use the right mouse button to drag, don't recall and can't check.

Please be specific as to exactly what you are doing in the source Sequential File stage and your settings in the XML Input. How did you import the xml metadata, i.e. your XPath Expressions?

Hashed file metadata is positional keys v. data, so I'm assuming it doesn't match exactly across jobs: column order and keys, specifically.

Sounds like you're close, you just need to move what you're doing either to the 'after job' area or a separate step post job. For the latter, you'll need to attach to the job in question rather than using DSJ.ME.

Since you didn't mention searching, that would be the first suggestion - search here for that error and see if any of the other posts on the subject help.

It is an Oracle error that could very well be caused by DataStage, hence the node question.

I can't imagine that all of those 17 parameters need to be passed to the job. If the default value is fine for that run, then leave them out of the dsjob command line.

I guess we shall see.

Hmmm... I meant the hashed file itself will need a VOC entry for the RESIZE command to work with it.

Your servers (and DataStage) will use all CPUs, so if you want to stay compliant you'll need to ensure you have no more installed than your license allows.

I was assuming 69 files. If so, that would be 69 new post-processing jobs which seems a little much compared to making a simple change to the original jobs. Hopefully, by now, they've actually done this and the question is moot.

Your parallel hint gets you a parallel query - parallelism within the database - but has no bearing on DataStage or the nodes it uses. 35 seems a wee bit high to me, btw.

So, what was the resolution?