DSXchange

If you are doing DataStage work for someone, they should be able to give you what you need.

No such thing.

Assuming your metadata matches the detail records and thus the others were silently dropping during import for metadata mismatches. Define the file as one long string instead.

paranoid wrote:Can the existing DS software that is installed in solaris be used on AIX box?

No. As has already been stated twice.

Need to? No, but that's certainly an option. You could also do this is one and then use a Funnel to bring the streams together.

Being 'very new' to XML is going to... complicate this, I'm afraid.

Try hash partitioning on the key field(s) used in the upsert.

You'll need to create them separately and then combine them in the proper order as the final step.

There's no magic here, you'll need whatever number of 'dsn' that you need to get this done. After that, seems like a multi-instance job would be the way to go. Out of curiousity, is this a one-time task or something you'll need to repeat on a regular basis?

Give that a shot, but I honestly don't believe it is as simple as that even though it seems like it should be. Let us know.

paranoid wrote:Does that mean that Solaris version of DS that is currently installed on Solaris box cannot be installed on new AIX box? Do we need to request AIX version from IBM?

Yes, of course you need to request/install the AIX version.

No, there's no shortcut there. Derivation Substitution in a Transformer would certainly be one, however.

As noted, the very important first step is to identify them. And that doesn't mean as junk or a 'box' but rather the actual ascii value of the character(s). Enlist the help of your SQL Server DBA if you are unsure how to do that. Then you'll be able to decide if/how they can be accomodated in the ch...

Have you used the Loop stages before? This would be a pretty simple design - run the job inside the Loop stages, tell it to loop three times and use triggers to check the job's completion status to decide what to do each iteration. Job ran ok? Exit the loop 'early' by branching to a stage after the ...

Now that one I can't answer. Problems in Server jobs are always reported by the active stage, hence the error from the transformer when the bulk loader went astray. You'd probably need to involve your official support provider if you really wanted to track that one down.