DSXchange

Not sure anyone will just hand over a working routine for you. How much coding experience do you have? It wouldn't be all that hard to write what you need in BASIC, but it will take a bit of research to know all of the functions and syntax you'd need. From a high level, I'd think you could do someth...

Is this scenario can be met with the datastage Just for the record, this particular scenario can't be met with any tool. As noted, comitted is comitted, so no rollback is available after the fact. Now, that doesn't mean you can't play fancy games with some kind of 'before image' as Ray notes, or wi...

You are trying to make 2 records out of each set of 7 records? Or do you only need to produce the 'heading' row once?

You would need to use stage variables to (ultimately) build a concatenated record (including field delimiters) and then only write it out when you have a 'complete record' as one long varchar. That could be when you read the "EndorsementSqn" record or after you've read 7 records, depending on if you...

Please verify that you will be using Server jobs for this and that you are indeed still running a 5.x version of DataStage. If that's the case, I'd think your answer will be to write out a single file with this unique value as the first field and then use some other mechanism to split it up into sep...

Cool. Like Arnd, I too ran into this problem several years ago and had a bitch of a time trying to figure out what in the heck was going on. Once you do, though, you never forget.

And as Arnd noted, it's not an issue that manifests itself using the 'test harness' code that the Manager generates.

Sure... but if you always want to start from 1 use the @OUTROWNUM system variable instead.

Without knowing what 'KeyGenerateNextValue' is, not sure how anyone can help. And to 'convert' in Server, just move it to a numeric field. Or you can do math on it as well.

Interesting. So perhaps for these tests you can ditch the 'mod 50' and log each and every iteration (with job name) to see if there is one particular job causing the problem?

I'll bet it's because you are attaching to the current job, which can be... problematical. You want to check the jobname and not even attach to it, not simply detach when you hit it without doing any processing.

Sheesh... that is a ridiculous value to use and falls into the 'bigger must be better' camp. Typically, you need to be aware of your network packet size and average record length then adjust your Array Size accordingly. In other words, there is no one magic value - it will vary from job to job. You'...

Well, as noted, it certainly can be done in one Server job. Rather easily. No magic required.

Exactly. Look for the 'XML PACK' pdf documentation in your Docs directory. Also go to Kim Duke's website and download the XML Best Practices document that is in his Tips section.

Sounds like what you are missing are the XPath Expressions that drive the XML parsing. Read the docs.

Array Size is neither a project level variable nor a magic bullet for performance issues. You'll need to get more specific if you want specific help.

Why not just substring the first character and check that?