DSXchange

Disable operator combination so you can learn exactly which operator (stage) is throwing the error. This may shed some light on what is (or is not) happening.

... unless you set ORLogging. However, I am assuming that the original question related to doing design work on the UNIX machine.

If you have Oracle, your organization already has the requisite Oracle client software. Get your DBA to install it on the DataStage server (and any other machines executing nodes).

cnjelita wrote:The Cleanup resources link is grayed out or not active from my director client.

It needs to be enabled on the General tab of project properties in the Administrator client.

OK that was a guess. But it still works - the level 01 item "contains" (in some sense) all the higher-numbered-level items. I'd be happy to learn, if anyone knows the real answer, especially if they can back it up with a reference.

Maybe you don't have SELECT privilege on the required system tables?

Not until you advise the blocking and matching columns you used in each pass. As an interim thought, phonetic columns (like NYSIIS of primary name) are better blocking fields then original columns.

handle_null(inputcolumn, replacementvalue)

The Parallel Job Developer's Guide is wrong to specify NullToValue(), which is for the Transformer stage.

If your company is using UNIX-based servers, then the authentication information will not be correct for your Windows-based server installation.

For urgent support sign up for priority service with your support provider. We don't do "asap", particularly not with "!!!!!!". DSXchange is an all-volunteer site whose members post as, when and if they can. "I'd love to help you out - which way did you come in?" (Grouc...

For urgent support sign up for priority service with your support provider. We don't do "urgent". DSXchange is an all-volunteer site whose members post as, when and if they can. One solution would be to throw more resources at the problematic job (maybe more nodes, maybe faster CPUs, maybe...

These are stage variables in a Transformer stage that presumably have not been declared to have an initial value. Then the job design makes reference to them.

Perhaps the importer has missed importing the initial value. Check on both projects. See if there are any differences.

And, indeed, please comment on why this question is posted as an Information Analyzer question. Information Analyzer typically does not run sequences of jobs at all.

It's possible but definitely not easy. Your script needs to get and read the xyz.fs descriptor file for the File Set, to get the individual data files (which may all be on separate machines). Then these have to be individually FTPd to appropriate locations. Finally you have to figure out how to edit...

Again, it shouldn't. It doesn't for me (8.0.1 on Windows).