DSXchange

It should be identical. You use the same instantiation string and it exposes the same objects.

There's no direct equivalent. First you need to know what terminal type your telnet client is emulating, for example VT100. The command to tell DataStage (dssh) about this is SET.TERM.TYPE SET.TERM.TYPE VT100 This command supports a number of options, which you can find out from HELP or from the Uni...

You are permitted to create environment variables, using the Administrator client. Not every environment variable mentioned in the manual is part of the product as shipped, though the ability to react to them is.

Posting the same question in multiple forums is annoying and wastes our time. Please refrain from doing so.

Does the new user's login script (for example .profile) cause them to be directed to a directory other than the DataStage project?

Welcome aboard. This question has been asked, and solutions offered, in the past. Search is your friend.

Is PXEngine visible to (copied to or mounted on) all nodes mentioned in your configuration file?

... or use a Remove Duplicates stage to take the final member of each group (which you would have found had you really done "enough" searching).

They may not have the same names on the input links.

As far as the database server is concerned, a DataStage job is just another client application. The machine on which that application runs has to be configured as for any other client application that accesses that database. There is nothing special for DataStage, though the dsenv script is a conven...

Enrol in a DataStage Essentials training class. DataStage is no longer a tool that can be used "out of the box" by a novice. At least not effectively.

You omitted generating the surrogate key value in a Transformer stage.

Yes. Change capture can be done externally to DataStage. Read about IBM's acquisition last year of Data Mirror, and search the web for other techniques such as processing transaction logs.

The Palette is fully customizable. Categories are totally artificial.

Or (in version 8.0 and later) you can store this information in a Connector in your repository, and use this in a Connector stage in your job. It amounts to the same thing as what Craig described, but you only have to design it once and you can test it before using it "in anger".