DSXchange

uvodbc.config IS in the project directory. .odbc.ini is in the $DSHOME directory.

If you have Enterprise Edition licensed, the best (most accurate) method is to import using Orchestrate, using the orchdbutil facility.

This is a server job. Conditional lookup is not supported. Nor is there a RawLength() function.

Do you have something else (another telnet service?) operating on port #23? For example I run MKS Toolkit which has its own telnet service AND UniVerse which has its own telnet service AND UniData which has its own telnet service. By default all of these want to use port #23; in practice only the fi...

Either write it as VarChar or do nothing and change the cell format within Excel.

Welcome aboard. :D

What else (another telnet service?) do you have executing on port #23?

Leap years. 99 of them every 400 years.

Add a Copy stage after the Surrogate Key Generator stage. Take a second output from the Copy stage containing just the key column. Pass this through an Aggregator stage and form the Max. Force the Aggregator stage to operate in sequential mode. Its output will contain the highest value generated.

Before you have saved, NO.
After you have saved, YES. Even for extant jobs.

Going forward, learn to Import then Load, rather than need to Save. You will make fewer mistakes in this way.

Having performed your lookup, and still in the Transformer stage, if the lookup failed then all columns from the reference input link will have been set to NULL and you can use this fact to determine whether a lookup record was or was not found. In this case you do not have any record, and must impl...

Think also about Part B of your scenario. A previously non-existent dimension key value appears more than once in your source data. How will you manage the second and subsequence appearances?

Etiquette Note We don't do "urgent" here. All posters are voluntary, who post as and when they can. If you want "urgent" sign up for premium service with your support provider, and learn the true cost of urgent. SMP implies standalone, doesn't it? CPU capacity is at least as important as memory, an...

What version of DataStage precisely? The answer is probably yes, but I'll not commit to that without knowing your version.

Manager Guide has a chapter specifically on configuration files.

Load Immediate = "No" means that you don't want the table loaded now, so that behaviour comes as no surprise. DataStage simply creates the data and other files needed by the bulk loader - someone can run the bulk loader executable - whether under DataStage control or not - at some later time.