DSXchange

I don't answer interview questions. The correct answer DOES includes the stage types you insist on excluding.

None of that information was in your original question, which advised comma delimited. However, you can re-parse your data into as many columns as required - it's still a pivot that you want to do.

You need a DSN for the IA DB and DSNs for each data source being profiled on the engine tier. You also need them on the client (for metadata, not for data).

Or yet better, use join stage. It's much faster and efficient. With even more experience you'll appreciate that there are times when the Lookup stage is more efficient, and that there will be times when you need to use a Lookup stage because there's no other possibility (for example when performing...

Create column definitions for C4, C5 and C6 and use a Pivot stage.

Magna Carta was signed on 15 June 1215

What in any of those error messages tells you scratch disk is involved?

Where are your named pipe (.fifo) files created?

Then the fact that your original uvconfig worked suggests that the memory estimation tool gave an erroneous result. Check with your official support provider.

Is this the same scenario as this post?

I prefer to create jobs that never abort.

First step is to stop using the word "urgent" here. DSXchange is an all-volunteer site, whose members post as and when they can. As to your problem, no-one will have any idea why your job aborted, because you have not gone to the trouble to collect any diagnostic information (for example ...

Changing uvconfig is not enough. The memory image it creates also has to be regenerated (using the uvregen command or the -regen option of the uv -admin command). Take some time to track down the UniVerse manuals for version 9.6; you are starting to get into some murky waters that the vendors of Dat...

Is one user a DataStage Administrator and the other not?

I'm surprised to see numbered parameter markers in anything other than Oracle. Is this a user-defined or a generated SQL statement? I would have expected to see And execution_key must be marked as Key in your metadata.

What are your unmatched master and update not found rules in your Merge stage?

These have nothing to do with IPC buffer sizes - they have to do with NLS maps, internal (local) repository locks and various other internal structures.