DSXchange

No, we can NOT help you out, because your design is not clear.
Wrap it in code tags, as requested.

For that matter, what is the question?

Or are you just sharing the joy?

Can you explain - for the benefit of future searchers - how to achieve this result using a batch? And why you would prefer a batch to a job control routine?

Or just cat file1 file2 file3 as the Filter command in a Sequential File stage (the stage reads stdout from the filter command). The next question is whether you are intending to read this file executing in parallel mode (that is, multiple readers per node), in which case the header and trailer line...

What is the record delimiter property set to? This needs to be "None".

Parallel jobs do not permit "passive" stages to have both input and output link (reject link is a special case).

You need two jobs, possibly run under control of a job sequence. It's the only way. Ranga's solution does not achieve the dependency of the third file upon the output from the second.

That you have sorted the key columns is irrelevant.

You must tell the Aggregator stage that this is the case, by setting the Options/Group property to Sort - its default value is Hash

Try using an in-band null (that is, a value that is legal for the domain but not present in the data, such as Str("\",30))

You can convert this to an out-of-band null (using SetNull()) in a column derivation expression.

Before/After stage subroutines are not supported in parallel jobs.

Before/After job subroutines are supported, because the conductor process (which looks after startup and shutdown) typically runs on the node where the DataStage BASIC run machine resides.

If the first five rows don't match the metadata of the remaining rows, you can get away with doing nothing, just allow the import to fail for these rows (though it will log a warning message). You can capture these import failures via a Reject link on the Sequential File stage, thus allowing your to...

Create a Rejects link on your Sequential File stage. This will capture any row that could not be imported because it does not match the metadata. Note also that the "representation of NULL" (the Null Field Value property) in a fixed width file must have the same number of characters as the field its...

Welcome aboard. :D

There's no need to re-quote everything - it just wastes space on the server.

Go to your job's Parameters grid (in job properties dialog).
Click Add Environment Variable.
Choose APT_DISABLE_COMBINATION from the list supplied.

When you run the job, set it to True.

What's changed? Involve your support provider. You need to determine whether the quotes have been stored in the job design. A quick test would be to create a trivial job that uses parameters of the same kind and see whether its parameters are quoted in the various environments. Check the job log for...

Nac Mac Feegle!

(The Wee Free Men)