DSXchange

Short answer is "probably" if the cube can be loaded with SQL statements.
As Craig suggests, please be more specific.

Welcome aboard.

You could look at Tools > Options in Designer, but I really think it's something to do with the Windows colour scheme (theme) on the client machine.

Do you have any strategy for identifying the corrupted jobs, and how they may have become corrupted?

Here are the conventions I prefer to use:

IN = stream source
REF = reference source

OUT = stream output (e.g. to file or Data Set)
INS = inserts
UPD = updates
DEL = deletes
UPS = upserts (rarely used)
REJ = rejects

Looks like you have double-pasted the class name ?

Source code for all the SDK routines is supplied. Make a copy of (= clone) the one in which you're interested (UtilityRunJob) and edit its code so that it returns as soon as the DSRunJob call has been made. Do not have it wait for the job to complete.

It's not. Who creates these weird files anyway? My suggestion, just for the metadata import, would be to grab a head sample of, say, 30 records and translate (tr) the delimiter so something "regular" (maybe tab) that's not in the data. Then, in the actual job, use the delimiter string prop...

Use a Reject link on the Lookup stage to capture the failed lookups.

Direct the regular output link to a sink of some kind, such as a Copy stage with no output link.

Check that RCP is enabled on the output link from the Transformer stage.

Yes, but cloning and adapting UtilityRunJob. But I would counsel against this approach - how will you determine the successful completion or otherwise of the asynchronously started jobs?

Out of the box? No.

You probably need some QualityStage processing too, to get around minor spelling variations and the like, and so that your master Customer record is standardized according to your company's preferred standards.

Whatever your data and requirements demand, you design that. There's nothing special about an audit table.

One of the restrictions against using Bulk/Load is that it can't be a Multiset table. By default the Teradata Connector, when you choose Replace or Create, specifies that Teradata should create its default table style, which in most sites IS Multiset. In the Create Table options section in the Conne...

Use another Lookup stage. It is permissible to have more than one.

Or set the Lookup "failed lookup" rule to Continue and filter for null valued column downstream of the Lookup stage.

Use the DataStage Estimation Tool (in Designer).