DSXchange

In the Lookup stage you can specify how you want failed lookups to be handled. Generating an error is only one possibility. There are four in all.

IBM web sites. Some of the white papers have good overall architecture diagrams. Another good source is IBM training materials.

QualityStage jobs in version 8 run using the parallel execution engine, which runs as osh. The first process to start is the conductor process (osh) which triggers section leader processes on each node (osh) which, in turn, fork player processes (osh) on each node. The engine is the "PX engine&...

Nothing is mandatory. But think about what will happen if you don't partition on the first sort key value. The same value of the first sort key may occur on more than one partition which means you will end up with more groups than there really are.

(no access to unix account)

I doubt that no log is produced. Have you checked the base job as well as the individual instances in Director?

Yes. You can create sequences that reset jobs automatically (if required) before they run.
You can create a routine that runs through all of the jobs in the project and resets any that require it. You could also create a job that performs that task.

Look in the Available Properties to Add pane in the Sequential File stage. There is a property that allows you to specify the column name that will receive the file name. Make sure that a column of that name occurs on the output link.

Design-time or run-time? The jobs that are actually executed from a sequence may be a subset of the job activities that occur in the sequence, depending upon the logic of the sequence. Design time: look on the Dependencies tab in the sequence design. Run time: look in the "summary of sequence r...

Convert from ustring to string within your DataStage job, using either a Modify stage or a Transformer stage.

A Data Set consists of a single ".ds" file which points to multiple data, or "segment" files. Even when you truncate a Data Set the ".ds" file retains its pointers to the data files; it is these that are emptied.

Not possible. But there is a computed GOTO statement.

NO, NO and NO.

The .ds file does not contain data, and thus must not be truncated.

Use orchadmin truncate, as Arnd advised, to empty the Data Set of records.

It usually suffices to partition on just the first grouping (sorting) key. If this has only a few values in its domain, add the second grouping (sorting) key to the partitioning algorithm too.

The empty set. With NLS not installed, DataStage uses ASCII internally. ASCII is a single-byte character set.