DSXchange

What job type? You marked it "parallel" yet you are talking about a hashed file, which is not supported in parallel jobs.

Use Code tags to make your design clearer.

If you're running the jobs from dsjob command, specify the NOLOG flag.

DS_AUDIT is probably the easiest. Imagine that it has a child table called UNNEST DS_AUDIT ON MODS containing DTM and MODIFIER columns. Then use regular SQL.

Please describe all the stages in your job and how they are connected. What kind of standardization are you performing, and what (input) columns are being standardized?

The stages you're using are OK. What partitioning are you using, and what are the lookup keys?

Check with your official support provider but I believe any 7.5.x can talk to any 7.5.x.

In a normal lookup you will never see SQL, because SQL is not used to perform the lookup. Instead, the entire reference data set (partitioned as specified) is loaded into memory and an in-memory index built "on the fly" on the designated key, whether one or more than one column. It is this...

Multiple database stages. Funnel stage.

Is this a normal Lookup or a sparse Lookup?

What stage type is feeding the reference link into the Lookup stage?

Ordinarily prototypes for all databases are supplied in .odbc.ini. Find out who edited .odbc.ini at your site, and where they saved the original version (for example they may have called it .odbc.ini.orig).

No light to shed. The documentation is correct. It can not be done in a Sequential File stage, unless that stage has a filter command to convert the three different delimiter characters to a consistent one. Otherwise read each line from the file as a single string (for example if you need to preserv...

In short, you are not correct. Scope of User Variables is limited to the sequence itself; indeed only to activities directly downstream of the User Variables activity.

Your dates are of data type Date or String? If they're of type Date (which I suspect they are), they are binary values within DataStage, only translated to your default date format where necessary for display (for example View Data) or when being converted to string (for example being exported to a ...

You can do it with one, but I believe your two-Lookup solutions is easier to understand and therefore to maintain.

I believe that the User Variable activity was introduced in version 8.x.

But a version 7.x solution would involve creating a parameter in the sequence itself, and passing that to the ten job activities.