DSXchange

MT wrote:(there is no 31-JAN-9999 etc.)

Yes there is. There's a 31-JAN in every year.

Search DSXchange for posts about problems that can occur with namespace identifiers. I suspect that this will help you resolve your issue.

This is easily handled using a server Sequential File stage, which has (among other things) a "Missing Column" rule for each column.

You can use this stage in a parallel job be encasing it in a server Shared Container.

Yes, if you're on two nodes and the Entire partitioning is on the stream input. It would be quadruple with four nodes.

Please clarify "not working".

@DATE is the system date that the query began executing, in internal format. So @DATE - 24 is the date 24 days before that (not 24 months).

You need a better technique, perhaps @YEAR - 2.

Add constraints on PS_TIME (and maybe PS_DATE) with the constants in the WHERE clause being job parameters. For example: WHERE PS_TIME > '#jpMinTime#' AND PS_TIME <= '#jpMaxTime#' If you need a period to run through midnight, construct a "timestamp" in the SELECT query and filter on that.

The conventions for shell and environment variables in UNIX is:

• when the variable is named, it is not preceded with $
  
  when the value of the variable is retrieved, the variable name is preceded with $

DataStage follows this convention for using environment variables.

Two nodes? Entire partitioning?

Does your server job still work? If so, keep it. Server jobs can be run even though you have Enterprise Edition.

Best practice is to use a dedicated project for Information Analyzer jobs, usually called ANALYZERPROJECT. The integration between the two products is only at the metadata level; published analysis results are available in shared table definitions accessed from DataStage. If you use RCP in DataStage...

Sigh.

Lots of differences, including the range of sources/targets that you can use and the range of transformational facilities available. You can research each using the internet if there's something in particular that you're interested in.

You can find a decryption algorithm by searching DSXchange, but I didn't tell you that it's there.

Select Oracle Enterprise Stage in the Stage Types > Parallel > Database folder of the Palette. Right click and request a Where Used analysis.

I don't recall there being a version 6.5. That said, any 6.x can, with the exception of Version Control components, be migrated to any higher version currently extant.