DSXchange

That is my understanding. Of course any developer can override them in a job.

You don't need a separate "job level" parameter. You can use the environment variable job parameter directly anywhere you can use a "job level" job parameter.

jaysheel wrote:It was working fine few weeks ago.

But suddenly its not working now and getting aborted with the following error message.

Added field has duplicate identifier()

OK, what has changed?

The initial value of COL4 is not a timestamp.

Oh, yuk. An old Pickie approach! You can do it all with SQL. The second table is a regular table, as noted. Dynamic normalization is accomplished by specifying the table name and the association name, separated by an underscore. For example: SELECT * FROM DS_AUDIT_MODS ; This will only yield the pri...

The error is from cpio. Get another copy of the software. Somewhere in the image you have a file header is invalid.

That's not the point. The NLS administrator is the user ID that started DataStage (and therefore owns the NLS shared memory segment).

Most of the sites where I've worked use sequential files to stage data; a few use database tables and a few use hashed files. There are advantages to each. I prefer sequential files, because they're fastest, but the other methods do allow you to set up things like security and integrity constraints ...

Unicode and UTF-8 are not the same.

Try using a "pure Unicode" map. Or save in UTF-8 format from Excel. Or use an appropriate Korean map in both applications.

Yes it's easy to do, but the fact that your audit table is editable would violate your SOX requirements, at least for a vigilant auditor. Therefore the separate table is probably better. It would need to be keyed by KEY, INSTANCE and DTM. Make use of dynamic normalization to find the appropriate row...

Please post the generated OSH so we can see what the generator operator is actually being asked to do.

This particular problem is very easily solved with a server job!

Looping in the Transformer stage is in the roadmap for DataStage. This means "beyond version 8.1" with no particular promise on when it will be available.

How about a simple expression to generate the last day of the month (perhaps in a stage variable), and build the new date from that? Don't forget to handle leap year (perhaps a separate stage variable).

Not possible. DataStage passes the row. If only this column is in that row, then that might be one way. Otherwise, your wrapped program could use the metadata to extract the particular column from the row.