DSXchange

You can use the uvwalk command to traverse the directory (folder) structure. If you have a common location or naming convention for your hashed files, then life becomes simpler. For example you might be able to use the -prefix option. Combine the results with dir commands in a BAT file to determine ...

Lock tables are hashed files (in memory). As I said, there's lots of kinds of hashed files. But I will not elaborate further, as it doesn't help anyone.

DataStage is not a codebreaker. It can only do what any other database client can do - issue SQL that includes appropriate SQL functions for performing particular tasks, and receive the results of such SQL.

You need to discover what constant your particular database (you chose not to reveal which it was) uses to represent the current date. Is it SYSDATE, CURRENT_DATE or some other formulation. It is not a DataStage problem. That said, the Sort stage does not have the capacity to execute SQL in any case...

Please think about what you've just asked. You should be able to answer your own question. A Job activity waits until its job finishes.

It's not per job, it's per connected client.

...none of which will work on the Repository when you get version 8.

But far, far better is to eliminate the cause.

Users are created using unskilled labour!

A man and a woman need to be involved at some point. And performance is slow. Using parallelism, however, nine women ought to be able to create a baby in one month.

Users are created using unskilled labour!

Wanna bet? DS.REINDEX ALL with connected users can make things worse - you can guarantee that the index is emptied!

Perhaps, but I've seen it at three separate sites.

That's because you're breaking the rules again. The operating system (not DataStage) only allows one writer to a file at one time. You must write to separate files, and assemble them into one afterwards with a TYPE command.

Prefer mput to send multiple files. The suffix ought to make no difference.

The ORABULK stage invokes sqlldr.

You may prefer to use a Sequential File stage and your own, custom, sqlldr control file (with tuned buffer sizes, etc.) You can also truncate your table with an OCI stage set to "truncate and insert" but which you send no rows.