DSXchange

Tail stage set to 1 row and executing on one node only (perhaps in sequential mode).

... with a leading slash: /opt/SUNWspro/bin/CC

Use categories. Segregate the parallel jobs using categories. Export only those categories.

You can drag/drop jobs between categories - you could even restore them to their original locations later. 70 is not a huge number.

MultiLoad locks the table. This is documented. Cautious approach would involve executing a BTEQ script to ensure that the table is not already locked. And you can not select or read from the same table while MultiLoad is running. Your job design must handle all possible abnormalities in data without...

You can not change the job parameter value (even if it is an environment variable) once the job is running, even by changing the environment variable. You are correct that the $PROJDEF, $ENV and $UNSET default values are not available in version 6. You need to manage this is the controlling job sequ...

1. No.
2. Yes.
3. Yes.
4. Maybe. But the Merge stage is limited to two inputs so you would need two jobs.

Strange, isn't it, how they're happy to configure - even over-configure - table spaces, but unwilling to configure work space (temp space, rollback segments, etc.). It's like they don't want anyone to touch their baby. (If it's any consolation, many mainframe administrators are like this too, only w...

A negative difference could be meaningful, for example difference between date required and date shipped. Negative might mean that the shipment is late, and the business rules (contract) may stipulate penalties in this case.

:D

I do my best. See my web site.

However I've never used the Command stage; it's my philosophy not to do anything platform-specific, in case the business decides to move platforms.

I trust that everyone noticed that SDK routines are catalogued with a "DSX." prefix (rather than "DSU.") and that the catalog name is all upper case.

You've marked this topic Resolved. What was the resolution? Help others who may encounter this issue in future by posting the solution here.

When you first deploy a job, the structure of the project will be created, provided that your run profile specifies the location of your master project directory (the parent directory on the server of project directories).

If you still have the old job log entries, you can look at the second event of any job run to view the environment variable values with which it ran. Do you archive your job logs before they are purged? If so, consult the archives. Unfortunately, restoring a job log from a system backup is a non-tri...

Further, a Data Set can have multiple physical files (segments) per node. So 2GB is not an issue. To perform more detailed diagnosis you will need to disable operator combination, either at the job level (via APT_DISABLE_COMBINATION environment variable) or at the stage level (via Advanced tab on st...

Design it well; you should never need to kill anything in a well-designed DataStage job.