DSXchange

Could it be that you only had read permissions to the soft mount? Or that the mount was not even there when the failures occurred?

The command dsjob -log perhaps invoked from an after-job subroutine is one way, or you can continue to invoke the Utility functions via Routine activities in job sequences. If you want to write your own, the DSCAPI is documented in the Parallel Job Advanced Developer's Guide - in particular the DSLo...

Another link into the database stage, with the rule "update existing rows only".

There are NO log files in DataStage. Logged events are stored in database tables (hashed files) called RT_LOGnnn, where nnn is the job number. You can archive the events from them by a number of techniques, all of which can be found via a Search of the forum.

Step 1: read the entire message. Which job generated the invalid job handle?

Step 2: check that job in Status view in Director. Discover that is it not in a runnable state. Correct that.

Step 3: fix the job sequence so that the rule is "reset if required, then run".

Somewhere in your Columns definitions, somewhere in the job (not necessarily in the Oracle Enterprise stage) you have a VarChar data type with no (=zero) precision. This is illegal.

string[max=n] is the Orchestrate Framework's way of representing VarChar(n) in a schema.

OETL (object ETL) is not a three-letter acronym, and would therefore be frowned upon by IBM, the TLA company.

SETL (spatial ETL) is too close an acronym to SETI, a project that is consuming humungous amounts of CPU.

This (from today's Home page) has to be preserved for posterity.


There are NO MORE guaranteed Charter Memberships left!

Hurry! Get your Premium Membership before we realize what happened!
When we do, the discount ends and the rate returns to US$99 per year!

Have you imported the table definitions for these tables? If so, how are these column names represented in the table definition? The short answer is that, whatever this is, that's the name that you need to use.

To access data in DB2 tables on mainframe or AS/400 systems you need to use the DB2/UDB API stage and communicate via DB2-Connect.

DataTypePicS9 handles zoned right signed decimal numbers but ignores the scale. The easiest approach is to divide the result by 100.

It's a little more complex than explained. A continuous process captures changed rows in SAP into a text file. A new file is opened every hour. The SAP records in this file may contain Chinese characters as well as English. DataStage initiates (via BAPI) an ABAP process that consolidates the hourly ...

There is no field called NUM in the record schema but you are making some reference to a field of that name in your job design. You need to find it and fix it.

That is mega-weird!!! I trust you've reported it as a likely bug. I am not aware of any use the DataStage startup makes of the PS1 environment variable (nor can I conceive of any!).

For Windows replace all /dev/null with .\NUL - there is a NUL file in every folder.