DSXchange

Are you running on Windows XP? If so, you may need to reinstall the XP patch that corrects this behaviour, given that you've reinstalled DataStage from its original media (without the patch).

These are not data types. They refer to the order in which pairs of bytes are stored/processed. More than anything this is a function of what kind of CPU chip is in use - over-simply Intel or otherwise. DataStage is automatically aware, because of the platform on which it is installed, of the correc...

No. Source code for DSD.StageRun is an IBM proprietary secret. It is the subroutine that invokes a server job Transformer stage. Your support provider will be able to ask the question, and someone in IBM can look at the source code, but you'll not be permitted to do so. In Appendix E of Administerin...

Yes. Do an all terms search for archive and log. This is one of the posts you will find.

If you're reading YYYY-MM-DD as a VarChar from a sequential file and loading it into a hashed file as VarChar, there is NO WAY it can be converted into internal format. Does your job look like this? SeqFile ------> HashedFile ------> ODBC I've been seeing a number of developers lately interposing ha...

Did you run the oracle.sql script (in the C:\Program Files\Ascential\DataStage\Documentation Tool folder) to create the tables in Oracle? If so, you've uncovered a previously undocumented feature (that's marketing-speak for a bug) that you should report via your support provider.

Refer to on-line help for full details.

Make the column VARCHAR in the hashed file.

Is the DataStage Operator role sufficient, with the "Operator can view full log" property enabled for the project? Then you only install Director client on the person's PC (use the xxxxx-DSDIR licence rather than the xxxxx-DSDES licence).

Have you read the manual for the CFF stage? It's a fairly comprehensive description.

The operating system will only allow one writer at a time to a sequential file. This is not a DataStage problem - it's a design issue. Have the shared container write to a different file in each instance - make the file name a container parameter.

Did you Save and close the workbook after adding the new lines?

You would use EVAL rather than WHEN construct but otherwise UniVerse SQL can do that kind of query. Indexing the constraining columns would help, as it would in any database. It would be more efficient in UniVerse to use Select Lists on which you can perform in-memory set operations (UNION, INTERSEC...

If it was working yesterday and it is not working today something has changed. We need to know that that change is.

I heard that every job is stored as an .XML in server and we can audit the data using this .XML file. You have been misinformed. Every job is stored as a number of objects in (Repository) database records on the server. Changes to most of these objects are recorded by DataStage automatically, with ...