DSXchange

A safer bet may be to copy and paste just the user defined section of the DSParams rather then the full set of values. Developers can change other settings at will such as turning on debug variables such as APT_DUMP_SCORE. There are also file location settings such as APT_ORCHHOME and APT_CONFIG_FIL...

I am not discounting the possibility that you found a bug however somebody may have experimented with different partitioning settings in the remove duplicates and sort stage and caused the problem, or the stages might have been copied in from another job with incorrect settings, and the unlinking an...

DataStage Manager Usage Analysis on saved table definitions can be unreliable as it depends on programmer discipline. If the developers only entered column information by loading the same table definition into all jobs that needed it then it will work. If they took a different route, such as enterin...

Test your code in a routine. Put your compare statement into a new routine and use the Test button to see if it works. Call the routine direct from your transformer if it is a complex statement that you want to use in multiple locations or copy it into your transformer if it is a one off.

It's on the Diagram menu - Show performance statistics.

The Aggregation stage makes a Group field mandatory which means you do have to create a column with a constant value in order to group by one field. The column generator should be the most efficient way to add this column unless you already have a transformer in your job. The underlying group operat...

Good on you Brad, I missed that thread first time around and it is very good to see real examples of buildop code. Anyone else has buildops examples out there please post. :D

Is this problem specific to this job or is it affecting all your parallel jobs? If it is just the one job you can use a process of elimination to find the culprit. Take a copy of the job and replace the target stage with a copy stage and run the job. If you still get the error replace the filter sta...

An upgrade from 6.x to 7.x is very easy and there are a few threads you can uncover in a forum search on the subject. You just export your projects to a backup location and run the upgrade. It helps to have a system integration test environment on another server - or turn your development server int...

I think you should use any type of SQL you are comfortable with from generated to SQL builders to user defined as long as you don't transform your data within the select statement. I always try to use generated first, when it becomes too complex I move to a SQL builder, my last resort is user define...

Some large organisations push software out to desktops overnight to keep them standardised. You can have a happy DataStage at the end of the day and come in the following morning to find OLE automation errors. The standardised changes being pushed out are not tested against DataStage client software...

The transformer stage method with stage variables is the fastest method if you can get your input data sorted quickly. It will also let you output discarded duplicates quite easily. See the remove duplicates FAQ.

A parallel job will automatically add sorting and partitioning to certain stages such as the remove duplicates and join stages. If you know the data is already sorted, let's say you have a sorted sequential file or a database source with an ORDER BY statement, you can turn off this sort insertion by...

A sequential file as a source or target needs to be repartitioned as it is (as the name suggests) a single sequential stream of data. A dataset can be saved across nodes using the partitioning method selected so it is always faster when used as a source or target.

Generally I prefer generated SQL up front with WHERE clause filters to restrict the number of rows coming back. I join tables where there is a simple foreign key link with proper indexing and the SQL performs well. The new SQL Builder let's you join tables together and automatically synchronises the...