DSXchange

OK, please show us the properties that you have set within the Surrogate Key Generator stage.

Welcome aboard. Get yourself on a DataStage training class, sooner rather than later. Provided the job is closed, you can right click on the job in the Repository toolbar of Designer and create a copy right there. There are two main export tools, one in Designer (the Export menu) and one called Info...

Thanks for that clarification. I'm not sure I can think of a use for them off the top of my head (ok, maybe one, central storage of connection parameter values - but that's what values files give us in a way).

Fix Pack 2 for Foundation Tools has been announced/released by IBM. It contains four fixes for FastTrack, nine for Business Glossary and a massive 34 fixes (count 'em) for Information Analyzer. Get them from Fix Central. The release notes can be had from InfoSphere Information Center and a list of f...

Check out "Shared Containers". You don't need to create your own stage (though that's possible too, though you would need to code it in C++), you just need to create a shared container. And you can do that by selecting the existing five stages and creating the container from them.

Sequences don't have after-job subroutines. But you could make the final activity in the sequence a Routine activity that does the same thing.

This precise question has been asked previously, and by you, and an answer was provided.

will I have any issues exporting parallel jobs from 7.X to 8.X using DataStage Manager? No, but yes. (Trivial really.) Version 8.x no longer has a Manager client - you import using the Designer client. Other than that it will all be good*. Do recompile all jobs and routines post-import. * I'm assum...

mavrick21 wrote:there is a requirement to Hash partition

Why?

You really aren't telling us much.

What do you understand to be the point of a key-based partitioning algorithm such as Hash or Modulus?

Maybe the user ID is registered in cron.allow but not in at.allow. Ask your UNIX administrator to check.

Please show the SQL you are using to insert the row, and indicate how many columns there are in your job design being sent to the database.

Probably easiest is an after-job subroutine to move the report file to a different location, possibly renaming the file to include a date or timestamp component in its name.

How about some business-like specifications for the derived fields? For example it seems to me that: Top_Flag = "has no parents" Bottom_Flag = "has no children" Child_Level = level in hierarchy and must be one larger than that of its direct parent except for self-referential rela...

The easiest solution, particularly if your skills are limited, is to use two jobs, one to extract (to a Data Set), the other to load (from the Data Set). That way you can have one set of active environment variables per job.

While that is true, it may introduce inefficiencies. For example, is the sort really necessary? (Data Sets preserve sorted order among other things.)