DSXchange

Modify stage does not generate new columns - for this you need stage types such as Column Generator or Transformer stage.

Default values are not set in a Lookup Stage. There are lots of other stages where it can be done, for example a Modify stage using the handle_null() function.

To run an instance from Director, fill in the Invocation ID field.

To run from an API or command line, use job.instanceID syntax.

For example (probably with different parameter settings).

The problem is "Available memory exceeded". It's reported by a Transformer stage (because errors are reported by active stages) but it's almost certainly generated by the Folder stage, which has the inexplicable behaviour of trying to cache the entire contents of the folder in memory. Don't worry ab...

That's the only official place it resides. Maybe someone would be good enough to post it in a downloadable area (such as ADN). I don't have access to a DataStage CD at the moment.

Welcome aboard. :D

When you're building the expression, use the graphical expression editor. There you will find the appropriate data type conversion function.

You may find that it is more efficient to use a Modify stage if this is all the transformation that you need to do.

What about libvmdsapi.so ?

You copied vmdsapi.so - that is not the same name.

The solution, in purely economic terms (supply and demand) is clear!

Iconv() is a BASIC function - you could use a BASIC Transformer stage but there is a performance penalty. Otherwise there are data type conversion functions available, which you can find in the expression editor (Transformer stage) or in on-line help for Modify stage or Transformer stage. You can al...

Getting the key should have been done upstream of the Change Capture stage via a lookup against a "business key to surrogate key" data set.

Varying the delete flag or end date then becomes a straightforward update.

The obvious one is getting the section leader processes started and talking to the conductor processes. There are many potential variables here especially on a cluster of machines. Connecting to database servers from multiple player processes also can take a lot of time. It's not just connection, it...

Hundreds of jobs simultaneously? Possibly your server is overloaded.

Search the forum for the meanings of these signals.

Is this a remote file or an explicit named pipe? The message informs you that the job times out waiting for a reader to connect. Is your system heavily loaded, or have you run out of processes in the process pool?

Well done, you solved it while I slept.

You can only avoid the warning by satisfying the condition about which it complains. You can easily create a message handler, however; just right click on the warning and define a message handler to demote the message to Informational.