DSXchange

That's exactly what I'm suggesting. Whether you double the number of scratch disk spaces or double the number of process nodes, ultimately you need to be able to provision sufficient scratchdisk space. Breaking the job into pieces plays havoc with your sorting - you would need to introduce measures ...

No idea. Where did you find it?

Do you have any columns defined on your reject link?

Is it possible to drag all the input columns for the stage onto the reject link?

Nothing special. LD_LIBRARY_PATH should automatically be adjusted for the process that executes the transform operator.

Using ExecTCL it is not the developers who access dssh, it's the user ID under which the job runs. This is likely to be dsapi_server if the job is run from one of the clients.

It is, of course, possible to set up a server so that only server jobs and sequences can be created. That's achieved by not licensing the parallel engine. But, as noted, it's not really possible at a project level. Why do you want to accomplish this? You could, of course, remove all the parallel sta...

Unless you get more specific than "doesn't work as expected" we can't help you. What did you expect, what did you actually get? Why do you think this is?

This is Unit Testing 101 stuff!

Not possible. For two files you need two Sequential File stages. Use a Copy stage, with a Column Generator stage on one stream to generate the constant 0 for the amount field.

Logic for the rest can be managed in a parent sequence.

So inspect the stage to let us know whether it's generated or user-defined SQL. We suspect it's the latter, and that whoever constructed the SQL did not know how to reference the ORCHESTRATE object in the VALUES clause.

Disable operator combination and run again, so that we (and you) can determine which stage, or operator, is throwing the error.

You need to make sure that LD_LIBRARY_PATH includes the parent directory of the library referred to in the error message.

Perhaps Finance_WS.Finance_Call.Ln_Ac_Bal was not found because of something in the job design (you will need to check this) or because no value for it was returned from the web service you invoked.

You can not use the same instance tag. You can give the same name as DataStage administrator. (You will also need an Information Server administrator and WebSphere Application Server administrator and an owner for the shared metadata repository database.)

Do all the VARCHAR fields have Precision set? Do all the VARCHAR fields have Precision set large enough for the data?

No idea. The logs from the other projects won't affect the job in the new project, but if the job is throwing a bazillion events into its log in the new project, then that will have an effect. Only you can see how large the log is, and it may not be the log that is the cause - I had to guess about t...