DSXchange

Please post the entire error message; that is, the detail of it. The text you have provided is insufficient for any reasonable diagnosis to be made.
Double click the log entry to open detail view, and copy/paste from there.

Welcome aboard! :D

There is a patch required for DataStage if you have SP2 installed. You can obtain it from your vendor. SP2 changed some things upon which DataStage depended.

Are you running Windows XP ? Server 2000 ? Server 2003?

If it's SIGINT it's almost certainly intermittent. DataStage has trapped a SIGINT signal; DataStage's signal handler aborts when this occurs. It would be useful if you could determine what other process on the system may have generated such a signal.

You might also search for "link variable" and "LASTERR".

At least two possibilities then. (1) You are not looking at the same Data Set. Check that the control file pathname is the same, and that APT_CONFIG_FILE has the same value. (2) You have the column filter on in the data browser. I don't think it's this because it's not possible to open the data brow...

Erk, I didn't mean to start a flame war! Hopefully the explanation of the script will help anyone new to the stream editor (sed) to understand how it's controlled.

Begin with a plan.

Exactly what is in the single source file? Exactly what do you want in the smaller files? How would you achieve this (in language, not in UNIX/DataStage terms)?

Then you can either convert this to appropriate commands and/or job designs.

Can you please post this question more clearly? It's perfectly possible to have ten rows and 25 columns.

Did you restart DataStage, so that it re-reads the .operator.adm and .developer.adm files?

Rather than the separate macros, why not use for stage variable initialization
(There may even be a DSJobStartTimeStamp macro - I don't have access to DataStage right now.)

Your shell script - or your dsenv shell script for that matter - could read them from DSParams.

Problem with that approach is that there's a separate DSParams file for each project, while your .profile and your dsenv are system-wide, irrespective of the project to which you're attaching.

Did you ask IBM why they recommend against installing the second version on the production box? If so please post their answer here. Unless it's to do with limited available resources, I can't think of a good reason not to keep all production pieces together.

What happens if you use a mapped network drive or a UNC pathname?

You obviously need a certain kind of mind to do that kind of sed stuff! David, it might be nice to explain what the sed script 's/\(|*\)\[[0-9][0-9]*,[0-9][0-9]*\]/\1/g' is doing, for those who haven't used the utility before. Particularly the need for the escape characters ("\"), the repeat...

You will probably not be able to pass "the test". This is because the test that is first released will be for DataStage EE, and have many questions about parallel jobs, configuration, environment and so on. That you've posted on the server forum suggests that you have only had server job experience.