"Scheduling" as in via the DataStage Director? They're not in your database, they are stored in Window's "at" system. So I would google around and see how one would dump the "at" schedule on one machine and load it on another.
Anything you can do from the command line can be done 'after job' - either literally there in a job using ExecSH or in a Sequence job's Execute Command stage. I wouldn't bother with a stored procedure.
The key is finding the differences between the two and it sounds like you have. We've seen odd Oracle issues with 10g on HP-UX Itanium boxes, so I'd probably open a support case to see if this is one of the known issues. What box is the development environment? I was told there's no 9i client availa...
The difference in layouts shouldn't be an issue as long as you can define what "sync" means between the two. And "real time" or batch won't change what the "sync" does, only the DataStage implementation.
You'd need to be more specific in your needs. Send an XML file somewhere? Call a notification web service with an XML packet? Something else? One needs "RTI" and the other doesn't.
What data type is this "spalte_a" column? And are you saying that when it works from the 7.5.2 server and doesn't work from the 7.5.3 server they are both connecting to the same instance? Using the same version of the Oracle client?
Yes, definitely chat with your DBA. How long did the job run before it blew? How busy was the database with other processes at the time it was running?
Search the forums for "milliseconds", this has been asked and answered before. Never mind the fact that it is documented for the StringToTimestamp() function.