DSXchange

"First line is column names" property mis-set is often the culprit when the error occurs in the first record. However the first record is usually numbered 0. As noted, you need to check for any kind of metedata mismatch. Use a reject link to capture the rejected rows and write these to a f...

Parallel routines are actually external to DataStage, so there are no special "DataStage" requirements. You refer to other functions from within your code exactly as you would if DataStage were not in the mix.

Use explicit Sort stage to sort the right input and allocate extra memory.

I would have a stored procedure in oracle to execute these statements and would be invoking it using a stored procedure stage in Datastage. SQL Statements can be passed in a single field to SP stage. An Oracle stored procedure to execute statements intended to be used against MySQL? A novel approac...

Welcome aboard.

There is a Search option in the menu at the top of this page.

You will find that this question has been asked, and resolved, in the past. The answer involves using a sequence and, possibly, the user status area of the first job as the "user defined variable".

When you generate test data, you need to construct expected results which you compare with actual results. Can you please describe the expected results for your two examples? I suspect Arnd's partial solution will get you most of the way there, but that you require not "Good"/"Bad&quo...

Use Super Operator role.

Solaris also has such a limit. Approximately 32K if memory serves.

Information Analyzer jobs don't actually get compiled (they generate osh directly) so this message is a bit of a mystery. It suggests that the osh has not been (is not being) properly generated. The problem is to find out why, which is why our first thought was inability to write the osh into the RT...

Try to discover which character(s) can not be mapped (is it into or out of?) the IBM-1051 character set.

Try using identical metadata (same order) on all links.

Just out of curiosity, what is the total number of entries in this particular job's log (you can find out in Director log view or by counting the entries in the log table - SELECT COUNT(*) FROM RT_LOGnnnn; where nnnn is the job number from DS_JOBS.

Well, for starters, I know of some Linux servers that haven't needed to be re-booted for nearly two years. You don't get that with Windows. Linux scales farther than Windows - there is still an upper limit, for example, for how much memory a Windows O/S can address. There's one for most Linux varian...

You can determine the limit by reviewing the Transformer stage properties in DS_STAGETYPES. Last time I looked the limit on links to a Transformer stage was 127. Hmm, might need to look again.

There is a limit, but it's not 100.

Your problem was the duplicate identifier, as the error message reported.