DSXchange

Do you mean TO your DataStage server from something external, or from your DataStage server to some data source?

The first would require an ODBC driver for DataStage Engine, and no such animal exists. Nor is ever likely to.

Well, some progress has been made. Now you have to work out why the IsValid() function is reporting your "numeric" data not to be valid as 32-bit integers. Look for non-printing characters in the date. Try the IsValid() function with "uint32" as its second argument.

.. or right-click and choose Save As to download the DSX file.

That is indeed a curious error message, particularly since it is being emitted by the IBM Oracle JDBC driver. Perhaps this uses port #50000 internally (but only IBM could verify this for sure). I assume that your firewall does not pass this port number.

This forum is not for explicit questions about parallel jobs. This forum is for suggestions for topics for the FAQ forum. Please post your question in the correct forum. Or Search first; you may even find your answer without even needing to post a question!

*Note: Topic Relocated - Content Editor*

There are other things happening than shared memory segments, and this only gets more complicated when you move to version 8. The main hazard with removing the shared memory segments (other than needing the root login ID) is making sure you get them all - but then what about the semaphores and other...

What if you use something like the following for an int8 result?

Click on the "Yes" - it's a link to a library of server job routines.

Regard all Transformer stage functions as intolerant of null. Please be specific about what you tried and what you mean by "doesn't work" (including any error message). The approach that was suggested will work, provided that the Else clause is complete. You don't need SetNull() in the The...

You can also find information about the effects of potassium cyanide on the net, but you wouldn't eat it. Be very, very careful with any form of "kill".

santhob, this is stock-standard SQL. (Yes, it's the concatenation operator.) It appears that undertaking some SQL training would be beneficial for you.

Have you taken any formal DataStage training? This is exactly the kind of thing you would learn on the DataStage Essentials class (or equivalent).

Trim() with only one argument trims leading, trailing and redundant white space characters, exactly as described in the documentation.

Pno and Mat does not give you a unique return result from the lookup. Did you enable multi-row return from the lookup? Otherwise, to get a unique result, include SDate as a lookup key, since this is also available in your source.

Square brackets indicate optional parts of the syntax. Trim() is a function, so the argument list is surrounded by parentheses, not square brackets. The third argument to Trim() must be a literal string (that is, quoted) or the name of a variable. In your case Trim(Expression, " ", "B...