DSXchange

Asit, The DataStage server engine only has one real datatype (the other is an internal one used to store file pointers), and it is not a contiguous string whose size can be measured in bytes. Everything is a string in Server.

I would try using "binary" instead of "ascii" transfer, that avoids any mapping of characters; and if you are going from two identical OS's that shouldn't cause you any proglems. I'm not sure where you draw the distinction between reading a remote file and transferring it (for purposes of reading). ...

Welcome to DSXchange, Ruud.

Have you actually set breakpoints and started the debug run? The debugger is a very powerful tool and I don't use it as much as I should, but it does take a bit of practice to get it working the way you expect.

Do you have a before-job subroutine call? Also, do any other jobs show this problem or just this one?

Have you checked your mail server logs to see if the request even made it that far?

What is the Database DDL for "KEYWORD" and what have you declared it as in your DS source stage?

Peter - concurrent reads are OK but not concurrent writes.

The odds are that your job is bottlenecked on CPU; especially if your source and target files are sequential ones as you've indicated. If you have a multi-cpu system or one with CPU capacity left, then I would highly recommend using 2 transform stages and splitting the CPU consuming processings acro...

Kevin - welcome to DSXchange! The server forum is more applicable to this question, but it isn't completely out of place in the General one. If you specify "initial value" for a stage variable, that will only get executed one time. If you specify a derivation for a stage variable, it will be re-comp...

The compiler is called using the "BASIC" command. It is better to create the folder as type 19 so it doesn't do funky things with your filename. BASIC commands do not need to be terminated with a semicolon.

...This isn't going to be very fast. I hate quoting myself, but I re-read this thread and realized that I've become infected by the renowned British understatement sickness. I should correct that - if you call "dsjob" from a trigger it will run INCREDIBLY slowly, even if you don't wait for the job ...

I think the poster might be referring to vectors, but they are adequately covered in the CFF stage documentation.

TRIM() can be used to removed trailing spaces. But I think you have a @FM in there as well - see my previous post.

This is not a datastage issue, but a trigger question for whatever database you are using. If your DB allows external calls, you can start a job using the OS command "dsjob" and appropriate options. This isn't going to be very fast.

Can you access the ODBC source on your DS server using MSQuery? I usually start with that when I have problems - if it is visible and works there, it should work in DS as well.