DSXchange

eze_ke- I don't know which parameters work for you; you are the one who has to try to get information. I am just saying that your problem is most likely coming because of different methods of calling your job, in one case the defaults (as defined in the designer) are being used and in the other run ...

eze_ke, most likely the reason that the job runs in one configuration and not in the other is because the parameters are different. That is why I've now asked 3 times for you to compare the parameters of a run that worked with one that didn't. How do you know that parameters are the same without hav...

What is a "force start" and how is it different from the first time you ran it? Please do check the parameters for both runs in the director log to see if they are identical.

This is a bit confusing. If your master input has 24 rows then going through a lookup should also bring back 24 rows - with the looked-up new columns having values or nulls depending on your keys.

Look into your director log to see what the parameter values were for the FTP stage and compare with the values when you started manually.

When you run your job does it find and return rows from the lookup? The only immediate reason I can think of for you getting null values returned is when the lookup is not successful.

What is the full key for file Z when you wrote it? If it is a composite key they you need to use the same columns to look it up. When you run the job, does it return any rows for lookups on Z?

Jags, that is not what I said, in fact I said You can create and access your own environment variables . As Ray has shown, you can use the DataStage environment variables. Another option is to write a small routine which SHells out to the operating system and gets the value of a runtime environment ...

Umberto, if your designer canvas has those 3 stages and just 2 links then it will do both loads to your table in parallel and using different processes. You can design and configure your job to do things differently, for instance by using a link collector stage for links 1 and 2 to make just one dat...

You cannot create your own system variables in DataStage.

You can create and access your own environment variables. You can create your own routines and functions which can be accessed as if they were system variables.

DataStage doesn't have one, but many file systems default to a 2Gb limit for sequential files. If your OS or installation doesn't have a limit then DataStage doesn't have one, either.

You first need to change to your project from the master account by using a "LOGTO {projectname}" command within TCL.

Meena - can you get it to work with an explicit port number to ANY machine? Right now your problem could be because you are specifying a port or because you are specifying a non-default port. Which is it?

I don't know about the "[29]" - what happens when you try it to a machine using the default port and you specify that number in [<port>] your call?

Meena - I don't see a call to "ftp" in your script. Also, what error are you getting?