DSXchange

Yes, you can use the stage, just the performance increase is not going to be as noticeable. All of the DB/2 access is standardized to use the Enterprise stage in your project and whatever tables you are accessing might be partitioned in the future so it is best to stick with the enterprise stage fro...

What about using VARCHAR() datatypes in the job as the result of a TRIM() and then letting the implicit conversion from varchar to char plus the output attributes take care of the padding?

One possible reason: partitioning

ganive, if you go to your output stage, choose your column and right-click to enter the "edit row" you will have some string options on the fill character to use. I'm not at a PX computer now, so I can't tell you the exact attribute to set, but that is the correct window to make this output in the f...

Koolnitz,

that is certainly good news. I think you should be using the faster DB/2 UDB stage since it is configured correctly to function in your installation; that will increase performance and also remove some of the difficulties you are having with the API you have chosen.

If your source is a sequential file, define it as having 1 column (with a very long string) and no delimeters, read it and perform the conversion of all "." to CHAR(128), write it back, then read it as before. Or use any one of many different UNIX level options to effect the same change to the file....

All jobs are imported into the server. The clients have no DataStage programs stored on them, they access the information on the server. You can connect any number of clients to the server, limited by your license. The clients must be PCs running Windows.

If "exp_EMPTY_STRING_FDM" is a transform stage I would look into that and see what functions you use in there. There is a problem in the GCI (general calling interface) and it might be that it is not finding a routine or program. A bit more information about the names referred in the error messages ...

Veni, that is the difference - the null in the defaults is not an oracle null value. You will need to put this conversion in a modify stage or a transform. I think it will be best to keep the period to 0x00 conversion and then use the handle_null in a modify stage. An alternative that is not quite a...

There is no direct to do this and ensure a correct list. Hashed files created in the project directory (and not in a remote path) are detectable, but hard to distinguish from the normal files used by DataStage. The builtin metadata and report generator can give some clue as to which files are used, ...

No, nothing to worry about, the project name is not part of anything in the .dsx file and one can easily export from one and import into another.

I concur with Craig. Using the builtin sort in DataStage for your volumes will .... hmmm, I wanted to use some witty simile but I can't think of an appropriate one - it is going to be very, very slow until it aborts. Using CoSort or SyncSort will certainly speed things up. Have you tried to see your...

Koti9, The fact that you cannot delete the job means that there are still locks open on it; even if you haven't located the appropriate lock. If you stop the DS engine and after the restart still cannot delete the job then you might have some sort of corruption in the project - in which case the DS....

Charuthareja, please read the documentation first, then if you test it and it remains unclear it is worth posting a question. The DSXchange forum is not meant to replace the documentation. I suggest you open up the DataStage Director Guide to the chapter Running DataStage jobs -> Multiple Job Invoca...

Each instance of a multi-instance job must have it's own distinct instance name, you are attempting to run with the same instance name (most likely an empty instance) and therefore you are getting this error.