DSXchange

There are a few things you can - should - do to clean up the DataStage environment occasionally. The main two are: Purge job logs of old entries (you can set this to automatic). The larger the log file (it's actually a table in the Repository), the slower it can be to update. Purge old entries from ...

Since DataStage can't handle BLOBs, you're seeking an approach that DataStage won't be able to implement. Transmit the BASE64-encoded text string unchanged, and decode it after DataStage has pushed it through. Or don't process it with DataStage at all; just use operating system facilities (copy, rem...

Oconv (and Iconv) can handle a multi-valued list of conversion specifications, which are applied one after another. So, even more brief: Convert(Convert("#$%","",Oconv(FieldIn,"MC/A":@VM:"MC/N"),"",FieldIn) This is a good way to ...

Is there a heavy-duty after-job subroutine being executed? (Or one in which there's an inadvertent infinite loop?) Use the DS.TOOLS menu "List All Processes" (or Cleanup Resources (Show All processes) in Director) to identify what the particular uvsh.exe pid is doing. That is, which command it is ex...

I agree. Particularly in a production environment you should ensure that there are no warnings. Then, when there are warnings, they are truly indicative that a problem has occurred that needs investigation. By this means I once discovered that "they" had changed the format of the source file without...

I doubt very much that you can write code that can call the GCI routines; their calling syntax is not published.

Which one(s) did you want to use?

If they're in the catdir directory, they're already globally cataloged, so you don't have to do anything. :D Anything with a "$" prefix in the CALL is what is known as a GCI subroutine. This means that it is written in C rather than BASIC. There is a quite different mechanism for getting these into ...

Link ordering can be altered in the Transformer stage properties page. You can get directly to the correct tab by clicking on the rightmost tool on the Transformer stage's toolbar.

I second that! If you really must keep a table updated, take a copy of it (only the actual columns that you need, which are probably the natural key and surrogate key columns) into a hashed file, and maintain this while your job is running (that is, if the natural key is not found in the hashed file...

The first thing you're going to need to find is a UNIX-based ODBC driver that can connect to MS Access. Do you have one of these? If so, the documentation that accompanies it will instruct you in the entries that are required in the odbc.ini file. You must also edit the uvodbc.config file in your pr...

I was referring specifically to stage types in mainframe jobs.

Welcome aboard! :D Thanks for posting the fix, hopefully this will help someone else from falling into the same trap.

Welcome aboard! :D If you reset the job after it aborts, is there an event logged called "from previous run"? If so, it may contain additional diagnostic information. Access violation usually means an attempt to access a piece of memory that the process does not own. It can be generated, for example...

1. You must have a COBOL compiler on the mainframe. DS390 generates COBOL source code plus JCL to get it compiled and executed. Nothing is executable until is it compiled; COBOL is not an interpreted language. 2. Use the appropriate stage types in your job design. Where are the data now? You get at ...

Please be more specific. Do you mean that the job design gets locked (this job is in use by another process)? Or do you mean locked in that it appears to be running but not processing any rows? Or something else again?