DSXchange

Somehow you've lost everything else that should be in LIBPATH. To see what I mean echo $LIBPATH in some other (non-DataStage) process. I suspect that, somehere in dsenv, there is LIBPATH=pathname instead of LIBPATH=$LIBPATH:pathname.

Do you also use LD_LIBRARY_PATH?

Have you tried uninstalling and re-installing the 7.5 client?

Do you have the DataStage patch for XP sp2 installed?

If you have the time it would be interesting to diagnose whether the problem is occurring in Exp() or Pwr() functions or in the final arithmetic. I have a suspicion that you are hitting a precision limit here, that something, somewhere, only works with numbers that can be described in 32 bits. Divid...

Definitely a routine. What are the rules for establishing the dates and premium amount in the individual output records? It would be a Very Good Idea to document these business rules before designing anything, and getting agreement (sign off) that the specification is correct.

UV compatibility is only about date-type data. Every hashed file can be accessed via a UV stage, though you may need to create a pointer to a directory-based hashed file in the VOC (search the forum for SETFILE).

Who's going to maintain this code into the future as the requirements increase - particularly after you've moved on? Can you guarantee to meet key performance indicators (whatever "performance" is) using a hand-coded approach? What testing regime exists? Do you know what all the possible values to b...

There is a uvodbc.config file in each project directory.

You will find, upon inspection, that the DSN localuv - the one you need to use the UniVerse stage to access hashed files in the project's account - is already defined in each uvodbc.config file on the server.

I'd use DataStage, a tool whose sole purpose is to move data from any point A to any point B, possibly re-organizing them on the way through.

You or any future searcher should be able to glean everything you need to know about the question from the incredibly specific subject the poster chose for this thread.
Not.


Another demand for the RMM stage.

If you create the sequential file with First Line is Column Names set to True then the column names will be in the file. FTP is simply a matter of invoking the appropriate command (I prefer to wrap it in a shell script) and invoking that from DataStage, perhaps via an after-job subroutine, perhaps v...

echo $LIBPATH and tell us what you see - in particular whether the directory containing liborchsun4.so is present in that list.

Try isolating the environment variables, and verify that APT_ORCHHOME is set. By isolating I mean using curly braces, for example ${APT_ORCHHOME}/lib.

Try importing it using the Orchestrate orchdbutil facility. This is known to be more accurate than other importers.

Import > Table Definitions > Orchestrate...

Please encase your design in Code tags - what you have posted is incomprehensible. Does your design work? The direction of sort, wherever it is specified, is a sub-property that does not accept a job parameter reference.

There's no need for sequential operation; why do you believe there is?

OutputColumn:int32=InputColumn or OutputColumn:nullable int32=InputColumn Input column is of data type int64. This will generate "possible loss of precision" warning. You can create an additional specification containing just the single token NOWARN to suppress the warnings: this is probably not a ...

Yes, he is. But it's not the whole story. Get a second licence and things will improve only all things being equal (you're not loading more compiles onto it). However, single compiles performed in isolation will not improve at all; you'll simply not be using one of the compiler licences.