Page 1 of 1

Routines issue while upgrading DS

Posted: Fri Apr 01, 2005 3:24 pm
by kollurianu
Hi all,

I heard there would be routine issues while upgrading DS (3.6 to 7.5)

but from the document i have seen that , under upgrade guide/upgrading issues/Transform Routine Arguments page no. (164/192), how this is been handled,

but iam not able to understand how to use this utility that is mentioned.

THE DOC SAYS THAT

"For example , input column variables are passed into Routines as arguments when called from a column derivation in the Transformer.

If an input column variable is passed into a Routine as an argument and then directly modified within the Routine, there is like to be corruption of that in put column data on returning from the routine.

In order to check that Routines contained in the project donot break this
restriction, the Datastage server installation has been modified to run Routine checker utility which will report on any Routine which appears to
be modifying it's input arguments. the utility is run against each project on installation and the results are output to a file named CheckerRoutines.log in the project directory."

So please if some one could help out with issue , that would be great!
Is it something done during installation if so what care should be taken to
to run that utility on the whole project.

Can you also please tell me datastage basic directories like project directory and dsenv directory .

Thank you all very much for ur time,

Posted: Fri Apr 01, 2005 3:39 pm
by ray.wurlod
It's not really an issue. All it means is that, in the newer versions, you will get a warning message generated by the Routine compiler if you assign a value to any variable whose name has been used as the name of an input argument, to warn you (implicitly) that DataStage uses pass-by-reference.

The DataStage "dsenv directory" (not a terminology you'll find in any documentation) is the administration directory (or "account"). Its location may be stored in a hidden file called .dshome in the root directory and/or may be specified by the environment variable DSHOME.

Each DataStage project exists in its own directory. The pathname of this directory is visible in the status bar of the Administrator client when the particular project name is selected. Location of project directories is determined when the project is created; the only restriction is that you are not permitted to create a project in the root directory.

Posted: Fri Apr 01, 2005 3:48 pm
by kollurianu
How do i check the log of the checker utility , which is mentioned that it would be under the project directory.

Thank you Ray for ur answers.

Posted: Fri Apr 01, 2005 4:23 pm
by ray.wurlod
The project directory is on the server machine. Log in on the server machine and cd to the project directory. The log is a text file; you can view it with any text viewer, such as more, cat, vi and so on.

Posted: Fri Apr 01, 2005 7:34 pm
by kollurianu
thank you very much Roy. :lol: