I'm almost at the end of a project doing something along these lines at the
moment. I found that Job Descriptions and code that had hard coded
libraries in it required a lot of work. You will need to take a look at the
system plumbing that comprises your application, like subsystems, jobqueues
and the like which tend to get hardcoded will give you some thought. You
will need to think long and hard about your library structure and library
I'd recommend checking that Vision can do what you need it to do (note that
I'm not suggesting it can't) and understand how it does it.
Remember that all your common applications (I'm talking about utilities and
fax ware and stuff along those lines must be kinda database Neutral. You
really want to have your databases and code isolated as much as possible in
If you have application code running in *SYSBAS then you are in for some
testing times. You would really want to have all your data in the IASPs.
If you have remote machines and use DDM then this will require some changes
especially if they are currently SNA based. Easiest way I could find to
accomplish this was to convert them to DRDA access which leads into some
bizarre alleyways with authentication entries to provide access to the
If you use any client connections that have sockets listeners on them (e.g.
Delphi or similar) check that they will work with the IASP. You may find
that you have to hack the job description to get them to see the IASP by
changing the INLASPGRP parameter of the job description they run under.
IBM's ODBC works fine as it sets all that stuff up according to the user
profile but other listeners or ODBC drivers may not be as ready for this
As a last thought, setting up an IASP for development seems to me to be
perhaps more work than it is worth. One rogue development job could screw
your production environment up big time as well.
Seems to me it would be easier to just rename the libraries and duplicate
the data into them for compilation testing and the like. Why do they need
to be the same names ?
There are some issues with disk arms and temp storage that IBM allude to in
the red book which we haven't been able to test yet but we put our IASP-ed
database under load in the next few days. Should be interesting :)
We want to create a mirrored machine with Vision and we want to develop on
this machine. However, for the development machine we need some production
libraries to compile and test against that are also on the the production
We want to create the libraries with the same name but in different IASP's.
I have read some of the older threads, but I wonder if there are any new
insights. What needs to be changed the most if we take the IASP approach?
As an Amazon Associate we earn from qualifying purchases.
This mailing list archive is Copyright 1997-2022 by midrange.com and David Gibbs as a compilation work. Use of the archive is restricted to research of a business or technical nature. Any other uses are prohibited. Full details are available on our policy page. If you have questions about this, please contact
Operating expenses for this site are earned using the Amazon Associate program and Google Adsense.