Phil Campaigne wrote:
> Hi Ron,
> I had a couple of questions on your instructions:
> 1. what is this for?
> >#make install-all-headers
According to the docs you need it if you are going to create your own
functions, however the documentation is a bit *vague*. "If you plan to
do any server-side program development (such as custom functions or data
types written in C), then you may want to install the entire PostgreSQL
include tree into your target include directory." I am working under the
assumption that "custom functions" include any functions you create in
plpgsql, sql, c, etc and not just C functions. I would recommend using
this if you will be working with the database a lot, because you will
probably eventually want to create some of your own functions.
>
>
> 2. is it advisable for me to use UNICODE for defalt encoding?
> > initdb -E UNICODE -D /usr/local/psql/data
Some of our databases were originally encoded using SQL-ASCII and we
still occasionally have problems when someone enters a character with an
accent (eg in cafe). The jdbc driver won't convert it and the app
'breaks' wrt that data set. So I would recommend that you UNICODE, I
don't believe that it really adds much in the way of overhead, etc.
>
>
> 3. My version of postgresql's linux script file instructs me to create
> symbolic links to:
> /etc/rc.d/rc2.d/K02postgresql
> /etc/rc.d/rc3.d/S98postgresql
> However the init script file for starting up the PostgreSQL server
> that I have is K15postgresql.
>
> Is it ok that I substitute a link to K15postgresql for rc2.d and rc3.d?
I suppose that's okay. My very limited understanding of these
directories is that they are used during startup/shutdown to ensure that
services are started and stopped correctly and in the proper order.
>
> thanks,
> Phil
Please respond to the list for more questions and cc me (or vice versa)
if you have more questions. This way if anyone else is in a similar
predicament they can read this, and others can contribute and/or correct me.
Ron
>
>
