RelationalDBDesignRelationalDBDesign


DB Creation   «Prev 

Config files

On many UNIX installations, if you look in the pfile directory for an Oracle instance you will find both an initXXXX.ora file and a configXXXX.ora file.
Why is this done?
Answer: This is a way that Oracle is using to separate initialization parameters into two groups. One group consists of parameters containing the name of a specific database. These are placed into the configXXXX.ora file. Here is an example of a config.ora file from one of the databases I manage:

$ cat configSEED.ora
control_files = (/orcl1/oradata/SEED/control01.ctl,
                /orcl2/oradata/SEED/control02.ctl,
                /orcl03/oradata/SEED/control03.ctl)
background_dump_dest = /ORACLE_BASE/admin/SEED/bdump
user_dump_dest = /ORACLE_BASE/admin/SEED/udump
db_name                 = SEED


When a config.ora file is used like this, the initXXXX.ora file will contain a line such as the following:
ifile = /ORACLE_BASE/admin/SEED/pfile/configSEED.ora

When you start an instance like this, Oracle will start by reading the init file. The ifile directive tells it to include the contents of the specified file, which in this example is the config file. You do not need to do anything differently for this situation; this is just an explanation of why you may notice two different files being used on some systems.

Windows

When you install the software on Windows, Oracle does not include config.ora files. If you are running Windows, all of the initialization parameters will be in the init.ora file. Regardless of what operating system you are using, you should do the same, and place all parameters into initCOIN.ora.