Re: Cloning a database

marko mikulicic <marko_at_seul.org> wrote in news:3B7F589A.5070701_at_seul.org:

> Hello,
>
> I have Oracle 9i on linux glib2.2.2 kernel 2.4.3
>
>
> I created a little (400MB) database with dbca with no templates
> and after hours of swapping (I have 512 RAM) and many failures
> (the installer did not copy the STANDARD package).
> Creation from templates was too big.
> Somehow I managed to get it installed and now I don't want
> pass again all this. I want to test replication on a single host, just
> to learn how to do it, but I had not disk space to waste. Template
> creation doesn't allow to change the tablespace dimensions, so I
> thought that cloning that one with another name would do the job.
>
> But wich files do I need to copy ? (I know how to setup net stuff)
> Is the databasename stored inside the database itself ?
>
> I have a slow machine with much ram, but oracle can be a hog :-)
>
> Thanks in advance
>
>

Below is a SQL script that generate a Korne Shell script which I use to clone any DB. FWIW "pandora" is my test box.

set pages 0 feedback off lines 255
spool clone-db.ksh
select 'rcp ' || file_name || ' pandora:' || file_name || ' &' from dba_data_files;
select 'rcp ' || name || ' pandora:' || name || ' &' from v $controlfile;
select 'rcp ' || member || ' pandora:' || member || ' &' from v $logfile;
spool off Received on Sun Aug 19 2001 - 18:59:43 CDT