Re: managing two instances on same server

I have a concern right at the point of creating the new instance from a script, because our standard creation script alternates between "internal" and "system" connections.
It hails from the v7 and v8.0 days.

First as "internal" you startup nomount with the new profile. Then disconnect-reconnect as internal. Then create database (blablabla) and catalog.sql.
Then again connected as "internal" we create the system rbs, online it, then the rbs tblspace and the temp one, the rbs segments, and quotas are given to sys and system on temp.
THEN you finally get to connect as system/manager to run catdbsyn. Going back to "internal" you run three sql scripts (standard, dbmsstdx, and catproc. You then connect as system/manager and run pupbld.sql

Now all of this "internal" stuff has me thinking 'How can we know which instance we are working with?' Should I put "set instance <newname>" with all these, and also change to
system/manager@<newname>? Is the instance name meaningful right after the startup with the new profile, even though there is no database created or catalogued yet?

Chris O'Connor

In article <86q6nm$mvh$1_at_nnrp1.deja.com>,   chrisoc_at_ans.net wrote:
>
>
> I want the maximum independence in taking down one
> instance, modifying its storage etc. without affecting
> the other one.
>
> If I build the second instance with entirely distinct
> tablespaces and datafiles (including system, temp, rbs,
> redologs) and run the catproc etc. for it, I am still
> not certain how I will be able to connect as system or
> sys to just that instance and manage it without
> affecting the first one.
>
> Point me to a doc if the answer is complicated.
>
> Thank you,
> Chris O'Connor
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
>

Sent via Deja.com http://www.deja.com/
Before you buy. Received on Thu Jan 27 2000 - 17:49:38 CST