Re: Conservation of OIDs - Mailing list pgsql-general

From
Subject Re: Conservation of OIDs
Date
Msg-id 64784.216.238.112.88.1068835070.squirrel@$HOSTNAME
Whole thread Raw
In response to Re: Conservation of OIDs  (Manfred Koizar <mkoi-pg@aon.at>)
Responses Re: Conservation of OIDs  (Peter Eisentraut <peter_e@gmx.net>)
Re: Conservation of OIDs  (Alvaro Herrera <alvherre@dcc.uchile.cl>)
List pgsql-general
>
> If you're really concerned, you can initdb separate clusters for QAT
> and DEV and run three postmasters using three different ports.
>

Follow-up question: Are different ports really necessary? I currently
have the three different databases defined all in the same cluster, and
differentiated by name, e.g., mydb, mydbqat, and mydbdev. If I have the
postmaster start these three instances in separate clusters,
respectively, using three different directories, such as

$ postmaster -D /usr/local/pgsql/data
$ postmaster -D /usr/local/pgsql/qat
$ postmaster -D /usr/local/pgsql/dev

and make no change regarding which port the daemon listens on, but DO
have a single, unique database name, one per cluster (mydb, mydbqat,
mydbdev) in the respective clusters, will connections requests to any one
of these databases on that same port find their way to the right place
(and back)?

Or maybe would be better to not used different database names at all to
differential PROD, QAT, and DEV, and instead differentiate solely which
cluster is to be used by means of your suggestion of using different
ports? I dunno. What do you think?

~Berend Tober




pgsql-general by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Storage consumption
Next
From: Jeff
Date:
Subject: Re: Updated Documentation