Re: Why dump/restore to upgrade?

From: "Dann Corbit" <DCorbit(at)connx(dot)com>
To: <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Why dump/restore to upgrade?
Date: 2002-02-08 03:28:49
Message-ID: D90A5A6C612A39408103E6ECDD77B8290FD4C3@voyager.corporate.connx.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

-----Original Message-----
From: mlw [mailto:markw(at)mohawksoft(dot)com]
Sent: Thursday, February 07, 2002 6:56 PM
To: PostgreSQL-development
Subject: [HACKERS] Why dump/restore to upgrade?

We all understand the reasons why one MUST dump and restore to upgrade
to 7.2,
but I'd like to make a general call to arms that this (or 7.3) should be
the
last release to require this.

It doesn't look good. One should be able to upgrade in place, and even
revert
to an older version, if nessisary.

What do you all think?
>>-------------------------------------------------------------------
I think sometimes it might be good to change the internal structure.

Rdb has made me do dump and load.
SQL*Server has made me do dump and load.

However, if there is a change to internal structure that requires such a
transition, it is traditional to change the MAJOR version number since
the change is traumatic.

Something that can make the transition a lot less painful is to
automatically output the SQL schema in a manner such that dependencies
are not violated. (e.g. don't try to create a view before the table is
defined, stuff like that).

To try to say "The internal structure will never need to change again"
will lead to either certain failure or stagnation because you can't
update the internals.
<<-------------------------------------------------------------------

Browse pgsql-hackers by date

  From Date Subject
Next Message Peter Bierman 2002-02-08 03:32:59 Re: Why dump/restore to upgrade?
Previous Message mlw 2002-02-08 02:56:22 Why dump/restore to upgrade?