Re: (A) native Windows port - Mailing list pgsql-hackers

From Lamar Owen
Subject Re: (A) native Windows port
Date
Msg-id 200207091104.15384.lamar.owen@wgcr.org
Whole thread Raw
In response to Re: (A) native Windows port  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-hackers
On Tuesday 09 July 2002 11:41 am, Hannu Krosing wrote:
> The old postmaster should not be built/distributed. As it is for
> _upgrading_ only, you just have to _keep_ it when doing an upgrade, not
> build a new "old" one ;)

Let me reiterate one thing about this.  In the midst of a total OS upgrade, 
during which PostgreSQL is being upgraded as well (the new OS release 
includes a 'better' PostgreSQL), you also get library upgrades.  If the 
upgrade is from an old enough version of the OS, the old postmaster/postgres 
may not even be able to execute AT ALL.

Some may say that this is a problem for the vendor.  Well I know of one vendor 
that has thrown up its hands in disgust over our lack of upgradability that 
they have now quit supporting even the kludgy semi-automatic upgrade process 
I did up three years ago.  They will refuse to support any mechanism that 
requires any portion of an old package to remain around.  The new package 
must be self-contained and must be able to upgrade the old data, or they will 
not accept it.

Their statement now is simply that PostgreSQL upgrading is broken; dump before 
upgrading and complain to the PostgreSQL developers.
-- 
Lamar Owen
WGCR Internet Radio
1 Peter 4:11


pgsql-hackers by date:

Previous
From: Hannu Krosing
Date:
Subject: Re: (A) native Windows port
Next
From: Bruce Momjian
Date:
Subject: Re: [GENERAL] CRC function? (fwd)