> -----Original Message-----
> From: Vince Vielhaber [mailto:vev@michvhf.com]
> Sent: 29 January 2003 16:45
> To: Dave Page
> Cc: Katie Ward; Tom Lane; Curtis Faith; pgsql-hackers@postgresql.org
> Subject: Re: [mail] Re: [HACKERS] Windows Build System
>
>
> On Wed, 29 Jan 2003, Dave Page wrote:
>
> > > hammering the betas is a far cry from an "industrial-strength
> > > solution".
> >
> > Have you a better suggestion? Seems a bit catch 22 if testing won't
> > prove it's good and we can't use it until we know it's
> good... Still,
> > industrial strength testing or not, it's more reliable than the SQL
> > 2000 and DB2 installations I have here.
>
> Well you have a beta running, load it up with data and let a
> few hundred clients loose on it. I've seen win2k BSOD with
> less stress than that.
That's what I was doing, loading it up with hundreds of connections from
other boxes, using osdb and pgbench. I'm not saying it's bug free, or
that Win2K won't crash under it, but it performed well for me - better
than 2 of the leading commercial databases.
I agree with Katie, dismissing a largely untested product because it
runs on Windows is not a good thing. Yes, Windows can BSOD, but when a
system is built on good hardware (for which good quality drivers are
available), and configured well it can be as reliable, if not more so
than some versions of Linux that have been released.
I would be interested to know how many windows servers those that are
against a windows port of PostgreSQL have or do manage, and how
experienced they are with that platform...
Regards, Dave.