RE: [Nolug] call for help: white papers to show my CIO regarding Linux, Unix & Oracle

How ironic that I sit here working on a SQL Server, trying to recover data
and fix a screwed up file system at the same time. Lessons learned MS SQL
7.0 was never built for clustering (improper failover), when MSCS fails it
really hurts (corrupt file system)! I would defiantly recommend avoiding
this if at all possible.

-----Original Message-----
From: Andrew S. Johnson [mailto:andy@asjohnson.com]
Sent: Monday, April 28, 2003 6:40 PM
To: nolug@joeykelly.net
Subject: Re: [Nolug] call for help: white papers to show my CIO
regarding Linux, Unix & Oracle

On Sunday 27 April 2003 08:53 pm, Dustin Puryear wrote:
> > > Is Oracle really as DBA-intensive as I've heard?

In the same way that a 747 is harder to get off the ground than
a Yugo is to get out of the driveway, Oracle might be harder.
But then, once at cruising altitude, it's basically effortless. If you
take your hands off the Yugo for more than a few seconds, you
will surely crash. Once my Oracle databases are set up the way
I want them, they stay up as long as the hardware holds on. I take
them for granted, like the sunrise. I sure don't feel that way about
any of my Windoze servers.

I have no doubt that anybody who can pass the MCSE test can
operate a SQL server database as well. It's like Windoze that
way. Just keep a firm hand on the wheel, and hope something
doesn't jump out in front of you. It hasn't changed much since
they raped Sybase for it. But then, that's true of many of
Micros~t's products. A fancier facade with each new iteration,
but little of substance underneath.

At least Oracle is improving with each new version. Most DBA's
find the new features worth the upgrade. I know I do. Anyway,
I've been using Oracle on Linux in production where I work for
three years now on the process control side. IT was sufficiently
impressed that they switched from Windoze to Linux as well.
Its a smallish database, maybe 50 gigs or so, with about 1200
transactions per minute. The _ONLY_ outages have been
hardware related: RAID controller, 15k SCSI drives, and
power (UPS's don't last forever).

If you are really looking for something for your CIO, send him the
articles about Oracle converting internally from Solaris to Linux.
If the second largest software company in the world is converting
to Linux, what else do you need. Maybe drop other names too,
like Google, Home Depot, and Toyota. Any of those sound familiar?

Andy Johnson

___________________
Nolug mailing list
nolug@nolug.org
___________________
Nolug mailing list
nolug@nolug.org
Received on 04/28/03