| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Usenet -> c.d.o.server -> Re: Oracle 10g on HP blade server
JEDIDIAH wrote:
> On 2006-11-08, Charles Hooper <hooperc2000_at_yahoo.com> wrote:
> > DA Morgan wrote:
> >> Niall Litchfield wrote:
> >> > DA Morgan wrote:
> >> >> If you can eliminate Windows and go to Linux you will, however, greatly
> >> >> improve the quality of the experience.
> >>
> >> > none of my windows admins understand linux,
> >>
> >> All *NIX admins, however, understand Windows ... which explains
> >> why they are working with *NIX. <g>
> >> --
> >> Daniel A. Morgan
> >> University of Washington
> >> damorgan_at_x.washington.edu
> >> (replace x with u to respond)
> >> Puget Sound Oracle Users Group
> >> www.psoug.org
> >
> > And some of us Windows admins have worked with Linux, having had to
> > patch and compile the kernel to add X.509 certificate capabilities in
> > order to build a VPN server from the ground up, and then finding that
>> > were susceptible to an exploit, and that development of that VPN server
> Oh really?
>
> > the patch was no good and having to start over with compiling the
> > Working Overloaded Kernel, and then finding three months later (having
> > spent three months building the VPN server) that all such VPN servers
>> kernel but in a userspace daemon. Either way, the original build
> That's a nice story.
>
> Except the vulnerability you're talking about is not in the
>> "problem". Although the gratis-ware ones offer the basic services.
> This goes for the kernel as well as the ipsec utilities.
>
> The commercial distributors specifically exist to avoid this
>> > outages), I wouldn't trust Linux to run an Oracle database. That said,
> >
> > We have a couple Linux boxes running here, but after seeing one of them
> > complain that there is severe file corruption on the hard drive, EVERY
> > time the server has been restarted after 90 days of uptime (darn power
>> a house to fall on them before replacing obviously broken hardware.
> It's nice to know my own admins aren't the only ones that need
>> > PC Support Specialist
> >
> > Charles Hooper
Based on the tone of your response, my post apparently can be read with a couple different interpretations. The post was not intended as an insult to Linux, Unix, or Windows operating systems, nor was it intended as an insult to the administrators of those system.
As stated in my opening statement, "And some of us Windows admins have worked with Linux" and the closing statement "That said, I put two more Linux servers in production last night", Linux has a lot of potential. In between the opening and closing remarks, I tried to demonstrate that at least one of the Windows admins is Linux aware, as I am sure many other Windows admins already are, or are striving to increase their knowledge. We use Windows systems where they make the most sense, and Linux where they make the most sense (roughly a 50/50 split for servers).
It is far easier for a person fresh out of college to be a Windows admin, compared to being a Linux, Unix, or even a Novell Netware 3.12 admin - Linux and Unix administration requires greater depth of skill than does being a Windows admin, as does Oracle administration compared to SQL Server administration. The greater depth of skill, and improved flexibility of the platforms allows things to happen on those platforms that can't be done on Windows.
If I built an Oracle database on a Windows 2003 x64 platform, and a similar one on a Linux system on the same hardware configuration, which setup would be better? Likewise, if a couple of the Unix and Linux experts on this group started with the same hardware, which of their systems would be better? The answer is, it depends. Does it matter, for instance, if a Unix system is 33% faster than a Windows system, or a Linux system is 25% faster than a Windows system, if that system is put into use in an environment that is 100% something else? I suggest that being so focused that one thing is ALWAYS better than something else is not only wrong, but dangeous.
Any statement that includes ALL, ALWAYS, or NONE only takes one contrary case to be untrue.
Charles Hooper
PC Support Specialist
K&M Machine-Fabricating, Inc.
Received on Wed Nov 08 2006 - 20:43:34 CST
 |
 |