Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
![]() |
![]() |
Home -> Community -> Mailing Lists -> Oracle-L -> Re: Oracle on Windows Vs. Linux
On 05/19/2006 03:33:30 PM, oracle-l-bounce_at_freelists.org wrote:
> We're rerunning all the scenarios today. I hope to have some stuff
> tomorrow. I will look into these settings.
>
As a clown prince of this group and known Linux True Believer(TM),
I believe that I'm expected to write something on this subject.
Basically, if it is even a question for you, if there is a trace of
doubt, then you should use Windows.
Linux is only for us, The True Believers(TM) who will gladly tolerate
endless patching, driver quirks, failed standards like I2O and crappy
slow things like FireWire and lack of basic provisions present in all
other operating systems, like per process accounting of I/O requests.
Let me quote a bit from another list I follow, only as a passive reader:
"[lwlan-user] Build fails on 2.6.16 Fedora kernel Solomon Peachy solomon at linux-wlan.com Thu Mar 30 13:34:04 EST 2006
On Thu, Mar 30, 2006 at 02:22:08PM +0100, Dave Jenkins wrote:
> make[4]: Entering directory `/usr/src/kernels/2.6.16-1.2069_FC4-i686'
> CC [M]
> /home/davej/linux-wlan/linux-wlan-ng-0.2.3/src/prism2/driver/prism2_usb.o
> In file included from
> /home/davej/linux-wlan/linux-wlan-ng-0.2.3/src/prism2/driver/prism2_usb.c:2:
> /home/davej/linux-wlan/linux-wlan-ng-0.2.3/src/prism2/driver/hfa384x_usb.c:117:26:
> error: wlan/version.h: No such file or directory
> /home/davej/linux-wlan/linux-wlan-ng-0.2.3/src/prism2/driver/hfa384x_usb.c:138:30:
> error: wlan/wlan_compat.h: No such file or directory
No p80211 build, and then these go nuts. It looks like the semantics of kbuild changed, yet again. sigh. I'm looking into this now. Normally I'm fairly adventureous when it comes to newer kernels, but I've been using 2.6.15[.x] for a while now.
The last kernel that his majesty, Linus the 1st, gave us in his infinite wisdom was 2.6.16.16. The two number sixteens are not there because I'm drunk and seeing double, they're there because the kernel 2.6.16 was patched 16 times and always re-released, which is definitely a confidence instilling release procedure."
That means that Linux is changing kernel driver interface FOR AN ACTIVE KERNEL
RELEASE! No other OS does that and every other OS manufacturer would take some
serious fire for it. Can you imagine the reactions if SUN or MS did that? That
would be like Oracle suddenly releasing new and incompatible release of libclntsh.so
and not supporting some unimportant old calls like "bind" or "parse". Bummer!
PHP would no longer be usable, and neither would Perl (not that DBD::Oracle is
very useful now. It's a masterfully screwed up piece of OSS), Python or Ruby.
Only the The True Linux Believers(TM) can tolerate this. Linux is a matter of
faith, not a matter of benchmarks. Running Linux is like thanking the divine
providence for the jerk that cut you off on the highway, without giving direction
signal, which would have revealed his intentions. It's only for the strongest
in faith.
I must say that my faith was swayed by the recent post from Michael Haddon about
using Solaris on AMD64. There is an alternative to Linux! There is another form of
Unix operating system on PC platform, this one decent, stable and well documented
and somebody is using it! Now that is what I call a good news! Linux, as it is now,
is almost buggy enough for Oracle to produce its own version. Then, we will all be
blessed by Linux 2.8.37.135.20048.FOOBAR version of Linux! There is a God and Linus
Torvalds is his prophet! There is no other explanation for the popularity of this
crap.
-- Mladen Gogala http://www.mgogala.com -- http://www.freelists.org/webpage/oracle-lReceived on Sat May 20 2006 - 13:31:40 CDT
![]() |
![]() |