Re: More benchmark bullshit, and Linux luser mating calls... (was Re: Linux betas NT in TPC testing, running Oracle8

From: Jarkko Hietaniemi <jhi_at_alpha.hut.fi>
Date: 30 Apr 1999 22:46:56 +0300
Message-ID: <oeeu2tx99hr.fsf_at_alpha.hut.fi>


Stephen Edwards <ja207030_at_primenet.com> writes:

> Leslie Mikesell <les_at_MCS.COM> wrote:
>
> : In article <7g55ts$sqh$2_at_nnrp03.primenet.com>,
> : Stephen Edwards <ja207030_at_primenet.com> wrote:
> : >
> : >Whether or not one needs to reboot a server for changes has _ZERO_ bearing
> : >on its stability. Stability is an issue of how good the hardware and the
> : >drivers using that hardware are.
> : >
> : >[] "No footnote for you!" -- Footnote Nazi
> : >--
> : >.-----.
>
> : Downtime is downtime. Stability means running. I take it you haven't
>
> No. Stability means that the operating system doesn't crash. Just
> because I reboot an OS, doesn't make it unstable. If I have to boot SCO
> UNIX 3 or four times in one month, that means "it has to be rebooted to
> relink the kernel because I added stuff to it, etc.", not "it's unstable."
> What you're saying does not make any sense.

Many of the commercial Unixes can configure a running kernel: change kernel parameters, link (maybe even unlink?) kernel modules on the fly. I'm pretty certain Solaris can do this, and I know AIX can. Ergo, no reboots needed even then.

-- 
$jhi++; # http://www.iki.fi/jhi/
        # There is this special biologist word we use for 'stable'.
        # It is 'dead'. -- Jack Cohen
Received on Fri Apr 30 1999 - 21:46:56 CEST

Original text of this message