From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jure Pecar Subject: Re: Date: Tue, 4 Jun 2002 23:55:18 +0200 Sender: linux-raid-owner@vger.kernel.org Message-ID: <20020604235518.2cb7d7d7.pegasus@telemach.net> References: <20020604154712.AD48D8347@phoenix.clouddancer.com> Mime-Version: 1.0 Content-Type: multipart/signed; protocol="application/pgp-signature"; boundary="=.7/gcCXELj+7KAp" Return-path: In-Reply-To: <20020604154712.AD48D8347@phoenix.clouddancer.com> To: Colonel Cc: linux-raid@vger.kernel.org List-Id: linux-raid.ids --=.7/gcCXELj+7KAp Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit On Tue, 4 Jun 2002 08:47:12 -0700 (PDT) klink@clouddancer.com (Colonel) wrote: > > True, I think that the point is that of the 5 possible 2 disk > failures, 2 of them (in striped mirrors, not mirrored stripes) kill > the array. For RAID5, all of them kill the array. But the fancy RAID > setups are for _large_ arrays, not 4 disks, unless you are after the > small write speed improvement (as I am). going offtopic here ... what kind of raid setup is the best for write intensive load like mail queues & co? > Plus any raid metadevice made of metadevices cannot autostart, which > means tinkering during startup, which is only worth it for those large > drive arrays. hm? it does for me. probalby the redhat's rc.sysinit does the right thing ... -- Jure Pecar --=.7/gcCXELj+7KAp Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) iD8DBQE8/TdMOHUMrcOSg+8RApAiAJ99W2c93ggFRMpIJ/5tvEvF0tlLAQCgjeWq +ia2+EQq1T2ASBva16GbU1U= =EzXG -----END PGP SIGNATURE----- --=.7/gcCXELj+7KAp--