From mboxrd@z Thu Jan 1 00:00:00 1970 From: Keld =?iso-8859-1?Q?J=F8rn?= Simonsen Subject: Re: Sharing disks amoung multiple software RAIDs Date: Mon, 5 May 2008 01:10:16 +0200 Message-ID: <20080504231016.GA28472@rap.rap.dk> References: <1209692359.16523.2.camel@localhost> <73000.21239.qm@web50212.mail.re2.yahoo.com> <1209696179.16523.13.camel@localhost> <72dbd3150805020006p7f976a04o487389840f03b4a2@mail.gmail.com> <1209715763.7827.2.camel@localhost> <20080502102500.GB16198@rap.rap.dk> <481E0726.1030501@harddata.com> <20080504212927.GB20650@rap.rap.dk> <481E3374.4070105@harddata.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <481E3374.4070105@harddata.com> Sender: linux-raid-owner@vger.kernel.org To: Maurice Hilarius Cc: vger majordomo for lists List-Id: linux-raid.ids On Sun, May 04, 2008 at 04:06:44PM -0600, Maurice Hilarius wrote: > Keld J=F8rn Simonsen wrote: > >.. > >Hmm, would raid10,n2, or raid10,f2 then have these flaws too? > > =20 > I do not know. > I am, however, about to do some testing to find out. > Our company routinely was using RAID1 on systems, until we discovered= this. > With no ability to boot from RAID5 and similar this leaves us in a bi= t=20 > of a bind. As raid10,n2 is data wise equivalent to RAID1, then grub/lilo should be able to boot from raid10,n2.=20 raid10,n2 should have better performance than RAID10, but raid10,f2 actually should have better performance than both of the others.=20 See http://linux-raid.osdl.org/index.php/Performance (which I wrote). That is why I advocated raid10,f2 for the system partitions, as far as possible, only having /boot as a RAID1.=20 I have a setup which implements this and it runs fine. I look forward to hear about your tests on this. Best regards keld -- To unsubscribe from this list: send the line "unsubscribe linux-raid" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html