From mboxrd@z Thu Jan 1 00:00:00 1970 From: Phil Turmel Subject: Re: 4x3TB raid5 - no autodetect Date: Mon, 22 Oct 2012 12:14:45 -0400 Message-ID: <508570F5.9090202@turmel.org> References: <441614417.20121022155507@oudeis.org> <508553FF.6020302@anonymous.org.uk> <123699751.20121022175600@oudeis.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <123699751.20121022175600@oudeis.org> Sender: linux-raid-owner@vger.kernel.org To: =?UTF-8?B?UmFpbmVyIEbDvGdlbnN0ZWlu?= Cc: John Robinson , linux-raid@vger.kernel.org List-Id: linux-raid.ids On 10/22/2012 11:56 AM, Rainer F=C3=BCgenstein wrote: >=20 > JR> In-kernel auto-assembly is only supported for metadata 0.90, and = even=20 > JR> then the feature is deprecated. If you have your root filesystem = on a md=20 > JR> array, use an initrd with mdadm in it to start it. >=20 > if I understand it correctly, metadata 0.90 is not supporting drives = > > 2TB? v0.9 supports drives to 4T, but there was a bug in several released kernels that triggered above 2T. There are also unsolvable misidentification issues for partitioned drives where the last partitio= n can be confused with the whole disk. > will in-kernel auto-assembly be supported sometimes in the future? > this was quite a useful feature. I'm sure Neil will pipe up if necessary, but he has been adamantly against extending the in-kernel autoassembly code in any way. Auto- assembly is also incompatible with GPT partitions. You should use v1.0 for mirrored boot partitions, and v1.2 for all othe= r arrays. And use an initramfs. I, too, squirmed when faced with the need to set up an initramfs, as I was quite comfortable with my legacy configuration. I'm glad I got ove= r it. HTH, Phil -- 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