From mboxrd@z Thu Jan 1 00:00:00 1970 From: John Robinson Subject: Re: Another take on replacing failed raid drives Date: Fri, 26 Mar 2010 18:59:26 +0000 Message-ID: <4BAD040E.2010304@anonymous.org.uk> References: <4BAA469B.3050001@tmr.com> <4BACCD8B.8040705@xs4all.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4BACCD8B.8040705@xs4all.nl> Sender: linux-raid-owner@vger.kernel.org To: Linux RAID List-Id: linux-raid.ids On 26/03/2010 15:06, John Hendrikx wrote: > Bill Davidsen wrote: >> I stumbled on this description of replacing a drive in a raid array. I >> share it because different takes on a subject are usesful, and because >> he had multiple arrays using partitions of the failed drive, rather >> than a partitioned array. >> >> http://blogs.techrepublic.com.com/opensource/?p=1368 >> > I do that too (multiple arrays using partitions). My reason for that is > to make future upgrades less painful as it is possible to just copy and > drop one array at a time. I do that too, but my reason is that the arrays are different types. For example, first partitions make RAID-1 for /boot, second partitions RAID-1 or RAID-10 swap, third partitions RAID-5 or RAID-6 filesystem (or several over LVM). Cheers, John.