From mboxrd@z Thu Jan 1 00:00:00 1970 From: Phil Turmel Subject: Re: Requesting assistance recovering RAID-5 array Date: Wed, 1 Apr 2020 14:32:18 -0400 Message-ID: References: <058b3f48-e69d-2783-8e08-693ad27693f6@youngman.org.uk> <1d6b3e00-e7dd-1b19-1379-afe665169d44@turmel.org> <061b695a-2406-fc00-dd6d-9198b85f3b1b@turmel.org> <5E8485DA.2050803@youngman.org.uk> <9a303b9b-52b8-f0c6-4288-120338c6572f@turmel.org> <1f4b8c74-4c38-6ea4-6868-b28f9e5c4a10@turmel.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Content-Language: en-US Sender: linux-raid-owner@vger.kernel.org To: Daniel Jones Cc: Wols Lists , linux-raid@vger.kernel.org List-Id: linux-raid.ids Hi Daniel, On 4/1/20 2:07 PM, Daniel Jones wrote: > Hi Phil, > > So far so good. Yes. > # mdadm --manage /dev/md0 --add /dev/mapper/sdb1 > mdadm: added /dev/mapper/sdb1 Don't do this. Overlays can't really handle the amount of data that would be involved, and you definitely don't want to rebuild yet. > 4: Summary > > The drives have had physical partitions written. > I think I've found the correct offset and device order to use --create > to restore the array to the degraded state it was in before the > superblocks were overwritten. Yes. > I'm not sure why the --add doesn't work. Don't do the --add operation until you've copied anything critical in the array to external backups (while running with 3 of 4). The reason is that any not-yet-discovered URE on those three will certainly crash the array during rebuild. It could still crash copying critical stuff, but you can repeatedly --assemble --force to keep going with the next items to backup. Only when you've backed up everything possible do you --add the fourth drive back into the array. > Thanks so much for your help this far. You're welcome. > Regards, > DJ Phil