From mboxrd@z Thu Jan 1 00:00:00 1970 From: Phil Turmel Subject: Re: RAID 6 Not Mounting (Block device is empty) Date: Sat, 7 Nov 2015 13:49:02 -0500 Message-ID: <563E479E.50409@turmel.org> References: Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: Sender: linux-raid-owner@vger.kernel.org To: Francisco Parada , linux-raid@vger.kernel.org List-Id: linux-raid.ids On 11/07/2015 12:05 PM, Francisco Parada wrote: > Hello, >=20 > I=E2=80=99m not sure if this is the right way to go about it, so let = me give > you my story. I had a 7 x 3TB RAID array (12TB in total), 6 drives > in the array and 1 spare that was also part of the array but simply a > spare waiting at the ready in case of a drive failure, is what I had > running before my array broke. I added two new arrays to my system > last night, in order to back up my current RAID 6. The first array > was a 3 x 3TB RAID 0, for a total of 9TB. Then a 2 x 1TB RAID 0 > array, for a total of 2TB. 9+2 =3D 11 and although I=E2=80=99m 1TB s= hy, I knew > I had a bunch of crap and redundancy to get rid of, I just really > needed a solid backup after I was going to clean up. Yes, this is the right way to go about it. Missed just a few items tha= t would help. Good report. > After creating the new arrays, I started transferring from my 12TB > array, 2TB worth of data to the 2TB RAID 0 array. At some point > during the transfer, rsync complained of an I/O error. It seemed to > have transferred 500GB worth of data before this mishap. The > following morning, I noticed that error, and saw that I couldn=E2=80=99= t > mount my 12TB array anymore. Mind you, I didn=E2=80=99t touch this o= riginal > array, but I think what happened was that the I/O error blew 2 of my > drives. >=20 > What I=E2=80=99m thinking of doing is the following, but I=E2=80=99m = just looking for > some advice in case I=E2=80=99m missing anything: >=20 > sudo mdadm create --assume-clean --level=3D6 --raid-devices=3D7 --siz= e=3D2930135040 /dev/md127 /dev/sde /dev/sdf /dev/sdg /dev/sdh missing missing /dev/sdd Absolutely not! Your array went from running to dead in minutes, so the variation of event counts doesn't matter that much. You should forcibly re-assemble with all devices. However, before you do *anything*, you need to figure out why so many devices were ejected from your array. Was it a controller glitch? A power supply failure? Or, most likely, Unrecoverable Read Errors being exposed by your first-ever backup, combined with timeout mismatch? Any attempt to reassemble/recreate/recovery will simply blow up again i= f the root cause isn't addressed. In your next reply, please paste: 1) the dmesg from the time around the event, +/- a few minutes. 2) the output of the following drive diagnostics: for x in /dev/sd[a-z] ; do echo $x ; smartctl -i -A -l scterc $x ; done Do *not* perform any --create operation on your array. *Do* read the list archives linked below -- if any part of it is unclear, please ask in your next reply. Phil [1] http://marc.info/?l=3Dlinux-raid&m=3D139050322510249&w=3D2 [2] http://marc.info/?l=3Dlinux-raid&m=3D135863964624202&w=3D2 [3] http://marc.info/?l=3Dlinux-raid&m=3D135811522817345&w=3D1 [4] http://marc.info/?l=3Dlinux-raid&m=3D133761065622164&w=3D2 [5] http://marc.info/?l=3Dlinux-raid&m=3D132477199207506 [6] http://marc.info/?l=3Dlinux-raid&m=3D133665797115876&w=3D2 [7] http://marc.info/?l=3Dlinux-raid&m=3D142487508806844&w=3D3 -- 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