On 17/06/18 13:56, Piero wrote: > Hello all, > first time that I write in this mailing list, I hope is the right place. > I need some help, and someone in the linux italian google group advices > me to write here, so excuse me in advance if this is not the right place. > > I have some skill in linux, but surely I'm not a "raid ninja". > > I have a old (very old) Thecus 3200PRO NAS, on wich I have installed, in > the far 2011, Ubuntu 10.04 server. The NAS have three 2TB HD configured > in RAID-5, mounted in /dev/md0. > Two or three weeks ago, I heard a strange noise from one of the hard disks. > > Checking the "situation" with mdadm -D /dev/md0 tells me that the RAID > is degraded, and one of the disks has been excluded from the RAID. > > Unfortunately, I have removed the wrong HD from the tray (damned > hurry!). Then I have re-inserted the good disk and removed the faulty > one, but now the RAID no longer starts > fdisk -l says the following (/dev/sdb is the drive removed by error, I > think): > So long as you didn't try to force a start, it shouldn't have messed things up. That said, I am puzzled by some of your results ... > Disco /dev/sdb: 2000.4 GB, 2000398934016 byte > 255 testine, 63 settori/tracce, 243201 cilindri > Unità = cilindri di 16065 * 512 = 8225280 byte > Sector size (logical/physical): 512 bytes / 4096 bytes > I/O size (minimum/optimal): 4096 bytes / 4096 bytes > Identificativo disco: 0x0b4f7f28 > > Dispositivo Boot Start End Blocks Id System > /dev/sdb1 1 243201 1953512001 83 Linux > Partition 1 does not start on physical sector boundary. > > and this for the good one: > > WARNING: GPT (GUID Partition Table) detected on '/dev/sdc'! The util > fdisk doesn't support GPT. Use GNU Parted. Okay. Boot using a modern rescue disk. fdisk should support GPT. And if you have any trouble you REALLY want to be using the absolute latest mdadm. > > Disco /dev/sdc: 2000.4 GB, 2000398934016 byte > 256 testine, 63 settori/tracce, 242251 cilindri > Unità = cilindri di 16128 * 512 = 8257536 byte > Sector size (logical/physical): 512 bytes / 512 bytes > I/O size (minimum/optimal): 512 bytes / 512 bytes > Identificativo disco: 0x00000000 > > Dispositivo Boot Start End Blocks Id System > /dev/sdc1 1 242252 1953514583+ ee GPT > > > I have buyed a new 2TB Hard Disk, but I haven't touched nothing since I > don't know what to do if not more damages. Brilliant!!! The number of people who mess things up by trying when they shouldn't ... You might want to buy a second spare 2TB disk (or not), but we'll see. > > As I have seen on "asking for help" page of the wiki, these are the > outputs that can help someone to help me: > > root@thecus:~# mdadm --examine /dev/sdb1 > /dev/sdb1: > Magic : a92b4efc > Version : 1.2 > Feature Map : 0x0 > Array UUID : 3115aed7:e283c89f:4407bf04:b1c52771 > Name : thecus:0 (local to host thecus) > Creation Time : Sun Nov 6 12:27:15 2011 > Raid Level : raid5 > Raid Devices : 3 > > Avail Dev Size : 3907023730 (1863.01 GiB 2000.40 GB) > Array Size : 7803486208 (3720.99 GiB 3995.38 GB) > Used Dev Size : 3901743104 (1860.50 GiB 1997.69 GB) > Data Offset : 272 sectors > Super Offset : 8 sectors > State : clean > Device UUID : 97707fac:738775ea:c9bc41e2:43f46b2c > > Update Time : Mon May 28 16:39:36 2018 > Checksum : 4a45df11 - correct > Events : 44021 Okay ... > > Layout : left-symmetric > Chunk Size : 4096K > > Array Slot : 3 (failed, 1, failed, 2, failed, failed, failed, > failed, failed, failed, failed, failed, failed, failed, failed, failed, <snip> > Array State : _uU 382 failed > > root@thecus:~# mdadm --examine /dev/sdc1 > /dev/sdc1: > Magic : a92b4efc > Version : 1.2 > Feature Map : 0x0 > Array UUID : 3f7c6079:b69c27eb:8e4344e4:a3403434 > Name : thecus:10 (local to host thecus) > Creation Time : Sun Nov 6 12:27:14 2011 > Raid Level : raid1 > Raid Devices : 3 > > Avail Dev Size : 4192256 (2047.34 MiB 2146.44 MB) > Array Size : 4192232 (2047.33 MiB 2146.42 MB) > Used Dev Size : 4192232 (2047.33 MiB 2146.42 MB) > Data Offset : 2048 sectors > Super Offset : 8 sectors > State : clean > Device UUID : 19dbe832:b6c96752:770ba051:d7569302 > > Update Time : Sun Jun 19 19:17:54 2016 > Checksum : 3a2e0b5a - correct > Events : 110 > This worries me ... this is WILDLY different from the other disk ... > > Array Slot : 1 (0, 1, 2, failed, failed, failed, failed, failed, > failed, failed, failed, failed, failed, failed, failed, failed, failed, <snip> > failed, failed, failed, failed, failed, failed, failed) > Array State : uUu 381 failed > > and similar for the /dev/sdc(2 and 3) but not for /dev/sdb(2 and 3) > > root@thecus:~# mdadm --detail /dev/md0 > mdadm: md device /dev/md0 does not appear to be active. > > > root@thecus:~# cat /proc/mdstat > Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] > [raid4] [raid10] > md10 : inactive sdc1[1](S) > 2096128 blocks super 1.2 > > md50 : inactive sdc3[1](S) > 524276 blocks super 1.2 > > md0 : inactive sdc2[1](S) sdb1[3](S) > 3904385465 blocks super 1.2 > > unused devices: <none> > > > hoping to have given all the necessary information, you think can I > recover the RAID and data stored in? > I'm worried ... Anyways, first things first. I'm guessing /dev/sda failed. Do you have another computer? Can you examine this disk there? Can you do a ddrescue to copy the contents onto your new hard disk? > How can I do? > > Sorry for the length of the message (and obviously for my english too), > and really really thanks in advance for your help > Dunno whether anybody can provide any other help but ... I'm worried that sdc has been trashed. That means that sdb is your only good disk. It might well be a good idea to get another 2TB drive and ddrescue that as well. If you can ddrescue sda, let us know what --examine says. Hopefully it's reasonably intact, and we can rebuild the array from the two copied disks. Cheers, Wol -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
