Re: looking for advice on raid0+raid5 array recovery with mdadm and sector offset
From: den Hoog <hidden>
Date: 2014-01-06 06:53:53
Hi Neil, hope you have had good holidays and appreciate your help! good to know that it is useless to go on the offset path for imsm. I know for sure the sdb disk was cleared by the win recovery as it created a new 2GB partition on it I will however try to re-create when I get home and keep you posted Probably will have to go the 'missing' way. Will that somehow figure out that it needs an offset? br Dennis On Mon, Jan 6, 2014 at 2:41 AM, NeilBrown [off-list ref] wrote:
On Thu, 2 Jan 2014 20:45:24 +0100 den Hoog [off-list ref] wrote:quoted
Hi Neil I apologize if I made mistakes with the first mail post but probably something went wrong, so this is a retry. I'm looking for advice on my plan to recover my raid5 volume with mdadm. I was in a hurry and made a stupid mistake when upgrading the MB bios. Forgot to turn on the Intel SATA raid, and Windows recovery erased the first disk of 4. It is an array of 4x4TB, in a matrix, having 1 RAID0 volume, and 1 RAID5 vol. Although the array displays a failed array in Windows, 3 disks are active, and 1 is missing and showing as a non-raid array being available. In (my) theory, I should still be able to recover the raid5 vol. with the remaining 3 disks, however I should specify the specific sector offset I guess. I read many articles on this, but none of them address the 'difficulty' of recovering a specific volume when multiple exist in an array. Although I've some backups, I really would appreciate your help in getting this recovered. sda is the SSD sdb is the 'missing' and erased drive (serial ending on P82C) sdc is the second drive in the array sdd is the 3rd drive in the array sde is the 4th drive in the array sdf is the usb stick I'm running Fedora live from What I've done so far : - Started Fedora 15 Live from a USB - installed the mdadm package data_offset and compiled My plan to work with an offset to recover the [HitR5] volume: - echo 1 > /sys/module/md_mod/parameters/ start_dirty_degraded - mdadm -C /dev/md/imsm -e imsm -n 4 /dev/sdb /dev/sdc /dev/sdd /dev/sde - mdadm -C /dev/md0 -l5 -n4 -c 128 /dev/sdb:1073746184s /dev/sdc:1073746184s /dev/sdd:1073746184s /dev/sde:1073746184sThis certainly won't work. You need "--data-offset=variable" for the "NNNNs" suffixes to be recognised, and even then it only works for 1.x metadata, not for imsm metadata. There isn't much support for sticking together broken IMSM arrays at present. Your best bet is to re-create the whole array. So: mdadm -C /dev/md/imsm -e imsm -n 4 /dev/sd[bcde] mdadm -C /dev/md0 -l0 -n4 -c 128K -z 512G /dev/md/imsm then check that /dev/md0 looks OK for the RAID0 array. If it does then you can continue to create the raid5 array mdadm -C /dev/md1 -l5 -n4 -c 128k --assume-clean /dev/md/imsm That *should* then be correct. If the RAID0 array doesn't look right, the possible sdb really was cleared rather than just having the metadata erased. In this case the RAID0 is definitely gone and it will be a bit harder to create the RAID5. It could be something like: mdadm -C /dev/md1 -l5 -n4 -c 128k missing /dev/sd[cde] but I'm not sure that 'missing' is works for imsm. If you need to go this way I can try to make 'missing' for for imsm. It shouldn't be too hard. NeilBrownquoted
I'm in doubt about working with missing disks first to start a degraded array, with -C and missing for the first drive. Or choosing assemble --auto, or as stated above and create the volume with an offset. Another thing I'm not certain of: do I need to build a new mdadm with data_offset, or is it already present in my 3.2.6 version? When I built a new version with Neils mdadm I ended up with a 3.2.5 18May2012 version. As I guess I have only one shot at this I have not executed anything yet. thanks many for your help, time and advice! best regards Dennis =======output mdadm -Evvvvs============= root@localhost ~]# mdadm -Evvvvs mdadm: No md superblock detected on /dev/dm-1. mdadm: No md superblock detected on /dev/dm-0. /dev/sdf1: MBR Magic : aa55 Partition[0] : 432871117 sectors at 3224498923 (type 07) Partition[1] : 1953460034 sectors at 3272020941 (type 16) Partition[3] : 924335794 sectors at 50200576 (type 00) /dev/sdf: MBR Magic : aa55 Partition[0] : 15769600 sectors at 2048 (type 0b) /dev/sde: Magic : Intel Raid ISM Cfg Sig. Version : 1.3.00 Orig Family : f3437c9b Family : f3437c9d Generation : 00002c5f Attributes : All supported UUID : 47b011c7:4a8531ea:7e94ab93:06034952 Checksum : 671f5f84 correct MPB Sectors : 2 Disks : 4 RAID Devices : 2 Disk03 Serial : PL1321LAG4RXEH State : active Id : 00000005 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) [HitR0]: UUID : 73ecf2cf:fcfd2598:d6523381:71e57931 RAID Level : 0 Members : 4 Slots : [_UUU] Failed disk : 1 This Slot : 3 Array Size : 4294967296 (2048.00 GiB 2199.02 GB) Per Dev Size : 1073742088 (512.00 GiB 549.76 GB) Sector Offset : 0 Num Stripes : 4194304 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean [HitR5]: UUID : 71626250:b8fc1262:3545d952:69eb329e RAID Level : 5 Members : 4 Slots : [_UU_] Failed disk : 3 This Slot : 3 (out-of-sync) Array Size : 20220831744 (9642.04 GiB 10353.07 GB) Per Dev Size : 6740279304 (3214.02 GiB 3451.02 GB) Sector Offset : 1073746184 Num Stripes : 26329208 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean Disk00 Serial : PL2311LAG1P82C:0 State : active Id : ffffffff Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk01 Serial : PL1321LAG4NMEH State : active Id : 00000003 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk02 Serial : PL1321LAG4TH4H State : active Id : 00000004 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) /dev/sdd: Magic : Intel Raid ISM Cfg Sig. Version : 1.3.00 Orig Family : f3437c9b Family : f3437c9d Generation : 00002c5f Attributes : All supported UUID : 47b011c7:4a8531ea:7e94ab93:06034952 Checksum : 671f5f84 correct MPB Sectors : 2 Disks : 4 RAID Devices : 2 Disk02 Serial : PL1321LAG4TH4H State : active Id : 00000004 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) [HitR0]: UUID : 73ecf2cf:fcfd2598:d6523381:71e57931 RAID Level : 0 Members : 4 Slots : [_UUU] Failed disk : 1 This Slot : 2 Array Size : 4294967296 (2048.00 GiB 2199.02 GB) Per Dev Size : 1073742088 (512.00 GiB 549.76 GB) Sector Offset : 0 Num Stripes : 4194304 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean [HitR5]: UUID : 71626250:b8fc1262:3545d952:69eb329e RAID Level : 5 Members : 4 Slots : [_UU_] Failed disk : 3 This Slot : 2 Array Size : 20220831744 (9642.04 GiB 10353.07 GB) Per Dev Size : 6740279304 (3214.02 GiB 3451.02 GB) Sector Offset : 1073746184 Num Stripes : 26329208 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean Disk00 Serial : PL2311LAG1P82C:0 State : active Id : ffffffff Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk01 Serial : PL1321LAG4NMEH State : active Id : 00000003 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk03 Serial : PL1321LAG4RXEH State : active Id : 00000005 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) /dev/sdc: Magic : Intel Raid ISM Cfg Sig. Version : 1.3.00 Orig Family : f3437c9b Family : f3437c9d Generation : 00002c5f Attributes : All supported UUID : 47b011c7:4a8531ea:7e94ab93:06034952 Checksum : 671f5f84 correct MPB Sectors : 2 Disks : 4 RAID Devices : 2 Disk01 Serial : PL1321LAG4NMEH State : active Id : 00000003 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) [HitR0]: UUID : 73ecf2cf:fcfd2598:d6523381:71e57931 RAID Level : 0 Members : 4 Slots : [_UUU] Failed disk : 1 This Slot : 1 Array Size : 4294967296 (2048.00 GiB 2199.02 GB) Per Dev Size : 1073742088 (512.00 GiB 549.76 GB) Sector Offset : 0 Num Stripes : 4194304 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean [HitR5]: UUID : 71626250:b8fc1262:3545d952:69eb329e RAID Level : 5 Members : 4 Slots : [_UU_] Failed disk : 3 This Slot : 1 Array Size : 20220831744 (9642.04 GiB 10353.07 GB) Per Dev Size : 6740279304 (3214.02 GiB 3451.02 GB) Sector Offset : 1073746184 Num Stripes : 26329208 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean Disk00 Serial : PL2311LAG1P82C:0 State : active Id : ffffffff Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk02 Serial : PL1321LAG4TH4H State : active Id : 00000004 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk03 Serial : PL1321LAG4RXEH State : active Id : 00000005 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) mdadm: No md superblock detected on /dev/sdb1. /dev/sdb: MBR Magic : aa55 Partition[0] : 4294967295 sectors at 1 (type ee) /dev/sda2: MBR Magic : aa55 Partition[0] : 1816210284 sectors at 1920221984 (type 72) Partition[1] : 1953653108 sectors at 1936028192 (type 6c) Partition[3] : 447 sectors at 27722122 (type 00) /dev/sda1: MBR Magic : aa55 Partition[0] : 1816210284 sectors at 1920221984 (type 72) Partition[1] : 1953653108 sectors at 1936028192 (type 6c) Partition[3] : 447 sectors at 27722122 (type 00) /dev/sda: MBR Magic : aa55 Partition[0] : 716800 sectors at 2048 (type 07) Partition[1] : 499396608 sectors at 718848 (type 07) mdadm: No md superblock detected on /dev/loop4. mdadm: No md superblock detected on /dev/loop3. mdadm: No md superblock detected on /dev/loop2. mdadm: No md superblock detected on /dev/loop1. mdadm: No md superblock detected on /dev/loop0. /dev/md127: Magic : Intel Raid ISM Cfg Sig. Version : 1.3.00 Orig Family : f3437c9b Family : f3437c9d Generation : 00002c5f Attributes : All supported UUID : 47b011c7:4a8531ea:7e94ab93:06034952 Checksum : 671f5f84 correct MPB Sectors : 2 Disks : 4 RAID Devices : 2 Disk02 Serial : PL1321LAG4TH4H State : active Id : 00000004 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) [HitR0]: UUID : 73ecf2cf:fcfd2598:d6523381:71e57931 RAID Level : 0 Members : 4 Slots : [_UUU] Failed disk : 1 This Slot : 2 Array Size : 4294967296 (2048.00 GiB 2199.02 GB) Per Dev Size : 1073742088 (512.00 GiB 549.76 GB) Sector Offset : 0 Num Stripes : 4194304 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean [HitR5]: UUID : 71626250:b8fc1262:3545d952:69eb329e RAID Level : 5 Members : 4 Slots : [_UU_] Failed disk : 3 This Slot : 2 Array Size : 20220831744 (9642.04 GiB 10353.07 GB) Per Dev Size : 6740279304 (3214.02 GiB 3451.02 GB) Sector Offset : 1073746184 Num Stripes : 26329208 Chunk Size : 128 KiB Reserved : 0 Migrate State : idle Map State : failed Dirty State : clean Disk00 Serial : PL2311LAG1P82C:0 State : active Id : ffffffff Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk01 Serial : PL1321LAG4NMEH State : active Id : 00000003 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) Disk03 Serial : PL1321LAG4RXEH State : active Id : 00000005 Usable Size : 7814030862 (3726.02 GiB 4000.78 GB) -- To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html