What was: RAID 1 with mdadm on whole disks say /dev/md0 from /dev/sda and sdb, GPT with one partition on md0 and ext4 on this partition (md0p1).
What I think happens: after changing motherboard, Linux detects problems with ext4 on md0p1.
I run fsck on md0p1 and say "yes" for all questions. It was some bad check sums, lots of extent tree could be narrower and some not empty journals.
It seems to end successfully and I try to mount /dev/md0p1
but have same error about bad filesystem.
I run fsck on md0p1 again but now it say "no superblock" and alternative superblock numbers doesn't help.
I reboot and now mdadm can't find its superblock on both sda and sdb. GPT partition still fine, but no signs of ext4 on both disks found by testdisk (only ms data).
# fdisk -l
GPT PMBR size mismatch (3907028991 != 3907029167) will be corrected by w(rite).
Disk /dev/sda: 1,8 TiB, 2000398934016 bytes, 3907029168 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: gpt
Disk identifier: D09686A6-2428-48EC-868B-D3C8CE5E0C23
Device Start End Sectors Size Type
/dev/sda1 34 3907024064 3907024031 1,8T Microsoft basic data
Partition 1 does not start on physical sector boundary.
...
GPT PMBR size mismatch (3907028991 != 3907029167) will be corrected by w(rite).
Disk /dev/sde: 1,8 TiB, 2000398934016 bytes, 3907029168 sectors
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 4096 bytes
I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Disklabel type: gpt
Disk identifier: D09686A6-2428-48EC-868B-D3C8CE5E0C23
Device Start End Sectors Size Type
/dev/sde1 34 3907024064 3907024031 1,8T Microsoft basic data
Partition 1 does not start on physical sector boundary.
# mdadm --examine /dev/sd*
/dev/sda:
MBR Magic : aa55
Partition[0] : 3907028991 sectors at 1 (type ee)
mdadm: No md superblock detected on /dev/sda1.
...
/dev/sde:
MBR Magic : aa55
Partition[0] : 3907028991 sectors at 1 (type ee)
mdadm: No md superblock detected on /dev/sde1.
gdisk /dev/sda
GPT fdisk (gdisk) version 1.0.1
Partition table scan:
MBR: protective
BSD: not present
APM: not present
GPT: present
Found valid GPT with protective MBR; using GPT.
Command (? for help): p
Disk /dev/sda: 3907029168 sectors, 1.8 TiB
Logical sector size: 512 bytes
Disk identifier (GUID): D09686A6-2428-48EC-868B-D3C8CE5E0C23
Partition table holds up to 128 entries
First usable sector is 34, last usable sector is 3907029134
Partitions will be aligned on 8-sector boundaries
Total free space is 5070 sectors (2.5 MiB)
Number Start (sector) End (sector) Size Code Name
1 34 3907024064 1.8 TiB 0700
Command (? for help): i
Using 1
Partition GUID code: EBD0A0A2-B9E5-4433-87C0-68B6B72699C7 (Microsoft basic data)
Partition unique GUID: E11B0DE3-9ABD-47B2-9F09-E993F76FBC6F
First sector: 34 (at 17.0 KiB)
Last sector: 3907024064 (at 1.8 TiB)
Partition size: 3907024031 sectors (1.8 TiB)
Attribute flags: 0000000000000000
Partition name: ''
So my questions:
- Is there a chance to recover filesystem? I can recover some files by foremost but it is 2 TB of trash and very few valuable files, so it is meaningless without at least filenames.
- More important: what was wrong? It seems I did all what was suggested (except no backup), and lost my data.
Because situation is weird, I'll tell whole story:
Linux runs on SSD and most of data (include home directory) lives on RAID over 2 HDD.
RAID works fine all time since 2011 or 2012.
6-8 month ago I upgraded computer: change processor from 2 to 8 cores, add RAM and insert SSD for Windows.
After this computer ended turning on from first attempt, I had to press reset button 1-2 within 10-20 seconds to turn it on. But all other systems works fine.
1-2 month ago 2 times all applications start crashing and I see I/O errors in console, but after reboot all works fine.
1 month ago a upgraded kubuntu to latest release.
2 weeks ago things gone bad
Linux won't start – some error on SSD. I bought other SSD, manage to save most of file system by ddrescue, but it would not start, so I installed fresh OS on empty partition on SSD. It assemble RAID after install mdadm but would not add partition to /dev: it was /dev/md127 but not md127p1.
I fixed GPT table in md127 by gdisck followed it suggestions (it has fine primary GTP table) and align backup GTP table that was corrupted. fsck on md127p1 (it changed to md0p1) was fine, I successfully mount it.
It works one or two days, then computer refused to start anyway.
I managed to lunch BIOS one time and it was no IDE devices, so I bought new motherboard (old was asrock 900FX Extreme3, new gigabyte 970-DS3P).
After motherboard change I run Linux, it starts in recovery mode
(/dev/md0p1 has problems with file system) and then it was what I wrote in begining of my question.
What was done wrong?
- no backup – of course. Now I understand that RAID isn't a backup.
- ignoring of IO errors? It crashes my SSD with system, so it just run me to install fresh system.
- is it bad idea to have partitions inside RAID? Is it better to assemble raid from sda1 and sdb1 then from sda and sdb?
addition:
lsblk
NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT
sda 8:0 0 1,8T 0 disk <-- raid
└─sda1 8:1 0 1,8T 0 part
sdb 8:16 1 957,9M 0 disk
└─sdb1 8:17 1 956,9M 0 part
sdc 8:32 0 59,6G 0 disk
├─sdc1 8:33 0 1M 0 part
├─sdc2 8:34 0 29,8G 0 part /old
└─sdc3 8:35 0 29,8G 0 part /
sdd 8:48 0 119,2G 0 disk
└─sdd1 8:49 0 119,2G 0 part
sde 8:64 0 1,8T 0 disk <-- raid
└─sde1 8:65 0 1,8T 0 part
sr0 11:0 1 2G 0 rom
