On Wed, Oct 11 2017, Joseba Ibarra wrote:

Hi Mikael,

I had ext4

and for commands:

root@grafico:/mnt# fsck -n /dev/md0
fsck de util-linux 2.29.2
e2fsck 1.43.4 (31-Jan-2017)
ext2fs_open2(): Bad magic number in superblock
fsck.ext2: invalid superblock, trying backup blocks...
fsck.ext2: Bad magic number in super-block while trying to open /dev/md0

The superblock could not be read or does not describe a ext2/ext3/ext4 
filesystem.
If the device is invalid and it really contains an ext2/ext3/ext4 filesystem
(and not swap or ufs or something else), then the superblock is corrupt; 
and you might try running e2fsck with an alternate superblock:
     e2fsck -b 8193 <device>
  o
     e2fsck -b 32768 <device>

A gpt partition table is found in /dev/md0

Mikael suggested:

quoted

try to run fsck -n (read-only) on md0 and/or md0p1.

But you only tried
   fsck -n /dev/md0
why didn't you also try
   fsck -n /dev/md0p1
??

NeilBrown

I'm getting more escared....... No idea what to do

Thanks

quoted

Mikael Abrahamsson <mailto:swmike@swm.pp.se>
11 de octubre de 2017, 16:01
On Wed, 11 Oct 2017, Joseba Ibarra wrote:


Do you know what file system you had? Looks like next step is to try 
to run fsck -n (read-only) on md0 and/or md0p1.

What does /etc/fstab contain regarding md0?

Joseba Ibarra <mailto:wajalotnet@gmail.com>
11 de octubre de 2017, 13:56
Hi Adam

root@grafico:/mnt# cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] 
[raid4] [raid10]
md0 : inactive sdd1[3] sdb1[1] sdc1[2]
      2929889280 blocks super 1.2

unused devices: <none>


root@grafico:/mnt# mdadm --manage /dev/md0 --stop
mdadm: stopped /dev/md0


root@grafico:/mnt# mdadm --assemble /dev/md0 /dev/sd[bcd]1
mdadm: /dev/md0 assembled from 3 drives - not enough to start the 
array while not clean - consider --force.



root@grafico:/mnt# cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] 
[raid4] [raid10]
unused devices: <none>

At this point I´ve followed the advise using --force

root@grafico:/mnt# mdadm --assemble --force /dev/md0 /dev/sd[bcd]1
mdadm: Marking array /dev/md0 as 'clean'
mdadm: /dev/md0 has been started with 3 drives (out of 4).


root@grafico:/mnt# cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] 
[raid4] [raid10]
md0 : active (auto-read-only) raid5 sdb1[1] sdd1[3] sdc1[2]
      2929889280 blocks super 1.2 level 5, 512k chunk, algorithm 2 
[4/3] [_UUU]
      bitmap: 0/8 pages [0KB], 65536KB chunk

unused devices: <none>


Now I see the RAID, however can't be mounted. So, I'm not sure how to 
backup the data. Gparted shows the partition /dev/md0p1 with the used 
and free space.


If I try

mount /dev/md0 /mnt

again the output is

mount: wrong file system, bad option, bad superblock in /dev/md0, 
missing codepage or helper program, or other error

    In some cases useful info is found in syslog - try dmesg | tail or 
something like that.

I do dmesg | tail

If I try root@grafico:/mnt# mount /dev/md0p1 /mnt
mount: /dev/md0p1: can't read superblock

And


root@grafico:/mnt# dmesg | tail
[ 3263.411724] VFS: Dirty inode writeback failed for block device 
md0p1 (err=-5).
[ 3280.486813]  md0: p1
[ 3280.514024]  md0: p1
[ 3452.496811] UDF-fs: warning (device md0): udf_fill_super: No 
partition found (2)
[ 3463.731052] JBD2: Invalid checksum recovering block 630194476 in log
[ 3464.933960] Buffer I/O error on dev md0p1, logical block 630194474, 
lost async page write
[ 3464.933971] Buffer I/O error on dev md0p1, logical block 630194475, 
lost async page write
[ 3465.928066] JBD2: recovery failed
[ 3465.928070] EXT4-fs (md0p1): error loading journal
[ 3465.936852] VFS: Dirty inode writeback failed for block device 
md0p1 (err=-5).


Thanks a lot for your time


Joseba Ibarra

Adam Goryachev <mailto:adam@websitemanagers.com.au>
11 de octubre de 2017, 13:29
Hi Rudy,

Please send the output of all of the following commands:

cat /proc/mdstat

mdadm --manage /dev/md0 --stop

mdadm --assemble /dev/md0 /dev/sd[bcd]1

cat /proc/mdstat

mdadm --manage /dev/md0 --run

mdadm --manage /dev/md0 --readwrite

cat /proc/mdstat


Basically the above is just looking at what the system has done 
currently, stopping/clearing that, and then trying to assemble it 
again, finally, we try to start it, even if it has one faulty disk.

At this stage, chances look good for recovering all your data, though 
I would advise to get yourself a replacement disk for the dead one so 
that you can restore redundancy as soon as possible.

Regards,Adam





Joseba Ibarra <mailto:wajalotnet@gmail.com>
11 de octubre de 2017, 13:14
Hi Rudy

1- Yes, with all 4 disk plugged in, system does not boot
2- Yes, with the broken disk unplugged, it boots
3 - Yes, raid does not assemble during boot. I assemble manually doing

root@grafico:/home/jose# mdadm --assemble --scan /dev/md0
root@grafico:/home/jose# mdadm --assemble --scan
root@grafico:/home/jose# mdadm --assemble /dev/md0

4 -When I try to mount

  mount /dev/md0 /mnt

mount: wrong file system, bad option, bad superblock in /dev/md0, 
missing codepage or helper program, or other error

    In some cases useful info is found in syslog - try dmesg | tail or 
something like that.

I do dmesg | tail

root@grafico:/mnt# dmesg | tail
[  705.021959] md: pers->run() failed ...
[  849.719439] EXT4-fs (md0): unable to read superblock
[  849.719564] EXT4-fs (md0): unable to read superblock
[  849.719589] EXT4-fs (md0): unable to read superblock
[  849.719616] UDF-fs: error (device md0): udf_read_tagged: read 
failed, block=256, location=256
[  849.719625] UDF-fs: error (device md0): udf_read_tagged: read 
failed, block=512, location=512
[  849.719638] UDF-fs: error (device md0): udf_read_tagged: read 
failed, block=256, location=256
[  849.719642] UDF-fs: error (device md0): udf_read_tagged: read 
failed, block=512, location=512
[  849.719643] UDF-fs: warning (device md0): udf_fill_super: No 
partition found (1)
[  849.719667] isofs_fill_super: bread failed, dev=md0, iso_blknum=16, 
block=32

Thanks a lot for your helping
Rudy Zijlstra <mailto:rudy@grumpydevil.homelinux.org>
11 de octubre de 2017, 12:42
Hi Joseba,



Let me see if i understand you correctly

- with all 4 disks plugged in, your system does not boot
- with the broken disk unplugged, it boots (and from your description 
it is really broken, no DISK recovery possible unless by specialised 
company)
- raid does not get assembled during boot, you do a manual assembly?
     -> please provide the command you are using

from the log above, you should be able to do a mount of /dev/md0 which 
would auto-start the raid.

If that works, the next step would be to check the health of the other 
disks. smartctl would be your friend.
Another useful action would be to copy all important data to a backup 
before you add a new disk to replace the failed disk.

Cheers

Rudy

-- 
<http://64bits.es/>
--
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