D-Link Forums

The Graveyard - Products No Longer Supported => D-Link Storage => DNS-320 => Topic started by: ritzdank on November 22, 2013, 12:48:14 AM

Title: DNS-320 HDD crashed can't remount it again
Post by: ritzdank on November 22, 2013, 12:48:14 AM

I was using my DNS-320 with fun_plug for over 1.5years now. All fine, no RAID setup, just two harddisk in standard mode. The problem is that the second harddisk doesn't appear anymore and the D-Link web-administration tool tells me that the Volume has crashed.

After looging into it via ssh and trying to manually mount it, I get following messages through dmesg:

Code: [Select]

sd 1:0:0:0: [sdb] Result: hostbyte=0x00 driverbyte=0x08
sd 1:0:0:0: [sdb] Sense Key : 0x3 [current] [descriptor]
Descriptor sense data with sense descriptors (in hex):
        72 03 00 00 00 00 00 0c 00 0a 80 00 00 00 00 00
        00 1f e0 31
sd 1:0:0:0: [sdb] ASC=0x0 ASCQ=0x0
end_request: I/O error, dev sdb, sector 2089009
EXT3-fs: can't read group descriptor 5
cat /proc/partitions

Code: [Select]

major minor  #blocks  name

   7     0      33662 loop0
  31     0       1024 mtdblock0
  31     1       5120 mtdblock1
  31     2       5120 mtdblock2
  31     3     104448 mtdblock3
  31     4      10240 mtdblock4
  31     5       5120 mtdblock5
   9     0     530048 md0
   8     0 1953514584 sda
   8     1     530112 sda1
   8     2 1951441688 sda2
   8     4     512040 sda4
   8    16  976762584 sdb
   8    17     530112 sdb1
   8    18  974670848 sdb2
   8    20     512040 sdb4
   8    32    3956736 sdc
   8    33    3955712 sdc1
fdisk -l

Code: [Select]

Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
256 heads, 63 sectors/track, 121126 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1               1  1953525167   976762583+  ee  GPT
Is there any recovery tool? What happened to the disk?

Title: Re: DNS-320 HDD crashed can't remount it again
Post by: JavaLawyer on November 22, 2013, 05:37:19 AM

Did you run the S.M.A.R.T. test through the DNS-320 web UI?

Depending on the physical and logical state of the HDD, a solution in the following post is something to consider: DNS-320 - Data Recovery (Windows PCs) (http://forums.dlink.com/index.php?topic=41400.0)

Title: Re: DNS-320 HDD crashed can't remount it again
Post by: ivan on November 22, 2013, 08:07:50 AM

It appears that your disk crashed and some sectors became unreadable.

Download the disk manufacturers disk test tools and install them on your PC. Put the disk in a USB/SATA caddy or adapter, plug it into the PC - if it is a windows PC don't let windows do anything to the disk - then run the disk tools with the USB drive as the drive to test.

This will give you all the information about the health of the disk and if you need to replace it or not.  If the disk passes the disk tests it appears that you will need to reformat it in the NAS and restore your data from your backup.  If you have no backup you might be able to get at some of your data using the tools mentioned in JavaLawyer's post.

Title: Re: DNS-320 HDD crashed can't remount it again
Post by: ritzdank on November 23, 2013, 10:16:16 AM

I took out the harddisk and tried to mount it on a live ubuntu system. There are apparently some bad sectors on the disk. Right now, I am not able to even mount it in the Linux system

I've tried to recover the superblock but to no avail!

Code: [Select]

e2fsck -C0 -p -f -v /dev/sdc2
e2fsck: Bad magic number in super-block while trying to open /dev/sdc2
/dev/sdc1:
The superblock could not be read or does not describe a correct ext2
filesystem.  If the device is valid and it really contains an ext2
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>

tried with several -b parameters, no success!

Code: [Select]

dumpe2fs /dev/sdc2 | grep -i superblock
dumpe2fs 1.42 (29-Nov-2011)
dumpe2fs: Bad magic number in super-block while trying to open /dev/sdc2
Couldn't find valid filesystem superblock.
Unfortunately, I wasn't smart enough to immediately make a disk dump to another drive, because that's what happened a little while later:

Code: [Select]

dd if=/dev/sdc of=/dev/sdb bs=32M
dd: reading `/dev/sdc': Input/output error
31+1 records in
31+1 records out
So, currently I am running ddrescue

Code: [Select]

ddrescue -n --force /dev/sdc /dev/sdb rescue.log

Press Ctrl-C to interrupt
Initial status (read from logfile)
rescued:         0 B,  errsize:       0 B,  errors:       0
Current status
rescued:    50462 MB,  errsize:   40960 B,  current rate:     109 MB/s
   ipos:    50462 MB,   errors:       1,    average rate:     103 MB/s
   opos:    50462 MB,     time from last successful read:       0 s
Copying non-tried blocks...
And hope to work further once it's copied.

What really confuses me is, that even though I had this disk as a standard volume in my DNS-320, it appears as Linux-raid (had this error before the disk slowly died away: mount: unknow filesystem type 'linux_raid_member'.)

Is there any other path/strategy I should follow considering that the Ext3 filesystem was part of a RAID? Tools like mdadm, i. e.??

Thanks to everyone who can provide some help!

Title: Re: DNS-320 HDD crashed can't remount it again
Post by: ritzdank on November 23, 2013, 01:03:32 PM

ddrescue should save my day: https://www.gnu.org/software/ddrescue/manual/ddrescue_manual.html

It is still copying, but detected only 4Mb of errors. Rest got copied to new drive.