LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Hardware (http://www.linuxquestions.org/questions/linux-hardware-18/)
-   -   External USB disk: repeated file system corruption (http://www.linuxquestions.org/questions/linux-hardware-18/external-usb-disk-repeated-file-system-corruption-576490/)

Sven Sandberg 08-12-2007 04:58 AM

External USB disk: repeated file system corruption
 
I have a new external USB disk with an ext3 file system. It appears that the file system is getting corrupted about every second time I reboot my laptop. It has happened twice that the file system was corrupted, and both times I repaired with with e2fsck (losing some files). The first time, I had to run e2fsck 5 times before all errors were gone. The second time, I had to run e2fsck 8 times. After the last 8 runs, I rescued a few of the files in /lost+found, unmounted the disk, unplugged it, switched off the laptop, and booted up again. When I tried to mount the disk, I got the following error again:

Code:

# mount /dev/sdb1 /media/x/
mount: wrong fs type, bad option, bad superblock on /dev/sdb1,
      missing codepage or other error
      In some cases useful info is found in syslog - try
      dmesg | tail  or so

So I ran:
Code:

# dmesg | tail -n 27
[  47.808000] eth0: no IPv6 routers present
[  133.112000] usb 4-1: new high speed USB device using ehci_hcd and address 2
[  133.244000] usb 4-1: configuration #1 chosen from 1 choice
[  133.484000] usbcore: registered new interface driver libusual
[  133.500000] Initializing USB Mass Storage driver...
[  133.504000] scsi2 : SCSI emulation for USB Mass Storage devices
[  133.504000] usbcore: registered new interface driver usb-storage
[  133.504000] USB Mass Storage support registered.
[  133.504000] usb-storage: device found at 2
[  133.504000] usb-storage: waiting for device to settle before scanning
[  139.504000] usb-storage: device scan complete
[  144.032000] scsi 2:0:0:0: Direct-Access    SAMSUNG  HD501LJ          0-10 PQ: 0 ANSI: 2 CCS
[  144.040000] SCSI device sdb: 976773168 512-byte hdwr sectors (500108 MB)
[  144.040000] sdb: Write Protect is off
[  144.040000] sdb: Mode Sense: 00 38 00 00
[  144.040000] sdb: assuming drive cache: write through
[  144.044000] SCSI device sdb: 976773168 512-byte hdwr sectors (500108 MB)
[  144.044000] sdb: Write Protect is off
[  144.044000] sdb: Mode Sense: 00 38 00 00
[  144.044000] sdb: assuming drive cache: write through
[  144.044000]  sdb: sdb1
[  144.052000] sd 2:0:0:0: Attached scsi disk sdb
[  144.052000] sd 2:0:0:0: Attached scsi generic sg2 type 0
[  149.496000] EXT3-fs error (device sdb1): ext3_check_descriptors: Block bitmap for group 790 not in group (block 3758096384)!
[  149.504000] EXT3-fs: group descriptors corrupted!
[  208.312000] EXT3-fs error (device sdb1): ext3_check_descriptors: Block bitmap for group 790 not in group (block 3758096384)!
[  208.316000] EXT3-fs: group descriptors corrupted!

Some more info:
Code:

# uname -r
2.6.20-16-generic

I have fresh new Ubuntu 7.04. My disk is a new Samsung Xtend AdStore XAS3UI500 with 500 GB and USB2. My laptop is a Compaq nx7010 and it supports USB2. `e2fsck -c -c` tells me that there is no physical damage.

A couple of thoughts:
  • Can there be data lying around in some cache that gets lost if I unplug the disk too soon after unmounting?
  • Should I avoid unplugging the disk when the computer is on?
  • (I first thought there could be a problem with putting the computer in suspend mode with the disk mounted. However, I did not suspend before the last file system corruption.)

Why is this happening? I have no clue what the error may be. Any help would be greatly appreciated.

Sven Sandberg 08-12-2007 05:02 AM

More info
 
I uploaded the output of my 13 runs of e2fsck here (zipped, 190 kB). An output of dumpe2fs from when the disk was clean is here (zipped, 147 kB) and a current output from dumpe2fs is here (zipped, 100 kB) (the latter fails prematurely with an error).


All times are GMT -5. The time now is 09:45 AM.