Find the answer to your Linux question:
Page 1 of 2 1 2 LastLast
Results 1 to 10 of 14
I have a 2TB WD MyBook 1110 plugged into my PC. Code: Bus 003 Device 006: ID 1058:1110 Western Digital Technologies, Inc. Every PC in the house accesses it through ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
  1. #1
    Linux User
    Join Date
    Dec 2011
    Location
    Turtle Island West
    Posts
    425

    2TB USB drive goes Read-Only occasionally.


    I have a 2TB WD MyBook 1110 plugged into my PC.
    Code:
    Bus 003 Device 006: ID 1058:1110 Western Digital Technologies, Inc.
    Every PC in the house accesses it through NFS and it gets used quite heavily.

    About twice a year something happens and the kernel remounts it read-only. I have to shut down nfs, umount it, and fsck it. This can take half a day on a drive of this size, especially when there's a ton of errors.

    /var/log/messages:
    Code:
    Mar 12 12:54:06 timberwolf kernel: SCSI error : <0 0 0 0> return code = 0x8000002
    Mar 12 12:54:06 timberwolf kernel: Current sda: sense key Hardware Error
    Mar 12 12:54:06 timberwolf kernel: Additional sense: Internal target failure
    Mar 12 12:54:06 timberwolf kernel: end_request: I/O error, dev sda, sector 113246271
    Mar 12 12:54:06 timberwolf kernel: EXT3-fs error (device sda1): read_block_bitmap: Cannot read block bitmap - block_group = 432, block_bitmap = 1
    4155776
    Mar 12 12:54:06 timberwolf kernel: Aborting journal on device sda1.
    Mar 12 12:54:08 timberwolf kernel: ext3_abort called.
    Mar 12 12:54:08 timberwolf kernel: EXT3-fs abort (device sda1): ext3_journal_start: Detected aborted journal
    Mar 12 12:54:08 timberwolf kernel: Remounting filesystem read-only
    Mar 12 12:54:28 timberwolf kernel: EXT3-fs error (device sda1) in ext3_prepare_write: IO failure
    Mar 12 12:54:28 timberwolf kernel: EXT3-fs error (device sda1) in start_transaction: Journal has aborted
    Mar 12 12:54:28 timberwolf last message repeated 7 times
    Same stuff in syslog.

    12:54 was when it got remounted RO, but it was behaving strangely a few hours before that. I was going through all my youtube crap and deleting most of it. All of a sudden it started taking a full minute to delete 2 files, when normally it would take less than a second.

    In /var/log/kernel/warnings we have this:
    Code:
    Mar 12 05:04:14 timberwolf kernel: SCSI error : <0 0 0 0> return code = 0x8000002
    Mar 12 05:04:14 timberwolf kernel: Current sda: sense key Hardware Error
    Mar 12 05:04:14 timberwolf kernel: Additional sense: Internal target failure
    Mar 12 05:04:14 timberwolf kernel: end_request: I/O error, dev sda, sector 123528519
    So obviously something was starting to go wrong at 5:04 in the morning.

    I can set up a cron job to watch /var/log/kernel/warnings for this kind of message, but what I'd really like is for a rude warning dialog to pop up on every workspace the instant this gets detected.

    Is there some fairly simple way to accomplish this?

    Peace and Cheer.

  2. #2
    Linux Newbie reginaldperrin's Avatar
    Join Date
    Oct 2010
    Location
    Christchurch, New Zealand
    Posts
    123
    I would suspect a possible hardware issue.
    The drive itself has controllers built into the external housing. These have a finite lifetime. Check the heat in the drive also.
    And, by its very nature, USB is inherently unreliable (even though it is pretty good most of the time).
    It's designed to be plugged and unplugged, and this means eventual physical damage.
    If the drive is plugged in and mounted, then it is probably in an always-on state, and in operation most the time. This will shorten its life as well.
    If it is a WD 'Green' drive, they have a significantly shorter lifespan than the more expensive ones.
    Could simply be that the drive is reaching the end of its life. Can you access the drives SMART data output? This mat give you a better idea about the drive.

    All that aside, you mention 12:54 am and 5:04 am. Was there some sort of electrical event at those times? You may want to consider investing in an inexpensive UPS that the drive plugs into, possibly also for the PC itself. This will smooth out electrical fluctuations. It will also eliminate these things as causes (or fix it, if it was the cause).

    Hope this helps.

  3. #3
    Linux User
    Join Date
    Dec 2011
    Location
    Turtle Island West
    Posts
    425
    Quote Originally Posted by reginaldperrin View Post
    I would suspect a possible hardware issue.
    The drive itself has controllers built into the external housing. These have a finite lifetime. Check the heat in the drive also.
    And, by its very nature, USB is inherently unreliable (even though it is pretty good most of the time).
    It's designed to be plugged and unplugged, and this means eventual physical damage.
    If the drive is plugged in and mounted, then it is probably in an always-on state, and in operation most the time. This will shorten its life as well.
    If it is a WD 'Green' drive, they have a significantly shorter lifespan than the more expensive ones.
    Could simply be that the drive is reaching the end of its life.
    I suspect you are correct. It took 14 hours for fsck to get to block 1300000, and that's out of about 2 billion blocks. I did the math. That would mean it would take 3.2 years to finish fsck'ing. I think it is getting to the end of it's life.

    Why do USB drives fail faster than normal HDs? It's sitting out and well ventilated. It's not hot. It's only about 5 years old. I have perfect working HDs that are more than 12 years old, still going strong, 24/7.
    Can you access the drives SMART data output? This mat give you a better idea about the drive.
    I don't know. How would I do this?
    All that aside, you mention 12:54 am and 5:04 am. Was there some sort of electrical event at those times? You may want to consider investing in an inexpensive UPS that the drive plugs into, possibly also for the PC itself. This will smooth out electrical fluctuations. It will also eliminate these things as causes (or fix it, if it was the cause).
    5:04 in the morning was when the first warnings appeared. It didn't get remounted RO until 12:54 in the afternoon. That's when I clued in something was wrong.

    I should buy a UPS. But I figured 6 miles of extension cords and power bars in a tangled snake pit would smooth out the juice

    After about 24 hours of playing with it, it's getting worse, not better. Even mounted RO it takes forever to read the contents of a directory, let alone copy 16000 tracks off it. Thank god my source code isn't on the thing.

  4. $spacer_open
    $spacer_close
  5. #4
    Linux Newbie reginaldperrin's Avatar
    Join Date
    Oct 2010
    Location
    Christchurch, New Zealand
    Posts
    123
    Quote Originally Posted by Miven View Post
    I don't know. How would I do this?
    Looks like smartmontools is a reasonable utility.
    Get the disk health status with SMART monitor tools on Debian and Ubuntu Linux FAQforge

    Might not be able to access usb drives though. Only one way to find out for sure...

  6. #5
    Linux User
    Join Date
    Dec 2011
    Location
    Turtle Island West
    Posts
    425
    Thanks for that link. I downloaded and compiled it. Here's the results:
    Code:
    [root@timberwolf miven]# smartctl -i /dev/sda
    smartctl 6.2 2013-07-26 r3841 [i686-linux-2.6.6] (local build)
    Copyright (C) 2002-13, Bruce Allen, Christian Franke, www.smartmontools.org
    
    === START OF INFORMATION SECTION ===
    Vendor:               WD
    Product:              My Book 1110
    Revision:             2003
    User Capacity:        1,999,696,297,984 bytes [1.99 TB]
    Logical block size:   512 bytes
    Device type:          disk
    Local Time is:        Tue Mar 18 12:03:01 2014 PDT
    SMART support is:     Unavailable - device lacks SMART capability.
    And that's about as fancy as this drive gets. I tried a few other options, nothing reveals anything interesting. Bummer.

  7. #6
    Linux Newbie reginaldperrin's Avatar
    Join Date
    Oct 2010
    Location
    Christchurch, New Zealand
    Posts
    123
    What a shame that the drive "lacks SMART capability". This probably also points somewhat to its age.
    If you have bad sector issues, then the drive is probably at the end of its life. A way to check this is to remove it from its external housing.
    Unless you can be very careful, this will probably mean the end of the housing. Nevertheless, your data is probably more valuable!
    Install the drive in a computer, and run a disk checker on it. I see you have run fsck on it. "It took 14 hours for fsck to get to block 1300000"... this may be because of the USB connection, rather than because of damage.
    I have used Spinrite a few times, and while nothing can fix damaged sectors, this program certainly does more to recover data from failing or near failing drives than almost anything else.
    You should run it only on drives connected directly, not via a USB connection, hence the need to remove it from the housing.
    If the drive is functioning well enough simply to copy all the data off it, then you should probably invest in a new backup drive asap, and use "ddrescue".
    There's a nice page here about it: Ddrescue - Forensics Wiki
    Hope this helps

  8. #7
    Linux Engineer
    Join Date
    Dec 2013
    Posts
    1,486
    Maybe this will help. Disassembling a Western Digital My Book

    I plan on doing this to a networked wd mybook.

  9. #8
    Linux Engineer docbop's Avatar
    Join Date
    Nov 2009
    Location
    Woodshed, CA
    Posts
    947
    Thinking out loud don't some of these drive go into Sleep mode and spin down, wonder if that could be it? Probably is configuable on the WD interface.

  10. #9
    Linux Engineer
    Join Date
    Dec 2013
    Posts
    1,486
    Quote Originally Posted by docbop View Post
    Thinking out loud don't some of these drive go into Sleep mode and spin down, wonder if that could be it? Probably is configuable on the WD interface.
    Mine does. PITA.

  11. #10
    Linux User
    Join Date
    Dec 2011
    Location
    Turtle Island West
    Posts
    425
    Quote Originally Posted by gregm View Post
    Thanks for the link. It looks simple enough. Right now I'm just keeping it RO until I can get a replacement. Does anyone know a good brand that has 2TB model? I thought WD has been really good to me over the years until this drive. Seagate? SSDs are still too small and expensive. Maybe better to buy an external housing and get a drive to fit?

Page 1 of 2 1 2 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •