11

Final Update:

So it turns out that the drive was indeed defective and I was able to get it replaced - the replacement works great!

What happens:

  1. I insert the flash drive.
  2. I try to copy 2GB of files.
  3. About 600MB (but not always the same amount) in to the transfer, it aborts and forcefully dismounts the drive.
  4. Plugging it in again mounts it as read only due to filesystem errors.

Information in /var/log/syslog:

Apr  4 20:19:09 my-laptop kernel: [10720.130076] usb 1-4: new high speed USB device using ehci_hcd and address 2
Apr  4 20:19:09 my-laptop kernel: [10720.284655] scsi6 : usb-storage 1-4:1.0
Apr  4 20:19:10 my-laptop kernel: [10721.280796] scsi 6:0:0:0: Direct-Access     Corsair  Survivor GTR     0.00 PQ: 0 ANSI: 2
Apr  4 20:19:10 my-laptop kernel: [10721.281920] sd 6:0:0:0: Attached scsi generic sg3 type 0
Apr  4 20:19:10 my-laptop kernel: [10721.284361] sd 6:0:0:0: [sdc] 63438848 512-byte logical blocks: (32.4 GB/30.2 GiB)
Apr  4 20:19:10 my-laptop kernel: [10721.284863] sd 6:0:0:0: [sdc] Write Protect is off
Apr  4 20:19:10 my-laptop kernel: [10721.284870] sd 6:0:0:0: [sdc] Mode Sense: 00 00 00 00
Apr  4 20:19:10 my-laptop kernel: [10721.284876] sd 6:0:0:0: [sdc] Assuming drive cache: write through
Apr  4 20:19:10 my-laptop kernel: [10721.287862] sd 6:0:0:0: [sdc] Assuming drive cache: write through
Apr  4 20:19:10 my-laptop kernel: [10721.287873]  sdc: sdc1
Apr  4 20:19:10 my-laptop kernel: [10721.301238] sd 6:0:0:0: [sdc] Assuming drive cache: write through
Apr  4 20:19:10 my-laptop kernel: [10721.301243] sd 6:0:0:0: [sdc] Attached SCSI removable disk
Apr  4 20:20:03 my-laptop kernel: [10774.494709] FAT: Filesystem error (dev sdc1)
Apr  4 20:20:03 my-laptop kernel: [10774.494714]     fat_free_clusters: deleting FAT entry beyond EOF
Apr  4 20:20:03 my-laptop kernel: [10774.494718] FAT: Filesystem has been set read-only
Apr  4 20:22:04 my-laptop kernel: [10895.160070] usb 1-4: reset high speed USB device using ehci_hcd and address 2
Apr  4 20:22:04 my-laptop kernel: [10895.700029] usb 1-4: device not accepting address 2, error -71
Apr  4 20:22:04 my-laptop kernel: [10895.820047] usb 1-4: reset high speed USB device using ehci_hcd and address 2
Apr  4 20:22:04 my-laptop kernel: [10895.950038] usb 1-4: device descriptor read/64, error -71
Apr  4 20:22:05 my-laptop kernel: [10896.191050] usb 1-4: device descriptor read/64, error -71
Apr  4 20:22:05 my-laptop kernel: [10896.420054] usb 1-4: reset high speed USB device using ehci_hcd and address 2
Apr  4 20:22:05 my-laptop kernel: [10896.840042] usb 1-4: device not accepting address 2, error -71
Apr  4 20:22:05 my-laptop kernel: [10896.962678] usb 1-4: reset high speed USB device using ehci_hcd and address 2
Apr  4 20:22:06 my-laptop kernel: [10897.380047] usb 1-4: device not accepting address 2, error -71
Apr  4 20:22:06 my-laptop kernel: [10897.380095] usb 1-4: USB disconnect, address 2
Apr  4 20:22:06 my-laptop kernel: [10897.380122] sd 6:0:0:0: [sdc] Unhandled error code
Apr  4 20:22:06 my-laptop kernel: [10897.380125] sd 6:0:0:0: [sdc] Result: hostbyte=DID_ERROR driverbyte=DRIVER_OK
Apr  4 20:22:06 my-laptop kernel: [10897.380130] sd 6:0:0:0: [sdc] CDB: Write(10): 2a 00 00 12 9e 00 00 00 f0 00
Apr  4 20:22:06 my-laptop kernel: [10897.380141] end_request: I/O error, dev sdc, sector 1220096
Apr  4 20:22:06 my-laptop kernel: [10897.382530] sd 6:0:0:0: [sdc] Unhandled error code
Apr  4 20:22:06 my-laptop kernel: [10897.382534] sd 6:0:0:0: [sdc] Result: hostbyte=DID_NO_CONNECT driverbyte=DRIVER_OK
Apr  4 20:22:06 my-laptop kernel: [10897.382539] sd 6:0:0:0: [sdc] CDB: Write(10): 2a 00 00 12 9e f0 00 00 f0 00
Apr  4 20:22:06 my-laptop kernel: [10897.382549] end_request: I/O error, dev sdc, sector 1220336
Apr  4 20:22:06 my-laptop kernel: [10897.385121] FAT: FAT read failed (blocknr 329)
Apr  4 20:22:06 my-laptop kernel: [10897.385150] FAT: FAT read failed (blocknr 322)
Apr  4 20:22:06 my-laptop kernel: [10897.385173] FAT: FAT read failed (blocknr 329)
Apr  4 20:22:06 my-laptop kernel: [10897.385188] FAT: FAT read failed (blocknr 322)
Apr  4 20:22:06 my-laptop kernel: [10897.390581] FAT: unable to read inode block for updating (i_pos 496133)
Apr  4 20:22:06 my-laptop kernel: [10897.830055] usb 1-4: new high speed USB device using ehci_hcd and address 3
Apr  4 20:22:06 my-laptop kernel: [10897.962671] usb 1-4: device descriptor read/64, error -71
Apr  4 20:22:07 my-laptop kernel: [10898.200060] usb 1-4: device descriptor read/64, error -71
Apr  4 20:22:07 my-laptop kernel: [10898.432548] usb 1-4: new high speed USB device using ehci_hcd and address 4
Apr  4 20:22:07 my-laptop kernel: [10898.560068] usb 1-4: device descriptor read/64, error -71
Apr  4 20:22:07 my-laptop kernel: [10898.802603] usb 1-4: device descriptor read/64, error -71
Apr  4 20:22:07 my-laptop kernel: [10899.030075] usb 1-4: new high speed USB device using ehci_hcd and address 5
Apr  4 20:22:08 my-laptop kernel: [10899.460032] usb 1-4: device not accepting address 5, error -71
Apr  4 20:22:08 my-laptop kernel: [10899.580064] usb 1-4: new high speed USB device using ehci_hcd and address 6
Apr  4 20:22:08 my-laptop kernel: [10900.000071] usb 1-4: device not accepting address 6, error -71
Apr  4 20:22:08 my-laptop kernel: [10900.000101] hub 1-0:1.0: unable to enumerate USB device on port 4
Apr  4 20:22:09 my-laptop kernel: [10900.322588] usb 4-2: new full speed USB device using uhci_hcd and address 2
Apr  4 20:22:09 my-laptop kernel: [10900.466510] usb 4-2: not running at top speed; connect to a high speed hub
Apr  4 20:22:19 my-laptop kernel: [10910.471597] usb 4-2: string descriptor 0 read error: -110
Apr  4 20:22:24 my-laptop kernel: [10915.471643] usb 4-2: can't set config #1, error -110

Other steps taken:

  • I tried plugging it into different USB ports.
  • I tried plugging it into one of the root ports on the PC.
  • I ran badblocks on the device and it came up clean.
  • I tried wiping the partition table before formatting - no difference.
  • I have an external USB 2.0 1TB hard drive that runs fine and can perform large transfers without incident.
  • smartctl -T normal /dev/sdX fails, saying "Terminate command early due to bad response to IEC mode page. A mandatory SMART command failed: exiting."

Note: thanks to djeikyb for helping me with some of the steps.

11
  • 6
    Some drives are fake. If you got it "new" for a price that was too good to be true, it was. A lot of "new" cheap large USB drives on "that auction site" have had problems. The fake drives are drives, but turn out to be older, smaller drives -- not as described. On the other hand, that last message about not running at top speed is curious. Is the drive plugged into a modern hub? As I don't have a solution, I'll leave this as a starting point for others to chime in.
    – Paul
    Mar 21, 2011 at 8:22
  • @George can you give us some details about the actual drive? Make and model, partitioning, etc?
    – Oli
    Mar 21, 2011 at 11:58
  • @Oli: It's 32GB and the entire thing is formatted as FAT32. Also, I think I tried badblocks on it and it was fine. Mar 21, 2011 at 16:30
  • @Paul: As for the hub... at first I was using a (USB 2) hub, but then I tried plugging it directly into the PC... same thing. Mar 21, 2011 at 16:31
  • 1
    @ali: When I tried a write benchmark with Ubuntu's Disk Utility, it consistently failed with an "I/O error" on the same location. Apr 6, 2011 at 3:00

7 Answers 7

13
+100

Pretty sure the drive is broken.

  1. Is this a file system problem? No.

  2. Is this a usb port/bus problem? No.

  3. Is this a usb driver problem? No.

    • Same problems in Windows Vista.
    • !!! It would be helpful to know from the driver's perspective if there is a difference between a usb flash disk and a usb sata disk.
  4. Is the flash drive broken? Likely.

    Evidence in favour:

    • I/O errors in /var/log/syslog
    • I/O errors happen at a consistent location.
    • The above steps.
5
  • That information should be part of the question. ;)
    – htorque
    Apr 5, 2011 at 7:33
  • @hto: It is now. Apr 5, 2011 at 17:42
  • Based on the IO error in the syslog, it sure looks like the drive is busted. Double check that you are using badblocks properly; it should get the same error. Smartctl does not work on flash sticks either.
    – psusi
    Apr 5, 2011 at 18:36
  • @psusi: I was just running 'badblocks /dev/sdc'... do I need to pass other parameters? Apr 6, 2011 at 3:02
  • @George Edison try dd if=/dev/sdc of=/dev/null bs=512 iflag=direct
    – psusi
    Apr 6, 2011 at 14:03
6

Try the drive on a windows machine.

If it has the same problem on windows then this is not a ubuntu problem and is likely a hardware or firmware fault.

In which case I would send it back.

1
  • 1
    It has the same problem in Windows. Apr 6, 2011 at 3:01
5

To me the following errors are the important ones.

[10897.385188] FAT: FAT read failed (blocknr 322) same error on blocknr 329

It shows me that their is either a failure of the device, or a corrupt FAT(file allocation table). Based on further information I would say the the actual hardware is at fault. This wiki section shows you the components of a flash drive. I would say that the flash memory controller or flash memory chip are at fault. If you can return the drive do so. If you still have the original packaging and warranty you should be able to get a replacement.

To get further information run the following to test for bad block and report verbosely put it into a file (badblocks.txt).

sudo fsck.vfat -Vtv /dev/[your_usb_dev] > badblocks.txt # was sdc in the log file

paste the badblocks.txt into your question so we can see. That command will mark all the bad block as unaccessible in the FAT and should stop the OS trying to write to those failed blocks. This should fix your problem.

To find your disk, that you need to put into the {/dev/[your_usb_dev]} run the command bellow. Please paste the contents of my_partitions.txt as well so we can see some more detail about your drive.

sudo fdisk -l > my_partitions.txt

One last note. 32bit FAT has a file limit of 4 GiB. Look right at the end of that section in the wiki article. So if you file is bigger then 4Gib you get a not enough space error. I don't think that is the problem here though.

4
  • I'll try the commands. Also, the files I was copying were no larger than 700MB or so (they were ISO images). Apr 6, 2011 at 3:03
  • I get an error when running the first command: "-t and -w require -a or -r" Apr 6, 2011 at 3:06
  • @george-edison I would then say the the flash drive is faulty at a hard ware level. I recently received a 16Gb drive from friend who traveled to Tai Land for a shopping trip. Needless to say the drive is actually 4GB. It reports being 16Gb, it will let you copy that number of files but it show up corrupt and unaccessible when I try to retrieve the files.
    – nelaaro
    Apr 6, 2011 at 6:18
  • mike-s, if you ever look back here, I removed your paragraph. (1) linux boxen are personal computers, and have supported files >2.15gb for ages (2) microsoft says fat32 supports ~4gb files.
    – djeikyb
    Oct 12, 2013 at 5:50
1

Am pretty sure you have now a fake usb. You just mention all the symptoms and the linux ouput looks very similar to the one it shows me for a fake 16mb and 128mb usb drives i have here. Am sorry to say but you got ripped off. There has been many cases, not just corsair but any other pen drive company, even extreme examples like this:

CASES THAT HAVE ALREADY HAPPEN -

Big Box of nothing - http://forum.corsair.com/forums/showthread.php?s=826f662d847dd2e8283552f17c161de5&t=94466 .

I went to the corsair site and seen your post here about it http://forum.corsair.com/forums/showthread.php?s=826f662d847dd2e8283552f17c161de5&t=94466

You can find some cases like the following: http://hardforum.com/showthread.php?t=1397409 where they mention the fake scam with some corsairs

A friend in Costa Rica bought this one: http://wn.com/PEN_DRIVE_64GB_Kingston_FAKE!!!! and as you can see in the video he was ripped off. We watch as he tried to copy anything beyong the 4GB that was the real size with no luck.

This has been happening for a long time now, see this link for 2006: http://linuxrevolution.blogspot.com/2006/08/fake-usb-drives-sleuthing.html

FAKE CHECK SOFTWARE

Here are ways to test if you have a FAKE pen drive: http://www.linuxquestions.org/questions/linux-hardware-18/fake-usb-stick-testing-tool-786415/

Another is this one: http://fixfakeflash.wordpress.com/2010/08/20/linux-h2testw-alternative-program-called-f3-by-michel%C2%A0machado/ To test for FAKE pen drives.

The site for F3 Tool: http://oss.digirati.com.br/f3/ where you can compile the program with gcc. Very simple tool that verifies the size of the flash drives. IF you read the webpage you will see this user ALSO had an experience as you and me.

In any case you might have a real 512MB to 1GB flash drive that is faked as a 32GB one.

I know you might not give a correct answer here to anyone since this are really bad news. I feel you since i bought one 32gb also, a friend a 128mb and my girlfriend a 16gb. all corsair or kingston. So please review all the info in this page with an open mind. Good luck friend.

RECOMMENDATION - For my experience i recommend to run the programs i mention above and when buying a flash drive test them the minute you get your hands on it. A quick test is to fill the flash drive full of files up to its maximun said size.

2
  • I kind of doubt that it was fake (it came with proper packaging and accessories that were identical to the proper model) - but I won't rule it out. One thing to note - the error was usually around 2 - 3 GB into the drive, so the actual capacity would have to be >3GB. Jun 3, 2011 at 5:58
  • @George Thank you. I also think the same thing. Jun 3, 2011 at 13:48
1

Remember that the FAT32 is a very delicate filesystem and it does not like the electricity being switched off when it is doing any work. So you have to always safely remove the drive from USB port. If you have removed it unsafely at least once in the past - then there may be something broken in the FAT structure - then run dosfsck before using it again.

Why does it unmount on error? Well it probably depends on default configuration - it just tries to remount as read-only so you can do backup of current data and then do all the checks. I suppose it works as if errors=remount-ro option is set.

I saw a similar problem here on linuxquestions.org.

1

Are you copying files to the USB drive's root directory that already has a lot of other files in it?

It's a limitation of FAT to not allow more than a certain number of files stored in its root directory (subdirectories can contain more files).

If that is the case, try copying the files to a subdirectory on the USB drive.

3
-1

If the file system of your USB is fat 32 or 16 Maximum file size of 4 GB (minus 1 byte) for fat32 for fat 16 less. try to format your USB to NTFS file system.

FAT16, FAT32, NTFS, or exFAT on USB Flash Drives?

7
  • 4
    Copying files on any file system does not unmount a drive. This is not a file system issue.
    – Takkat
    Mar 21, 2011 at 8:48
  • 1
    For ubuntu, reiserfs, extfs and such are far more appropriate (users, permissions). Mar 21, 2011 at 11:12
  • Well first of all, none of the files are over 10MB. Secondly, it needs to be accessible on Windows machines as well as a Mac perhaps. Mar 21, 2011 at 16:32
  • I'd use ntfs in that case installing necessary compatibility apps/packets on linux and mac os x machines
    – Pitto
    Apr 5, 2011 at 9:12
  • if you try and reset all of the info by using the startup disk creator, does it unmount then? Because if it's a fake then it should be shown in the file size right?
    – Alvar
    Apr 5, 2011 at 10:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .