[gentoo-user] USB drive dead? (Commands to check?)

[gentoo-user] USB drive dead? (Commands to check?)

[Mark Knecht]

Hi,
   I seem to have lost an external USB drive I've been using on my
MythTV backend server for video storage. What commands can I try to
get it to wake up or at least show me what's working and what isn't?
The drive shows under fdisk /dev/sda. I can see the large partition,
and it seems to be the right size, according to fdisk anyway, but I
cannot mount it using mount, and so far I cannot get e2fsck to do
anything. Both of these fail:


mount -t ext2 /dev/sda1 /video
mount -t ext3 /dev/sda1 /video

   I have power cycled the drive and I've rebooted the MacMini. Nothing changed.

Thanks in advance,
Mark

MacMini ~ # fdisk /dev/sda

The number of cylinders for this disk is set to 19929.
There is nothing wrong with that, but this is larger than 1024,
and could in certain setups cause problems with:
1) software that runs at boot time (e.g., old versions of LILO)
2) booting and partitioning software from other OSs
   (e.g., DOS FDISK, OS/2 FDISK)

Command (m for help): p

Disk /dev/sda: 163.9 GB, 163928604672 bytes
255 heads, 63 sectors/track, 19929 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xa9b5c6b5

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1       19929   160079661   83  Linux

Command (m for help):



MacMini ~ # df
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/hda4             75890040  37169416  34865560  52% /
udev                     10240       156     10084   2% /dev
shm                     257396         0    257396   0% /dev/shm
MacMini ~ # fsck -t ext3 /dev/sda
fsck 1.41.3 (12-Oct-2008)
e2fsck 1.41.3 (12-Oct-2008)
fsck.ext3: Superblock invalid, trying backup blocks...
fsck.ext3: Bad magic number in super-block while trying to open /dev/sda

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>

MacMini ~ # e2fsck -b 8193 /dev/sda
e2fsck 1.41.3 (12-Oct-2008)
e2fsck: Bad magic number in super-block while trying to open /dev/sda

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>

MacMini ~ # fsck -t ext3 /dev/sda1
fsck 1.41.3 (12-Oct-2008)
e2fsck 1.41.3 (12-Oct-2008)
fsck.ext3: Attempt to read block from filesystem resulted in short
read while tr            ying to open /dev/sda1
Could this be a zero-length partition?
MacMini ~ # e2fsck -b 8193 /dev/sda1
e2fsck 1.41.3 (12-Oct-2008)
e2fsck: Bad magic number in super-block while trying to open /dev/sda1

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>

MacMini ~ # e2fsck -b 8193 /dev/sda
e2fsck 1.41.3 (12-Oct-2008)
e2fsck: Bad magic number in super-block while trying to open /dev/sda

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>

MacMini ~ #

[gentoo-user] Re: USB drive dead? (Commands to check?)

[Mark Knecht]

On Sat, Sep 19, 2009 at 3:21 PM, Mark Knecht <markknecht@gmail.com> wrote:
> Hi,
>   I seem to have lost an external USB drive I've been using on my
> MythTV backend server for video storage. What commands can I try to
> get it to wake up or at least show me what's working and what isn't?
> The drive shows under fdisk /dev/sda. I can see the large partition,
> and it seems to be the right size, according to fdisk anyway, but I
> cannot mount it using mount, and so far I cannot get e2fsck to do
> anything. Both of these fail:
>
>
> mount -t ext2 /dev/sda1 /video
> mount -t ext3 /dev/sda1 /video
>
>   I have power cycled the drive and I've rebooted the MacMini. Nothing changed.
>
> Thanks in advance,
> Mark
>
> MacMini ~ # fdisk /dev/sda
>
> The number of cylinders for this disk is set to 19929.
> There is nothing wrong with that, but this is larger than 1024,
> and could in certain setups cause problems with:
> 1) software that runs at boot time (e.g., old versions of LILO)
> 2) booting and partitioning software from other OSs
>   (e.g., DOS FDISK, OS/2 FDISK)
>
> Command (m for help): p
>
> Disk /dev/sda: 163.9 GB, 163928604672 bytes
> 255 heads, 63 sectors/track, 19929 cylinders
> Units = cylinders of 16065 * 512 = 8225280 bytes
> Disk identifier: 0xa9b5c6b5
>
>   Device Boot      Start         End      Blocks   Id  System
> /dev/sda1               1       19929   160079661   83  Linux
>
> Command (m for help):
>
>
>
> MacMini ~ # df
> Filesystem           1K-blocks      Used Available Use% Mounted on
> /dev/hda4             75890040  37169416  34865560  52% /
> udev                     10240       156     10084   2% /dev
> shm                     257396         0    257396   0% /dev/shm
> MacMini ~ # fsck -t ext3 /dev/sda
> fsck 1.41.3 (12-Oct-2008)
> e2fsck 1.41.3 (12-Oct-2008)
> fsck.ext3: Superblock invalid, trying backup blocks...
> fsck.ext3: Bad magic number in super-block while trying to open /dev/sda
>
> 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>
>
> MacMini ~ # e2fsck -b 8193 /dev/sda
> e2fsck 1.41.3 (12-Oct-2008)
> e2fsck: Bad magic number in super-block while trying to open /dev/sda
>
> 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>
>
> MacMini ~ # fsck -t ext3 /dev/sda1
> fsck 1.41.3 (12-Oct-2008)
> e2fsck 1.41.3 (12-Oct-2008)
> fsck.ext3: Attempt to read block from filesystem resulted in short
> read while tr            ying to open /dev/sda1
> Could this be a zero-length partition?
> MacMini ~ # e2fsck -b 8193 /dev/sda1
> e2fsck 1.41.3 (12-Oct-2008)
> e2fsck: Bad magic number in super-block while trying to open /dev/sda1
>
> 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>
>
> MacMini ~ # e2fsck -b 8193 /dev/sda
> e2fsck 1.41.3 (12-Oct-2008)
> e2fsck: Bad magic number in super-block while trying to open /dev/sda
>
> 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>
>
> MacMini ~ #
>

A little more info:

Buffer I/O error on device sda, logical block 12
sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
end_request: I/O error, dev sda, sector 96
Buffer I/O error on device sda, logical block 12
sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
end_request: I/O error, dev sda, sector 96
Buffer I/O error on device sda, logical block 12
sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
end_request: I/O error, dev sda, sector 96

MacMini ~ # fsck -t ext3 /dev/sda1
fsck 1.41.3 (12-Oct-2008)
e2fsck 1.41.3 (12-Oct-2008)
fsck.ext3: Attempt to read block from filesystem resulted in short
read while trying to open /dev/sda1
Could this be a zero-length partition?
MacMini ~ #


end_request: I/O error, dev sda, sector 95
__ratelimit: 58 callbacks suppressed
Buffer I/O error on device sda1, logical block 16
Buffer I/O error on device sda1, logical block 17
Buffer I/O error on device sda1, logical block 18
Buffer I/O error on device sda1, logical block 19
Buffer I/O error on device sda1, logical block 20
Buffer I/O error on device sda1, logical block 21
Buffer I/O error on device sda1, logical block 22
Buffer I/O error on device sda1, logical block 23
Buffer I/O error on device sda1, logical block 24
Buffer I/O error on device sda1, logical block 25
sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
end_request: I/O error, dev sda, sector 95
__ratelimit: 86 callbacks suppressed
Buffer I/O error on device sda1, logical block 16
Buffer I/O error on device sda1, logical block 17
Buffer I/O error on device sda1, logical block 18
Buffer I/O error on device sda1, logical block 19
Buffer I/O error on device sda1, logical block 20
Buffer I/O error on device sda1, logical block 21
Buffer I/O error on device sda1, logical block 22
Buffer I/O error on device sda1, logical block 23
Buffer I/O error on device sda1, logical block 24
Buffer I/O error on device sda1, logical block 25
MacMini ~ #


I could try reinstalling the file system but I had a few backups on
this drive for other systems. (non-critical, but possibly useful) I'd
like to be sure I cannot recover them before I blow everything away.

Thanks,
Mark

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Volker Armin Hemmann]

On Sonntag 20 September 2009, Mark Knecht wrote:
> On Sat, Sep 19, 2009 at 3:21 PM, Mark Knecht <markknecht@gmail.com> wrote:
> > Hi,
> >   I seem to have lost an external USB drive I've been using on my
> > MythTV backend server for video storage. What commands can I try to
> > get it to wake up or at least show me what's working and what isn't?
> > The drive shows under fdisk /dev/sda. I can see the large partition,
> > and it seems to be the right size, according to fdisk anyway, but I
> > cannot mount it using mount, and so far I cannot get e2fsck to do
> > anything. Both of these fail:
> >
> >
> > mount -t ext2 /dev/sda1 /video
> > mount -t ext3 /dev/sda1 /video
> >
> >   I have power cycled the drive and I've rebooted the MacMini. Nothing
> > changed.
> >
> > Thanks in advance,
> > Mark
> >
> > MacMini ~ # fdisk /dev/sda
> >
> > The number of cylinders for this disk is set to 19929.
> > There is nothing wrong with that, but this is larger than 1024,
> > and could in certain setups cause problems with:
> > 1) software that runs at boot time (e.g., old versions of LILO)
> > 2) booting and partitioning software from other OSs
> >   (e.g., DOS FDISK, OS/2 FDISK)
> >
> > Command (m for help): p
> >
> > Disk /dev/sda: 163.9 GB, 163928604672 bytes
> > 255 heads, 63 sectors/track, 19929 cylinders
> > Units = cylinders of 16065 * 512 = 8225280 bytes
> > Disk identifier: 0xa9b5c6b5
> >
> >   Device Boot      Start         End      Blocks   Id  System
> > /dev/sda1               1       19929   160079661   83  Linux
> >
> > Command (m for help):
> >
> >
> >
> > MacMini ~ # df
> > Filesystem           1K-blocks      Used Available Use% Mounted on
> > /dev/hda4             75890040  37169416  34865560  52% /
> > udev                     10240       156     10084   2% /dev
> > shm                     257396         0    257396   0% /dev/shm
> > MacMini ~ # fsck -t ext3 /dev/sda
> > fsck 1.41.3 (12-Oct-2008)
> > e2fsck 1.41.3 (12-Oct-2008)
> > fsck.ext3: Superblock invalid, trying backup blocks...
> > fsck.ext3: Bad magic number in super-block while trying to open /dev/sda
> >
> > 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>
> >
> > MacMini ~ # e2fsck -b 8193 /dev/sda
> > e2fsck 1.41.3 (12-Oct-2008)
> > e2fsck: Bad magic number in super-block while trying to open /dev/sda
> >
> > 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>
> >
> > MacMini ~ # fsck -t ext3 /dev/sda1
> > fsck 1.41.3 (12-Oct-2008)
> > e2fsck 1.41.3 (12-Oct-2008)
> > fsck.ext3: Attempt to read block from filesystem resulted in short
> > read while tr            ying to open /dev/sda1
> > Could this be a zero-length partition?
> > MacMini ~ # e2fsck -b 8193 /dev/sda1
> > e2fsck 1.41.3 (12-Oct-2008)
> > e2fsck: Bad magic number in super-block while trying to open /dev/sda1
> >
> > 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>
> >
> > MacMini ~ # e2fsck -b 8193 /dev/sda
> > e2fsck 1.41.3 (12-Oct-2008)
> > e2fsck: Bad magic number in super-block while trying to open /dev/sda
> >
> > 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>
> >
> > MacMini ~ #
> 
> A little more info:
> 
> Buffer I/O error on device sda, logical block 12
> sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
> sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
> sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
> end_request: I/O error, dev sda, sector 96
> Buffer I/O error on device sda, logical block 12
> sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
> sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
> sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
> end_request: I/O error, dev sda, sector 96
> Buffer I/O error on device sda, logical block 12
> sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
> sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
> sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
> end_request: I/O error, dev sda, sector 96
> 
> MacMini ~ # fsck -t ext3 /dev/sda1
> fsck 1.41.3 (12-Oct-2008)
> e2fsck 1.41.3 (12-Oct-2008)
> fsck.ext3: Attempt to read block from filesystem resulted in short
> read while trying to open /dev/sda1
> Could this be a zero-length partition?
> MacMini ~ #
> 
> 
> end_request: I/O error, dev sda, sector 95
> __ratelimit: 58 callbacks suppressed
> Buffer I/O error on device sda1, logical block 16
> Buffer I/O error on device sda1, logical block 17
> Buffer I/O error on device sda1, logical block 18
> Buffer I/O error on device sda1, logical block 19
> Buffer I/O error on device sda1, logical block 20
> Buffer I/O error on device sda1, logical block 21
> Buffer I/O error on device sda1, logical block 22
> Buffer I/O error on device sda1, logical block 23
> Buffer I/O error on device sda1, logical block 24
> Buffer I/O error on device sda1, logical block 25
> sd 0:0:0:0: [sda] Result: hostbyte=0x00 driverbyte=0x08
> sd 0:0:0:0: [sda] Sense Key : 0x3 [current]
> sd 0:0:0:0: [sda] ASC=0x4b ASCQ=0x0
> end_request: I/O error, dev sda, sector 95
> __ratelimit: 86 callbacks suppressed
> Buffer I/O error on device sda1, logical block 16
> Buffer I/O error on device sda1, logical block 17
> Buffer I/O error on device sda1, logical block 18
> Buffer I/O error on device sda1, logical block 19
> Buffer I/O error on device sda1, logical block 20
> Buffer I/O error on device sda1, logical block 21
> Buffer I/O error on device sda1, logical block 22
> Buffer I/O error on device sda1, logical block 23
> Buffer I/O error on device sda1, logical block 24
> Buffer I/O error on device sda1, logical block 25
> MacMini ~ #
> 
> 
> I could try reinstalling the file system but I had a few backups on
> this drive for other systems. (non-critical, but possibly useful) I'd
> like to be sure I cannot recover them before I blow everything away.
> 
> Thanks,
> Mark
> 

you might be lucky and able to use smartctl on that device. Sadly most usb 
converters don't support smart even when the drive does.

That could tell you some more, but a quick glance says:

your drive is hosed. 
Block errors... ugh... maybe the heads had a bit of platter contact.

You can try using magicrescue - or even better photorec.

they won't repair your filesystems - but they might be able to get the data off 
you are after.

Glück Auf
Volker

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Mark Knecht]

On Sat, Sep 19, 2009 at 3:59 PM, Volker Armin Hemmann
<volkerarmin@googlemail.com> wrote:
> On Sonntag 20 September 2009, Mark Knecht wrote:
>> On Sat, Sep 19, 2009 at 3:21 PM, Mark Knecht <markknecht@gmail.com> wrote:
>> > Hi,
>> >   I seem to have lost an external USB drive I've been using on my
>> > MythTV backend server for video storage. What commands can I try to
>> > get it to wake up or at least show me what's working and what isn't?
>> > The drive shows under fdisk /dev/sda. I can see the large partition,
>> > and it seems to be the right size, according to fdisk anyway, but I
>> > cannot mount it using mount, and so far I cannot get e2fsck to do
>> > anything. Both of these fail:
>> >
>> >
>> > mount -t ext2 /dev/sda1 /video
>> > mount -t ext3 /dev/sda1 /video
>> >
>> >   I have power cycled the drive and I've rebooted the MacMini. Nothing
>> > changed.
>> >
>> > Thanks in advance,
>> > Mark
>> >
<SNIP>
>> I could try reinstalling the file system but I had a few backups on
>> this drive for other systems. (non-critical, but possibly useful) I'd
>> like to be sure I cannot recover them before I blow everything away.
>>
>> Thanks,
>> Mark
>>
>
> you might be lucky and able to use smartctl on that device. Sadly most usb
> converters don't support smart even when the drive does.
>
> That could tell you some more, but a quick glance says:
>
> your drive is hosed.
> Block errors... ugh... maybe the heads had a bit of platter contact.
>
> You can try using magicrescue - or even better photorec.
>
> they won't repair your filesystems - but they might be able to get the data off
> you are after.
>
> Glück Auf
> Volker
>

Thanks. I'll check these apps out.

So far the only life I've found is with e2fsck -b 32768 /dev/sda1.
This starts finding some things that make sense name-wise but then
starts complaining about other things.

Again, it's just mythtv video files so those can probably just be
rerecorded at some point. The backups were for my Windows machines
which are working right now so as long as I make some new backups
elsewhere I should be reasonably safe unless I find one day that
something I need then isn't on my machine right now.

Sad when your backup strategy goes against you...

Again, thanks for the pointers.

- Mark

[gentoo-user] Re: USB drive dead? (Commands to check?)

[walt]

On 09/19/2009 03:38 PM, Mark Knecht wrote:
> On Sat, Sep 19, 2009 at 3:21 PM, Mark Knecht<markknecht@gmail.com>  wrote:
>> Hi,
>>    I seem to have lost an external USB drive I've been using on my
>> MythTV backend server for video storage. What commands can I try to
>> get it to wake up or at least show me what's working and what isn't?
>> The drive shows under fdisk /dev/sda...

I hate to be the party poop, but I suspect disk hardware failure.

Can you use dd to read the raw disk?  e.g.
# dd if=/dev/sda of=/tmp/ddout bs=512 count=1024(or whatever)

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Volker Armin Hemmann]

On Sonntag 20 September 2009, walt wrote:
> On 09/19/2009 03:38 PM, Mark Knecht wrote:
> > On Sat, Sep 19, 2009 at 3:21 PM, Mark Knecht<markknecht@gmail.com>  wrote:
> >> Hi,
> >>    I seem to have lost an external USB drive I've been using on my
> >> MythTV backend server for video storage. What commands can I try to
> >> get it to wake up or at least show me what's working and what isn't?
> >> The drive shows under fdisk /dev/sda...
> 
> I hate to be the party poop, but I suspect disk hardware failure.
> 
> Can you use dd to read the raw disk?  e.g.
> # dd if=/dev/sda of=/tmp/ddout bs=512 count=1024(or whatever)
> 

he has block errors - dd won't help him much. ddrescue is a better choice in 
such circumstances.

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Mark Knecht]

On Sat, Sep 19, 2009 at 4:12 PM, walt <w41ter@gmail.com> wrote:
> On 09/19/2009 03:38 PM, Mark Knecht wrote:
>>
>> On Sat, Sep 19, 2009 at 3:21 PM, Mark Knecht<markknecht@gmail.com>  wrote:
>>>
>>> Hi,
>>>   I seem to have lost an external USB drive I've been using on my
>>> MythTV backend server for video storage. What commands can I try to
>>> get it to wake up or at least show me what's working and what isn't?
>>> The drive shows under fdisk /dev/sda...
>
> I hate to be the party poop, but I suspect disk hardware failure.
>
> Can you use dd to read the raw disk?  e.g.
> # dd if=/dev/sda of=/tmp/ddout bs=512 count=1024(or whatever)
>

Hi Walt,
   Don't worry about it. If it's dead it's dead.

   I've actually managed to make some headway. After fiddling around
with e2fsck -b 32768 /dev/sda1 the drive is now mountable but running
e2fsck after a reboot says the drive still has errors:

MacMini ~ # e2fsck /dev/sda1
e2fsck 1.41.3 (12-Oct-2008)
/dev/sda1 contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
^C/dev/sda1: e2fsck canceled.

/dev/sda1: ********** WARNING: Filesystem still has errors **********

MacMini ~ #

If I mount the drive I can actually see all the MythTV files and
amazingly they still seem to play so I don't think the drive is dead.
I got a few messages about my backup directory being hosed so I
attempted to delete it. Now the drive mounts but the sizes and things
are messed up:

MacMini ~ # mount
/dev/hda4 on / type ext3 (rw,noatime)
proc on /proc type proc (rw,nosuid,nodev,noexec)
sysfs on /sys type sysfs (rw,nosuid,nodev,noexec)
udev on /dev type tmpfs (rw,nosuid,size=10240k,mode=755)
devpts on /dev/pts type devpts (rw,nosuid,noexec,gid=5,mode=620)
shm on /dev/shm type tmpfs (rw,noexec,nosuid,nodev)
usbfs on /proc/bus/usb type usbfs (rw,noexec,nosuid,devmode=0664,devgid=85)
/dev/sda1 on /video type ext3 (rw)

MacMini ~ # df /video
Filesystem           1K-blocks      Used Available Use% Mounted on
/dev/sda1            157566568  -1551528 151114116   -  /video
MacMini ~ #

MacMini ~ # ls /video/
1003_20090531163000.mpg.png  1042_20090611190000.mpg.png
1189_20090617183000.mpg.png
1003_20090603173000.mpg.png  1042_20090911210000.mpg
1189_20090619183000.mpg
1003_20090621163000.mpg.png  1042_20090911210000.mpg.png

<SNIP>

1017_20090817193000.mpg.png  1189_20090617183000.mpg               lost+found
1042_20090611170000.mpg.png  1189_20090617183000.mpg.100x75.png
MacMini ~ #


Now, my question is how can I use the file system tools to fix all the
tables on this drive?

I see Volker is suggesting ddrescue. As it seems I've already lost the
Windows backup files but have good MYthTV files is there a way to
repair the disk tables and only end up with the existing Myth files
and clean tables? I don't have another disk to copy all this stuff to
and the MacMini is headless and PPC-based so it's difficult to use
gparted or anything like that.

I think the existing Myth files are possibly small enough that I could
store them temporarily on the Mac while I rebuild the USB drive.

Thanks,
Mark

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Volker Armin Hemmann]

On Sonntag 20 September 2009, Mark Knecht wrote:
> On Sat, Sep 19, 2009 at 4:12 PM, walt <w41ter@gmail.com> wrote:
> > On 09/19/2009 03:38 PM, Mark Knecht wrote:
> >> On Sat, Sep 19, 2009 at 3:21 PM, Mark Knecht<markknecht@gmail.com> 
 wrote:
> >>> Hi,
> >>>   I seem to have lost an external USB drive I've been using on my
> >>> MythTV backend server for video storage. What commands can I try to
> >>> get it to wake up or at least show me what's working and what isn't?
> >>> The drive shows under fdisk /dev/sda...
> >
> > I hate to be the party poop, but I suspect disk hardware failure.
> >
> > Can you use dd to read the raw disk?  e.g.
> > # dd if=/dev/sda of=/tmp/ddout bs=512 count=1024(or whatever)
> 
> Hi Walt,
>    Don't worry about it. If it's dead it's dead.
> 
>    I've actually managed to make some headway. After fiddling around
> with e2fsck -b 32768 /dev/sda1 the drive is now mountable but running
> e2fsck after a reboot says the drive still has errors:
> 
> MacMini ~ # e2fsck /dev/sda1
> e2fsck 1.41.3 (12-Oct-2008)
> /dev/sda1 contains a file system with errors, check forced.
> Pass 1: Checking inodes, blocks, and sizes
> ^C/dev/sda1: e2fsck canceled.
> 
> /dev/sda1: ********** WARNING: Filesystem still has errors **********
> 
> MacMini ~ #
> 
> If I mount the drive I can actually see all the MythTV files and
> amazingly they still seem to play so I don't think the drive is dead.
> I got a few messages about my backup directory being hosed so I
> attempted to delete it. Now the drive mounts but the sizes and things
> are messed up:
> 
> MacMini ~ # mount
> /dev/hda4 on / type ext3 (rw,noatime)
> proc on /proc type proc (rw,nosuid,nodev,noexec)
> sysfs on /sys type sysfs (rw,nosuid,nodev,noexec)
> udev on /dev type tmpfs (rw,nosuid,size=10240k,mode=755)
> devpts on /dev/pts type devpts (rw,nosuid,noexec,gid=5,mode=620)
> shm on /dev/shm type tmpfs (rw,noexec,nosuid,nodev)
> usbfs on /proc/bus/usb type usbfs (rw,noexec,nosuid,devmode=0664,devgid=85)
> /dev/sda1 on /video type ext3 (rw)
> 
> MacMini ~ # df /video
> Filesystem           1K-blocks      Used Available Use% Mounted on
> /dev/sda1            157566568  -1551528 151114116   -  /video
> MacMini ~ #
> 
> MacMini ~ # ls /video/
> 1003_20090531163000.mpg.png  1042_20090611190000.mpg.png
> 1189_20090617183000.mpg.png
> 1003_20090603173000.mpg.png  1042_20090911210000.mpg
> 1189_20090619183000.mpg
> 1003_20090621163000.mpg.png  1042_20090911210000.mpg.png
> 
> <SNIP>
> 
> 1017_20090817193000.mpg.png  1189_20090617183000.mpg              
>  lost+found 1042_20090611170000.mpg.png  1189_20090617183000.mpg.100x75.png
> MacMini ~ #
> 
> 
> Now, my question is how can I use the file system tools to fix all the
> tables on this drive?
> 
> I see Volker is suggesting ddrescue. As it seems I've already lost the
> Windows backup files but have good MYthTV files is there a way to
> repair the disk tables and only end up with the existing Myth files
> and clean tables? I don't have another disk to copy all this stuff to
> and the MacMini is headless and PPC-based so it's difficult to use
> gparted or anything like that.
> 
> I think the existing Myth files are possibly small enough that I could
> store them temporarily on the Mac while I rebuild the USB drive.
> 
> Thanks,
> Mark
> 

seriously, I think you should try to get off everything you want to keep - and 
then replace the disk with a new one. If a disk starts throwing block errors 
it will only become worse. Don't worry about 'repairing' the file system as 
long as there is hardware damage. Try to get off the disk as much as possible - 
and then scrap it.

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Mark Knecht]

On Sat, Sep 19, 2009 at 4:48 PM, Volker Armin Hemmann
<volkerarmin@googlemail.com> wrote:
<SNIP>
>>
>
> seriously, I think you should try to get off everything you want to keep - and
> then replace the disk with a new one. If a disk starts throwing block errors
> it will only become worse. Don't worry about 'repairing' the file system as
> long as there is hardware damage. Try to get off the disk as much as possible -
> and then scrap it.
>
>

I suspect you're right. It's just another $100 to go buy a new one....

Anyway, I'll see what I can set up to save the files that are still there.

Cheers,
Mark

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Stroller]

On 20 Sep 2009, at 01:01, Mark Knecht wrote:

> On Sat, Sep 19, 2009 at 4:48 PM, Volker Armin Hemmann
> <volkerarmin@googlemail.com> wrote:
> <SNIP>
>>>
>>
>> seriously, I think you should try to get off everything you want to  
>> keep - and
>> then replace the disk with a new one. If a disk starts throwing  
>> block errors
>> it will only become worse. Don't worry about 'repairing' the file  
>> system as
>> long as there is hardware damage. Try to get off the disk as much  
>> as possible -
>> and then scrap it.
>
> I suspect you're right. It's just another $100 to go buy a new one....
>
> Anyway, I'll see what I can set up to save the files that are still  
> there.


Although he seems to be demonstrating in this thread an inability to  
snip long sections of quoted text, leaving the reader with many lines  
of irrelevance to scroll through, I agree with Volker.

If the drive fails you're going to be spending $100, anyway. If it  
fails without having been replaced & your data pulled off it then you  
could find yourself floating down Effluent River and unable to start  
your outboard.

I think this drive is quite likely to fail catastrophically, from my  
experience of having seen similar errors in the past. I really  
wouldn't trust this drive with important data right now. If you get  
your data off it and replace it in it's current capacity, there's  
nothing stopping you using it as a secondary drive in the future; I  
wouldn't trust it with important data right now, but if it's still  
chugging away in 6 months time then you can probably begin to have  
faith in it. Once you've gotten your data off the drive it wouldn't do  
any harm to format it nice with a clean filesystem; and writing a  
bunch of big unimportant files on the drive (e.g. `dd if=/dev/zero of=/ 
mnt/sda1/foo`) might allow it to map away a bunch of bad sectors.

But right now you should probably act like the drive is definitely  
hosed. I don't think you should be saying "oh, this might cost me  
$100, I hope it doesn't" - you should be saying "s#!t d@mn! I had to  
buy a new hard-drive. But at least my data's ok".

But maybe data isn't as important to you as it is to me. Relying on  
this drive for the backups of your Windows machines right now would be  
a mockery; around here sod's law would conspire for me to need one of  
those backups, were I to continue using a drive showing errors like  
this. I write as a guy you just bought yet another 500gb drive on  
Friday; I would too have preferred not to spend that money, but  
experience shows that frugality can sometimes be a mistake.

Stroller.

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Mark Knecht]

On Sun, Sep 20, 2009 at 3:24 AM, Stroller
<stroller@stellar.eclipse.co.uk> wrote:
>
> On 20 Sep 2009, at 01:01, Mark Knecht wrote:
>
>> On Sat, Sep 19, 2009 at 4:48 PM, Volker Armin Hemmann
>> <volkerarmin@googlemail.com> wrote:
>> <SNIP>
>>>>
>>>
>>> seriously, I think you should try to get off everything you want to keep
>>> - and
>>> then replace the disk with a new one. If a disk starts throwing block
>>> errors
>>> it will only become worse. Don't worry about 'repairing' the file system
>>> as
>>> long as there is hardware damage. Try to get off the disk as much as
>>> possible -
>>> and then scrap it.
>>
>> I suspect you're right. It's just another $100 to go buy a new one....
>>
>> Anyway, I'll see what I can set up to save the files that are still there.
>
>
> Although he seems to be demonstrating in this thread an inability to snip
> long sections of quoted text, leaving the reader with many lines of
> irrelevance to scroll through, I agree with Volker.
>
> If the drive fails you're going to be spending $100, anyway. If it fails
> without having been replaced & your data pulled off it then you could find
> yourself floating down Effluent River and unable to start your outboard.
>
> I think this drive is quite likely to fail catastrophically, from my
> experience of having seen similar errors in the past. I really wouldn't
> trust this drive with important data right now. If you get your data off it
> and replace it in it's current capacity, there's nothing stopping you using
> it as a secondary drive in the future; I wouldn't trust it with important
> data right now, but if it's still chugging away in 6 months time then you
> can probably begin to have faith in it. Once you've gotten your data off the
> drive it wouldn't do any harm to format it nice with a clean filesystem; and
> writing a bunch of big unimportant files on the drive (e.g. `dd if=/dev/zero
> of=/mnt/sda1/foo`) might allow it to map away a bunch of bad sectors.
>
> But right now you should probably act like the drive is definitely hosed. I
> don't think you should be saying "oh, this might cost me $100, I hope it
> doesn't" - you should be saying "s#!t d@mn! I had to buy a new hard-drive.
> But at least my data's ok".
>
> But maybe data isn't as important to you as it is to me. Relying on this
> drive for the backups of your Windows machines right now would be a mockery;
> around here sod's law would conspire for me to need one of those backups,
> were I to continue using a drive showing errors like this. I write as a guy
> you just bought yet another 500gb drive on Friday; I would too have
> preferred not to spend that money, but experience shows that frugality can
> sometimes be a mistake.
>
> Stroller.
>

Turns out none of the backup or Myth files were savable in any
practical manner of speaking. Myth could play them, or at least start
playing them - I don't know if it could get to the end of any of them,
but if I tried to copy them off to another drive the machine just
started hanging with lots of dmesg drive errors. None of the previous
windows backups were savable. I've taken new windows backups starting
last night.

The failing drive is now off line and the family will just have to
live with less Myth recording time. I've switched to an old, slow 80GB
1394 drive vs the newer 160GB USB2 drive that failed.

I guess this now comes down to having no backup for my backup system.
;-) Saving anything on hard drives always results with this risk I
suppose. My daily backups are 1-2 GB in total so I guess I could start
writing DVDs or something like that.

Thanks to all for your inputs and ideas.

Cheers,
Mark

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Neil Bothwick]

[-- Attachment #1: Type: text/plain, Size: 883 bytes --]

On Sun, 20 Sep 2009 11:41:09 -0700, Mark Knecht wrote:

> Turns out none of the backup or Myth files were savable in any
> practical manner of speaking. Myth could play them, or at least start
> playing them - I don't know if it could get to the end of any of them,
> but if I tried to copy them off to another drive the machine just
> started hanging with lots of dmesg drive errors. None of the previous
> windows backups were savable. I've taken new windows backups starting
> last night.

Did you try photorec as previously suggested? When a drive starts
behaving like this, retrieving the data should be the first thing you
try. Repairing the filesystem should be the last because you run the risk
of causing further damage. Photorec can retrieve files that are
inaccessible through the filesystem.


-- 
Neil Bothwick

WinErr 013: Unexpected error - Huh ?

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-user] Re: USB drive dead? (Commands to check?)

[Mark Knecht]

On Mon, Sep 21, 2009 at 1:21 AM, Neil Bothwick <neil@digimed.co.uk> wrote:
> On Sun, 20 Sep 2009 11:41:09 -0700, Mark Knecht wrote:
>
>> Turns out none of the backup or Myth files were savable in any
>> practical manner of speaking. Myth could play them, or at least start
>> playing them - I don't know if it could get to the end of any of them,
>> but if I tried to copy them off to another drive the machine just
>> started hanging with lots of dmesg drive errors. None of the previous
>> windows backups were savable. I've taken new windows backups starting
>> last night.
>
> Did you try photorec as previously suggested? When a drive starts
> behaving like this, retrieving the data should be the first thing you
> try. Repairing the filesystem should be the last because you run the risk
> of causing further damage. Photorec can retrieve files that are
> inaccessible through the filesystem.
>
>
> --
> Neil Bothwick

I haven't tried it yet. I don't like installing extra softwrae on the
Myth backend so I disconnected the drive, used this old 1394 drive as
a replacement and now have the old drive back here in my office. I
figure I'll install Photorec et all next weekend on my AMD64 box and
see what it can find.

Cheers,
Mark

Re: [gentoo-user] USB drive dead? (Commands to check?)

[Paul Hartman]

On Sat, Sep 19, 2009 at 5:21 PM, Mark Knecht <markknecht@gmail.com> wrote:
> Hi,
>   I seem to have lost an external USB drive I've been using on my
> MythTV backend server for video storage. What commands can I try to
> get it to wake up or at least show me what's working and what isn't?
> The drive shows under fdisk /dev/sda. I can see the large partition,
> and it seems to be the right size, according to fdisk anyway, but I
> cannot mount it using mount, and so far I cannot get e2fsck to do
> anything.

I had similar errors with an external USB drive recently and it turned
out to be related to the USB port on the computer. (I suspect the USB
controller was overloaded). I plugged it into a port on a different
controller and it started working normally again. So some easy things
I would suggest trying before messing with data:

A) diferent USB port on the same computer
B) plug it into a different computer entirely
C) try a different USB cable

Re: [gentoo-user] USB drive dead? (Commands to check?)

[Mark Knecht]

On Mon, Sep 21, 2009 at 12:07 PM, Paul Hartman
<paul.hartman+gentoo@gmail.com> wrote:
> On Sat, Sep 19, 2009 at 5:21 PM, Mark Knecht <markknecht@gmail.com> wrote:
>> Hi,
>>   I seem to have lost an external USB drive I've been using on my
>> MythTV backend server for video storage. What commands can I try to
>> get it to wake up or at least show me what's working and what isn't?
>> The drive shows under fdisk /dev/sda. I can see the large partition,
>> and it seems to be the right size, according to fdisk anyway, but I
>> cannot mount it using mount, and so far I cannot get e2fsck to do
>> anything.
>
> I had similar errors with an external USB drive recently and it turned
> out to be related to the USB port on the computer. (I suspect the USB
> controller was overloaded). I plugged it into a port on a different
> controller and it started working normally again. So some easy things
> I would suggest trying before messing with data:
>
> A) diferent USB port on the same computer
> B) plug it into a different computer entirely
> C) try a different USB cable
>
>

By definition this will be true when I move it from the PPC-based Myth
backend to the AMD64-based MythTV frontend here in the office. If it
'magically' starts working then that sort of cause may well be the
reason.

I'll report back on this but won't likely touch it before the weekend.

Thanks!

Cheers,
Mark

Re: [gentoo-user] USB drive dead? (Commands to check?)

[Mark Knecht]

On Mon, Sep 21, 2009 at 2:39 PM, Mark Knecht <markknecht@gmail.com> wrote:
> On Mon, Sep 21, 2009 at 12:07 PM, Paul Hartman
> <paul.hartman+gentoo@gmail.com> wrote:
>> On Sat, Sep 19, 2009 at 5:21 PM, Mark Knecht <markknecht@gmail.com> wrote:
>>> Hi,
>>>   I seem to have lost an external USB drive I've been using on my
>>> MythTV backend server for video storage. What commands can I try to
>>> get it to wake up or at least show me what's working and what isn't?
>>> The drive shows under fdisk /dev/sda. I can see the large partition,
>>> and it seems to be the right size, according to fdisk anyway, but I
>>> cannot mount it using mount, and so far I cannot get e2fsck to do
>>> anything.
>>
>> I had similar errors with an external USB drive recently and it turned
>> out to be related to the USB port on the computer. (I suspect the USB
>> controller was overloaded). I plugged it into a port on a different
>> controller and it started working normally again. So some easy things
>> I would suggest trying before messing with data:
>>
>> A) diferent USB port on the same computer
>> B) plug it into a different computer entirely
>> C) try a different USB cable
>>
>>
>
> By definition this will be true when I move it from the PPC-based Myth
> backend to the AMD64-based MythTV frontend here in the office. If it
> 'magically' starts working then that sort of cause may well be the
> reason.
>
> I'll report back on this but won't likely touch it before the weekend.

So last night my replacement drive - 1394-bsed, not USB - went
off-line during recording time so sll the late evening recordings were
hosed.

What's with Linux support of external drives? Is it just not reliable
enough to depend on? This was not a drive failure but just a bunch of
sense code message problems and everything quit. I probably could have
spent time removing drivers, etc, and then restarting it but I just
rebooted and everything came back.

I used to use this drive for weeks at a time on one of my Windows
boxes. No problems at that time so I have no strong reason to suspect
the drive when this is the second drive issue in a few days wit this
system.

I wonder how I determine if it's a drive problem or a kernel/driver problem?

- Mark

Re: [gentoo-user] USB drive dead? (Commands to check?)

[Mick]

[-- Attachment #1: Type: text/plain, Size: 1178 bytes --]

On Tuesday 22 September 2009, Mark Knecht wrote:

> What's with Linux support of external drives? Is it just not reliable
> enough to depend on? This was not a drive failure but just a bunch of
> sense code message problems and everything quit. I probably could have
> spent time removing drivers, etc, and then restarting it but I just
> rebooted and everything came back.
>
> I used to use this drive for weeks at a time on one of my Windows
> boxes. No problems at that time so I have no strong reason to suspect
> the drive when this is the second drive issue in a few days wit this
> system.
>
> I wonder how I determine if it's a drive problem or a kernel/driver
> problem?

I wonder if you have a memory problem with that box?  I don't know what errors 
you've been getting in the logs, but it is a thought when the common 
denominator is the box.  Have you tried running memtest86+ overnight to see 
what gives?

Another reason might be the physical location.  If the drives in question are 
submitted to physical vibration (e.g. next to a door; staircase, etc) then 
the failures could be due to mechanical reasons.

HTH.
-- 
Regards,
Mick

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]