Hiding Hard Drives

[Nicole Massey]

I'm new to the list, and I have an interesting little problem.

First, the environment:
Intel Pentium 4 3.4GHz single core processor
Gigabyte motherboard
4GB of RAM
32-bit Windows XP Professional running Service Pack 3
Jaws 12 screen reader

The environment also includes a Windows 2003 Server machine for file
storage, backup, printer sharing, and other tasks. It's connected using
100Mbit ethernet

I also have four external hard drive boxes. Two of them detect just fine
when active in Computer Management on both the Device Manager and Disk
Management lists. The other two do not. Both are IDE drives in the boxes,
and both are Western Digital drives. One of the boxes is a ByteCC 5.25 form
factor external hard drive box, the other one is much smaller and I haven't
gotten a sighted person to look it over and find a name and model number. My
sighted roommate was unable to find a model number on the ByteCC box.
I'm running USB 2 and Driver Detective finds no drivers that need updating.
The two drives in question don't recognize in My Computer, and they also
don't show up in Disk Management at all.
Now, for a touch of background. Thanks to an expired virus protection
program I got hit with a virus in the late winter of 2011. I had to do a
complete reinstall of the OS to clean out its residue. When I did that I
copied all my valuable files to the ByteCC drive, as it was recognized with
no problem then. After doing the reinstall the system wouldn't recognize the
drive. Nothing had changed except for the OS -- same cable, port, drive box,
drive, etc.
I managed to get some of the most important files off the drive using
R-Studio. But I'd like to get this drive working again. I'd like to get the
rest of the files off it, and then use it to backup music files and such.
(Since I'm blind I rip everything I buy so I can access it from the
computer)
I'm having this problem with another drive box too.
Hints it may be a driver problem: In Device Manager the drives show up as
"Generic USB 2.0 Device" (for the ByteCC drive) and "Usb Device" for the
other one. Going into properties and looking at the volume information
returns a blank. Hitting the Populate button does nothing but mess up the
maneuvering, though if I hit alt+tae once, then again I can navigate to OK
or Cancel. Then the Populate button is inaccessible, too.
But driver detective doesn't return anything.

There are some additional steps I can take, though they will require sighted
help. For one thing, these may be set as slave drives with the jumpers.
Another possibility is going into safe mode and deleting all of the USB
drive drivers and letting Plug and Pray reinstall and re-detect them.
One more thing ... I went to the ByteCC site, and there are drivers, but
they're by model number. So that's a concern there.

Suggestions? Know of anything that will detect a manufacturer and model
number of a USB drive box?and all updates

 

[Flasherly]

I'm new to the list, and I have an interesting little problem.

First, the environment:
Intel Pentium 4 3.4GHz single core processor
Gigabyte motherboard
4GB of RAM
32-bit Windows XP Professional running Service Pack 3
Jaws 12 screen reader

The environment also includes a Windows 2003 Server machine for file
storage, backup, printer sharing, and other tasks. It's connected using
100Mbit ethernet

I also have four external hard drive boxes. Two of them detect just fine
when active in Computer Management on both the Device Manager and Disk
Management lists. The other two do not. Both are IDE drives in the boxes,
and both are Western Digital drives. One of the boxes is a ByteCC 5.25 form
factor external hard drive box, the other one is much smaller and I haven't
gotten a sighted person to look it over and find a name and model number. My
sighted roommate was unable to find a model number on the ByteCC box.
I'm running USB 2 and Driver Detective finds no drivers that need updating.
The two drives in question don't recognize in My Computer, and they also
don't show up in Disk Management at all.
Now, for a touch of background. Thanks to an expired virus protection
program I got hit with a virus in the late winter of 2011. I had to do a
complete reinstall of the OS to clean out its residue. When I did that I
copied all my valuable files to the ByteCC drive, as it was recognized with
no problem then. After doing the reinstall the system wouldn't recognize the
drive. Nothing had changed except for the OS -- same cable, port, drive box,
drive, etc.
I managed to get some of the most important files off the drive using
R-Studio. But I'd like to get this drive working again. I'd like to get the
rest of the files off it, and then use it to backup music files and such.
(Since I'm blind I rip everything I buy so I can access it from the
computer)
I'm having this problem with another drive box too.
Hints it may be a driver problem: In Device Manager the drives show up as
"Generic USB 2.0 Device" (for the ByteCC drive) and "Usb Device" for the
other one. Going into properties and looking at the volume information
returns a blank. Hitting the Populate button does nothing but mess up the
maneuvering, though if I hit alt+tae once, then again I can navigate to OK
or Cancel. Then the Populate button is inaccessible, too.
But driver detective doesn't return anything.

There are some additional steps I can take, though they will require sighted
help. For one thing, these may be set as slave drives with the jumpers.
Another possibility is going into safe mode and deleting all of the USB
drive drivers and letting Plug and Pray reinstall and re-detect them.
One more thing ... I went to the ByteCC site, and there are drivers, but
they're by model number. So that's a concern there.

Suggestions? Know of anything that will detect a manufacturer and model
number of a USB drive box?and all updates

I have either three or eight external drives next to me. Eight is the
number of physical hard drives. Three is the number of USB docking
stations to hold them, one I which I bought on sale for $9US last
week. Twenty eight drive letters is the limit of the operating
system, which I should now be able to exceed.

All run hot over time, so I also have a 120V 4" fan to keep things
cool. Never run into a problem, but then I wouldn't consider
disadvantageous not being at a local restaurant to plug one in for
storing wifi access. At least not as easily.

 

[Paul]

I'm new to the list, and I have an interesting little problem.

First, the environment:
Intel Pentium 4 3.4GHz single core processor
Gigabyte motherboard
4GB of RAM
32-bit Windows XP Professional running Service Pack 3
Jaws 12 screen reader

The environment also includes a Windows 2003 Server machine for file
storage, backup, printer sharing, and other tasks. It's connected using
100Mbit ethernet

I also have four external hard drive boxes. Two of them detect just fine
when active in Computer Management on both the Device Manager and Disk
Management lists. The other two do not. Both are IDE drives in the boxes,
and both are Western Digital drives. One of the boxes is a ByteCC 5.25 form
factor external hard drive box, the other one is much smaller and I haven't
gotten a sighted person to look it over and find a name and model number. My
sighted roommate was unable to find a model number on the ByteCC box.
I'm running USB 2 and Driver Detective finds no drivers that need updating.
The two drives in question don't recognize in My Computer, and they also
don't show up in Disk Management at all.
Now, for a touch of background. Thanks to an expired virus protection
program I got hit with a virus in the late winter of 2011. I had to do a
complete reinstall of the OS to clean out its residue. When I did that I
copied all my valuable files to the ByteCC drive, as it was recognized with
no problem then. After doing the reinstall the system wouldn't recognize the
drive. Nothing had changed except for the OS -- same cable, port, drive box,
drive, etc.
I managed to get some of the most important files off the drive using
R-Studio. But I'd like to get this drive working again. I'd like to get the
rest of the files off it, and then use it to backup music files and such.
(Since I'm blind I rip everything I buy so I can access it from the
computer)
I'm having this problem with another drive box too.
Hints it may be a driver problem: In Device Manager the drives show up as
"Generic USB 2.0 Device" (for the ByteCC drive) and "Usb Device" for the
other one. Going into properties and looking at the volume information
returns a blank. Hitting the Populate button does nothing but mess up the
maneuvering, though if I hit alt+tae once, then again I can navigate to OK
or Cancel. Then the Populate button is inaccessible, too.
But driver detective doesn't return anything.

There are some additional steps I can take, though they will require sighted
help. For one thing, these may be set as slave drives with the jumpers.
Another possibility is going into safe mode and deleting all of the USB
drive drivers and letting Plug and Pray reinstall and re-detect them.
One more thing ... I went to the ByteCC site, and there are drivers, but
they're by model number. So that's a concern there.

Suggestions? Know of anything that will detect a manufacturer and model
number of a USB drive box?and all updates

I've had a strange case here involving an external drive, but it
might not be what you're seeing.

I have a USB hard drive enclosure with Cypress USB chip inside it.
The Cypress chip, converts IDE hard drive protocol, to USB Mass Storage
protocol. I didn't even know it used a Cypress chip, until the day it
failed to work. And when that happened, I knew right away which
brand was involved. (I'd read of the Cypress problem before.)

One day, I decided to download Seagate "Seatools For Windows". It includes
the ability to scan for USB hard drives (unlike the DOS version of the
same tool). It scanned my USB hard drive... And then the hard drive disappeared.
The hardware identity of the Cypress chip had changed. It was no longer
declaring itself as a USB Mass Storage device. (The word "Cypress" does not
normally appear in Device Manager. Only when the chip has lost its
identity, does it say it is a Cypress brand chip.)

I managed to find a "repair procedure", which re-flashes some storage
that keeps class information for the hardware. That got my enclosure
detected again.

If that were to happen, you'd see a new (errant) USB entry. (I have to use
the archived copy of this site, since the original entry is gone.)

http://web.archive.org/web/20080507...m=knowledgebase&_a=viewarticle&kbarticleid=74

"Cypress At2LP RC58 driver problem"

So a declaration of "Cypress AT2LP" in Device Manager, is the symptom.

(Note - I can't view this page in Firefox. My copy of IE with Flash
was needed, to make the text visible.)

http://www.cypress.com/?id=4&rID=38494

Basically, I used a repair tool, that loads generic, unbranded info
into the storage chip. Which doesn't hurt anything. The company that
made the USB hard drive enclosure, would not think of providing that
patching tool, so using the tool provided by another manufacturer
got things running again. First you install a driver that works
with AT2LP (so the repair tool can "reach" the drive), then you
run the repair tool, and the next time the drive is seen,
it has its old identity back.

USB enclosures can disappear because:

1) Power supply is bad.
2) Disk drive is dead, and cannot be read. The USB device
may not show up at all in Windows, because it can't get the
drive to work.
3) USB chip config info is damaged. Sometimes the erasure
seems to be spontaneous (hardware problem). In other cases,
like in my case, Seatools did it.

Only a few of those cases, leave something in Device Manager for
you to work with and diagnose. For power, you can place your ear
on the USB enclosure casing, and listen for "spinup" and "vibration"
of a working hard drive. If the casing is equipped with a cooling
fan, that fan may hide the easy-to-detect acoustic signature.

USB enclosures for 2.5" drives, may not have their own power adapter.
In that case, a failure mode is for the self-resetting fuse on the
computer motherboard to open temporarily, as an overload is detected.
Some USB 2.5" drives, come with a "two headed", Y-shaped USB cable,
to get power from two USB ports to run the hard drive.

On 3.5" drives, the power may come from a 12V supply that plugs
into the wall. Or, the 12V supply can be inside the enclosure itself.
What's really needed, is both 12V and 5V. But enclosures like the one
I own, make the 5V out of the supplied 12V signal. Only more ancient
enclosures, have an actual dual power supply, four pin connector, and
make both 12V and 5V. That's a good way to do it, but it costs more
money to make them that way. As a matter of fact, my 12V adapter, the
rating on the label makes it "underpowered", but it has never
failed to power anything I've put in the enclosure. So I can't
really say I've been cheated or anything.

If you were sighted, you'd take the enclosure apart, unplug the
IDE ribbon and Molex power, then transfer the drive inside to your
desktop computer. And test, to see if the drive could be detected.
If the drive showed up, you'd know the problem was with the
enclosure. Some brands of enclosures, are notorious for power
supply failures. And the funny thing was, the company with that
problem, also charged the most for their products. So price
in that case, did not equal quality.

Every Windows OS from Win2K onwards, should have a built-in
USB Mass Storage driver. So I doubt that's the problem.
There should be enough drivers in Windows, to make it work.
Microsoft, in fact, doesn't allow third parties to re-package
those drivers. While you're right, that the USB stack
occasionally needs to be reset (delete a bunch of entries
from Device Manager), I don't see an incentive to go on
such a mission at the moment.

Another way to detect USB hard drives, is via the BIOS
"popup boot menu". When the computer starts, if I press F8
on the keyboard, I'm presented with a BIOS-based boot menu.
It lists *all* the storage devices. USB flash sticks are
listed. USB hard drives are listed. I can use the cursor key,
to select a particular internal or external drive or
even an optical drive or floppy to boot from. So that
is yet another way, to develop a list of devices. And since
it does not rely on a Windows driver (just BIOS capabilities),
it can't be a driver problem if the drive doesn't show up there.
(On my other computer, you press F11 to access that menu. The
key used, varies from brand to brand. I don't know if
your screen reader, will be able to communicate the contents
of that menu to you.)

I have had cases, where when I run Linux from a LiveCD, during
shutdown Linux writes something to the USB devices, that prevents
them from being detected. I have to remove power from the
affected device to recover. For a USB flash, I unplug and
replug, to get rid of that state information. I still haven't
determined, why or how that one happens. You've probably had
your computer turned off at least once since the problem showed
up, so it's probably not a Linux induced problem. That only
started showing up, on LiveCDs within the last year or so.
Older versions of Linux, didn't do that, and all devices
could be detected on a reboot.

Paul

 

[Loren Pechtel]

There are some additional steps I can take, though they will require sighted
help. For one thing, these may be set as slave drives with the jumpers.
Another possibility is going into safe mode and deleting all of the USB
drive drivers and letting Plug and Pray reinstall and re-detect them.
One more thing ... I went to the ByteCC site, and there are drivers, but
they're by model number. So that's a concern there.

Suggestions? Know of anything that will detect a manufacturer and model
number of a USB drive box?and all updates

I have had a *LOT* of trouble with external drives simply dying. It's
always been the controller, not the physical drive. Every single time
I've been able to get the data off without difficulty by mounting it
somewhere else.

 

[Nicole Massey]

I've had a strange case here involving an external drive, but it
might not be what you're seeing.

I have a USB hard drive enclosure with Cypress USB chip inside it.
The Cypress chip, converts IDE hard drive protocol, to USB Mass Storage
protocol. I didn't even know it used a Cypress chip, until the day it
failed to work. And when that happened, I knew right away which
brand was involved. (I'd read of the Cypress problem before.)

One day, I decided to download Seagate "Seatools For Windows". It includes
the ability to scan for USB hard drives (unlike the DOS version of the
same tool). It scanned my USB hard drive... And then the hard drive
disappeared.
The hardware identity of the Cypress chip had changed. It was no longer
declaring itself as a USB Mass Storage device. (The word "Cypress" does
not
normally appear in Device Manager. Only when the chip has lost its
identity, does it say it is a Cypress brand chip.)

I managed to find a "repair procedure", which re-flashes some storage
that keeps class information for the hardware. That got my enclosure
detected again.

If that were to happen, you'd see a new (errant) USB entry. (I have to use
the archived copy of this site, since the original entry is gone.)

http://web.archive.org/web/20080507...m=knowledgebase&_a=viewarticle&kbarticleid=74

"Cypress At2LP RC58 driver problem"

So a declaration of "Cypress AT2LP" in Device Manager, is the symptom.

(Note - I can't view this page in Firefox. My copy of IE with Flash
was needed, to make the text visible.)

http://www.cypress.com/?id=4&rID=38494

Basically, I used a repair tool, that loads generic, unbranded info
into the storage chip. Which doesn't hurt anything. The company that
made the USB hard drive enclosure, would not think of providing that
patching tool, so using the tool provided by another manufacturer
got things running again. First you install a driver that works
with AT2LP (so the repair tool can "reach" the drive), then you
run the repair tool, and the next time the drive is seen,
it has its old identity back.

USB enclosures can disappear because:

1) Power supply is bad.
2) Disk drive is dead, and cannot be read. The USB device
may not show up at all in Windows, because it can't get the
drive to work.
3) USB chip config info is damaged. Sometimes the erasure
seems to be spontaneous (hardware problem). In other cases,
like in my case, Seatools did it.

Only a few of those cases, leave something in Device Manager for
you to work with and diagnose. For power, you can place your ear
on the USB enclosure casing, and listen for "spinup" and "vibration"
of a working hard drive. If the casing is equipped with a cooling
fan, that fan may hide the easy-to-detect acoustic signature.

USB enclosures for 2.5" drives, may not have their own power adapter.
In that case, a failure mode is for the self-resetting fuse on the
computer motherboard to open temporarily, as an overload is detected.
Some USB 2.5" drives, come with a "two headed", Y-shaped USB cable,
to get power from two USB ports to run the hard drive.

On 3.5" drives, the power may come from a 12V supply that plugs
into the wall. Or, the 12V supply can be inside the enclosure itself.
What's really needed, is both 12V and 5V. But enclosures like the one
I own, make the 5V out of the supplied 12V signal. Only more ancient
enclosures, have an actual dual power supply, four pin connector, and
make both 12V and 5V. That's a good way to do it, but it costs more
money to make them that way. As a matter of fact, my 12V adapter, the
rating on the label makes it "underpowered", but it has never
failed to power anything I've put in the enclosure. So I can't
really say I've been cheated or anything.

If you were sighted, you'd take the enclosure apart, unplug the
IDE ribbon and Molex power, then transfer the drive inside to your
desktop computer. And test, to see if the drive could be detected.
If the drive showed up, you'd know the problem was with the
enclosure. Some brands of enclosures, are notorious for power
supply failures. And the funny thing was, the company with that
problem, also charged the most for their products. So price
in that case, did not equal quality.

Every Windows OS from Win2K onwards, should have a built-in
USB Mass Storage driver. So I doubt that's the problem.
There should be enough drivers in Windows, to make it work.
Microsoft, in fact, doesn't allow third parties to re-package
those drivers. While you're right, that the USB stack
occasionally needs to be reset (delete a bunch of entries
from Device Manager), I don't see an incentive to go on
such a mission at the moment.

Another way to detect USB hard drives, is via the BIOS
"popup boot menu". When the computer starts, if I press F8
on the keyboard, I'm presented with a BIOS-based boot menu.
It lists *all* the storage devices. USB flash sticks are
listed. USB hard drives are listed. I can use the cursor key,
to select a particular internal or external drive or
even an optical drive or floppy to boot from. So that
is yet another way, to develop a list of devices. And since
it does not rely on a Windows driver (just BIOS capabilities),
it can't be a driver problem if the drive doesn't show up there.
(On my other computer, you press F11 to access that menu. The
key used, varies from brand to brand. I don't know if
your screen reader, will be able to communicate the contents
of that menu to you.)

I have had cases, where when I run Linux from a LiveCD, during
shutdown Linux writes something to the USB devices, that prevents
them from being detected. I have to remove power from the
affected device to recover. For a USB flash, I unplug and
replug, to get rid of that state information. I still haven't
determined, why or how that one happens. You've probably had
your computer turned off at least once since the problem showed
up, so it's probably not a Linux induced problem. That only
started showing up, on LiveCDs within the last year or so.
Older versions of Linux, didn't do that, and all devices
could be detected on a reboot.

Definitely not a power supply problem. First, the larger box has a fan
because its power supply is internal, while the smaller one gets warm fairly
quickly. (It's a metal case) I can also hear both drives spinning up -- my
hearing is rather good. And I can read both drives in R-Studio, a program
for recovering data on failing hard drives. So that's not a part of the
problem.
Moving the box to the laptop or the server gets similar results, so I don't
think it's a computer problem. It seems to be something inherent in the
drive box somewhere. So a BIOS problem might be the problem. That said, they
show up in device manager under hard drives, just with no volume
information.
In the past it'd be easy. Run Fdisk /mbr and I'd be good to go. But I never
had to do such things to NTFS partitions, so I don't know how to make it
show up.
I guess I'm going to have to do the ugly thing and remove the drive and drop
it in another drive box. I loathe to do that, as getting the drive secured
in the larger box is a major pain in the backside without being able to see
it -- last time it took me about four hours to actually find the right screw
hole on the drive through the screw slots in the case, all told.

 

[Nicole Massey]

I have had a *LOT* of trouble with external drives simply dying. It's
always been the controller, not the physical drive. Every single time
I've been able to get the data off without difficulty by mounting it
somewhere else.

I'll give that a try, then. What a pain it's going to be, though.

 

[Paul]

Definitely not a power supply problem. First, the larger box has a fan
because its power supply is internal, while the smaller one gets warm fairly
quickly. (It's a metal case) I can also hear both drives spinning up -- my
hearing is rather good. And I can read both drives in R-Studio, a program
for recovering data on failing hard drives. So that's not a part of the
problem.
Moving the box to the laptop or the server gets similar results, so I don't
think it's a computer problem. It seems to be something inherent in the
drive box somewhere. So a BIOS problem might be the problem. That said, they
show up in device manager under hard drives, just with no volume
information.
In the past it'd be easy. Run Fdisk /mbr and I'd be good to go. But I never
had to do such things to NTFS partitions, so I don't know how to make it
show up.
I guess I'm going to have to do the ugly thing and remove the drive and drop
it in another drive box. I loathe to do that, as getting the drive secured
in the larger box is a major pain in the backside without being able to see
it -- last time it took me about four hours to actually find the right screw
hole on the drive through the screw slots in the case, all told.

OK, if they show up in Device Manager (devmgmt.msc), you can
also check in Disk Management (diskmgmt.msc) and see if
the partitions on the disk are showing up.

If the partitions show up there, but not in Computer, then the
partition might not be mounting properly. As in, the OS doesn't
recognize the file system. Or, there is a drive letter
conflict, Perhaps if a network share mounts first and takes
a particular letter, the drive can't use that letter.

Another set of concepts are "Take Ownership" and "Foreign File System".
Take Ownership is where the account that owned the drive, isn't the same
as the current operator of the computer. And that search term
"Take Ownership", may address the commands necessary to fix it.
There is even a .reg file (with registry commands), which
will add a right-click menu item, so you can do a Take Ownership
all in one shot.

The Foreign File System, has similar mechanics, only it involves
Dynamic Disks. When you set up a disk in Disk Management, the OS
tries to trick you into making the disks Dynamic. That's part of
a Logical Volume Management scheme. The idea is, if your disk
is "owned" by that kind of driver, you can span a couple disks
together, and make a larger virtual disk. Whereas, if you make
disks Basic, they have no special properties. For least headaches,
all my disks are normally set to Basic.

When using Dynamic disks, if you bring a Dynamic from another
system, and plug it in, they make you go through a little
ceremony. You have to import the "foreign" disk, to make it
part of the pool of dynamic disks on the current machine. And
then, if you take it back to the original machine, you do the
same thing again. After you've done that, it becomes visible.

Those are some examples of "permission-like" issues with disks.

Based on the feedback you've provided, I'd look in Disk Management next.

Paul

 

[Flasherly]

I have had a *LOT* of trouble with external drives simply dying. It's
always been the controller, not the physical drive. Every single time
I've been able to get the data off without difficulty by mounting it
somewhere else.

And I've been hearing that from you as well from others like you. Not
so much controller issues, but, from my interests, drive(s) comprising
the packaged storage capacity. Seems to me, a drive model ID, not to
mention the manufacturer, may become salient to a reception of
reported failure rates when way off skew. Supposedly not to worry,
though, if Dell or HP is marketing one wrapped up in ABS plastic,
having taken a token effort apparently to seal it beyond the cursory
surface inspection. Then, again, why wouldn't an under-2" fan be
sufficient to cool one, so long as it's so unobtrusively quiet hardly
to notice a difference should it fail;- although a minor, insufficient
query perhaps to those that don't see any better compulsion to bother
cooling them.

Thermaltake, with some reputability, came out first with a presence
for docking stations, a cumbersome relative by way of external USB/
SATA HD "half enclosures". Shortly thereafter followed by NewEgg's
entrepreneurial offshoot, the Rosewill brand and it's
counterproduction. At a cost difference being between $50US and
nothing -- what I paid for my first on a 600G HD sale with a free
Rosewill unit thrown in for sweetening -- I believed the argument
quite fairly stated. They're also a happy lot with Rosewill's unit,
for the most reviewing them, although I couldn't offhand say how
Thermaltake's integrity bears upon that, not having researched a
component assembly level. 600G would place mine, my oldest unit of
three, around three years functional, only if for efficient backup and
storage purposes -- how often, really to say, then is it going to be
heavily run. . .for a simple answer readily to qualify: at anything
other and under than bone-grinding sessions with DVDs, I'm ashamed to
admit in some accountability, might anyone then find a docking station
for up terabyte drives all the more accommodating to reason at less
than a heartache.

Although I'd wonder about the future of boutique enclosed drives with
up and coming solid state advancements;- really, what with those
tricky foxes and inventing new and enticing product applications.

 

[John Doe]

OK, if they show up in Device Manager (devmgmt.msc), you can
also check in Disk Management (diskmgmt.msc) and see if the
partitions on the disk are showing up.

From the original post.

"they don't show up in Disk Management at all"

 

[Nicole Massey]

OK, if they show up in Device Manager (devmgmt.msc), you can
also check in Disk Management (diskmgmt.msc) and see if
the partitions on the disk are showing up.

If the partitions show up there, but not in Computer, then the
partition might not be mounting properly. As in, the OS doesn't
recognize the file system. Or, there is a drive letter
conflict, Perhaps if a network share mounts first and takes
a particular letter, the drive can't use that letter.

Another set of concepts are "Take Ownership" and "Foreign File System".
Take Ownership is where the account that owned the drive, isn't the same
as the current operator of the computer. And that search term
"Take Ownership", may address the commands necessary to fix it.
There is even a .reg file (with registry commands), which
will add a right-click menu item, so you can do a Take Ownership
all in one shot.

The Foreign File System, has similar mechanics, only it involves
Dynamic Disks. When you set up a disk in Disk Management, the OS
tries to trick you into making the disks Dynamic. That's part of
a Logical Volume Management scheme. The idea is, if your disk
is "owned" by that kind of driver, you can span a couple disks
together, and make a larger virtual disk. Whereas, if you make
disks Basic, they have no special properties. For least headaches,
all my disks are normally set to Basic.

When using Dynamic disks, if you bring a Dynamic from another
system, and plug it in, they make you go through a little
ceremony. You have to import the "foreign" disk, to make it
part of the pool of dynamic disks on the current machine. And
then, if you take it back to the original machine, you do the
same thing again. After you've done that, it becomes visible.

Those are some examples of "permission-like" issues with disks.

Based on the feedback you've provided, I'd look in Disk Management next.

Yeah, mentioned that in the original post. They show up in Device Manager,
but not in Disk Management. And in Device Manager the volumes have no
volume information, and hitting the populate button just makes the populate
button go away.

 

[Paul]

Yeah, mentioned that in the original post. They show up in Device Manager,
but not in Disk Management. And in Device Manager the volumes have no
volume information, and hitting the populate button just makes the populate
button go away.

OK, I tried that in my Device Manager, and I can see the function
you're referring to.

The primary partitions on a drive, the information pointing to
them is stored in sector 0. If the entire drive was zeroed out,
I would still expect a hard drive item to be in Device Manager.
But, I would also expect the hard drive to be in Disk Management,
with no partitions showing, and just a single rectangle representing
unallocated space. In fact, if the disk was completely zeroed,
Disk Management would be attempting to "write a signature" to the
thing (some kind of numeric identifier).

Since you're not seeing the disks in Disk Management, that means the
OS is ignoring them for some reason. I don't know enough about
the whims of Disk Management, to guess why that might be.

You can get information from sector 0, with the free copy of
PTEDIT32 offered here. In Windows 7, you "Run as Administrator"
so you won't get an "error 5" from the program. On other
OSes, it should work OK without assistance. It lists the
(up to) four primary partitions.

ftp://ftp.symantec.com/public/english_us_canada/tools/pq/utilities/PTEDIT32.zip

You unzip that, and ptedit32.exe is inside. It was originally
a program in the Utilities section of Partition Magic. Partitions
which are non-zero, may have contained a file system. The signature
bytes at the end of sector zero (0xAA55) signify that some OS
has attempted to write valid values in there. If the entire thing
was zeroed, you'd assume a new disk from the factory, or an
"erasing accident".

Now, since Disk Management won't show the drives, then I doubt
the PTEDIT32 menu will list the drives either. So this utility
may be "a bust" in terms of being useful.

Another tool you can try, is TestDisk. This tool takes some
getting used to, as it uses a mostly text based (gotoxy) style
interface. It's an acquired taste. (Note - to quit at any
time, try typing control-C, which is the quit command in another
OS. Any time you don't like the looks of what it's doing,
you can try that option, even if the menu doesn't have "quit"
as an option at that level.)

http://www.cgsecurity.org/wiki/TestDisk_Step_By_Step

What that program can do for you, is scan a hard drive looking
for partitions. It would look on, say, cylinder boundaries,
for things it recognizes as file system headers. If sector
zero was wiped out, TestDisk can recompute the numbers that
belong in the four primary partition entries. And in theory,
that gives access to the data. Most of the time, when I use
TestDisk, I don't allow the tool to write anything. To use
TestDisk, it helps a lot to have "pre-knowledge" of the disk
contents. For example, if TestDisk detects four partitions,
and you know that only three of them are valid, you'd answer
"No" to any attempt to update sector zero with the information.
I've had that happen - TestDisk detected a partition I deleted
long ago, and the definition of that partition overlaps with
another partition. So I could see, by examining the numbers,
that the computed table would be a disaster area. I could
still use the info, by going back and using PTEDIT, and typing
in some numbers for three of the partition entries.

At this point in time, I don't really know what's going on
with your disk. So the two named utilities above, are only
"for a look", to see what you can see. And from a logic
perspective, I don't see how the missing disks, will even
be menu options in them.

Note that, TestDisk also has a "file viewer" feature. If
it locates a partition, there is an option in the interface,
to have it display the files under the root of the partition.
So if I was having trouble with C:, it might show me the
boot.ini file under C: if I was looking there. Being able
to see files that way, is a form of proof that the file
system metadata is intact and makes sense. And that the
origin of the partition, has been correctly deduced.

So the thing that still has me puzzled, is what would stop
an entry from showing in Disk Management. If the disks show
up in the disk drive section of Device Manager, then at least
their identity information is known. And that takes
successful communications with the disk. Reading sector 0
shouldn't be any more stressful, than doing that much.
Windows does keep lots of various identifiers for the disks,
but I fail to see how that would stop Disk Management from
displaying something. Even if the disks had virtually
identical contents, the hardware serial numbers of the disks
will be different enough, to give the disks a unique identity.

Just for kicks (since we're collecting dumb utilities ),
you could also get a copy of dd-0.5 from here. You unzip that,
and run "dd --list" from a command prompt, while viewing the
directory that dd.exe executable is in. That should list
hard drive and partitions, but uses raw device naming conventions.
I'm curious whether the dd command can see the disks or not.
All the above utilities, should be in agreement as to how many
disks exist. I'd be even more shocked, if they differed.

http://www.chrysocome.net/downloads/dd-0.5.zip

If all of this is a dead end, my next stop would be a Linux
LiveCD. Because Linux has equivalents of all these utilities as
well. And Linux wouldn't be quite as picky about displaying stuff.
If the hardware can be detected, you can work on it there.
If the disk is dead, then Linux can't help. But we'll give
Windows another shot for now. I don't know if your
screen reader would work in such an environment. And
if you're not familiar with Linux at all, the learning
curve would be steep.

Paul

 

[Loren Pechtel]

I'll give that a try, then. What a pain it's going to be, though.

It doesn't have to be that much of a pain.

I have a little dohickey I picked up at Fry's. There are two parts to
it, a USB cable that's connected to a square about 2" across and a
power brick feeding a standard Molex connector.

For SATA drives or 2.5" PIDE drives (laptop) you plug the Molex into
the square and plug the correct side of the square into the drive. For
3.5" PIDE drives (desktop) you plug the Molex directly into the drive
instead.

There's no enclosure (this isn't meant as a permanent solution, it's
meant as an easy way to do it when you're going to be changing things
around) so you have to be aware of static and not go bumping it while
you're using it but it makes a very easy way to connect virtually any
drive (it's three connectors support all drives that have been
normally used in anything short of server class machines in at least
10 years) to the system.

 

[Nicole Massey]

The primary partitions on a drive, the information pointing to
them is stored in sector 0. If the entire drive was zeroed out,
I would still expect a hard drive item to be in Device Manager.
But, I would also expect the hard drive to be in Disk Management,
with no partitions showing, and just a single rectangle representing
unallocated space. In fact, if the disk was completely zeroed,
Disk Management would be attempting to "write a signature" to the
thing (some kind of numeric identifier).

Since you're not seeing the disks in Disk Management, that means the
OS is ignoring them for some reason. I don't know enough about
the whims of Disk Management, to guess why that might be.

You can get information from sector 0, with the free copy of
PTEDIT32 offered here. In Windows 7, you "Run as Administrator"
so you won't get an "error 5" from the program. On other
OSes, it should work OK without assistance. It lists the
(up to) four primary partitions.

ftp://ftp.symantec.com/public/english_us_canada/tools/pq/utilities/PTEDIT32.zip

You unzip that, and ptedit32.exe is inside. It was originally
a program in the Utilities section of Partition Magic. Partitions
which are non-zero, may have contained a file system. The signature
bytes at the end of sector zero (0xAA55) signify that some OS
has attempted to write valid values in there. If the entire thing
was zeroed, you'd assume a new disk from the factory, or an
"erasing accident".

Now, since Disk Management won't show the drives, then I doubt
the PTEDIT32 menu will list the drives either. So this utility
may be "a bust" in terms of being useful.

Another tool you can try, is TestDisk. This tool takes some
getting used to, as it uses a mostly text based (gotoxy) style
interface. It's an acquired taste. (Note - to quit at any
time, try typing control-C, which is the quit command in another
OS. Any time you don't like the looks of what it's doing,
you can try that option, even if the menu doesn't have "quit"
as an option at that level.)

http://www.cgsecurity.org/wiki/TestDisk_Step_By_Step

What that program can do for you, is scan a hard drive looking
for partitions. It would look on, say, cylinder boundaries,
for things it recognizes as file system headers. If sector
zero was wiped out, TestDisk can recompute the numbers that
belong in the four primary partition entries. And in theory,
that gives access to the data. Most of the time, when I use
TestDisk, I don't allow the tool to write anything. To use
TestDisk, it helps a lot to have "pre-knowledge" of the disk
contents. For example, if TestDisk detects four partitions,
and you know that only three of them are valid, you'd answer
"No" to any attempt to update sector zero with the information.
I've had that happen - TestDisk detected a partition I deleted
long ago, and the definition of that partition overlaps with
another partition. So I could see, by examining the numbers,
that the computed table would be a disaster area. I could
still use the info, by going back and using PTEDIT, and typing
in some numbers for three of the partition entries.

At this point in time, I don't really know what's going on
with your disk. So the two named utilities above, are only
"for a look", to see what you can see. And from a logic
perspective, I don't see how the missing disks, will even
be menu options in them.

Note that, TestDisk also has a "file viewer" feature. If
it locates a partition, there is an option in the interface,
to have it display the files under the root of the partition.
So if I was having trouble with C:, it might show me the
boot.ini file under C: if I was looking there. Being able
to see files that way, is a form of proof that the file
system metadata is intact and makes sense. And that the
origin of the partition, has been correctly deduced.

So the thing that still has me puzzled, is what would stop
an entry from showing in Disk Management. If the disks show
up in the disk drive section of Device Manager, then at least
their identity information is known. And that takes
successful communications with the disk. Reading sector 0
shouldn't be any more stressful, than doing that much.
Windows does keep lots of various identifiers for the disks,
but I fail to see how that would stop Disk Management from
displaying something. Even if the disks had virtually
identical contents, the hardware serial numbers of the disks
will be different enough, to give the disks a unique identity.

Just for kicks (since we're collecting dumb utilities ),
you could also get a copy of dd-0.5 from here. You unzip that,
and run "dd --list" from a command prompt, while viewing the
directory that dd.exe executable is in. That should list
hard drive and partitions, but uses raw device naming conventions.
I'm curious whether the dd command can see the disks or not.
All the above utilities, should be in agreement as to how many
disks exist. I'd be even more shocked, if they differed.

http://www.chrysocome.net/downloads/dd-0.5.zip

If all of this is a dead end, my next stop would be a Linux
LiveCD. Because Linux has equivalents of all these utilities as
well. And Linux wouldn't be quite as picky about displaying stuff.
If the hardware can be detected, you can work on it there.
If the disk is dead, then Linux can't help. But we'll give
Windows another shot for now. I don't know if your
screen reader would work in such an environment. And
if you're not familiar with Linux at all, the learning
curve would be steep.

As I mentioned in my first message, I know the data is there -- R-Studio
finds most of it in its scan. This tells me the files are still there, just
not being read by the machine. So it's not a problem with files.
Remember also that this was a functioning drive up to the point where an OS
reinstall of XP happened, then XP lost the drives. That's a big question
right there.

 

[Nicole Massey]

It doesn't have to be that much of a pain.

I have a little dohickey I picked up at Fry's. There are two parts to
it, a USB cable that's connected to a square about 2" across and a
power brick feeding a standard Molex connector.

For SATA drives or 2.5" PIDE drives (laptop) you plug the Molex into
the square and plug the correct side of the square into the drive. For
3.5" PIDE drives (desktop) you plug the Molex directly into the drive
instead.

There's no enclosure (this isn't meant as a permanent solution, it's
meant as an easy way to do it when you're going to be changing things
around) so you have to be aware of static and not go bumping it while
you're using it but it makes a very easy way to connect virtually any
drive (it's three connectors support all drives that have been
normally used in anything short of server class machines in at least
10 years) to the system.

I have something similar with my newest drive box -- it leaves those
connectors when you take the back off, so I can do it that way. But getting
it hooked up to the drive means pulling the drive because of clearances, so
it's a minor pain in the backside.

 

[Nicole Massey]

Update: Moving the drive to a known good drive box accomplished nothing --
no read of the drive, no appearance in Disk Management.

 

[Paul]

Update: Moving the drive to a known good drive box accomplished nothing --
no read of the drive, no appearance in Disk Management.

Do you have any means to test, other than Windows ?

Your options would include:

1) BIOS detection via popup boot menu.

USB drives will appear in a popup boot menu (F8, or F11 key) but
the drives may not be detected in the regular BIOS screen. Some BIOS
have a USB specific screen, where drive devices do get listed. But
on my systems here, I'm not really that used to seeing USB devices
while in the BIOS.

The popup boot menu, is where I detect my USB enclosure
(holding either a hard drive or an optical drive, depending
on what I'm doing - I'm too cheap to own multiple enclosures).

2) Detection from a Linux LiveCD.

Some of the recent Linux LiveCDs are of less value for
forensics, due to their tablet like interfaces. So it's hard
to recommend a particular 700MB download, that won't be a pain
to use.

You could try a GParted CD I suppose. There is the program itself,
but there is also a LiveCD version. With the LiveCD version, it
boots Linux and then runs GParted on the screen for you. The pull down
menu should list the disks, and each disk will show partitions. The
sole purpose in this case, is to be able to "detect" a disk, using
something other than Windows. Making changes in there, isn't a
particular objective at this time.

http://en.wikipedia.org/wiki/Gparted

Maybe that's focused enough, to be a worthwhile alternative.
This one is 127MB, so no copies of LibreOffice on it Takes
about seven minutes on my slow broadband connection.

http://gparted.sourceforge.net/download.php

"GParted live is based on Debian live, and the default account
is "user", with password "live". There is no root password,
so if you need root privileges, login as "user", then run "sudo"
to get root privileges."

I test Linux LiveCDs in a virtual machine, and that 127MB one
has a couple problems (problems you wouldn't see on a real machine).
I got around the Xorg window system segfaulting, by using the forcevideo
mode, vesa driver, 1024x768 at 16 bit video mode, and then I could see
the screen. But the paravirtualization support built into Linux,
prevents any hard drives from being seen! What a stupid "innovation".
That spells the end of new releases of Linux in VMs. My virtual environment
is based on VPC2007, but Linux declares it has detected "Hyper-V" which is
not true.

An alternative, is Knoppix, with the Adrienne speech output version.
Then run "lshw" or "gparted" from there, to get some info on disks.

http://mirrors.kernel.org/knoppix/knoppix-cd/ADRIANE_KNOPPIX_V7.0.4CD-2012-08-20-EN.iso

Download won't be finished for a bit, so I'll post back later
as to whether that's a workable option. I expect the same hassles,
but Klaus Knopper is a pretty competent distro builder, and perhaps
he won't "copy and paste" every mistake the Linux community makes :-(

I wish I had a simple "hardware inventory disc", but don't know
of any off hand. There are likely commercial versions, but I'd
be looking for a freebie.

HTH,
Paul

 

[John Doe]

Update: Moving the drive to a known good drive box accomplished
nothing -- no read of the drive, no appearance in Disk
Management.

In accomplished something. It suggests that your drive is bad.
Putting another drive in the same box would add to that.

 

[John Doe]

Connecting the drive directly would be more evidence.

Windows doesn't just decide that a specific hard drive shouldn't
work, regardless of your installation.

 

[John Doe]

Maybe the hard drive jumper was changed or removed?

 

[Paul]

Do you have any means to test, other than Windows ?

Your options would include:

1) BIOS detection via popup boot menu.

2) Detection from a Linux LiveCD.

I burned a Gparted CD and booted the computer and that worked OK.
But the graphical nature of the interface, is going to be a problem
for you. The "Exit" icon which reboots the system, has never
been fixed after all these years. You click on it a few times,
there's no graphical feedback, and you can't tell if the computer
is rebooting or not.

The GParted application, draws a display similar to Disk Management,
and the pull-down menu allows selecting 1 of N drives. But this
would work better for someone who was sighted.

*******

I next burned the Knoppix CD (704MB).

To get the visually impaired interface to run, you can try

adriane fromhd=/dev/sr0

The command line argument I'm using there, solves a problem specific
to my machine. I have a Linux partition, and if I don't tell the
stupid CD based distro where to find Linux, it goes off and grabs
my old 5.3 partition. Telling it where the CDROM is located (/dev/sr0)
fixes that. I've had problems with Knoppix like this in the past,
only instead the problem was an inability to find that partition.
Now, it's gone the other way, grabbing the wrong partition.

Anyway, "adriane" gives a text based menu, and you can use cursor
keys.

The first problem with it, is the speech output volume is zero at
startup. So unless you had a screenreader that can automatically
read that text, you're in trouble. If you cursor down the right
number of times, you get to a settings, and in there is a volume
control. I made a one step change in the volume (from "72" to "73"
or so) and suddenly I was hearing speech output in English,
with a slight German accent. But the material spoken, tends to be
infuriating. On a screen full of text, only certain parts
of the screen get spoken. Which means you may not get a very
good "image" of what's there.

I also noticed an item for some kind of screen reader support,
but didn't look into details.

As far as software contents on the 704MB disc, it's missing a
copy of "lshw" and a copy of "disktype", two tools I might have
used for listing the disks. And proving a viable file system
was present.

Without lshw, you can use the "stab in the dark" approach to reading
out the disks. In a terminal, you'd type

sudo fdisk /dev/sda
p <--- this prints the partition list
q <--- this quits the program

If there wasn't an actual /dev/sda, then presumably the program would
instead exit with an error. You would then move on to the next disk.

sudo fdisk /dev/sdb

Disks can exist as /dev/sda or as /dev/hda. And the letter on the end
is bumped up for each successive disk. Individual partitions are
designated by a number added to the end. So /dev/sdb4, would be the
second disk drive, and fourth partition on the end. Partitions in that
case, can be primary or logical.

*******

I'm going to try this one next. See if it has any additional tools.

https://sourceforge.net/projects/sy...6/3.0.0/systemrescuecd-x86-3.0.0.iso/download

Paul