USB drive gets assigned a letter sometimes and sometimes not

[Yousuf Khan]

I have a 500GB external drive, connected through USB2. Recently it's
started not showing up as a drive letter under XP, until you restart it
a few times. This only happens if the drive has been turned off and you
turn it on while the computer is running. If you turn it on before
turning the computer on, then it always works properly.

When I turn it on while the computer is already operating, the device
does show up in the USB peripherals list properly as a "mass storage
device", it just doesn't get assigned a drive letter sometimes. I use
the Microsoft UVCView utility to determine if it shows up in the USB chain.

Yousuf Khan

 

[Rod Speed]

I have a 500GB external drive, connected through USB2. Recently it's
started not showing up as a drive letter under XP, until you restart it a few times. This only happens if the drive
has been turned off and
you turn it on while the computer is running. If you turn it on before
turning the computer on, then it always works properly.

When I turn it on while the computer is already operating, the device
does show up in the USB peripherals list properly as a "mass storage
device", it just doesn't get assigned a drive letter sometimes. I use the Microsoft UVCView utility to determine if it
shows up in the USB chain.

I dont get that effect with my 1TB externals with XP
which I always plug in after the system is running.

What happens with a linux live CD ?

 

[Yousuf Khan]

I dont get that effect with my 1TB externals with XP
which I always plug in after the system is running.

What happens with a linux live CD ?

I haven't really tested the effect recently under Linux yet, as this
external drive has mostly stuff that's used under Windows more than it's
used under Linux. In the past, this drive used to show up as a drive
under Linux pretty quickly. But then again, in the past this drive used
to show up as a drive under XP pretty quickly too.

Yousuf Khan

 

[Rod Speed]

Yousuf Khan wrote

Rod Speed wrote

I haven't really tested the effect recently under Linux yet, as this
external drive has mostly stuff that's used under Windows more than
it's used under Linux. In the past, this drive used to show up as a
drive under Linux pretty quickly. But then again, in the past this
drive used to show up as a drive under XP pretty quickly too.

If you try it with a linux live CD, that distinguishes between the xp install
getting corrupted and producing the problem and a hardware problem.

 

[mscotgrove]

Yousuf Khan wrote


If you try it with a linux live CD, that distinguishes between the xp install
getting corrupted and producing the problem and a hardware problem.

I have the feeling that mapped network drives can sometime confuse the
system if they have potentially the same drive letter.

Normally though, I plus a USB drive into XP and it can be seen.

Michael

 

[Rod Speed]

(e-mail address removed) wrote

I have the feeling that mapped network drives can sometime
confuse the system if they have potentially the same drive letter.

Its usually this problem
http://support.microsoft.com/kb/925196

 

[Yousuf Khan]

If it doesn't show up then you may have to force
a rescan for new USB devices.

Nothing seems to be working with this anymore. I'll try the drive on a
different computer and see if it shows up there. I'm suspecting that I'm
coming close to the limits of the Nvidia chipset on my motherboard. I've
had problems with internal hard drives and now external hard drives on
this system. I've already replaced the power supply and it seems to only
push the problems further down the chain.

Yousuf Khan

 

[Yousuf Khan]

Yousuf Khan wrote


If you try it with a linux live CD, that distinguishes between the xp install
getting corrupted and producing the problem and a hardware problem.

Well, I've got a Linux partition already installed on the system, so I
don't need to do it through a live CD. Anyways, I tried it today under
Linux and the problem occurs under Linux as well.

In fact, starting today, the drive is having a hell of a time even
starting at boot time. I'm going to go try it on a different machine to
see if works there.

Yousuf Khan

 

[Rod Speed]

Yousuf Khan wrote

Ato_Zee wrote

Nothing seems to be working with this anymore. I'll try the drive on a
different computer and see if it shows up there. I'm suspecting that
I'm coming close to the limits of the Nvidia chipset on my motherboard.

Its more likely that something has died.

That certainly isnt something that everyone with Nvidia chipsets sees.

I've had problems with internal hard drives and now external hard drives on this system. I've already replaced the
power supply and it seems to only push the problems further down the chain.

Not clear what you mean by that last.

 

[Rod Speed]

Yousuf Khan wrote

Rod Speed wrote

Well, I've got a Linux partition already installed on the system, so I
don't need to do it through a live CD. Anyways, I tried it today under
Linux and the problem occurs under Linux as well.

In fact, starting today, the drive is having a hell of a time even
starting at boot time. I'm going to go try it on a different machine
to see if works there.

Yeah, thats the best thing to try in those circumstances.

They arent that reliable.

 

[Franc Zabkar]

I have a 500GB external drive, connected through USB2. Recently it's
started not showing up as a drive letter under XP, until you restart it
a few times. This only happens if the drive has been turned off and you
turn it on while the computer is running. If you turn it on before
turning the computer on, then it always works properly.

When I turn it on while the computer is already operating, the device
does show up in the USB peripherals list properly as a "mass storage
device", it just doesn't get assigned a drive letter sometimes. I use
the Microsoft UVCView utility to determine if it shows up in the USB chain.

Yousuf Khan

Is it possible that UVCView is reporting the existence of the external
USB-SATA/PATA bridge, and could it be that this occurs before the hard
drive has spun up to speed?

Does the drive's SMART report suggest any degradation in its spin-up
time?

- Franc Zabkar

 

[Yousuf Khan]

Yousuf Khan wrote


Its more likely that something has died.

That certainly isnt something that everyone with Nvidia chipsets sees.


Not clear what you mean by that last.

My system is pretty well loaded with drives. It's got 4 hard drives and
two optical drives internally.

Yousuf Khan

 

[YKhan]

Yousuf Khan wrote

Yeah, thats the best thing to try in those circumstances.

They arent that reliable.

Well, I've now tried it on my laptop and it's not showing up in there
either. So it looks like it's well and truly gone. So perhaps those
early indications when it refused to startup under a running OS were
the first indications of an imminent full-blown death. Funny thing is
that this was one of the few drives that returned SMART info through
USB (at least to HD Sentinel), and there were zero issues recorded in
SMART for this drive. It said everything was perfect up until the end.
Yet another drive that died before my drive that's been giving SMART
warnings for years that is still running.

Yousuf Khan

 

[Rod Speed]

My system is pretty well loaded with drives. It's got 4 hard drives
and two optical drives internally.

It wasnt clear what you meant by the 'it seems to only push the problems further down the chain' bit.

 

[Rod Speed]

YKhan wrote

Well, I've now tried it on my laptop and it's not showing
up in there either. So it looks like it's well and truly gone.

The drive itself may well be fine tho. You can get new interfaces very cheaply.

So perhaps those early indications when it refused to startup under
a running OS were the first indications of an imminent full-blown death.

Yes, very likely.

Funny thing is that this was one of the few drives that returned
SMART info through USB (at least to HD Sentinel), and there
were zero issues recorded in SMART for this drive.
It said everything was perfect up until the end.

Its not very likely the drive itself has died, its most likely
just the external interface or just the external power.

Yet another drive that died before my drive that's been
giving SMART warnings for years that is still running.

SMART is just extra information on the state of the drive,
it was never claimed to be a perfect predictor of drive
failure. That isnt even possible. There are plenty of drive
failures that give no warning of failure, like power supply failure.

There are also plenty of drive failures that do give an indication of
the drive going bad before it become unusable, most obviously with
increased spin up times and with pending and reallocated sectors.

 

[YKhan]

YKhan wrote


The drive itself may well be fine tho. You can get new interfaces very cheaply.

I may try to remove the drive from its casing and try to attach it
directly to an IDE or SATA cable (not sure what it is internally). I
remember from a few years back that a friend of mine who had the exact
same external drive, which konked out much earlier that it wasn't an
easy thing getting the drive out of the enclosure.

Its not very likely the drive itself has died, its most likely
just the external interface or just the external power.

Actually no, the external interface is still visible. Scanning the USB
info through UVCView shows the "LaCie" string identifier, and all
parameters are visible too. Just no communication with the drive
behind it.

Now it's possible that the external interface is only broken on the
drive end of it, but is still working perfectly from the USB connector
end of it, but it seems a little unlikely that the problem would be so
peculiar.

Now in the olden days they used to put drives inside freezers to try
to get them to work. Does that still have any relevance anymore?

Yousuf Khan

 

[Rod Speed]

YKhan wrote

I may try to remove the drive from its casing and try to attach it
directly to an IDE or SATA cable (not sure what it is internally).

I remember from a few years back that a friend of mine who had
the exact same external drive, which konked out much earlier that
it wasn't an easy thing getting the drive out of the enclosure.

It can be a bit tricky opening the case, but
other than that its as easy as falling off a log.

I dont even bother with external enclosures myself, I use a USB/eSATA
docking station and just drop the bare drive into that like dropping in a
flash card. In fact the docking station takes drives and flash cards too.

I use my external drives mostly for overflow from the PVR so its
a lot cheaper to just buy another bare 1.5TB drive and drop it into
the docking station than it is to buy an external drive and that way
you get the full 3 or 5 year warranty on the drive itself too.

Actually no, the external interface is still visible.

OK, then it may just be the power supply thats died.

Scanning the USB info through UVCView shows the "LaCie"
string identifier, and all parameters are visible too. Just no
communication with the drive behind it.

That may be because the 12V rail from the power
supply has died so the drive cant spin up anymore.

Can you feel the drive spinning up ?

Now it's possible that the external interface is only broken on the drive
end of it, but is still working perfectly from the USB connector end of it,
Yep.

but it seems a little unlikely that the problem would be so peculiar.

Its one obvious way for it to fail.

Now in the olden days they used to put drives inside freezers to try
to get them to work. Does that still have any relevance anymore?

Yes. That was done when the drive has a cracked trace
or dry joint. The much lower temperature can see that
cracked trace or dry joint work again while its still very cold.

Important to put it in a plastic bag so you dont get ice or condensation on the drive etc tho.

 

[Yousuf Khan]

It can be a bit tricky opening the case, but
other than that its as easy as falling off a log.

Not this one, it was one of those special sleek designs. Just barely
enough space to put the drive inside and that's it. Doesn't even have a
fan. Not that a fan would really matter, I've seen drives die horrible
deaths inside fan enclosures too.

I dont even bother with external enclosures myself, I use a USB/eSATA
docking station and just drop the bare drive into that like dropping in a
flash card. In fact the docking station takes drives and flash cards too.

Normally I wouldn't have either, except this was given as a birthday
present, by the same friend whose own drive had died. Of course he gave
me this as a present before he had his problems with his own. Mine
lasted 3 times longer than his, but at least he got a warranty-based
replacement, which I'm not going to get.

The guy still hasn't learned his lesson from this fiasco, he still
prefers external drives, even got himself an external DVD writer.

OK, then it may just be the power supply thats died.


That may be because the 12V rail from the power
supply has died so the drive cant spin up anymore.

It's one of those power converter bricks like the type you see attached
to most laptops.

Can you feel the drive spinning up ?

No, but I think I've heard it clicking. Not sure though, noisy environment.

Yes. That was done when the drive has a cracked trace
or dry joint. The much lower temperature can see that
cracked trace or dry joint work again while its still very cold.

I'll give it another shot then.

Important to put it in a plastic bag so you dont get ice or condensation on the drive etc tho.

Well, it's already inside an enclosure though isn't it?

Yousuf Khan

 

[Yousuf Khan]

Is it possible that UVCView is reporting the existence of the external
USB-SATA/PATA bridge, and could it be that this occurs before the hard
drive has spun up to speed?

Yeah, well that's exactly what it's doing. After a few minutes of
retrying, the interface comes up with the yellow bang in device manager,
saying there was a problem. Two yellow bangs come up actually, one under
the USB tree where an error code 10 shows up against the USB mass
storage drive device. And another yellow bang shows up under SCSI & RAID
controllers tree, where some virtual SCSI controller shows up with error
code 37.

Does the drive's SMART report suggest any degradation in its spin-up
time?

I've been monitoring this drive intently for a long time. SMART doesn't
even report anything now, since it's not working. In the past when it
was working, SMART claimed everything was perfect on it. However, I was
suspicious about this drive in the past because I could see that its
temperature was sky-high, reaching into the 50°C range after long usage.

Yousuf Khan

 

[Rod Speed]

Yousuf Khan wrote

Rod Speed wrote

Not this one, it was one of those special sleek designs. Just barely enough space to put the drive inside and that's
it.

Its still just a matter of knowing how to open it. Surely
you can see that with that other one that did get opened ?

Doesn't even have a fan. Not that a fan would really matter, I've seen drives die horrible deaths inside fan
enclosures too.

A properly designed fanned enclosure works fine. Thats what the PC case is.

Normally I wouldn't have either, except this was given as a birthday
present, by the same friend whose own drive had died. Of course he
gave me this as a present before he had his problems with his own.

Yeah, they are sexy enough looking that they do appeal to some.

I wont use them for 3.5" drives, only for 2.5" drives.

Mine lasted 3 times longer than his, but at least he got a warranty-based replacement, which I'm not going to get.

Yeah, thats the other reason I always use bare drives, to get the full warranty.

The guy still hasn't learned his lesson from this fiasco, he still
prefers external drives, even got himself an external DVD writer.

They arent so much of a problem because heat isnt a problem with them.

I use externals just for the convenience. Much easer to drop another
drive into the docking station than fart around with internal drives.

I dont like internal racks, the better ones cost more than
even the 1.5TB drive you are putting in the damned things.

It's one of those power converter bricks like the type you see attached to most laptops.

OK, you should be able to check its outputs with a multimeter.

No, but I think I've heard it clicking. Not sure though, noisy environment.

I'll give it another shot then.

Well, it's already inside an enclosure though isn't it?

Condensation can still get in there.