D: Drive Disappeared

[~misfit~]

I have had corrupt disks a lot more than most people, and I
have always recovered.

Heh! That sentence says a lot.

Probably because most people don't use removable bays.

 

[~misfit~]

not necessarily, think "intermittent connection".

Heh! I'm glad I read your reply before hitting reply myself. That is exactly
what I was going to say.

I think you're right, he's in love with his tray/bay set up. If his account
of the history of his OS is anything to go by, these are probably old
devices anyway. I threw away a box full of them a while back as they were
giving intermittant problems. They are old, outdated technology. Most are
only internally wired with 40 wires and have insufficient cooling for the
HDD (if any).

Before USB2/Firewire, if you wanted a removable drive, these were the way to
go. However, in the last 4 or 5 years they've been replaced by far more
relaible systems.

 

[Bob]

Heh! That sentence says a lot.

Probably because most people don't use removable bays.

If this keeps up then I will put the boot disk on directly and use the
removables for backup only.

The problem I have with the hypothesis that the removable bay is the
cause of the problems I am experiencing is that these problems only
happen when I shut down Windows. Otherwise I have had absolutely NO
problems whatsoever.

You would expect that if the removable bays were at fault, I would see
data corruption at any time, not just at shutdown.

However, although I am reluctant to believe that the removable bay is
the culprit, I have not ruled it out. Since I have about exhausted
every other possibility, and the corruption does still occur albeit
not as frequently, I will probably eliminate the removable bay for the
boot disk.

 

[Bob]

The file you mention is a Pinnacle one having to do with a
PCI card of theirs.

I think I installed Pinnacle once but I uninstalled it. Why would this
be left over?

I did not install any PCI card.

However, since you get the same error
which you apparently fix with chkdsk doesn't that mean the
drive is heading out?

I have three identical WD 80GB hard drives in three removable trays
which I rotate for backup purposes. They all act the same.

 

[Bob]

Heh! I'm glad I read your reply before hitting reply myself. That is exactly
what I was going to say.

What you two are overlooking is the simple fact that if it were an
"intermittent connection" they why is it corrupting the disk only when
I shut down Windows? It should cause disk corruption anytime.

I think you're right, he's in love with his tray/bay set up.

I use it to rotate three identical drives for backup purposes.

If his account of the history of his OS is anything to go by, these are probably old
devices anyway.

You are very wrong. I have three identical WD 80GB drives which I
bought last year. They are less than 1 year old.

internally wired with 40 wires and have insufficient cooling for the
HDD (if any).

I am using ATA-133 80-wire ribbon cables. The trays are rated by
Kingwin as ATA-133.

MBM5 tells me the drive runs at 29C.

Before USB2/Firewire, if you wanted a removable drive, these were the way to
go. However, in the last 4 or 5 years they've been replaced by far more
relaible systems.

What might those be?

 

[kony]

If this keeps up then I will put the boot disk on directly and use the
removables for backup only.

The problem I have with the hypothesis that the removable bay is the
cause of the problems I am experiencing is that these problems only
happen when I shut down Windows. Otherwise I have had absolutely NO
problems whatsoever.

You would expect that if the removable bays were at fault, I would see
data corruption at any time, not just at shutdown.

There is some logic to your argument, but how far has it
gotten you towards resolving this? The obvious problems are
easily rectified but when the problem is not so obvious then
a methodical reduction of variables can be extremely useful.

If the solution was going to reach out and grab you, it
would've done so already. We're not telling you to throw
the removable bays away (yet), only to disconnect them for
the time being.

However, although I am reluctant to believe that the removable bay is
the culprit, I have not ruled it out. Since I have about exhausted
every other possibility, and the corruption does still occur albeit
not as frequently, I will probably eliminate the removable bay for the
boot disk.

Did you ever determine whether your system properly supports
48 bit LBA? It's kinda important for us to know
comprehensively what you've done to troubleshoot so far.

 

[kony]

I think I installed Pinnacle once but I uninstalled it. Why would this
be left over?

Because it's not uncommon for uninstallers to not uninstall.
Further, if you install a driver that copies a file to any
given location then just pull the card, if the driver didn't
have an uninstaller (or at least a thorough one), the files
it installed remain... and depending on where they are, may
be loaded each time windows boots due to their location, or
registry entry.

I did not install any PCI card.


I have three identical WD 80GB hard drives in three removable trays
which I rotate for backup purposes. They all act the same.

So you are certain you have been setting the jumpers
correctly for single drive vs master w/slave?

 

[kony]

What you two are overlooking is the simple fact that if it were an
"intermittent connection" they why is it corrupting the disk only when
I shut down Windows? It should cause disk corruption anytime.

Well since you won't do a clean install of Win2kSP4, we
might as well focus on the other parts and rule them out.

however, you really do need to try a clean install. It'll
probably speed up the system too, like a free hardware
upgrade.

I use it to rotate three identical drives for backup purposes.

backups are very important, but on the other hand, your
backups wouldn't be nearly so important if you didn't keep
losing entire drive volumes. If you need to keep swapping
'em, perhaps just leave the cable sitting in the open bay...
put a spot of velcro on it and on the bay frame so it
doesn't fall backwards into the abyss when disconnected from
drives.

You are very wrong. I have three identical WD 80GB drives which I
bought last year. They are less than 1 year old.

I can't help but wonder something though, we're you claiming
WD drives were great previously? It just strikes me as a
little odd, I mean, what could you be basing that on??

I am using ATA-133 80-wire ribbon cables. The trays are rated by
Kingwin as ATA-133.

I'm pretty sure I told you several months ago that Kingwin
makes questionable products. Maybe they're fine, maybe
they're even a great value IF they work fine, but you really
need to get those bays out of the system... and likewise,
eliminate all the other variables possible too.

MBM5 tells me the drive runs at 29C.

You might have MBM5 set wrong, no drive idles that cool in a
removable bay unless it has relatively noisey (high RPM)
fans running. Perhaps yours does, but (LOL), that'd be yet
another reason to get rid of them, to reduce noise.

What might those be?

Anything that doesn't rely on a low cost slide-in mating
connector. IOW, where the mechanical connections aren't
subject to repeated insertion, removal, or have contacts
better suited to it. High precision locking 40 pin
connectors would put the Kingwin product out of budgetary
bounds, I'm sure they did what they could at the price point
but they are a concern... even if they work ok right now,
the repeated insertion/removal cycles can be a problem
later.

 

[Timothy Daniels]

I'm pretty sure I told you several months ago that Kingwin
makes questionable products. Maybe they're fine, maybe
they're even a great value IF they work fine, but you really
need to get those bays out of the system... and likewise,
eliminate all the other variables possible too.

I have two Kingwin removable trays for my backup
hard drives and they haven't had a problem in the
2 1/2 years that I've had them. Who says they're
"questionable products" - you?

As for removable hard drive trays, various student
computer labs at UCLA use them routinely for classes
in which they are removed and inserted for each class
so that a student can have his/her own hard drive for
the duration of the course, and the techs tell me that
they've had no problems with them in the years that
they've been in use.

*TimDaniels*

 

[Bob]

If the solution was going to reach out and grab you, it
would've done so already.

I wish that were the case.

We're not telling you to throw
the removable bays away (yet), only to disconnect them for
the time being.

I am trying to isolate what I do that causes the boot disk to become
corrupt. I can go days and nothing happens. Then it becomes apparently
becomes corrupted for no apparent reason - but always when I reboot,
according to Event Viewer. Most of the time CHKDSK reports nothing
having been fixed.

The only time I have seen the disk become corrupt other than at reboot
is when I run either Defrag (Win2K version) or CHKDSK /F /R. Then it
will sometimes BSOD after running for a while. The tray has 3 40mm
fans and I have never seen it get the least bit hot, not even with
CPU/Disk intensive apps like DVD Shrink.

Did you ever determine whether your system properly supports
48 bit LBA?

Yes.

The mainboard is less than a year old. It is a Microstar Intel
chipset.

One thing I have yet to determine is whether there are more instances
of corrupt disks with one particular disk out of the three I use.

Another thing is I have all sorts of residue crap left over from
installs of junk software. I really should do at least an IPU (In
Place Upgrade) before I do anything. But I keep hoping I will find the
culprit directly.

The reason I lean towards a Win2K problem is I drastically reduced the
incidence of corruptions when I told Win2K to clean up the pagefile.
Some fool app must have changed that because somewehre along the way I
noticed that the shutdown procedure was way quicker than usual. I
thought nothing of it until recently when I changed the Registry entry
back to its default. Now the corruption only occurs infrequently.

Maybe there is another shutdown parameter that needs to be reset and
the problem goes away. Supposedly an IPU will take care of that
because the system part of the Registry is supposedly rebuilt with an
IPU.

 

[Bob]

Because it's not uncommon for uninstallers to not uninstall.
Further, if you install a driver that copies a file to any
given location then just pull the card, if the driver didn't
have an uninstaller (or at least a thorough one), the files
it installed remain... and depending on where they are, may
be loaded each time windows boots due to their location, or
registry entry.

I do not have any cards - everything I need is integrated into the
main board.

So you are certain you have been setting the jumpers
correctly for single drive vs master w/slave?

I am using Cable Select on all of them. I have the correct ribbon
cables too. I was going to use round cables but decided against it
because I wanted the most robust solution for anything that was
associated with the removable bay.

I have both removable drives on IDE channel zero, so the disks must be
cable selected if I am going to swap them around without changing
jumpers.

One other thing I have not considered and that is I could have a
virus. I have swept the entire system with Avast Home Edition and
Ad-Aware and nothing shows up.

 

[Bob]

Well since you won't do a clean install of Win2kSP4, we
might as well focus on the other parts and rule them out.

I have done IPUs before and may do one again. In the past when
something screwed up Win2K, the IPU cured it.

however, you really do need to try a clean install.

If Vista is going to be delayed for the usual 1-2 years that Microsoft
is notorious for, then I may just do that.

With removable bays it is not all that difficult. I take some screen
shots of things like Add/Remove Programs to remind me what I had on
the old system. Also the utility RegCleaner gives a manifest of
intalled items.

backups are very important, but on the other hand, your
backups wouldn't be nearly so important if you didn't keep
losing entire drive volumes.

It's not all that bad. It only happens once in a while and then only
when I reboot (except when I defrag)

I can't help but wonder something though, we're you claiming
WD drives were great previously? It just strikes me as a
little odd, I mean, what could you be basing that on??

My experience and that of employers.

I'm pretty sure I told you several months ago that Kingwin
makes questionable products.

Not me. If you have anything on that, I need to see it. I have assumed
that because I could not find anything wrong with Kingwin, it was not
a problem.

How it could be these ATA-133 drives put a strain on removable bays
that has not yet been reported.

I can easily move one drive to a direct connection and use the other
two for backups.

You might have MBM5 set wrong, no drive idles that cool in a
removable bay unless it has relatively noisey (high RPM)
fans running.

Everest reports the same SMART temp.

Perhaps yours does, but (LOL), that'd be yet
another reason to get rid of them, to reduce noise.

The removable bays/trays are not noisy. I can hear them but they are
not noisy.

High precision locking 40 pin
connectors would put the Kingwin product out of budgetary
bounds

The connectors are the Centronics kind. The tray is cammed into the
bay by the front lever and then locked in place with a keylock. Only
when the lock is tight will power be applied to the tray.

 

[Bob]

I have two Kingwin removable trays for my backup
hard drives and they haven't had a problem in the
2 1/2 years that I've had them. Who says they're
"questionable products" - you?

As for removable hard drive trays, various student
computer labs at UCLA use them routinely for classes
in which they are removed and inserted for each class
so that a student can have his/her own hard drive for
the duration of the course, and the techs tell me that
they've had no problems with them in the years that
they've been in use.

That's good to hear. And it is typical of the reports I have seen.

The only possible issue is the use of the latest ATA-133 hard disks.

 

[Bob]

Here's something else. If I reboot at least once every day, typically
in the morning, I "never" get any disk corruption - at least none that
I can recall.

I am going to test that this week by booting once in the morning and
once in the afternoon.

 

[~misfit~]

Everest reports the same SMART temp.

I would still suspect that temp reading. SMART is a closed, proprietary
format and is implimented differently by different drive manufacturers.

I bought my first new WD HDD for many years last week, a 200GB EIDE that was
a clearance product.

I've seen HDD failure, short life-span and data loss from over-heating HDDs
so I have forced cooling and constantly monitor all drive's temp, even have
it plotted on a graph so I can see the last 12 hours at any one time.

I have three drives in this system, a Maxtor, a Seagate and the WD. All
7,200 rpm, all "ATA133". Since I've installed the WD I've changed it's
physical position a few times as I'm convinced that the reported temp is
wrong. However, it's consistantly wrong and does increase when defragging
the drive or doing big file copies. It increases by 10°C+ at these times
where the other drives only go up a few degrees. It's the only drive that
exhibits large spikes in the graph.

The reason I'm convinced the WD drive is reporting the wrong temp? According
to the on-drive sensor it idles at below ambient! With the fans I have, the
Seagate will idle at ambient +5° and the Maxtor at ambient +8°. The WD idles
at ambient -2°. I might have to buy a few more of them and use them to cool
my case. <g>

I have tried at least 5 different utilities to read the reported drive temp
and they all report the WD HDD as doing the impossible. I think I'll stick
to Seagate in future, they've been good to me for many years. The only
reason I bought the WD was my local supplier was getting rid of WD PATA
stock at heavilly discounted prices as they weren't moving.

I know that this is one drive only and had basically considered it an
abberation, the reported temp is useful as a guide-line as it *does* report
increases during periods of heavy useage that accuratelty correspond with
expected increases. However, on hearing that your drives are reporting 29°
in trays I'm inclined to think that the whole WD range may have a
significantly different implimentation of SMART and under-report temps.
Unless you live in Iceland and keep your boxen outside?

 

[kony]

I have two Kingwin removable trays for my backup
hard drives and they haven't had a problem in the
2 1/2 years that I've had them. Who says they're
"questionable products" - you?

As for removable hard drive trays, various student
computer labs at UCLA use them routinely for classes
in which they are removed and inserted for each class
so that a student can have his/her own hard drive for
the duration of the course, and the techs tell me that
they've had no problems with them in the years that
they've been in use.

*TimDaniels*

There's more than one model, pretty important to clarify
this. Which ones are they using?

I wonder though if you're just making this up as you go
along, as it seems pretty likely... who goes around polling
all the techs about their external enclosures? Never mind,
I'm sure the tool that you are, you would do something like.
Ironically enough I'll agree that is evidence of that
particular model of external bay seeming to do well. Now
all we need is similar reports of all their other products
else we can only buy that one particular model.

I hope it was the cheapest one they sell, because those are
the products that sell in the highest volumes. Then again,
if they've had "no problems with them in the years...", then
that is an old product, not necessarily what anyone who
bought today or in the near past would have.

 

[kony]

I have done IPUs before and may do one again. In the past when
something screwed up Win2K, the IPU cured it.

So what relevance does that have, when you still have the
problem?

If Vista is going to be delayed for the usual 1-2 years that Microsoft
is notorious for, then I may just do that.

So if Vista were released today, then what? You have a
reasonable expectation that you have something to gain from
Vista, or are you just dying to try it, or ???

For all we know, Vista could have quite a few bugs when
first released. Significant ones. Win2k is mature and has
several service packs out, it (or XP for those who need a
friendly helpful bloated OS) will still be the best choice
right after Vista is released and until Vista has matured
through several series of patches.

With removable bays it is not all that difficult. I take some screen
shots of things like Add/Remove Programs to remind me what I had on
the old system. Also the utility RegCleaner gives a manifest of
intalled items.

I suspect your 3rd party software is only hurting. There is
no need to use a Regcleaner.

It's not all that bad. It only happens once in a while and then only
when I reboot (except when I defrag)

Average failure rate for a HDD is a few years, is it that
seldom? If your data is lost in 1/2 as many years, that is
a substantial problem. The way it reads, you're losing it
several times in the past few months.

How it could be these ATA-133 drives put a strain on removable bays
that has not yet been reported.

Are they ATA133? I would've expected WD 80GB drives to be
ATA100.

I can easily move one drive to a direct connection and use the other
two for backups.


Everest reports the same SMART temp.

Very curious, perhaps there is an external sensor so it's
not taking the temp of the same (parts) as with other
drives. Either that, or they're both wrong.

The removable bays/trays are not noisy. I can hear them but they are
not noisy.

Ok, so long as you're happy.
It would have to be relative though, hearing something is
certainly more noisey than not hearing it... and many cannot
hear any HDD cooling fans due to lower RPM or only a passive
intake.

The connectors are the Centronics kind. The tray is cammed into the
bay by the front lever and then locked in place with a keylock. Only
when the lock is tight will power be applied to the tray.

Well then your idea above about "not yet reported" is
wrong, see;

http://groups.google.com/group/24ho...s+problems&rnum=2&hl=en&#doc_9130fab954abb260

or if that wraps, use http://tinyurl.com/qspxd

What would you expect though, that if a lot of people
reported a problem to Kingwin that they'd be looking to
spread the word that people should avoid their product? I
don't think so.

 

[Timothy Daniels]

I wonder though if you're just making this up as you go
along, as it seems pretty likely... who goes around polling
all the techs about their external enclosures?

No one mentioned external enclosures. We were
discussing removable hard drive trays, remember?

As for polling "all the techs", I polled 2 techs - the lead
lab tech and his manager. They both said that there hadn't
been any problems with the removable trays or the drives
in them. They've used 2 brands that I've seen, depending
on which lab they were used in. One brand was Kingwin.
The trays relied on the case fan for airflow. The HD's
rotational speed was 7,200 rpm, but I can't recall whether
the channel rate was 100MHz or 133MHz. It doesn't
matter, though, as the cable and connector are the same.

Bottom line - only *you* are of the opinion that Kingwin's
removable HD trays (what Kingwin calls "mobile trays")
are "questionable".

*TimDaniels*

 

[~misfit~]

No one mentioned external enclosures. We were
discussing removable hard drive trays, remember?

As for polling "all the techs", I polled 2 techs - the lead
lab tech and his manager. They both said that there hadn't
been any problems with the removable trays or the drives
in them. They've used 2 brands that I've seen, depending
on which lab they were used in. One brand was Kingwin.
The trays relied on the case fan for airflow. The HD's
rotational speed was 7,200 rpm, but I can't recall whether
the channel rate was 100MHz or 133MHz. It doesn't
matter, though, as the cable and connector are the same.

Actually it *does* matter. It shows that you don't really know what you're
talking about. A "Channel rate" for HDDs measured in MHz? Ther's no such
beastie. The IDE controller runs at a nominal 33MHz but that can't be what
you're talking about. Perhaps you mean ATA100 or ATA133? The transfer rate?
Measured in MB/s? Perhaps that's what you mean. Hard to say.

I'd hate to run a modern HDD running at 7,200 rpm in a passively cooled
"mobile tray". In fact, I simply wouldn't do it unless the tray/bay system
was made out of heavy aluminium and made good enough contact with the
HDD/the other component to act as a heatsink.

The biggest cause of failure with modern HDDs is lack of cooling. This
especially applies to people upgrading and sitting their shiney new 7,200rpm
drive in the same place as their old 5,200rpm drive that only gave off 40%
of the heat the new drive does. Then they complain, rubbish the drive and
the manufacturer when the drive only just lasts out it's warranty period.

 

[yuenkitmun]

Did you try scandisk? Remember to select "Scan for and attemp
recovery of bad sectors".

Do it a few times and see if the bad sector count keeps on increasing
If it does, trash the disk

Also, try booting from a Live CD such as Knoppix and see if it ca
detect the drive