Bad block

[Terry Pinnell]

A routine check of my Event Viewer shows the entry "The device,
\Device\Harddisk0\D, has a bad block."

Only a day before, I'd run a chkdsk d:, and that didn't report
anything. WinXP desktop was up when I returned to PC after chkdsk had
finished. I deliberately did not use the /f switch, as I wanted to see
if there were any problems.

So:

Assuming that 'bad block' hadn't just arisen, how do I square the two?

And what if anything should I now do please?

 

[Richard Urban]

After you are done running chkdsk c: /f, upon reboot look in the event
viewer for an entry called winlogon. That is the result of the chkdsk, right
there for your pleasure.

--
Regards,

Richard Urban
Microsoft MVP Windows Shell/User

If you knew as much as you think you know,
You would realize that you don't know what you thought you knew!

 

[Ted Zieglar]

Go to the web site of the manufacturer of your hard disk, then download and run their disk diagnostics. If there is a problem with your disk, the manufacturer will replace the disk under warranty.

Ted Zieglar

 

[Terry Pinnell]

After you are done running chkdsk c: /f, upon reboot look in the event
viewer for an entry called winlogon. That is the result of the chkdsk, right
there for your pleasure.

Thanks both.

Neat tip re Winlogon. Presumably it doesn't work for plain chkdsk
though (no fix), as I can't find any entry for the recent check of D
that I described.

The most recent one is over 3 weeks old. I assume it corresponds to a
chkdsk d: /f I did then. FWIW, here is its summary:

Checking file system on D:
The type of the file system is NTFS.
Volume label is Data.

A disk check has been scheduled.
Windows will now check the disk.
Cleaning up minor inconsistencies on the drive.
Cleaning up 7 unused index entries from index $SII of file 0x9.
Cleaning up 7 unused index entries from index $SDH of file 0x9.
Cleaning up 7 unused security descriptors.

46283233 KB total disk space.
15181140 KB in 111190 files.
40756 KB in 10295 indexes.
4 KB in bad sectors.
233081 KB in use by the system.
65536 KB occupied by the log file.
30828252 KB available on disk.

Does chkdsk even find/report 'Bad blocks'?

 

[Richard Urban]

Bad sectors!

--
Regards,

Richard Urban
Microsoft MVP Windows Shell/User

If you knew as much as you think you know,
You would realize that you don't know what you thought you knew!

 

[cquirke (MVP Windows shell/user)]

On Sun, 23 Oct 2005 08:56:45 +0100, Terry Pinnell

A routine check of my Event Viewer shows the entry "The device,
\Device\Harddisk0\D, has a bad block."

Only a day before, I'd run a chkdsk d:, and that didn't report
anything. WinXP desktop was up when I returned to PC after chkdsk had
finished. I deliberately did not use the /f switch, as I wanted to see
if there were any problems.

ChkDsk and ChkDsk /f may trip over a bad block if it's still in use,
but they don't test the disk surface for these; that's ChkDsk /r (I
think - I don't use ChkDsk for this).

Instead, I'd get HD Tune from www.hdtune.com and do three things:
- check the S.M.A.R.T. history detail
- check temperature before and after the surface test
- do the "slow" surface test; any defects is Bad

Unlike some HD vendor's utilities that just give you a glib "OK" after
checking S.M.A.R.T., tools like HD Tune can show you the raw data, as
well as "you aren't dead yet" interpretation. It's not that easy to
understand, but Google is your friend

See also 22 September (second newest - I'm a slow blogger!) at
http://cquirke.blogspot.com on this topic, heh.

--------------- ----- ---- --- -- - - -

I *am* a power user!
I have electricity bills to prove it!

 

[Michael W. Ryder]

On Sun, 23 Oct 2005 08:56:45 +0100, Terry Pinnell




ChkDsk and ChkDsk /f may trip over a bad block if it's still in use,
but they don't test the disk surface for these; that's ChkDsk /r (I
think - I don't use ChkDsk for this).

Instead, I'd get HD Tune from www.hdtune.com and do three things:
- check the S.M.A.R.T. history detail
- check temperature before and after the surface test
- do the "slow" surface test; any defects is Bad

Unlike some HD vendor's utilities that just give you a glib "OK" after
checking S.M.A.R.T., tools like HD Tune can show you the raw data, as
well as "you aren't dead yet" interpretation. It's not that easy to
understand, but Google is your friend

I prefer Gibson Research's Spinrite 6 personally, but I have been using
Spinrite since version 1. I just upgraded a few months ago when I had a
disk drive suddenly start showing errors all over. I couldn't use Ghost
to backup the drive because of the bad sectors. Running Spinrite on the
drive cleared up the problems long enough for me to clone the drive and
replace it.

 

[cquirke (MVP Windows shell/user)]

cquirke (MVP Windows shell/user) wrote:

I prefer Gibson Research's Spinrite 6 personally, but I have been using
Spinrite since version 1. I just upgraded a few months ago when I had a
disk drive suddenly start showing errors all over. I couldn't use Ghost
to backup the drive because of the bad sectors. Running Spinrite on the
drive cleared up the problems long enough for me to clone the drive and
replace it.

I also have Spinrite (I bought Spinrite 5) but haven't had a success
case as yet (i.e. where it recovered access to lost material). I
don't feel bad about the value equation, as what I paid is more than
paid back by value Steve G has provided for free (Fix CIH, Trouble In
Paradise utility for Zip drives, etc.).

The feature sets are quite different, too...
- Spinrite accesses OS volumes, HD Tune is OS-agnostic
- Spinrite optionally writes to HD to fix, HD Tune doesn't
- HD Tune shows S.M.A.R.T. and temperature, Spinrite doesn't

So I don't see one as substituting for the other - I'd use HD Tune
first (starting with Temperature and S.M.A.R.T. and bailing out to
data recovery without stressing the HD via surface test if these
results are scary) then I'd recover data if bad, then I'd Spinrite and
do a second recovery while keeping the first.

What's really nice about HD Tune is you can watch the temperature
while the surface scan is in progress, and you can switch to the SMART
tab and watch the S.M.A.R.T. counters change as the surface scan is in
progress as well. I usually take a digital photo of the S.M.A.R.T.
tab at start and end of surface test to compare.

The temperature is always visible, no matter what tab (benchmarks,
S.M.A.R.T., surface etc.) is selected. Nice!

-- Risk Management is the clue that asks:

"Why do I keep open buckets of petrol next to all the
ashtrays in the lounge, when I don't even have a car?"

 

[Terry Pinnell]

On Sun, 23 Oct 2005 08:56:45 +0100, Terry Pinnell



ChkDsk and ChkDsk /f may trip over a bad block if it's still in use,
but they don't test the disk surface for these; that's ChkDsk /r (I
think - I don't use ChkDsk for this).

Instead, I'd get HD Tune from www.hdtune.com and do three things:
- check the S.M.A.R.T. history detail
- check temperature before and after the surface test
- do the "slow" surface test; any defects is Bad

Unlike some HD vendor's utilities that just give you a glib "OK" after
checking S.M.A.R.T., tools like HD Tune can show you the raw data, as
well as "you aren't dead yet" interpretation. It's not that easy to
understand, but Google is your friend

Thanks, sounds good. But that link fails, with a timeout error. Any
idea if it's just a temporary problem, or has the site gone? Is there
an alternative source to download this please?

FWIW, my worries are growing. On top of the evidence from chkdsk
reporting bad block and bad sectors, I've just run HDTach (which I had
around from a couple of years ago) and it failed during the final of
the 4-part 'Quick Test', the Sequential Read. Message like "Error
reported..."

See also 22 September (second newest - I'm a slow blogger!) at
http://cquirke.blogspot.com on this topic, heh.

Thanks, reading now. Some of it technically over my head, but very
interesting nevertheless.

 

[Richard Urban]

Terry,

E-Mail me
..

--
Regards,

Richard Urban
Microsoft MVP Windows Shell/User

Quote from George Ankner:
If you knew as much as you think you know,
You would realize that you don't know what you thought you knew!

 

[Terry Pinnell]

I prefer Gibson Research's Spinrite 6 personally, but I have been using
Spinrite since version 1.

I just took a look at that. But for a solo home user I hesitate at
spending over 100USD ($89 + VAT)!

I have just found and run another one, HDDLife. As you see from these
it *seems* to be reporting I'm OK.
http://www.terrypin.dial.pipex.com/Images/HDDLife-HD0CD.jpg
http://www.terrypin.dial.pipex.com/Images/HDDLife-HD1EF.jpg
http://www.terrypin.dial.pipex.com/Images/HDDLife-HD2J.jpg
(I've scrolled the lower window to show the pertinent partitions:
0=C&D; 1=E&F; 2=J)

Do you agree with that reassuring report from the program (which
reached its conclusions in only a few minutes, BTW)? To my
inexperienced eyes it seems worrying that the temperature of HD0, the
HD under discussion, is so much higher than the other two. The
relative ages of these 3 is HD0 = 3 yrs, HD1 = 2 yrs, HD2 = 1 month.

BTW, if you'll excuse my going off at a tangent, I'm disappointed that
HD2, which I had assumed would be similar to my other Maxtors, is
treated as a 'SCSI' HD, and therefore outside the scope of this
'SMART' stuff. That's presumably due to the PCI card I had to install
first, which is a 'HighPoint Rocket 133 2CH'?. WinXP Home Device
manager now includes another category, 'SCSI & RAID Controller', under
which I have the entry '-HPT302 UDMA/ATA133 Controller'. Anyway, it's
new and healthy so I'll forget about that now.

 

[Michael W. Ryder]

I just took a look at that. But for a solo home user I hesitate at
spending over 100USD ($89 + VAT)!

I guess it depends how much your data and time are worth. As it can
cost well over $300.00 to recover data from a defective hard drive, not
to mention the lost time, the price is a bargain. I have used Spinrite
since the days of DOS 3.x and it has saved me a lot of time and grief.
And it is a nice way to really test a hard drive. I realize that isn't
as important with today's disposable hard drives, but when you are
spending $400 for a 40 MB hard drive it is important.

I have just found and run another one, HDDLife. As you see from these
it *seems* to be reporting I'm OK.
http://www.terrypin.dial.pipex.com/Images/HDDLife-HD0CD.jpg
http://www.terrypin.dial.pipex.com/Images/HDDLife-HD1EF.jpg
http://www.terrypin.dial.pipex.com/Images/HDDLife-HD2J.jpg
(I've scrolled the lower window to show the pertinent partitions:
0=C&D; 1=E&F; 2=J)

Do you agree with that reassuring report from the program (which
reached its conclusions in only a few minutes, BTW)? To my
inexperienced eyes it seems worrying that the temperature of HD0, the
HD under discussion, is so much higher than the other two. The
relative ages of these 3 is HD0 = 3 yrs, HD1 = 2 yrs, HD2 = 1 month.

Spinrite's test will take several hours to check every single byte on
the drive, not just do a quick test like Scandisk. I would be ready to
replace HD0 as my last disk that failed was that age. One day it was
"fine" and the next day it was unusable. Fortunately for me the new
version of Spinrite worked with NTFS partitions and was able to fix the
problems long enough to Ghost the drive to a new one. So I spent a
couple of hours fixing the drive instead of several weeks.
If you are looking for an easier way to monitor drive temperatures take
a look at Speedfan. It will basically give you the same information you
are getting from HDDLife and will also monitor the other temperatures in
your computer and the voltages. From this you can see if there is
another problem that may be affecting the hard drives.

 

[Terry Pinnell]

Here's an update of current (bad) status.

It's now definite that HD0, the 4 (not 3) year old MAXTOR 6L060J3 60
GB HD that came with my PC, is at risk of failing.

The evidence so far includes
- chkdsk reporting bad block/bad sectors, as already described
- HDTach failing during its Sequential Read (part of Quick Test)
- HDTune:
HD Tune: MAXTOR 6L060J3 Health
------------------------------
(HD0, partitions C & D. This is the suspect drive.)

ID Current Worst ThresholdData Status
(01) Raw Read Error Rate 100 253 20 0 Ok
(03) Spin Up Time 67 64 20 4174 Ok
(04) Start/Stop Count 100 100 8 321 Ok
(05) Reallocated Sector Count 99 99 20 9 Failed
(07) Seek Error Rate 100 1 23 0 Failed
(09) Power On Hours Count 60 60 1 26756 Ok
(0A) Spin Retry Count 100 100 0 0 Ok
(0B) Calibration Retry Count 100 100 20 0 Ok
(0C) Power Cycle Count 100 100 8 284 Ok
(0D) Soft Read Error Rate 100 77 23 0 Ok
(C2) Temperature 82 75 42 47 Ok
(C3) Hardware ECC Recovered 100 1 0 39294398 Ok
(C4) Reallocated Event Count 100 100 20 0 Ok
(C5) Current Pending Sector 99 99 20 9 Ok
(C6) Offline Uncorrectable 100 253 0 0 Ok
(C7) Ultra DMA CRC Error Count 193 193 0 7 Ok
Power On Time : 26756
Health Status : Failed

For comparison, here's the report on HD1 (partitions E & F) which is
physically identical to above, about 3 yrs old.

HD Tune: MAXTOR 6L060J3 Health
-------------------------------

ID Current Worst ThresholdData Status
(01) Raw Read Error Rate 100 253 20 0 Ok
(03) Spin Up Time 67 64 20 4131 Ok
(04) Start/Stop Count 100 100 8 242 Ok
(05) Reallocated Sector Count 100 100 20 0 Ok
(07) Seek Error Rate 100 100 23 0 Ok
(09) Power On Hours Count 67 67 1 22104 Ok
(0A) Spin Retry Count 100 100 0 0 Ok
(0B) Calibration Retry Count 100 100 20 0 Ok
(0C) Power Cycle Count 100 100 8 231 Ok
(0D) Soft Read Error Rate 100 100 23 0 Ok
(C2) Temperature 84 78 42 43 Ok
(C3) Hardware ECC Recovered 100 100 0 3805 Ok
(C4) Reallocated Event Count 100 100 20 0 Ok
(C5) Current Pending Sector 100 100 20 0 Ok
(C6) Offline Uncorrectable 100 253 0 0 Ok
(C7) Ultra DMA CRC Error Count 200 200 0 0 Ok
Power On Time : 22104
Health Status : Ok

I also did an error scan on HD0 and it reports 4 bad blocks.

And here's the Inf' tab for HD0:
http://www.terrypin.dial.pipex.com/Images/HDTune-HD0-Info.gif

(A side issue is that Active entry, bottom right, which is the same
for both HD0 and HD1, which says I'm only using
ATA/100 not 133.)

I've also downloaded Powermax and made the floppy, but haven't yet run
it. To be honest, I'm now nervous about running *anything* that might
risk complete loss, before I get my act together. Without truly
understanding it, I read stuff about 'stressing' the HD, and fret that
maybe PowerMax might do that. Sort of tip it over the edge! In any
case, my action plan is now clear. I'll shop for another HD today
(probably not a Maxtor this time) and aim to install it asap.

Meanwhile, I need to get myself some security fast. Can't face the
prospect of a re-install! Luckily I added a new HD only a couple of
months ago, so I do have plenty of space.

My HDs and partitions are:

HD0 (the failing disk)
---------------------
C: my Win XP Home OS and about half my program files
D: most of my data and the other half of my program files.

HD1 (60 GB; identical to HD0)
-----------------------------
E: Bootable copy of OS, but now 2 months old
F: Backups (Most data, and most settings, but not programs.)

HD2 (200 GB Maxtor, recently installed, via new PCI card)
--------------------------------------------------------
J: Pictures and Music mainly

My data backups are pretty good. But I haven't been backing up
C:\Program Files and D:\Program Files. I don't use OE except when
forced to for some reason. My Agent email/newsreader folders are all
backed up nightly.

I have PQ PM 7.0 and PQ DI 2002. Rarely used, and have to re-study
manuals when I do so! I made a direct copy (not image) of Win XP Home
onto E: a couple of months ago, with PowerQuest Drive Image 2002.
Apart from an initial 'reassurance test', I haven't used it since. But
one of my jobs today will be to update that copy on E: That's assuming
these bad sectors/blocks don't screw that up?

Can't get enough hand-holding!

 

[cquirke (MVP Windows shell/user)]

On Fri, 28 Oct 2005 11:33:35 +0100, Terry Pinnell

Here's an update of current (bad) status.

It's now definite that HD0, the 4 (not 3) year old MAXTOR 6L060J3 60G
HD that came with my PC, is at risk of failing.

Not "at risk"; it IS failing!

The evidence so far includes
- chkdsk reporting bad block/bad sectors, as already described
- HDTach failing during its Sequential Read (part of Quick Test)
HD Tune: MAXTOR 6L060J3 Health

ID Current Worst ThresholdData Status
(05) Reallocated Sector Count 99 99 20 9 Failed
(07) Seek Error Rate 100 1 23 0 Failed
(0D) Soft Read Error Rate 100 77 23 0 Ok
(C2) Temperature 82 75 42 47 Ok
(C5) Current Pending Sector 99 99 20 9 Ok
(C7) Ultra DMA CRC Error Count 193 193 0 7 Ok

Raw data suggests "fixed" and "failed to fix" sectors at the firmware
level; the latter would be visible to ChkDsk/AutoChk.

Is that temperature 82C? Worst-recorded 75C? Those are high
temperatures, if those are indeed the values. Getting hot and falling
on its ass is something even the lamest single-platter 40G model
Maxtors seem to do with aplomb.

For comparison, here's the report on HD1 (partitions E & F) which is
physically identical to above, about 3 yrs old.

HD Tune: MAXTOR 6L060J3 Health

Maybe those temperatures are massaged and/or in F?

To be honest, I'm now nervous about running *anything* that might
risk complete loss

....such as "load a ton of files, run them as code, and constantly
write to HD" Windows, for starters...

I read stuff about 'stressing' the HD, and fret that maybe
PowerMax might do that. Sort of tip it over the edge!

Of course - what would you rather have, 10 pages of impeccable
diagnostic proof of the HD's death throes, or your data?

I'd say you've had more than the amount of warning you can expect -
hell, even the vendor's S.M.A.R.T. summary is telling you the HD is
dying - and Darwin's teeth must surely be close at hand.

Wake up and smell the blood :-(

------------ ----- ---- --- -- - - - -

The most accurate diagnostic instrument
in medicine is the Retrospectoscope