Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.sys.ibm.pc.hardware.storage > #2612 > unrolled thread

chkdsk all partitions at once, including those without drive letters

Back to article view | Back to comp.sys.ibm.pc.hardware.storage

Contents

  chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-11 10:59 -0800
    Re: chkdsk all partitions at once, including those without drive letters Franc Zabkar <fzabkar@iinternode.on.net> - 2012-02-12 06:46 +1100
    Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-12 07:15 +1100
      Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-11 13:00 -0800
        Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-12 08:53 +1100
          Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-11 21:54 -0800
            Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-12 19:31 +1100
              Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-12 07:04 -0800
    Re: chkdsk all partitions at once, including those without drive letters Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-02-11 19:02 -0500
      Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-12 13:21 +1100
        Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-11 22:05 -0800
          Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-12 19:39 +1100
            Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-12 19:24 -0800
              Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-12 19:37 -0800
              Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-13 19:52 +1100
                Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-13 22:20 -0800
                  Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-13 22:31 -0800
                    Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-15 02:40 +1100
                      Re: chkdsk all partitions at once, including those without drive letters AndyHancock <andymhancock@gmail.com> - 2012-02-14 20:51 -0800
                        Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-15 16:12 +1100
                  Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-15 02:36 +1100
        Re: chkdsk all partitions at once, including those without drive letters Rob <mesamine@gmail.com> - 2012-02-16 03:15 +1100
          Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-16 07:43 +1100
            Re: chkdsk all partitions at once, including those without drive letters Rob <mesamine@gmail.com> - 2012-02-16 14:29 +1100
              Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-16 16:05 +1100
                Re: chkdsk all partitions at once, including those without drive letters Yousuf Khan <bbbl67@spammenot.yahoo.com> - 2012-02-16 12:40 -0500
                  Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-17 07:59 +1100
                    Re: chkdsk all partitions at once, including those without drive letters Rob <mesamine@gmail.com> - 2012-02-17 08:55 +1100
                      Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-17 09:41 +1100
                        Re: chkdsk all partitions at once, including those without drive letters Rob <mesamine@gmail.com> - 2012-02-17 10:14 +1100
                          Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-17 11:20 +1100
                            Re: chkdsk all partitions at once, including those without drive letters Rob <mesamine@gmail.com> - 2012-02-17 23:37 +1100
                              Re: chkdsk all partitions at once, including those without drive letters "Rod Speed" <rod.speed.aaa@gmail.com> - 2012-02-18 01:02 +1100

Page 1 of 2  [1] 2  Next page →

#2612 — chkdsk all partitions at once, including those without drive letters

Toshiba computers come with 4 partitions.  I've looked up the
partition characteristics using the following:
 - Disk Management
 - PowerQuest Partition Table Editor
 - Partition Type codes at
   http://www.win.tue.nl/~aeb/partitions/partition_types-1.html

The partitions are:

1. Type: 0x27 (Windows Recover Environment hidden partition)
   Size: 1.46 GiB
   Disk Management labels: Healthy (Active, Recover partition)

2. Type: 0x07 NTFS
   Size: Hundreds of GiB
   Disk Management labels:
      Healthy (Boot, Page File, Crash Dump, Primary Partition)

3. Type: 0x17 (Hidden Internal File System
   Size: 22.37 GiB
   Disk Management labels: Healthy (Primary Partition)

4. Type: 0x17 (Hidden Internal File System
   Size: 12.83 GiB
   Disk Management labels: Healthy (Primary Partition)

I also have a paltry 1.x GiB of unallocated space after partition 4.

Partitions 2,3,4 are of types 0x07 and 0x17, which are well described
at http://fixunix.com/help/9334-what-hidden-ifs-e-g-hpfs-0x17.html .
Both are Internal File systems (IFSs), but 0x07 is unhidden while 0x17
is hidden.  IFSs seem to be OS dependent:
 - HPFS (OS/2)
 - NTFS (Windows)
 - Advanced unix or QUNX2.X(pre-1988)

From exchanges here and elsewhere, the feeling is that partition 1 is
boot & SRP, while partitions 3 & 4 are for creating recovery discs
(which don't come with the laptop), for disc-less recovery in case you
don't have (or have access to) recovery discs, and for other-language
versions of recovery.  Partition 2 is simply C-drive, and contains the
OS.  When Ghosting, I use SRP switches for partition 1, and the clone
satisfies the tests of booting properly, creating recovery discs, and
doing disc-less recovery.

Partitions 1,3,4 do not have letter drives.  So I was wondering
whether I can still use chkdsk (and if I can do that for all the
partitions at once, so much the better).  I've surfed and found info
on assigning letters to the partitions, but that option is grayed out
in Disk Management.  I am also concerned about possible side effects
of assigning letters on the intended functions of those drives.  Disk
management, partitions, volumes, file systems -- stuff I never really
knew about before having to clone (motivated by a boot failure scare).

[toc] | [next] | [standalone]

#2613

On Sat, 11 Feb 2012 10:59:31 -0800 (PST), AndyHancock
<andymhancock@gmail.com> put finger to keyboard and composed:

>Toshiba computers come with 4 partitions.
>
>The partitions are:
>
>1. Type: 0x27 (Windows Recover Environment hidden partition)
>   Size: 1.46 GiB
>   Disk Management labels: Healthy (Active, Recover partition)
>
>2. Type: 0x07 NTFS
>   Size: Hundreds of GiB
>   Disk Management labels:
>      Healthy (Boot, Page File, Crash Dump, Primary Partition)
>
>3. Type: 0x17 (Hidden Internal File System
>   Size: 22.37 GiB
>   Disk Management labels: Healthy (Primary Partition)
>
>4. Type: 0x17 (Hidden Internal File System
>   Size: 12.83 GiB
>   Disk Management labels: Healthy (Primary Partition)

>Partitions 1,3,4 do not have letter drives.  So I was wondering
>whether I can still use chkdsk (and if I can do that for all the
>partitions at once, so much the better).

You could change the partition IDs from 0x17 to 0x07 (assuming they
are NTFS volumes), and then run CHKDSK against them, in readonly mode.

- Franc Zabkar
-- 
Please remove one 'i' from my address when replying by email.

[toc] | [prev] | [next] | [standalone]

#2614

AndyHancock wrote:

> Toshiba computers come with 4 partitions.  I've looked
> up the partition characteristics using the following:
> - Disk Management
> - PowerQuest Partition Table Editor
> - Partition Type codes at
>   http://www.win.tue.nl/~aeb/partitions/partition_types-1.html

> The partitions are:

> 1. Type: 0x27 (Windows Recover Environment hidden partition)
>   Size: 1.46 GiB
>   Disk Management labels: Healthy (Active, Recover partition)

> 2. Type: 0x07 NTFS
>   Size: Hundreds of GiB
>   Disk Management labels:
>      Healthy (Boot, Page File, Crash Dump, Primary Partition)

> 3. Type: 0x17 (Hidden Internal File System
>   Size: 22.37 GiB
>   Disk Management labels: Healthy (Primary Partition)

> 4. Type: 0x17 (Hidden Internal File System
>   Size: 12.83 GiB
>   Disk Management labels: Healthy (Primary Partition)

> I also have a paltry 1.x GiB of unallocated space after partition 4.

> Partitions 2,3,4 are of types 0x07 and 0x17, which are well described
> at http://fixunix.com/help/9334-what-hidden-ifs-e-g-hpfs-0x17.html .
> Both are Internal File systems (IFSs), but 0x07 is unhidden while 0x17
> is hidden.  IFSs seem to be OS dependent:
> - HPFS (OS/2)
> - NTFS (Windows)
> - Advanced unix or QUNX2.X(pre-1988)

> From exchanges here and elsewhere, the feeling is that partition 1 is
> boot & SRP, while partitions 3 & 4 are for creating recovery discs
> (which don't come with the laptop), for disc-less recovery in case you
> don't have (or have access to) recovery discs, and for other-language
> versions of recovery.  Partition 2 is simply C-drive, and contains the
> OS.  When Ghosting, I use SRP switches for partition 1, and the clone
> satisfies the tests of booting properly, creating recovery discs, and
> doing disc-less recovery.

> Partitions 1,3,4 do not have letter drives.  So
> I was wondering whether I can still use chkdsk

I wouldnt, even if you change the partition type so Win will give it a letter.

Too much risk of chkdsk stuffing up the partition if it sees something it doesnt like.

> (and if I can do that for all the partitions at once, so much the better).

> I've surfed and found info on assigning letters to the partitions,
> but that option is grayed out in Disk Management.

Because of the partition type.

> I am also concerned about possible side effects of
> assigning letters on the intended functions of those drives.

Yeah, could be dangerous if chkdsk gets to run auto on it.

> Disk management, partitions, volumes, file systems -- stuff I never really
> knew about before having to clone (motivated by a boot failure scare).

Yeah, it can be quite complicated. 

[toc] | [prev] | [next] | [standalone]

#2616

On Feb 11, 3:15 pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
> AndyHancock wrote:
> > Toshiba computers come with 4 partitions.  I've looked
> > up the partition characteristics using the following:
> > - Disk Management
> > - PowerQuest Partition Table Editor
> > - Partition Type codes at
> >  http://www.win.tue.nl/~aeb/partitions/partition_types-1.html
> > The partitions are:
> > 1. Type: 0x27 (Windows Recover Environment hidden partition)
> >   Size: 1.46 GiB
> >   Disk Management labels: Healthy (Active, Recover partition)
> > 2. Type: 0x07 NTFS
> >   Size: Hundreds of GiB
> >   Disk Management labels:
> >      Healthy (Boot, Page File, Crash Dump, Primary Partition)
> > 3. Type: 0x17 (Hidden Internal File System
> >   Size: 22.37 GiB
> >   Disk Management labels: Healthy (Primary Partition)
> > 4. Type: 0x17 (Hidden Internal File System
> >   Size: 12.83 GiB
> >   Disk Management labels: Healthy (Primary Partition)
> > I also have a paltry 1.x GiB of unallocated space after partition 4.
> > Partitions 2,3,4 are of types 0x07 and 0x17, which are well described
> > athttp://fixunix.com/help/9334-what-hidden-ifs-e-g-hpfs-0x17.html.
> > Both are Internal File systems (IFSs), but 0x07 is unhidden while 0x17
> > is hidden.  IFSs seem to be OS dependent:
> > - HPFS (OS/2)
> > - NTFS (Windows)
> > - Advanced unix or QUNX2.X(pre-1988)
> > From exchanges here and elsewhere, the feeling is that partition 1 is
> > boot & SRP, while partitions 3 & 4 are for creating recovery discs
> > (which don't come with the laptop), for disc-less recovery in case you
> > don't have (or have access to) recovery discs, and for other-language
> > versions of recovery.  Partition 2 is simply C-drive, and contains the
> > OS.  When Ghosting, I use SRP switches for partition 1, and the clone
> > satisfies the tests of booting properly, creating recovery discs, and
> > doing disc-less recovery.
> > Partitions 1,3,4 do not have letter drives.  So
> > I was wondering whether I can still use chkdsk
>
> I wouldnt, even if you change the partition type so Win will give it a letter.
>
> Too much risk of chkdsk stuffing up the partition if it sees something it doesnt like.
>
> > (and if I can do that for all the partitions at once, so much the better).
> > I've surfed and found info on assigning letters to the partitions,
> > but that option is grayed out in Disk Management.
>
> Because of the partition type.
>
> > I am also concerned about possible side effects of
> > assigning letters on the intended functions of those drives.
>
> Yeah, could be dangerous if chkdsk gets to run auto on it.
>
> > Disk management, partitions, volumes, file systems -- stuff I never really
> > knew about before having to clone (motivated by a boot failure scare).
>
> Yeah, it can be quite complicated.

Franc, Rod,

Thanks for that info on the effect of partition type on changability
of drive letter and hazards of chkdsk changing the content.  Since
this is a restoration from a clone, I can afford to take a chance. I
was more concerned with breaking the functionality of partitions 3 & 4
from changes that are caused by assigning and de-assigning drive
letters. I'm not sure what are the possible changes that can result
in, either within those partitions or outside of them. I vaguely seem
to recall that some bookkeeping info about individual partitions are
not stored in the partitions themselves -- some are stored in a
"master" partition.

I can test the partitions by creating recovery discs and trying disc-
less recovery, but I've created so many sets that I'm  running out of
physical storage. I'd also have to restore my non-factory-default
state from a clone. The process will take days. If I can get enough
confidence that there are no function-breaking changes in drive letter
[de]assigning, I won't bother.

Rod, in response to what caused the BUMP on the replacement drive, I
found that I was swapping HDDs in & out of both the computer and
enclosure in back-cloning, figuring out non-hot-swappability, etc. So
I simply left the HDD in the enclosure without screwing it together.
At the very least, it is probably safer from static.

Did I think of the possibility that I'd forget that the enclosure was
not bolted together? I was all too aware of the hazard, but it was
such an obvious and big one that there was no way to forget. Well it
wasn't the first time that I've been wrong about something like that.
Now I will screw the enclosure together all the time, despite the
uncooperative threading and added risk of misthreading. However, in
learning that lesson, I also have to thoroughly test the HDD. Chkdsk,
Hitachi DFT (which fortunately sees this HDD), SeaTools, Toshiba
diagnostics, SMART report, etc.

[toc] | [prev] | [next] | [standalone]

#2617

AndyHancock wrote:
> On Feb 11, 3:15 pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
>> AndyHancock wrote:
>>> Toshiba computers come with 4 partitions. I've looked
>>> up the partition characteristics using the following:
>>> - Disk Management
>>> - PowerQuest Partition Table Editor
>>> - Partition Type codes at
>>> http://www.win.tue.nl/~aeb/partitions/partition_types-1.html
>>> The partitions are:
>>> 1. Type: 0x27 (Windows Recover Environment hidden partition)
>>> Size: 1.46 GiB
>>> Disk Management labels: Healthy (Active, Recover partition)
>>> 2. Type: 0x07 NTFS
>>> Size: Hundreds of GiB
>>> Disk Management labels:
>>> Healthy (Boot, Page File, Crash Dump, Primary Partition)
>>> 3. Type: 0x17 (Hidden Internal File System
>>> Size: 22.37 GiB
>>> Disk Management labels: Healthy (Primary Partition)
>>> 4. Type: 0x17 (Hidden Internal File System
>>> Size: 12.83 GiB
>>> Disk Management labels: Healthy (Primary Partition)
>>> I also have a paltry 1.x GiB of unallocated space after partition 4.
>>> Partitions 2,3,4 are of types 0x07 and 0x17, which are well
>>> described
>>> athttp://fixunix.com/help/9334-what-hidden-ifs-e-g-hpfs-0x17.html.
>>> Both are Internal File systems (IFSs), but 0x07 is unhidden while
>>> 0x17
>>> is hidden. IFSs seem to be OS dependent:
>>> - HPFS (OS/2)
>>> - NTFS (Windows)
>>> - Advanced unix or QUNX2.X(pre-1988)
>>> From exchanges here and elsewhere, the feeling is that partition 1
>>> is boot & SRP, while partitions 3 & 4 are for creating recovery
>>> discs (which don't come with the laptop), for disc-less recovery in
>>> case you don't have (or have access to) recovery discs, and for
>>> other-language versions of recovery. Partition 2 is simply C-drive,
>>> and contains the
>>> OS. When Ghosting, I use SRP switches for partition 1, and the clone
>>> satisfies the tests of booting properly, creating recovery discs,
>>> and doing disc-less recovery.
>>> Partitions 1,3,4 do not have letter drives. So
>>> I was wondering whether I can still use chkdsk
>>
>> I wouldnt, even if you change the partition type so Win will give it
>> a letter.
>>
>> Too much risk of chkdsk stuffing up the partition if it sees
>> something it doesnt like.
>>
>>> (and if I can do that for all the partitions at once, so much the
>>> better). I've surfed and found info on assigning letters to the
>>> partitions,
>>> but that option is grayed out in Disk Management.
>>
>> Because of the partition type.
>>
>>> I am also concerned about possible side effects of
>>> assigning letters on the intended functions of those drives.
>>
>> Yeah, could be dangerous if chkdsk gets to run auto on it.
>>
>>> Disk management, partitions, volumes, file systems -- stuff I never
>>> really knew about before having to clone (motivated by a boot
>>> failure scare).
>>
>> Yeah, it can be quite complicated.
>
> Franc, Rod,

> Thanks for that info on the effect of partition type on changability
> of drive letter and hazards of chkdsk changing the content.  Since
> this is a restoration from a clone, I can afford to take a chance. I
> was more concerned with breaking the functionality of partitions 3 & 4
> from changes that are caused by assigning and de-assigning drive
> letters. I'm not sure what are the possible changes that can result
> in, either within those partitions or outside of them. I vaguely seem
> to recall that some bookkeeping info about individual partitions are
> not stored in the partitions themselves -- some are stored in a
> "master" partition.

Thats really only true of bootable partitions.

Corse those recovery partitions are that, but its not very likely
that there will be any problem produced by changing the drive
type so the partition gets a letter, and as you say, its safe
enough with a clone where you can always just reclone.

> I can test the partitions by creating recovery discs and trying disc-
> less recovery, but I've created so many sets that I'm  running out
> of physical storage. I'd also have to restore my non-factory-default
> state from a clone. The process will take days. If I can get enough
> confidence that there are no function-breaking changes in drive letter
> [de]assigning, I won't bother.

> Rod, in response to what caused the BUMP on the replacement drive,
> I found that I was swapping HDDs in & out of both the computer and
> enclosure in back-cloning, figuring out non-hot-swappability, etc. So
> I simply left the HDD in the enclosure without screwing it together.
> At the very least, it is probably safer from static.

> Did I think of the possibility that I'd forget that the enclosure was
> not bolted together? I was all too aware of the hazard, but it was
> such an obvious and big one that there was no way to forget. Well it
> wasn't the first time that I've been wrong about something like that.
> Now I will screw the enclosure together all the time, despite the
> uncooperative threading and added risk of misthreading. However, in
> learning that lesson, I also have to thoroughly test the HDD. Chkdsk,
> Hitachi DFT (which fortunately sees this HDD), SeaTools, Toshiba
> diagnostics, SMART report, etc.

Trouble is that with a serious BUMP the effect can show up months later.

Its probably fine tho, since it presumably happened with the drive not
spinning and so with the heads retracted. 

[toc] | [prev] | [next] | [standalone]

#2620

On Feb 11, 4:53 pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
> AndyHancock wrote:
>> Franc, Rod,
>> Thanks for that info on the effect of partition type on changability
>> of drive letter and hazards of chkdsk changing the content.  Since
>> this is a restoration from a clone, I can afford to take a chance. I
>> was more concerned with breaking the functionality of partitions 3 & 4
>> from changes that are caused by assigning and de-assigning drive
>> letters. I'm not sure what are the possible changes that can result
>> in, either within those partitions or outside of them. I vaguely seem
>> to recall that some bookkeeping info about individual partitions are
>> not stored in the partitions themselves -- some are stored in a
>> "master" partition.
>
> Thats really only true of bootable partitions.
>
> Corse those recovery partitions are that, but its not very likely
> that there will be any problem produced by changing the drive
> type so the partition gets a letter, and as you say, its safe
> enough with a clone where you can always just reclone.

True dat. I perceived the problem as not knowing whether the
partitions need to be recloned.  But you know, the simplest solution
to that is to assume that they're no good after the letter drive
assignment and chkdsk, then reclone everything back.

Heck for that matter, blow away all partitions, create one big one
with NTFS formatting.  The formatting should uncover bad sectors and
such, right?

> Trouble is that with a serious BUMP the effect can show up months later.

Would you know the (theoretical) cause and effect that delays the
manifestation?  Just curious.

> Its probably fine tho, since it presumably happened with the drive not
> spinning and so with the heads retracted.

I was hoping, but not hopeful.  Because in mid-last decade, I had
asked a university IT guy about parking the heads on laptops.  A term
from the nineties...maybe the eighties.  He laughed.  Apparently, that
was a quaint notion, but laptop HDDs are too small for that.  The
heads just bounce along.  Not saying he was necessarily right, or that
things don't change.  I just assumed that he knew how things worked --
at least back then.  Are you sure that the heads retract in laptop
HDDs?

I seem to recall that the SMART report also had a measure of how much
the spindle axis is displaced.  It indicates whether the HDD has been
physcially impacted.  So I wonder if there are factors other than
impacted heads that affect the HDD in the case of an impact.

[toc] | [prev] | [next] | [standalone]

#2622

AndyHancock wrote
> Rod Speed <rod.speed....@gmail.com> wrote
>> AndyHancock wrote

>>> Franc, Rod,
>>> Thanks for that info on the effect of partition type on changability
>>> of drive letter and hazards of chkdsk changing the content.  Since
>>> this is a restoration from a clone, I can afford to take a chance. I
>>> was more concerned with breaking the functionality of partitions 3
>>> & 4 from changes that are caused by assigning and de-assigning drive
>>> letters. I'm not sure what are the possible changes that can result
>>> in, either within those partitions or outside of them. I vaguely
>>> seem to recall that some bookkeeping info about individual
>>> partitions are not stored in the partitions themselves -- some are
>>> stored in a "master" partition.

>> Thats really only true of bootable partitions.

>> Corse those recovery partitions are that, but its not very likely
>> that there will be any problem produced by changing the drive
>> type so the partition gets a letter, and as you say, its safe
>> enough with a clone where you can always just reclone.

> True dat. I perceived the problem as not knowing whether the
> partitions need to be recloned.  But you know, the simplest
> solution to that is to assume that they're no good after the letter
> drive assignment and chkdsk, then reclone everything back.

> Heck for that matter, blow away all partitions, create one big one with NTFS formatting.

True.

> The formatting should uncover bad sectors and such, right?

Only if you dont enable the quick format.

>> Trouble is that with a serious BUMP the effect can show up months later.

> Would you know the (theoretical) cause and
> effect that delays the manifestation?  Just curious.

Basically the mechanical damage can show up later with a cracked printed
circuit board trace or with debis that is floating around in the sealed chamber.
That debris can end up getting jammed under the heads, they fly very low.

>> Its probably fine tho, since it presumably happened with
>> the drive not spinning and so with the heads retracted.

> I was hoping, but not hopeful.  Because in mid-last decade, I had
> asked a university IT guy about parking the heads on laptops.  A
> term from the nineties...maybe the eighties.  He laughed.  Apparently,
> that was a quaint notion, but laptop HDDs are too small for that.

He's wrong.

> The heads just bounce along.

Nope, because laptop drives do see rather more physical
stress than normal internal drives in desktop systems.

> Not saying he was necessarily right,

He's wrong.

> or that things don't change.  I just assumed that
> he knew how things worked -- at least back then.

He didnt.

> Are you sure that the heads retract in laptop HDDs?

Yep, and you can see that in the SMART stats too.

> I seem to recall that the SMART report also had a measure
> of how much the spindle axis is displaced.  It indicates
> whether the HDD has been physcially impacted.

There are G sensors with some drives too.

> So I wonder if there are factors other than impacted
> heads that affect the HDD in the case of an impact.

Yes, you can get cracked traces and even loose debris in the sealed enclosure too. 

[toc] | [prev] | [next] | [standalone]

#2624

On Feb 12, 3:31 am, "Rod Speed" <rod.speed....@gmail.com> wrote:
> AndyHancock wrote:
> > Rod Speed <rod.speed....@gmail.com> wrote:
> >> AndyHancock wrote:
> >>> Franc, Rod,
> >>> Thanks for that info on the effect of partition type on changability
> >>> of drive letter and hazards of chkdsk changing the content.  Since
> >>> this is a restoration from a clone, I can afford to take a chance. I
> >>> was more concerned with breaking the functionality of partitions 3
> >>> & 4 from changes that are caused by assigning and de-assigning drive
> >>> letters. I'm not sure what are the possible changes that can result
> >>> in, either within those partitions or outside of them. I vaguely
> >>> seem to recall that some bookkeeping info about individual
> >>> partitions are not stored in the partitions themselves -- some are
> >>> stored in a "master" partition.
> >> Thats really only true of bootable partitions.
> >> Corse those recovery partitions are that, but its not very likely
> >> that there will be any problem produced by changing the drive
> >> type so the partition gets a letter, and as you say, its safe
> >> enough with a clone where you can always just reclone.
> > True dat. I perceived the problem as not knowing whether the
> > partitions need to be recloned.  But you know, the simplest
> > solution to that is to assume that they're no good after the letter
> > drive assignment and chkdsk, then reclone everything back.
> > Heck for that matter, blow away all partitions, create one big one with NTFS formatting.
>
> True.
>
> > The formatting should uncover bad sectors and such, right?
>
> Only if you dont enable the quick format.
>
> >> Trouble is that with a serious BUMP the effect can show up months later.
> > Would you know the (theoretical) cause and
> > effect that delays the manifestation?  Just curious.
>
> Basically the mechanical damage can show up later with a cracked printed
> circuit board trace or with debis that is floating around in the sealed chamber.
> That debris can end up getting jammed under the heads, they fly very low.
>
> >> Its probably fine tho, since it presumably happened with
> >> the drive not spinning and so with the heads retracted.
> > I was hoping, but not hopeful.  Because in mid-last decade, I had
> > asked a university IT guy about parking the heads on laptops.  A
> > term from the nineties...maybe the eighties.  He laughed.  Apparently,
> > that was a quaint notion, but laptop HDDs are too small for that.
>
> He's wrong.
>
> > The heads just bounce along.
>
> Nope, because laptop drives do see rather more physical
> stress than normal internal drives in desktop systems.
>
> > Not saying he was necessarily right,
>
> He's wrong.
>
> > or that things don't change.  I just assumed that
> > he knew how things worked -- at least back then.
>
> He didnt.
>
> > Are you sure that the heads retract in laptop HDDs?
>
> Yep, and you can see that in the SMART stats too.
>
> > I seem to recall that the SMART report also had a measure
> > of how much the spindle axis is displaced.  It indicates
> > whether the HDD has been physcially impacted.
>
> There are G sensors with some drives too.
>
> > So I wonder if there are factors other than impacted
> > heads that affect the HDD in the case of an impact.
>
> Yes, you can get cracked traces and even loose debris in the sealed enclosure too.

Thanks for the insight, Rod.  Thank goodness for plush carpets.

[toc] | [prev] | [next] | [standalone]

#2618

On 11/02/2012 1:59 PM, AndyHancock wrote:
> Partitions 1,3,4 do not have letter drives.  So I was wondering
> whether I can still use chkdsk (and if I can do that for all the
> partitions at once, so much the better).  I've surfed and found info
> on assigning letters to the partitions, but that option is grayed out
> in Disk Management.  I am also concerned about possible side effects
> of assigning letters on the intended functions of those drives.  Disk
> management, partitions, volumes, file systems -- stuff I never really
> knew about before having to clone (motivated by a boot failure scare).

Looks like we had the same issues recently. I bought a new Toshiba 
laptop too. What I did was changed the Type 17 partitions to Type 07, 
and then rebooted. They were automatically added with their own drive 
letters after the reboot. I don't see any reason why just mounting them 
with drive letters is going to corrupt them in any way.

I imaged both partitions, and then blew them away. I also used the 
Toshiba imaging utility to create recovery disks before blowing them 
away. They took about 3 DL-DVD's or 5 SL-DVD's.

	Yousuf Khan

[toc] | [prev] | [next] | [standalone]

#2619

Yousuf Khan wrote
> AndyHancock wrote

>> Partitions 1,3,4 do not have letter drives.  So I was wondering
>> whether I can still use chkdsk (and if I can do that for all the
>> partitions at once, so much the better).  I've surfed and found info
>> on assigning letters to the partitions, but that option is grayed out
>> in Disk Management.  I am also concerned about possible side effects
>> of assigning letters on the intended functions of those drives.  Disk
>> management, partitions, volumes, file systems -- stuff I never really
>> knew about before having to clone (motivated by a boot failure scare).

> Looks like we had the same issues recently. I bought a new Toshiba
> laptop too. What I did was changed the Type 17 partitions to Type 07,
> and then rebooted. They were automatically added with their own drive
> letters after the reboot. I don't see any reason why just mounting
> them with drive letters is going to corrupt them in any way.

> I imaged both partitions, and then blew them away. I also used the
> Toshiba imaging utility to create recovery disks before blowing them
> away. They took about 3 DL-DVD's or 5 SL-DVD's.

Fark. Wonder why its so big when Win7 isnt ? 

[toc] | [prev] | [next] | [standalone]

#2621

On Feb 11, 9:21 pm, "Rod Speed" <rod.speed....@gmail.com> wrote:
> Yousuf Khan wrote
>> AndyHancock wrote
>>> Partitions 1,3,4 do not have letter drives.  So I was wondering
>>> whether I can still use chkdsk (and if I can do that for all the
>>> partitions at once, so much the better).  I've surfed and found info
>>> on assigning letters to the partitions, but that option is grayed out
>>> in Disk Management.  I am also concerned about possible side effects
>>> of assigning letters on the intended functions of those drives.
>>
>> Looks like we had the same issues recently. I bought a new Toshiba
>> laptop too. What I did was changed the Type 17 partitions to Type 07,
>> and then rebooted. They were automatically added with their own drive
>> letters after the reboot. I don't see any reason why just mounting
>> them with drive letters is going to corrupt them in any way.
>> I imaged both partitions, and then blew them away. I also used the
>> Toshiba imaging utility to create recovery disks before blowing them
>> away. They took about 3 DL-DVD's or 5 SL-DVD's.

Imagers have been described in the past as "intelligent" in that they
recognize the file system and create a smaller image by leaving out
empty space.  I suppose that if the imager doesn't recognize the file
system, it can do a blind copy of everything.  However, 0x07 and 0x17
can have three different file systems.  I suspect that imagers rely on
more than the partition type code to determine the file system.

About side-effects, Rod says it is unlikely as well.  Did you verify
the functionality of partitions 3 & 4 by restoring them from their
images and attempting to create recovery discs, as well as attempting
disc-less recovery?  It's a pain, I know, but just wondering if you
might have.

> Fark. Wonder why its so big when Win7 isnt ?

All those drivers and apps that are meant for the laptop model,
maybe?  And the other-language stuff.

BTW, I didn't say this in my last follow-up, but I also want to chkdsk
partition 1, which is letterless and (I'm told) is the boot
partition.  Would that be wise?  I realize that I can always restore
partition 1 from a clone, but just wondering.

[toc] | [prev] | [next] | [standalone]

#2623

AndyHancock wrote
> Rod Speed <rod.speed....@gmail.com> wrote
>> Yousuf Khan wrote
>>> AndyHancock wrote

>>>> Partitions 1,3,4 do not have letter drives.  So I was wondering
>>>> whether I can still use chkdsk (and if I can do that for all the
>>>> partitions at once, so much the better).  I've surfed and found
>>>> info on assigning letters to the partitions, but that option is grayed
>>>> out in Disk Management.  I am also concerned about possible side
>>>> effects of assigning letters on the intended functions of those drives.

>>> Looks like we had the same issues recently. I bought a new Toshiba
>>> laptop too. What I did was changed the Type 17 partitions to Type
>>> 07, and then rebooted. They were automatically added with their own
>>> drive letters after the reboot. I don't see any reason why just mounting
>>> them with drive letters is going to corrupt them in any way.

>>> I imaged both partitions, and then blew them away. I also used
>>> the Toshiba imaging utility to create recovery disks before blowing
>>> them away. They took about 3 DL-DVD's or 5 SL-DVD's.

> Imagers have been described in the past as "intelligent" in that they recognize
> the file system and create a smaller image by leaving out empty space.

Yes, only the crudest imagers dont do that.

> I suppose that if the imager doesn't recognize the
> file system, it can do a blind copy of everything.

Yes, and that is what the better imagers do.

> However, 0x07 and 0x17 can have three different file systems.

More than that, actually.

> I suspect that imagers rely on more than the
> partition type code to determine the file system.

I havent tested that myself.

> About side-effects, Rod says it is unlikely as well.  Did you verify
> the functionality of partitions 3 & 4 by restoring them from their images
> and attempting to create recovery discs, as well as attempting disc-less
> recovery?  It's a pain, I know, but just wondering if you might have.

>> Fark. Wonder why its so big when Win7 isnt ?

> All those drivers and apps that are meant for the laptop model, maybe?

Dunno, thats a hell of a lot more than the standard Win7 distribution.

> And the other-language stuff.

The standard Win7 distribution has that, tho I spose it might include
various chinese versions etc that the standard Win7 distribution doesnt.

> BTW, I didn't say this in my last follow-up, but I also want to
> chkdsk partition 1, which is letterless and (I'm told) is the boot
> partition.  Would that be wise?

No, I dont even chkdsk normal NTFS partitions unless I know
they arent too bad, it can comprehensively stuff up a partition.

> I realize that I can always restore partition 1 from a clone, but just wondering.

It should be fine, you can always do a read only chkdsk to be safe. 

[toc] | [prev] | [next] | [standalone]

#2625

On Feb 12, 3:39 am, "Rod Speed" <rod.speed....@gmail.com> wrote:
> AndyHancock wrote
>> Rod Speed <rod.speed....@gmail.com> wrote
>>> Yousuf Khan wrote
>>>> AndyHancock wrote
>>>>> Partitions 1,3,4 do not have letter drives.  So I was wondering
>>>>> whether I can still use chkdsk (and if I can do that for all the
>>>>> partitions at once, so much the better).  I've surfed and found
>>>>> info on assigning letters to the partitions, but that option is
>>>>> grayed out in Disk Management.  I am also concerned about
>>>>> possible side effects of assigning letters on the intended
>>>>> functions of those drives.
>>>> Looks like we had the same issues recently. I bought a new
>>>> Toshiba laptop too. What I did was changed the Type 17 partitions
>>>> to Type 07, and then rebooted. They were automatically added with
>>>> their own drive letters after the reboot. I don't see any reason
>>>> why just mounting them with drive letters is going to corrupt
>>>> them in any way.

<...snip...>

>> However, 0x07 and 0x17 can have three different file systems.
>
> More than that, actually.
>
>> I suspect that imagers rely on more than the partition type code to
>> determine the file system.

<...snip...>

>> BTW, I didn't say this in my last follow-up, but I also want to
>> chkdsk partition 1, which is letterless and (I'm told) is the boot
>> partition.  Would that be wise?
>
> No, I dont even chkdsk normal NTFS partitions unless I know they
> arent too bad, it can comprehensively stuff up a partition.
>
>> I realize that I can always restore partition 1 from a clone, but
>> just wondering.
>
> It should be fine, you can always do a read only chkdsk to be safe.

Well, I used PowerQuest Partition Table Editor to change partitions 3
& 4 from 0x17 to 0x07 (unhidden).  They got drive letters after a
reboot.  Chkdsk  on them was uneventful.  I changed them back to 0x17.

I then changed partition 1 from 0x27 to 0x07.  It did not receive a
driver letter after a reboot, but it did reboot.  I could use Disk
Management to assign a drive letter.  Chkdsk issued the warning
"WARNING!  F parameter not specified.".  The chkdsk then completed
without any interesting messages.  Partition 1 showed up in Disk
Management as "System (E:)" and had the tags "Healthy (System, Active,
Primary Partition)".  I changed it back to 0x27 and am rebooting.  If
I seem to have fallen off the face of the earth, it means things
didn't go well after the reboot (though I may come back using the
clone).

Rod, what did you mean by stuff up a partition?

[toc] | [prev] | [next] | [standalone]

#2626

On Feb 12, 10:24 pm, AndyHancock <andymhanc...@gmail.com> wrote:
> Well, I used PowerQuest Partition Table Editor to change partitions 3
> & 4 from 0x17 to 0x07 (unhidden).  They got drive letters after a
> reboot.  Chkdsk  on them was uneventful.  I changed them back to 0x17.
>
> I then changed partition 1 from 0x27 to 0x07.  It did not receive a
> driver letter after a reboot, but it did reboot.  I could use Disk
> Management to assign a drive letter.  Chkdsk issued the warning
> "WARNING!  F parameter not specified.".  The chkdsk then completed
> without any interesting messages.  Partition 1 showed up in Disk
> Management as "System (E:)" and had the tags "Healthy (System, Active,
> Primary Partition)".  I changed it back to 0x27 and am rebooting.  If
> I seem to have fallen off the face of the earth, it means things
> didn't go well after the reboot (though I may come back using the
> clone).
>
> Rod, what did you mean by stuff up a partition?

I had to use Disk Management to de-assign the drive letter before
PowerQuest Partition Table Editor would let me change partition 1 back
to type 0x27.  It now shows up in Disk Management exactly as it did in
my original post.  Now I know that the SRP (or Toshiba's equivalent)
and the recovery partitions can be assigned a drive letter for read-
only chkdsk without harm.  Thanks, Yousuf & Rod.

Now it's time to check the recovery by going through the motions of
creating recovery discs (though I won't actually follow through, since
I have too many sets of those discs at this point).  I will also go
through the motions of disc-less recovery, though I will try to
interrupt and terminate the process before it's completion.  All I
want to know is that the process works.

After that, I'll blow away all partitions, format the drive to catch
any tenacious errors or problems, then recover from the clone.

Hours of fun  ;)   ......

[toc] | [prev] | [next] | [standalone]

#2627

AndyHancock wrote
> Rod Speed <rod.speed....@gmail.com> wrote
>> AndyHancock wrote
>>> Rod Speed <rod.speed....@gmail.com> wrote
>>>> Yousuf Khan wrote
>>>>> AndyHancock wrote

>>>>>> Partitions 1,3,4 do not have letter drives.  So I was wondering
>>>>>> whether I can still use chkdsk (and if I can do that for all the
>>>>>> partitions at once, so much the better).  I've surfed and found
>>>>>> info on assigning letters to the partitions, but that option is
>>>>>> grayed out in Disk Management.  I am also concerned about
>>>>>> possible side effects of assigning letters on the intended
>>>>>> functions of those drives.
>>>>> Looks like we had the same issues recently. I bought a new
>>>>> Toshiba laptop too. What I did was changed the Type 17 partitions
>>>>> to Type 07, and then rebooted. They were automatically added with
>>>>> their own drive letters after the reboot. I don't see any reason
>>>>> why just mounting them with drive letters is going to corrupt
>>>>> them in any way.

>>> However, 0x07 and 0x17 can have three different file systems.

>> More than that, actually.

>>> I suspect that imagers rely on more than the partition type code to
>>> determine the file system.

>>> BTW, I didn't say this in my last follow-up, but I also want to
>>> chkdsk partition 1, which is letterless and (I'm told) is the boot
>>> partition.  Would that be wise?

>> No, I dont even chkdsk normal NTFS partitions unless I know they
>> arent too bad, it can comprehensively stuff up a partition.

>>> I realize that I can always restore partition 1 from a clone, but
>>> just wondering.

>> It should be fine, you can always do a read only chkdsk to be safe.

> Well, I used PowerQuest Partition Table Editor to change partitions 3
> & 4 from 0x17 to 0x07 (unhidden).  They got drive letters after a
> reboot.  Chkdsk  on them was uneventful.  I changed them back to 0x17.

> I then changed partition 1 from 0x27 to 0x07.  It did not receive a
> driver letter after a reboot, but it did reboot.  I could use Disk
> Management to assign a drive letter.  Chkdsk issued the warning
> "WARNING!  F parameter not specified.".  The chkdsk then completed
> without any interesting messages.  Partition 1 showed up in Disk
> Management as "System (E:)" and had the tags "Healthy (System, Active,
> Primary Partition)".  I changed it back to 0x27 and am rebooting.  If
> I seem to have fallen off the face of the earth, it means things didn't
> go well after the reboot (though I may come back using the clone).

> Rod, what did you mean by stuff up a partition?

That was a bit carelessly described, I meant that it can see some files go missing. 

[toc] | [prev] | [next] | [standalone]

#2628

On Feb 13, 3:52 am, "Rod Speed" <rod.speed....@gmail.com> wrote:
> AndyHancock wrote:
> > Rod, what did you mean by stuff up a partition?
>
> That was a bit carelessly described, I meant that it can see some files go missing.

I suppose those are the files that contain sectors with irrecoverable
data?  Would those files be goners anyway?  Or maybe they might not be
totally useless if the corrupted parts of the files are not always
used.

[toc] | [prev] | [next] | [standalone]

#2629

On Feb 14, 1:20 am, AndyHancock <andymhanc...@gmail.com> wrote:
> On Feb 13, 3:52 am, "Rod Speed" <rod.speed....@gmail.com> wrote:
>
> > AndyHancock wrote:
> > > Rod, what did you mean by stuff up a partition?
>
> > That was a bit carelessly described, I meant that it can see some files go missing.
>
> I suppose those are the files that contain sectors with irrecoverable
> data?  Would those files be goners anyway?  Or maybe they might not be
> totally useless if the corrupted parts of the files are not always
> used.

Excuse me.  I forgot to add the update.  After letter drive
assignment, read-only chkdsk, and letter de-assignment, I confirmed
that recovery disc creation works by starting the process on one disc
(which I promptly threw out, since I already have a couple of sets).
I also confirmed that disc-less recovery worked by initiating the
process.  It was in the midst of installing drivers when I cut it
short in order to further test the drive through explicitly formatting
the HDD.  My understanding from Toshiba tech support (for what it's
worth) is that the recovery does formatting anyway, which would report
disk problems.  But there's a lot of implied meanings of formatting
these days, and Toshiba tech support is....well, it is what it is.

Boy, what a confusion.  Low-level formatting isn't really formatting
these days.  But I'll take what I can get: Diskpart's "clear all",
which writes zeros everywhere.  If I remember correctly, I deleted all
partitions first.  Partition 1 couldn't be done via Disk management --
it required DiskPart's delete command using the "force" parameter.
Now in the process of creating a primary partition and formatting an
NTFS file system.

[toc] | [prev] | [next] | [standalone]

#2631

AndyHancock wrote
> AndyHancock <andymhanc...@gmail.com> wrote
>> Rod Speed <rod.speed....@gmail.com> wrote
>>> AndyHancock wrote

>>>> Rod, what did you mean by stuff up a partition?

>>> That was a bit carelessly described, I meant that it can see some files go missing.

>> I suppose those are the files that contain sectors with irrecoverable
>> data? Would those files be goners anyway? Or maybe they might not be
>> totally useless if the corrupted parts of the files are not always used.

> Excuse me.  I forgot to add the update.  After letter drive
> assignment, read-only chkdsk, and letter de-assignment, I confirmed
> that recovery disc creation works by starting the process on one disc
> (which I promptly threw out, since I already have a couple of sets).
> I also confirmed that disc-less recovery worked by initiating the
> process.  It was in the midst of installing drivers when I cut it
> short in order to further test the drive through explicitly formatting
> the HDD.  My understanding from Toshiba tech support (for what it's
> worth) is that the recovery does formatting anyway, which would report
> disk problems.  But there's a lot of implied meanings of formatting
> these days, and Toshiba tech support is....well, it is what it is.

> Boy, what a confusion.  Low-level formatting isn't really
> formatting these days.  But I'll take what I can get:
> Diskpart's "clear all", which writes zeros everywhere.

And in fact thats what most modern drives do if you tell them to low level format now.

> If I remember correctly, I deleted all partitions first.  Partition 1 couldn't
> be done via Disk management -- it required DiskPart's delete command
> using the "force" parameter. Now in the process of creating a primary
> partition and formatting an NTFS file system. 

[toc] | [prev] | [next] | [standalone]

#2632

On Feb 14, 10:40 am, "Rod Speed" <rod.speed....@gmail.com> wrote:
> AndyHancock wrote
>> Boy, what a confusion.  Low-level formatting isn't really
>> formatting these days.  But I'll take what I can get: Diskpart's
>> "clear all", which writes zeros everywhere.
>
> And in fact thats what most modern drives do if you tell them to low
> level format now.

So when Toshiba said that problems with the media would be revealed in
a formating, would it be right to interpret "formatting" that as "low
level formating" (writing zeros), not the old-school formatting of
scribing tracks and marking sectors?  I'm putting forth the former as
the possibly right interpretation because it wouldn't make sense for
them to say the latter if modern HDDs do not do the latter.

When I formatted, I got no errors, and I'm guessing that it is because
writing zeros isn't a very rigorous test...I seem to recall that
sectors could be re-allocated even in a brand new HDD.

Then again, maybe I *do* have no bad sectors.  The SMART report says 0
raw Reallocated Sector Count.

[toc] | [prev] | [next] | [standalone]

#2633

AndyHancock wrote
> Rod Speed <rod.speed....@gmail.com> wrote
>> AndyHancock wrote

>>> Boy, what a confusion.  Low-level formatting isn't really
>>> formatting these days.  But I'll take what I can get:
>>> Diskpart's "clear all", which writes zeros everywhere.

>> And in fact thats what most modern drives do if you tell them to low level format now.

> So when Toshiba said that problems with the media would be
> revealed in a formating, would it be right to interpret "formatting"
> that as "low level formating" (writing zeros), not the old-school
> formatting of scribing tracks and marking sectors?

Dunno. I was quite surprised by Toshiba's comment along those lines and
find it very hard to believe that they do use that approach to diagnosing
a possible hard drive fault, if only because that approach is destructive
of the user's settings and content and can comprehensively stuff up a
system that can have the user's files and settings saved before doing
something that drastic.

I think its rather unlikely that the Toshiba droid that said that to you
is correct, but you can see some operations with the silliest ideas about
what makes sense at times, to it is possible, albeit rather unlikely IMO.

> I'm putting forth the former as the possibly right interpretation because it wouldn't
> make sense for them to say the latter if modern HDDs do not do the latter.

Yeah, they cant even do that.

> When I formatted, I got no errors, and I'm guessing
> that it is because writing zeros isn't a very rigorous test...

In fact modern drives just silently reallocate sectors in that situation
and the only evidence you get that the drive has a problem is that
the number of reallocated sectors increases after the format.

> I seem to recall that sectors could be re-allocated even in a brand new HDD.

While thats theoretically correct, in practice it isnt seen.

> Then again, maybe I *do* have no bad sectors.
> The SMART report says 0 raw Reallocated Sector Count.

Thats not what it said when you first did a SMART report. You had 696.

Presumably they replaced the drive, thats why its now 0. 

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | comp.sys.ibm.pc.hardware.storage

csiph-web