MFT in two places

[Bill Cunningham]

The tool nfi.exe seems to show that the MFT is not together would that
be the word. The MFT mirror is fine. And NTFS's repairing of the filesystem
has been run. I have several partitions. But if the MFT supposed to be in
two different places? If not IDK how to repair this.

Bill

[Paul]

It works perfectly fine when "fragmented".

Only in extreme cases, does it need help.

It just looks ugly in a block display.
And you're not supposed to let that bother you.

*******

Low-end defragmenter tools will not defragment
the $MFT.

The more modern versions of Windows, the builtin
defragmenter library, has the ability to handle
the $MFT. Yet, not every tool uses the capability.

So it's rather difficult to track down an actual,
working, $MFT defragmenter.

You can see in this thread, there is a lot of
"yes it does, no it doesn't". The behavior of the
tools could change, depending on which OS you're
running at the time.

http://superuser.com/questions/316003/how-do-you-defragment-the-mft-on-an-ntfs-disk

The commercial tool I bought, didn't touch the
$MFT. But then, I didn't test it on Win10 or
anything, to be sure it wasn't a "crusty OS" issue.

Your best bet, is to get a 30 day eval of each
candidate tool, and see which does the job.
And, use a decent OS for the job. WinXP does
have some sort of defragmenter API, but it won't
have quite all the same features as a later OS.

Even Win10 isn't perfect - witness the behavior
of the "shrink" function in Disk Management. There
is at least one piece of metadata they should be
able to move out of the way, which they do not. Now,
the staff at Microsoft have strong opinions on the
design of NTFS. And I suspect a reason for not
moving the pesky thing that is in the way, is
it is "by design" to be left at the 50% mark
in the partition layout. So pride prevents
them from moving it :-) A commercial tool easily
moved it to the left, as would any Partition Manager
that needs to shrink a partition to the bare minimum.

Microsoft does make an API, but even they don't
make effective usage of it.

Paul

[Bill Cunningham]

"Paul" <nos...@needed.invalid> wrote in message
news:nv8cec$6h1$1...@dont-email.me...

Ok I guess all's fine. Thanks Paul.

[VanguardLH]

Defragmentation of the $MFT is normal but you want to keep the fragment
count low. 2 is pretty low.

Don't know what is the "nfi.exe". Don't expect others to know a
utility, even a well-known one, by its filename. Is it the one included
in the OEM Support Tools package from Microsoft at:

https://support.microsoft.com/en-us/kb/253066

If so, you might want a newer tool that explicitly states support for
Windows XP. Win2000 tools often work okay on WinXP but with something
doing major surgery on the OS then I want something that says it works
with that OS; else, I would have to test using it on a test platform,
not a critical one or my only one. There are plenty of 3rd-party tools
to defrag the $MFT on Windows boot, such as:

SysInternals' Contig
https://technet.microsoft.com/en-us/sysinternals/bb897428.aspx
Hopefully it supports the NTFS version 3.1 in Windows XP; see
https://en.wikipedia.org/wiki/NTFS#Versions. I've read where some users
have noted NTFS V5.1 in Windows Server 2003 (spring 2003), NTFS V5.2 in
Windows Vista (mid-2005), and NTFS V6.0 in Windows Server 2008 & Windows
7 but I don't know where that get their info (users rarely cite their
sources). "fsutil fsinfo ntfsinfo c:" on my Windows 7 Home x64 SP-1
reports 3.1 for the NTFS version.

Piriform Defraggler
I've seen users claim this does $MFT defrag; however, Piriform says
otherwise; see THEIR list of what they defrag at boot time at
https://www.piriform.com/docs/defraggler/technical-information/boot-time-defrag.
Doesn't look like a boot-time defrag is required as noted at
https://www.piriform.com/docs/defraggler/using-defraggler/understanding-the-drive-map
("Defraggler can defragment the MFT as well, but it cannot move the
MFT's starting position."). The starting position of the MFT is fixed.
The OS has to know where to start looking for it (well, its bootloader
in the boot sector for the partition has to know where is the MFT) but
its extents (fragments) can be elsewhere. The first 16 elements of the
MFT are fixed; see http://ntfs.com/ntfs-mft.htm and
http://www.easeus.com/resource/ntfs-disk-structure.htm under "MFT and
its structure". You can see the Mft Start Lcn by using "fsutil fsinfo
ntfsinfo <drive>". I have never tried to alter the start sector for the
MFT (so that's something you can research if you're curious).

The first cluster of the $MFT is non-movable. Some system files cannot
be moved, like for journaling ($Usn.Jrnl), so they may force
fragmentation. I hit that with huge video files that I could not get
into 1 fragment because there wasn't enough contiguous space between
journal files to fit the entire huge video file. I saved a backup,
reformatted the partition, and restored (which laid down the files
sequentially as each was copied to the partition). Later found I could
have deleted the journals (fsutil usn deletejounal /d <drive>) and then
done the defrag. However, immediately after deleting the journals they
start to get recreated.

https://msdn.microsoft.com/en-us/library/aa363798(v=VS.85).aspx
"... the NTFS file system maintains an update sequence number (USN)
change journal. When any change is made to a file or directory in a
volume, the USN change journal for that volume is updated with a
description of the change and the name of the file or directory."

https://msdn.microsoft.com/en-us/library/aa363877(v=vs.85).aspx
"To delete a change journal, use the FSCTL_DELETE_USN_JOURNAL control
code. When you use this operation, it walks through all of the files on
the volume and resets the USN for each file to zero. The operation then
deletes the existing change journal. This operation persists across
system restarts until it completes. Any attempt to read, create, or
modify the change journal during this process fails with the error code
ERROR_JOURNAL_DELETE_IN_PROGRESS."
and
"Change journals are not necessarily created at startup. To create a
change journal, an administrator may do so explicitly or start another
service that requires a change journal."

So that last part might give you time to do a defrag before the change
journals show up. However, that means disabling services (e.g.,
indexing) that affect journaling before you delete the journals.

https://www.microsoft.com/msj/0999/journal/journal.aspx
"A Change Journal can be disabled on a given volume, preventing the
system from logging file and directory changes. By default, an NTFS
volume will have its Change Journal disabled. Some application must
explicitly activate the journal. Also note that any application can
activate or disable the volume's journal at any time. An application
must be able to gracefully handle the situation when a journal is
disabled while the first application is still using the journal."

When you delete the journals, you had better hope that the OS doesn't
crash. There would be any change logs (files and folders) in the
journals because there are none. Journaling works by recording the
changes before committing the changes. For example, when you delete a
file, the file is not deleted immediately. Instead a change log for
that change gets recorded in the journal (like "file X is to be
deleted"). Only after that is recorded is the file deleted from the
file system (it's removed from the MFT and its clusters become available
for reallocation). If the OS crashes when the file is only partially
deleted, the OS can use the change journal to replay the commit to
finish the task.

https://en.wikipedia.org/wiki/Journaling_file_system

Some folks think journaling is not needed for SSDs. They want to
eliminate all those little writes when files or folders are modified
(created, deleted, changed). That's not what journaling is for.
Regardless of using an SSD, the state of file system update can still be
corrupted by an OS crash (like simply pulling the power - you kicked out
the cord or there was a power outage). SSDs are faster so the window of
opportunity is smaller but it still exists. You never mentioned the
type of your mass storage media (HDD or SDD). Is the state of the OS
(yes, it does file changes for itself) and the safety of your data worth
less than the endurance of the SSD? This is like folks that massively
into overclocking resulting in a less stable platform. I'd rather have
stability (of OS and data) over durablity (of hardware). If you backup
then hardware is replaceable, and you should expect it to eventually
fail, anyway.

Some folks will say defrag of SSDs is pointless. That depends on the
SSD device and the OS. Windows XP does not support the TRIM feature so
you may want to occasionally defrag an SSD used under that OS. Some 3rd
party defraggers, like Defraggler, will use the TRIM feature instead of
doing the old style of defrag instead of moving lots of files around
which incurs a lot of writes that would be superfluous on SSDs; see
http://www.piriform.com/docs/defraggler/technical-information/defraggler-and-ssds.
Although Windows XP has no TRIM support, most SSDs have their own
internal TRIM-on-idle function; i.e., their firmware handles the idle
garbage collection. That is retroactive TRIM (which can be slow). For
proactive TRIM, you need to upgrade to Windows 7+.

[Bill Cunningham]

"Paul" <nos...@needed.invalid> wrote in message
news:nv8cec$6h1$1...@dont-email.me...

> Bill Cunningham wrote:

Neither contig nor ultimate defragger touches it. I use defraggler
mostly.

Bill

[Bill Cunningham]

"VanguardLH" <V...@nguard.LH> wrote in message
news:e7prcd...@mid.individual.net...

> Bill Cunningham wrote:
>
>> The tool nfi.exe seems to show that the MFT is not together would that
>> be the word. The MFT mirror is fine. And NTFS's repairing of the
>> filesystem has been run. I have several partitions. But if the MFT
>> supposed to be in two different places? If not IDK how to repair
>> this.
>
> Defragmentation of the $MFT is normal but you want to keep the fragment
> count low. 2 is pretty low.
>
> Don't know what is the "nfi.exe". Don't expect others to know a
> utility, even a well-known one, by its filename. Is it the one included
> in the OEM Support Tools package from Microsoft at:
>
> https://support.microsoft.com/en-us/kb/253066
>
> If so, you might want a newer tool that explicitly states support for
> Windows XP.

Yes. So these tools are older than XP x64. Well I have defraggler,
ultimate defrag and contig.exe none of them seem to touch the MFT. I use
nfi.exe from that zip package. The first entry is the MFT. Two groups of
logical addresses are displayed.

Bill

[VanguardLH]

Bill Cunningham wrote:

> I have defraggler, ultimate defrag and contig.exe none of them seem to
> touch the MFT.

Probably an unmovable system files (e.g., journal) between the fixed
starting location for the MFT and the rest of its extent(s). Does nfi
tell you what is occupying the space between the 2 MFT fragments?

[B00ze]

On 2016-10-31 17:34, Bill Cunningham <nos...@nspam.invalid> wrote:

> Yes. So these tools are older than XP x64. Well I have defraggler,
> ultimate defrag and contig.exe none of them seem to touch the MFT. I use
> nfi.exe from that zip package. The first entry is the MFT. Two groups of
> logical addresses are displayed.
>
> Bill

MyDefrag will defrag the MFT if you have the right commands in your
defrag script:

FileSelect
SelectNtfsSystemFiles(yes)
FileActions
PlaceNtfsSystemFiles(Ascending, VolumeSize * 0.015) // NTFS reserved
area 1.5% of disk
PlaceNtfsSystemFiles(Ascending, MftSize * 2) // NTFS reserved area is
200% of the MFT

Btw, where did you get NFI? We can't download kb253066 anymore :-(

Regards,

--
! _\|/_ Sylvain / B00...@hotmail.com
! (o o) Member:David-Suzuki-Fdn/EFF/Red+Cross/SPCA/Planetary-Society
oO-( )-Oo Autocorrect has become my worst enema.

[JJ]

For Ultimate Defragger, you'll need the boot-time defragmentation to frag
MFT.

[JJ]

On Mon, 31 Oct 2016 23:43:29 -0400, B00ze wrote:
>
> Btw, where did you get NFI? We can't download kb253066 anymore :-(
>
> Regards,

This is why archive.org is worth bookmarking.

<http://wayback.archive.org/web/20141117095714/http://download.microsoft.com/download/win2000srv/utility/3.0/nt45/en-us/oem3sr2.zip>

[Bill Cunningham]

"B00ze" <B00...@hotmail.com> wrote in message
news:nv930a$stk$1...@dont-email.me...

> Btw, where did you get NFI? We can't download kb253066 anymore :-(
>

I have had it for a long time. I better upload it to my online backup.
If anyone wants a copy I can post a link so they can download it.

Bill

[Bill Cunningham]

"JJ" <jj4p...@vfemail.net> wrote in message
news:mg1sis17h6hz$.ixri47ae8rww$.dlg@40tude.net...

> This is why archive.org is worth bookmarking.
>
> <http://wayback.archive.org/web/20141117095714/http://download.microsoft.com/download/win2000srv/utility/3.0/nt45/en-us/oem3sr2.zip>

Yes I see this.Not wuite sure how it functions but very good finding.

Bill

[Bill Cunningham]

"JJ" <jj4p...@vfemail.net> wrote in message

news:fsd3b4yvbv38$.13c12la8wuolr.dlg@40tude.net...

> For Ultimate Defragger, you'll need the boot-time defragmentation to frag
> MFT.

Got it. Done it. It still seems to be in two places. Fragmented. As much
as it's accessed I think being contigous is well worth it.

Bill

[Bill Cunningham]

"B00ze" <B00...@hotmail.com> wrote in message
news:nv930a$stk$1...@dont-email.me...

> FileSelect
> SelectNtfsSystemFiles(yes)
> FileActions
> PlaceNtfsSystemFiles(Ascending, VolumeSize * 0.015) // NTFS reserved area
> 1.5% of disk
> PlaceNtfsSystemFiles(Ascending, MftSize * 2) // NTFS reserved area is 200%
> of the MFT
>
> Btw, where did you get NFI? We can't download kb253066 anymore :-(

There seems to be an error somewhere in that script.

Bill

[VanguardLH]

Bill Cunningham wrote:

> B00ze wrote...

>
>> FileSelect
>> SelectNtfsSystemFiles(yes)
>> FileActions
>> PlaceNtfsSystemFiles(Ascending, VolumeSize * 0.015) // NTFS reserved area
>> 1.5% of disk
>> PlaceNtfsSystemFiles(Ascending, MftSize * 2) // NTFS reserved area is 200%
>> of the MFT
>>
>> Btw, where did you get NFI? We can't download kb253066 anymore :-(
>
> There seems to be an error somewhere in that script.

His NNTP client wrapped the lines. The // is a comment delimiter. His
comments wrapped to the next line.

[Bill Cunningham]

"VanguardLH" <V...@nguard.LH> wrote in message

news:e7s5g8...@mid.individual.net...

It says the command "FileSelect" is invalid <shrug>

Bill

[VanguardLH]

Bill Cunningham wrote:

> VanguardLH wrote...

>
>> Bill Cunningham wrote:
>>
>>> B00ze wrote...
>>>
>>>> FileSelect
>>>> SelectNtfsSystemFiles(yes)
>>>> FileActions
>>>> PlaceNtfsSystemFiles(Ascending, VolumeSize * 0.015) // NTFS reserved area 1.5% of disk
>>>> PlaceNtfsSystemFiles(Ascending, MftSize * 2) // NTFS reserved area is 200% of the MFT
>>>

>>> There seems to be an error somewhere in that script.
>>
>> His NNTP client wrapped the lines. The // is a comment delimiter. His
>> comments wrapped to the next line.
>
> It says the command "FileSelect" is invalid <shrug>

My interpretation of Booze's post is that it's a script for MyDefrag.
Do you have MyDefrag so it can load that script?

MyDefrag (mydefrag.com) went belly up. The author abandoned that
product back in June 2015 and the forums shutdown in Oct 2015. That
doesn't mean it stopped being usable. First it was open sourced
JkDefrag (JK = Jeroen Kessels, the program's author). It became closed
source MyDefrag when Kessels adding scripting support to his program.

Because the author closed his mydefrag.com web site, the doc for his
scripting (www.mydefrag.com/Manual-Scripts.html) also disappeared. I
was able to find an archived copy at:

https://web.archive.org/web/20150911043817/http://www.mydefrag.com/Manual-Scripts.html

A sample script there shows use of the FileSelect command. However, it
looks like that command is valid only after the VolumeSelect scope has
been selected. This is similar to selecting a partition in the diskpart
program but first you have to tell diskpart the volume/disk to select so
it knows where to find the partition.

https://web.archive.org/web/20150811002140/http://www.mydefrag.com/Scripts-FileSelect.html

Yep, I was right. First you have to select the volume scope. Then you
can use the file scope. So Booze's script is incomplete unless MyDefrag
automatically selects a volume (disk) scope, like he picks a drive shown
in the GUI and then a subsequent script already has a volume scope.

[Bill Cunningham]

"VanguardLH" <V...@nguard.LH> wrote in message

news:e7sahn...@mid.individual.net...

I have version 4.3.1. And I have it saved as mftdefrag.myd and it is in
the scripts folder. It shows up on the GUI that pops up.

Bill

[VanguardLH]

Bill Cunningham wrote:

> VanguardLH wrote ...

>
>> Bill Cunningham wrote:
>>
>>> VanguardLH wrote...
>>>

>>> B00ze wrote...
>>>
>>>> FileSelect
>>>> ...

>>>
>>> It says the command "FileSelect" is invalid <shrug>
>>

>> https://web.archive.org/web/20150911043817/http://www.mydefrag.com/Manual-Scripts.html
>>
>> A sample script there shows use of the FileSelect command. However, it
>> looks like that command is valid only after the VolumeSelect scope has
>> been selected. This is similar to selecting a partition in the diskpart
>> program but first you have to tell diskpart the volume/disk to select so
>> it knows where to find the partition.
>>
>> https://web.archive.org/web/20150811002140/http://www.mydefrag.com/Scripts-FileSelect.html
>>
>> Yep, I was right. First you have to select the volume scope. Then you
>> can use the file scope. So Booze's script is incomplete unless MyDefrag
>> automatically selects a volume (disk) scope, like he picks a drive shown
>> in the GUI and then a subsequent script already has a volume scope.
>
> I have version 4.3.1. And I have it saved as mftdefrag.myd and it is in
> the scripts folder. It shows up on the GUI that pops up.

Did you correct the scope of the FileSelect command as mentioned?

[B00ze]

Ohhh goody - I think of archive.org when I look for a dead website, I
somehow always forget that it also has files. Thanks m8!

--
! _\|/_ Sylvain / B00...@hotmail.com
! (o o) Member:David-Suzuki-Fdn/EFF/Red+Cross/SPCA/Planetary-Society

oO-( )-Oo "Where are we?" -Arthur, "Er... somewhere green." -Ford

[B00ze]

On 2016-11-01 17:26, VanguardLH <V...@nguard.LH> wrote:

[snip]

>>>> It says the command "FileSelect" is invalid <shrug>
>>>
>>> https://web.archive.org/web/20150911043817/http://www.mydefrag.com/Manual-Scripts.html
>>>
>>> A sample script there shows use of the FileSelect command. However, it
>>> looks like that command is valid only after the VolumeSelect scope has
>>> been selected. This is similar to selecting a partition in the diskpart
>>> program but first you have to tell diskpart the volume/disk to select so
>>> it knows where to find the partition.
>>>
>>> https://web.archive.org/web/20150811002140/http://www.mydefrag.com/Scripts-FileSelect.html
>>>
>>> Yep, I was right. First you have to select the volume scope. Then you
>>> can use the file scope. So Booze's script is incomplete unless MyDefrag
>>> automatically selects a volume (disk) scope, like he picks a drive shown
>>> in the GUI and then a subsequent script already has a volume scope.
>>
>> I have version 4.3.1. And I have it saved as mftdefrag.myd and it is in
>> the scripts folder. It shows up on the GUI that pops up.
>
> Did you correct the scope of the FileSelect command as mentioned?

Ah sorry guys, I assumed you would program around my example, it is not complete! Indeed you need a VolumeSelect, and ALSO, pick only ONE of the two PlaceNtfsSystemFiles - those were just examples of what you can do. Assuming you have a BIG hard drive, then reserving twice the MFT is better than 1.5% of the free space (which could be huge) so this is what I have below. Also, you want a gap first, otherwise your MFT will end-up at the very beginning of the disk. Here is a full script for defragmenting the MFT; you need to reboot after, for NTFS to start using the new MFT reserved area...

VolumeSelect
CommandlineVolumes()
VolumeActions
MakeGap(RoundUp(VolumeUsed * 0.50, VolumeSize * 0.02), DoNotVacate)
FileSelect
SelectNtfsSystemFiles(yes)
FileActions
PlaceNtfsSystemFiles(Ascending, MftSize * 2)
FileEnd
VolumeEnd

Regards,

--
! _\|/_ Sylvain / B00...@hotmail.com
! (o o) Member:David-Suzuki-Fdn/EFF/Red+Cross/SPCA/Planetary-Society

oO-( )-Oo Ensign Smith, why are you standing behind me?

[Bill Cunningham]

"B00ze" <B00...@hotmail.com> wrote in message

news:nvbclj$g3v$1...@dont-email.me...

> Ah sorry guys, I assumed you would program around my example, it is not
> complete! Indeed you need a VolumeSelect, and ALSO, pick only ONE of the
> two PlaceNtfsSystemFiles - those were just examples of what you can do.
> Assuming you have a BIG hard drive, then reserving twice the MFT is better
> than 1.5% of the free space (which could be huge) so this is what I have
> below. Also, you want a gap first, otherwise your MFT will end-up at the
> very beginning of the disk. Here is a full script for defragmenting the
> MFT; you need to reboot after, for NTFS to start using the new MFT
> reserved area...
>
> VolumeSelect
> CommandlineVolumes()
> VolumeActions
> MakeGap(RoundUp(VolumeUsed * 0.50, VolumeSize * 0.02), DoNotVacate)
> FileSelect
> SelectNtfsSystemFiles(yes)
> FileActions
> PlaceNtfsSystemFiles(Ascending, MftSize * 2)
> FileEnd
> VolumeEnd

I see. Well this one did run. The MFT is still in two pieces but as Paul
said it's really not hurting anything. I wanted it contiguous but I tried
anyway. :)

Bill

[B00ze]

Ah Darn, it /looks/ like it's defragmenting the MFT, and I know
rebooting makes it use the new reserved area since when I defragment
again afterwards, I can see where it has moved. Maybe if you try
SysInternal's CONTIG.EXE ? It's designed to defragment single files. Let
us know if you try this and it works :-)

Regards,

--
! _\|/_ Sylvain / B00...@hotmail.com
! (o o) Member:David-Suzuki-Fdn/EFF/Red+Cross/SPCA/Planetary-Society

oO-( )-Oo ALERT! ALERT! No, wait! It's TWO lerts!

[Paul]

If you don't lose hair trying this,
you're not trying hard enough.

The sad part, is you see so many promises
it's going to work, you have a look and
it's still in a couple pieces.

As for Contig, it's a pure defragmenter with
no optimization policy. It doesn't "push to the
left". If the file system is active (it's your C:
drive), a call to Contig doesn't even have to
defragment a file. It can end up fragmented again.
It can take several tries to succeed. So while
Contig is a "demo" of the defragmenter API, it doesn't
strain itself while attempting to defrag stuff.
Contig is more likely to succeed on a "quiet" file
system (since no other program attempts to "write"
the file system while Contig is on the playing field).

And I doubt it can handle the $MFT, unless you load
the call into BootExecute or something. Maybe it would
work with an OS which has a discrete call for handling
the $MFT, but then Contig doesn't have the logic to use
it, since Contig uses the simplest of the API calls.

At one time, Sysinternals offered PageDefrag, but
that program has been removed from the collection.
The only one I've got, is dated 2009. So that's
an example of a defragmenter, which runs around
BootExecute time or something. The screenshot
does not show it handling any of the "$" metadata
file types. There's no reason to get excited about
this one, except as a demo of an approach (attacking
the file system before the OS is "running").

(For historical reference only...)
http://web.archive.org/web/20120208151704/http://technet.microsoft.com/en-us/sysinternals/bb897426.aspx

Paul

[Bill Cunningham]

"Paul" <nos...@needed.invalid> wrote in message

news:nve6a2$1lh$1...@dont-email.me...

>> Ah Darn, it /looks/ like it's defragmenting the MFT, and I know rebooting
>> makes it use the new reserved area since when I defragment again
>> afterwards, I can see where it has moved. Maybe if you try SysInternal's
>> CONTIG.EXE ? It's designed to defragment single files. Let us know if you
>> try this and it works :-)
>>
>> Regards,

I've already tried contig.exe. I believe I reported there was not change. I
did reboot. It maybe defragging. It is still showing up as two groups of
blocks under nfi.exe. Contiguous allocation is what I am looks for.

[Bill Cunningham]

"Paul" <nos...@needed.invalid> wrote in message
news:nve6a2$1lh$1...@dont-email.me...

...

> At one time, Sysinternals offered PageDefrag, but
> that program has been removed from the collection.
> The only one I've got, is dated 2009. So that's
> an example of a defragmenter, which runs around
> BootExecute time or something. The screenshot
> does not show it handling any of the "$" metadata
> file types. There's no reason to get excited about
> this one, except as a demo of an approach (attacking
> the file system before the OS is "running").
>
> (For historical reference only...)
> http://web.archive.org/web/20120208151704/http://technet.microsoft.com/en-us/sysinternals/bb897426.aspx
>

Yeah I have downloaded that one before. It says it lacks a driver. This
could be because I have XP x64. It's based on the winserver 2003 codebase so
things can be a little tricky in looking for XP stuff.

Bill

[Bill Cunningham]

"B00ze" <B00...@hotmail.com> wrote in message

news:nvdv29$g97$1...@dont-email.me...

> Ah Darn, it /looks/ like it's defragmenting the MFT, and I know rebooting
> makes it use the new reserved area since when I defragment again
> afterwards, I can see where it has moved. Maybe if you try SysInternal's
> CONTIG.EXE ? It's designed to defragment single files. Let us know if you
> try this and it works :-)

I've tried contig.exe before and it doesn't make anything contiguous. It
said already defragg'd or something like that. nfi still is showing two
group's of MFT code.

Bill

[Bill Cunningham]

"Paul" <nos...@needed.invalid> wrote in message
news:nve6a2$1lh$1...@dont-email.me...

[...]

Are the calls you're mentioning the win32 API? I have never did anything
with it. I have a 64 bit gcc Compiler. I am more familiar with C than C++.
Is the API written in C? Should be able to get it somewhere. IF there's
application access to the filesystem.

Bill

[B00ze]

Lol, tough nut to crack. Maybe this is all because your filesystem is
live? I myself use MyDefrag offline (I boot WinPE,) but I always thought
it was good enough to defrag the MFT on a live disk :-(

Regards,

--
! _\|/_ Sylvain / B00...@hotmail.com
! (o o) Member:David-Suzuki-Fdn/EFF/Red+Cross/SPCA/Planetary-Society

oO-( )-Oo Aha! Another "undocumented feature"!

[Bill Cunningham]

"B00ze" <B00...@hotmail.com> wrote in message

news:nvgvbe$ur3$1...@dont-email.me...

> Lol, tough nut to crack. Maybe this is all because your filesystem is
> live? I myself use MyDefrag offline (I boot WinPE,) but I always thought
> it was good enough to defrag the MFT on a live disk

I agree. It should be rebooted and then repair. Contig.exe isn't going to do
that. I myself like defraggler. It reboots and defrags the pagefile.

Bill

[David G. Horsman]

Hi folks. I will be releasing a script library for MyDefrag that may be of interest.
It's quite powerful and will be fully in the background one that feature is added.
It's called TaylorDo Disk Optimizer. You can find the beat release here:

https://github.com/DHorse2/TaylorDo

It's an open source project if anyone wants to get involved.

BTW I have no problems placing and fully defragging the MFT however there are a few system files that can be very problematic, the $??? directories included. On a full disk it can create a real mess that prevents proper volume management. I am added commands to address this as well but it can alter user's settings.

Regards, Dave Horsman

[David G. Horsman]

I can only guarantee that this product works on Vista and above based on the MyDefrag documentation and it's use of the new Windows API. Can someone let me know if XP runs MyDefrag?

[John Dulak]

David:

I've run MyDefrag v4.3.1 on XP.

HTH & GL

John

--
\\\||///
------------------o000----(o)(o)----000o----------------
----------------------------()--------------------------
'' Madness takes its toll - Please have exact change. ''

[Shadow]

https://web.archive.org/web/20150218174231/https://mydefrag.com/Manual-DownloadAndInstall.html

MyDefrag was originally designed for Win 2000, Win XP or
worse(as JKDefrag)..
I use v4.3.1 on my XP system with a larger "gap" than the
default. I also added some very large data files to the "System Disk
Monthly" script so they will be placed "last".
[]'s

--
Don't be evil - Google 2004
We have a new policy - Google 2012
Google Fuchsia - 2021