Is there a way to defrag the MFT file and inode data?

[Antonio Perez]

In a NTFS filesystem there is a file called Master File Table (MFT) that
starts with 12.5% of the disk and grows as needed.

In Windows XP Pro, I tested:
Diskeeper: Total failure, it fakes results.
O&O Defrag: Does actually try to defrag MFT on boot.
PerfectDisk: Does the best job. But inodes are split.

But as reported by the Defrag Utility of Windows the MFT is still in 3
parts. PerfectDisk reports that the MFT was corrected but there are parts
on the inode file.

Bottom line:

Is copying the files to another disk/partition the only "real" solution?

 

[JS]

Read this article: http://www.tweakxp.com/article37043.aspx

Basically your are going to make changes to the registry
(so create a restore point and/or registry backup)
Copy your files to another drive or partition.
Reformat the partition.
Copy the files back.

JS

 

[SoCalCommie]

In a NTFS filesystem there is a file called Master File Table (MFT) that
starts with 12.5% of the disk and grows as needed.

In Windows XP Pro, I tested:
Diskeeper: Total failure, it fakes results.
O&O Defrag: Does actually try to defrag MFT on boot.
PerfectDisk: Does the best job. But inodes are split.

But as reported by the Defrag Utility of Windows the MFT is still in 3
parts. PerfectDisk reports that the MFT was corrected but there are parts
on the inode file.

Bottom line:

Is copying the files to another disk/partition the only "real" solution?

No... the 'real solution' is to stop worring about 'fragments'. Does your
system run faster after using the various 'tools'. If you answer yes...
you're suffering from the placibo effect.

 

[Antonio Perez]

SoCalCommie (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

No... the 'real solution' is to stop worring about 'fragments'. Does your
system run faster after using the various 'tools'. If you answer yes...
you're suffering from the placibo effect.

Thanks, your comment is really helpfull <grin>

 

[Antonio Perez]

JS @ wrote previously in microsoft.public.windowsxp.hardware:

Read this article: http://www.tweakxp.com/article37043.aspx

Basically your are going to make changes to the registry
(so create a restore point and/or registry backup)
Copy your files to another drive or partition.
Reformat the partition.
Copy the files back.

JS

Thanks, I was afraid that was the only solution :-(

 

[Antonio Perez]

SoCalCommie (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

"Antonio Perez" <[email protected]> wrote in message

No... the 'real solution' is to stop worring about 'fragments'. Does your
system run faster after using the various 'tools'. If you answer yes...
you're suffering from the placibo effect.

My previous post was incomplete.

What/who/where does it say that I want/need/wish to make my computer faster?

The reasons for this request are only mine, if you were so kind to ask I
might have told you.

But the arrogance to assume what I need/should/must do is yours only.

What power/status/condition makes my opinion less
valuable/important/relevant than yours?

Is it just plain arrogance to believe/think/feel that you have the truth?

I asked correctly for help. if you want to help, do so, if not do leave.

And, in the end, It's MY computer, I'll do as I see fit.

 

[Gerry]

JS

Have you ever done this to achieve a completely contiguous MFT? The link
is about sizing the MFT disk space allocation. I am uncertain as to
whether the Article fully takes into account the changes that were made
to the NTFS file system with the introduction of Windows XP?

In the past I have tried to find out why there are two fragments on some
partitions and three on others. The information does not seem to be
available. A lot of misinformation on defragmenting the MFT file seems
to have been posted as a result. Some of the claims made by third party
Defragmenter programme suppliers would appear to be spin to promote
sales.



~~~~


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

 

[JS]

No I have not, but have seen this problem and the fix mentioned in
several articles from time to time.

I also used Diskeeper and Norton's Speed Disk and ran some comparative
testing. Was under the impression that Diskeeper solved the
MTF fragmentation but according Antonio but apparently it does not.

JS

 

[Antonio Perez]

JS @ wrote previously in microsoft.public.windowsxp.hardware:

No I have not, but have seen this problem and the fix mentioned in
several articles from time to time.

I also used Diskeeper and Norton's Speed Disk and ran some comparative
testing. Was under the impression that Diskeeper solved the
MTF fragmentation but according Antonio but apparently it does not.

Just check with "Windows Defrag Utility Report". Yes, you need to un-install
Diskeeper to access that, this program hides that tool.

It will report as many fragments as you had before!!!.

Diskeeper just "fudge" about it.

 

[JS]

Have you noticed the similarity between Diskeeper's display and the Windows
Defrag utility.

JS

 

[Antonio Perez]

Not quite, the colors are quite different, at least the version I tried.

Your point being?

AP

JS @ wrote previously in microsoft.public.windowsxp.hardware:

 

[JS]

Well, the method of displaying the files is virtually identical and I think
that the built in Windows defragmenter is the old "Diskeeper Lite" version
they had years ago.

JS

 

[DL]

Go for it, and explore the hidden depths, who knows what you will discover

 

[Antonio Perez]

VanguardLH (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

Once the free sectors beyond the reserved MFT space gets consumed,
additional files will start consuming the "reserved" MFT space....

[big snip]

You are missing completely the point here, the explanation i've read in
painful detail somewhere else.

The point is: What to do _after_ is fragmented...

 

[Antonio Perez]

VanguardLH (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

No one asks a question for no reason.

And nobody posts a response without a reason. What is it for you?
Just make fun on others?

There is always an impetus.
Knowing the impetus allows a respondent to focus on that instead of
wandering all over their entire gamut of MFT knowledge.

[ a lot of yabba dabba]

The question was plain simple. Example:
Q: How to press a button?
A: Put your finger on it an exert pressure till it moves downward.

Could you stop being complex and exotic and give a simple answer to the
question?

No, I suppose that's not your nature.

There was a reason for your query. You choose not to divulge.

You never asked!!!!!!!!!!

You got prodded to divulge your reason. You still refused.

I never refused, I was never asked, but now I choose to specifically and
completelly not tell it to _you_

Respondents can GUESS why you asked but it could help immensely in knowing
WHY you asked. Otherwise, why bother asking here when you could've
Googled

Oh, I googled it in a lot of ways, don't assume incompetence, would you?

for every article discussing the topic if you didn't want to focus on WHY
you would need to defragment the MFT or HOW it might affect any
performance measure (which is still a vague topic without knowing what
measurements you expected)?

Ok, if you absolutely need a reason: because I want a unfragmented MFT.

Is it so alien to your way of thinking?

Presumably you asked here to get something
more than what you found through casual searches at Google and perhaps
at Microsoft's support knowledgebase.

Yes, absolutely, i wanted a clear and knowledged answer.
Something like:
sorry that's not possible.
or:
Do this and then do that.

Not a lot of dumbo-jumbo.

For an example look at JS first answer, clear and to the point, that's it.

 

[Antonio Perez]

VanguardLH (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

Use a hex editor on defrag.exe included in Windows. You'll find the
string "LegalCopyright 2001 Microsoft Corp. and Executive Software".
Executive Software changed their name for Diskeeper Corp in July 2005
(http://redmondmag.com/news/article.asp?editorialsid=6792).

Microsoft gets lots of their utilities from 3rd parties, or they end up
acquiring them.

If both tools are from the same source. Why do they report conflicting info?

Which is wrong here? Both?

 

[Antonio Perez]

JS @ wrote previously in microsoft.public.windowsxp.hardware:

Read this article: http://www.tweakxp.com/article37043.aspx

Basically your are going to make changes to the registry
(so create a restore point and/or registry backup)
Copy your files to another drive or partition.
Reformat the partition.
Copy the files back.

JS

Ok, tried that, MFT in only one string. Thanks.

So, there is no other way to do it?

 

[Gerry]

How often is the MFT file a really significant size in terms of the size
of modern hard drives.

The size of the of the MFT file on my 24 gb windows partitition is 79
mb!


~~~~


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

VanguardLH (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

Once the free sectors beyond the reserved MFT space gets consumed,
additional files will start consuming the "reserved" MFT space....
[big snip]

You are missing completely the point here, the explanation i've read
in painful detail somewhere else.

The point is: What to do _after_ is fragmented...

After all that work, and assuming you increased the
NtfsMftZoneReservation before reformatting the partition, when you run
defrag.msc and run Analyze to look at the report, what is the value
for "Percent MFT in use"?

Was all this effort for a data-only partition? Or did you somehow do
all this for the partition containing Windows?

 

[JS]

It's not.

And if you were using FAT32 instead of NTFS and had the same files as Gerry
has on his partition then you would really see some significant
fragmentation (not reported but the fragmentation is there) because of the
folders splattered all over the partition and the way some defragmentation
tools ignore folder clusters and sandwiched a single file between three of
four folders.

JS

How often is the MFT file a really significant size in terms of the size
of modern hard drives.

The size of the of the MFT file on my 24 gb windows partitition is 79 mb!


~~~~


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

VanguardLH (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

Once the free sectors beyond the reserved MFT space gets consumed,
additional files will start consuming the "reserved" MFT space....
[big snip]

You are missing completely the point here, the explanation i've read
in painful detail somewhere else.

The point is: What to do _after_ is fragmented...

After all that work, and assuming you increased the
NtfsMftZoneReservation before reformatting the partition, when you run
defrag.msc and run Analyze to look at the report, what is the value
for "Percent MFT in use"?

Was all this effort for a data-only partition? Or did you somehow do
all this for the partition containing Windows?

 

[Gerry]

JS

Not many users of Windows XP would find choosing FAT 32 over NTFS a
better choice. There is no MFT file in FAT32. MFT is a product of NTFS!
I am not sure why you have introduced FAT32 to this debate.

The originator of this thread refers to "inode data".
"A data structure holding information about files in a Unix file system.
There is an inode for each file and a file is uniquely identified by the
file system on which it resides and its inode number on that system.
Each inode contains the following information: the device where the
inode resides, locking information, mode and type of file, the number of
links to the file, the owner's user and group ids, the number of bytes
in the file, access and modification times, the time the inode itself
was last modified and the addresses of the file's blocks on disk. A Unix
directory is an association between file leafnames and inode numbers. A
file's inode number can be found using the "-i" switch to ls."
source: tldp.org/LDP/sag/html/glossary.html

Inode data seems to be specific to Unix not Windows! Odd that it should
be mentioned in the Subject of this thread.



~~~~


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

It's not.

And if you were using FAT32 instead of NTFS and had the same files as
Gerry has on his partition then you would really see some significant
fragmentation (not reported but the fragmentation is there) because
of the folders splattered all over the partition and the way some
defragmentation tools ignore folder clusters and sandwiched a single
file between three of four folders.

JS

How often is the MFT file a really significant size in terms of the
size of modern hard drives.

The size of the of the MFT file on my 24 gb windows partitition is
79 mb! ~~~~


Gerry
~~~~
FCA
Stourport, England
Enquire, plan and execute
~~~~~~~~~~~~~~~~~~~

Antonio Perez wrote:

VanguardLH (e-mail address removed) wrote previously in
microsoft.public.windowsxp.hardware:

Once the free sectors beyond the reserved MFT space gets consumed,
additional files will start consuming the "reserved" MFT space....
[big snip]

You are missing completely the point here, the explanation i've
read in painful detail somewhere else.

The point is: What to do _after_ is fragmented...

After all that work, and assuming you increased the
NtfsMftZoneReservation before reformatting the partition, when you
run defrag.msc and run Analyze to look at the report, what is the
value for "Percent MFT in use"?

Was all this effort for a data-only partition? Or did you somehow
do all this for the partition containing Windows?