On 01/12/2024 14:10, Ed Cryer wrote:

I produced an image of my C disk today, and got an initial message that
I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

Ed

If you are imaging the C: drive while the system is running then not everything is copied because some files are protected by the operating
system while they are used.

You can create a rescue media which can be used to boot-up the machine
to clone/image the system disk.

You asked how long this has been going on. The short answer is as long
as "MOST" imaging software were created. I never backup the system disk
while the machine is running. Rescue Media is what I use 100% of the time.


--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: To protect and to server (3:633/280.2@fidonet)

Keywords: VanguardLH,VLH

Ed Cryer <ed@somewhere.in.the.uk> wrote:

I produced an image of my C disk today, and got an initial message that
I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

I've not seen that message. Image backup, or logical (file) backup?

Some causes could be interrupted backup process, not enough space for
the backup on the destination drive, power outage even if momentary,
write error on the destination drive, interference by anti-virus
software, running high or critical priority processes that hog the CPU
so Reflect cannot function. Did you check the backup log?

Did you get a Windows update, but have not rebooted yet to ensure the
fileset was in sync that got changed?

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: Usenet Elder (3:633/280.2@fidonet)

On 01/12/2024 15:10, Ed Cryer wrote:

I produced an image of my C disk today, and got an initial message that
I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

Ed

Really strange.
I use Macrium Reflect for many years, and have never seen anything like
this. Never had a problem.

Fokke

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: ---:- FTN<->UseNet Gate -:--- (3:633/280.2@fidonet)

On Sun, 12/1/2024 9:18 AM, LiverpoolFC Fan wrote:

On 01/12/2024 14:10, Ed Cryer wrote:

I produced an image of my C disk today, and got an initial message that
I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

Ed

If you are imaging the C: drive while the system is running then not everything is copied because some files are protected by the operating system while they are used.

You can create a rescue media which can be used to boot-up the machine
to clone/image the system disk.

You asked how long this has been going on. The short answer is as long
as "MOST" imaging software were created. I never backup the system disk while the machine is running. Rescue Media is what I use 100% of the time.

Incorrect.

Most all of the backup softwares use VSS during a backup.
This is a mechanism for doing "hot" backups of the C: partition
in particular.

The way this works, is softwares with issues (like the OS), they
have to have a "provider" to quiesce certain files. When VSS is
triggered and a shadow copy made (a "frozen" snapshot of the OS partition), there is a ten second interval, where the providers are told it is
time to quiesce. And that's when the shadow is made.

Some files have not been getting backed up. Perhaps the Windows.db or Windows.edb of the SearchIndexer did not get backed up. I don't
think the Pagefile.sys gets backed up. Even though the traffic level
on that file is a *lot* different than in past OSes, it is unlikely
to be a good candidate for the design.

Thus, you will find things missing from a backup. But these are
also not "critical" items. If the pagefile.sys was replaced with
a like-sized file of zeros, you'd never notice at boot time.

And the Windows.edb can be regenerated. As a derived file, it's
not a critical file.

As the system gets more features involving virtualization,
it's hard to comment on the "things you can't see". Maybe
something like that is causing the OPs problem.

But generally speaking, the OS and the applications
are familiar with VSS and the shadow copy system and this
is why we're not freaking out about making backups
while the computer is running.

*******

Now, *if* a users computer has a problem, such as the VSS subsystem
is down, then on Macrium at least, it switches to PSS and that
is its own invention. That likely would be missing some additional files.

I would interpret the OPs message as an indication that
something is wrong with VSS, as a first approximation.

Shadows can be persistent or non-persistent. Macrium may use
persistent shadows for "incremental forever" or "incremental".
There have definitely been cases of some competing backup
products, using up all 64 shadow slots and running out! And
that would represent a failure of sorts. A design failure
by the backup software maker.

vssadmin list shadows

On WinXP, the difference was supposed to be, it did not support
persistent shadows, and you could have a shadow long enough to
make a full backup without a problem. I wasn't always convinced
that it did not have persistent shadows, as sometimes the shadow
list had an entry or two that should not have been there.

Paul

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: A noiseless patient Spider (3:633/280.2@fidonet)

On Sun, 12/1/2024 9:10 AM, Ed Cryer wrote:

I produced an image of my C disk today, and got an initial message that I've not seen before; a warning that parts of the operating system wouldn't be saved.
How long has this been going on?

Ed

Macrium usually makes a log file of some sort.

The error message could be slightly more detailed there,
about the issue.

Something is wrong with VSS on your setup, but I don't
have a nice testbench for you to run or anything.

The following command, is merely to get you some breadcrumbs
for a Google search. The log is where you really need to look.

vssadmin list shadows # as administrator

In services.msc , check the Volume Shadow Copy service.
Startup type is listed as "Manual" and mine is stopped right now.
This suggests we won't know how "healthy" VSS is, until Macrium
tries to start it. And if there is a fault, it will not be
breathing oxygen for very long. Maybe the vssadmin invocation
will start the service running ?

Paul

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: A noiseless patient Spider (3:633/280.2@fidonet)

Keywords: VanguardLH,VLH

Ed Cryer <ed@somewhere.in.the.uk> wrote:

VanguardLH wrote:

Ed Cryer <ed@somewhere.in.the.uk> wrote:

I produced an image of my C disk today, and got an initial message that
I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

I've not seen that message. Image backup, or logical (file) backup?

Some causes could be interrupted backup process, not enough space for
the backup on the destination drive, power outage even if momentary,
write error on the destination drive, interference by anti-virus
software, running high or critical priority processes that hog the CPU
so Reflect cannot function. Did you check the backup log?

Did you get a Windows update, but have not rebooted yet to ensure the
fileset was in sync that got changed?

I've been doing this same C backup for years. Suddenly the message appears. I've often restored from the image; always successfully.
Today's backup finished ok. The log just says image completed successfully.

Maybe you're right about some critical priority process running. Just a
very infrequent coincidence. I'll see what happens next time.

Ed

I figured to look at the log for the backup job that issued the warning
that system files couldn't be backed up.

I'd probably also run "chkdsk /r" on the drive to give some assurance
all clusters can be read.

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: Usenet Elder (3:633/280.2@fidonet)

Ed Cryer wrote:

I produced an image of my C disk today, and got an initial message that
I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

Ed

PaulNoSpam suspects a VSS error.
The Reflect folder contains file "vssfixx64.exe".
Run that and see if it helps.

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: A noiseless patient Spider (3:633/280.2@fidonet)

Keywords: VanguardLH,VLH

Paul in Houston TX <Paul@Houston.Texas> wrote:

Ed Cryer wrote:

I produced an image of my C disk today, and got an initial message that
I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

PaulNoSpam suspects a VSS error.

The Reflect folder contains file "vssfixx64.exe".
Run that and see if it helps.

The OP hasn't look at the Reflect log for the failed backup job, so the
error code noted in the article below may not be what he got:

https://knowledgebase.macrium.com/display/KNOWX/VSS+Error%3A+0x8004230F

I have Reflect, and it already has the vssfixx64.exe in its folder. The properties for mine says it is version 1.0.2.1. The downloaded x64 file
from the above article is the same version.

Besides the possible causes already mentioned, I didn't about the OP
using another VSS product, like ShadowProtect.

https://knowledgebase.macrium.com/display/KNOWX/VSS+fails+due+to+modification+by+3rd+party+software

"vssadmin list writers" will show what others there are. Nothing popped
out as obvious for Reflect's VSS writer. The names aren't all obvious. "vssadmin list providers" only shows me the one from Microsoft. At HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\VSS\Providers in
the registry, there is only 1 entry (showing the MS provider). Perhaps
a VSS provider can register themself at the time they are called.

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: Usenet Elder (3:633/280.2@fidonet)

Keywords: VanguardLH,VLH

VanguardLH <V@nguard.LH> wrote:

Paul in Houston TX <Paul@Houston.Texas> wrote:

Ed Cryer wrote:

I produced an image of my C disk today, and got an initial message that >>> I've not seen before; a warning that parts of the operating system
wouldn't be saved.
How long has this been going on?

PaulNoSpam suspects a VSS error.

The Reflect folder contains file "vssfixx64.exe".
Run that and see if it helps.

The OP hasn't look at the Reflect log for the failed backup job, so the
error code noted in the article below may not be what he got:

https://knowledgebase.macrium.com/display/KNOWX/VSS+Error%3A+0x8004230F

I have Reflect, and it already has the vssfixx64.exe in its folder. The properties for mine says it is version 1.0.2.1. The downloaded x64 file
from the above article is the same version.

Besides the possible causes already mentioned, I didn't about the OP
using another VSS product, like ShadowProtect.

https://knowledgebase.macrium.com/display/KNOWX/VSS+fails+due+to+modification+by+3rd+party+software

"vssadmin list writers" will show what others there are. Nothing popped
out as obvious for Reflect's VSS writer. The names aren't all obvious. "vssadmin list providers" only shows me the one from Microsoft. At HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\VSS\Providers in
the registry, there is only 1 entry (showing the MS provider). Perhaps
a VSS provider can register themself at the time they are called.

And for an even more esoteric cause:

There are spurious errors that remain unresolved that logs won't reveal
the causes. They'll reveal an error code, but not the causes. For
example, gamma radiation can significantly disrupt the circuitry in a
computer: ionizing atoms, flipping bits in memory, causing random
errors. I didn't bother to check solar history to see if there was a
recent gamma burst from the Sun, or if Earth happened to be in the
sweeping path of a pulsar. I haven't seen "radiation hardened
computers" for sale to consumers, but then I never looked since I wasn't
going to spend $200K to $300K on one.

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: Usenet Elder (3:633/280.2@fidonet)

On Sun, 12/1/2024 8:15 PM, VanguardLH wrote:

And for an even more esoteric cause:

There are spurious errors that remain unresolved that logs won't reveal
the causes. They'll reveal an error code, but not the causes. For
example, gamma radiation can significantly disrupt the circuitry in a computer: ionizing atoms, flipping bits in memory, causing random
errors. I didn't bother to check solar history to see if there was a
recent gamma burst from the Sun, or if Earth happened to be in the
sweeping path of a pulsar. I haven't seen "radiation hardened
computers" for sale to consumers, but then I never looked since I wasn't going to spend $200K to $300K on one.

Normally, I would cast scorn on this idea, but I did have
my *wrist watch* disturbed while it was sitting on my desk
here. I wasn't watching it, but the settings were "all screwy"
and the battery was still good.

Normal energy cosmic rays, shouldn't be able to do that
on 3 micron CMOS. CMOS has high noise margins to begin
with, as logic goes, and large geometry means that
a cosmic ray, at most, should upset one bit kind of thing,
not blow the bloody doors off it.

So something energetic happened, at a guess.

Maybe a cosmic ray went through the watch sideways,
but even that would not flip everything.

The metal of the watch, should give some protection
from an RF attack (say, a microwave beam).

That's the reason I now have two wrist watches sitting
on the desk. Yet another experiment, to see if there is
ever a next time for one of these, both watches are
kicked around or not. Other devices in the room, were
not affected. It's not like all sorts of stuff dropped
at the same time.

*******

With regard to the OPs device, it has 50 bytes of error
syndrome to protect 512 bytes of data. Even if hit with
a chain saw, it should be able to correct the error
and reconstitute the data. It should still be able to boot.

If you subjected a flash device to an XRay source, you
might be able to erase part of it that way. But how many
portable XRay sources do you have around the house ?
I don't think I have anything here, to give me 40kV to work
with. The best I can do is around 15kV.

Paul


--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: A noiseless patient Spider (3:633/280.2@fidonet)

Paul wrote:

On Sun, 12/1/2024 8:15 PM, VanguardLH wrote:

And for an even more esoteric cause:

There are spurious errors that remain unresolved that logs won't reveal
the causes. They'll reveal an error code, but not the causes. For
example, gamma radiation can significantly disrupt the circuitry in a
computer: ionizing atoms, flipping bits in memory, causing random
errors. I didn't bother to check solar history to see if there was a
recent gamma burst from the Sun, or if Earth happened to be in the
sweeping path of a pulsar. I haven't seen "radiation hardened
computers" for sale to consumers, but then I never looked since I wasn't
going to spend $200K to $300K on one.

Normally, I would cast scorn on this idea, but I did have
my *wrist watch* disturbed while it was sitting on my desk
here. I wasn't watching it, but the settings were "all screwy"
and the battery was still good.

Normal energy cosmic rays, shouldn't be able to do that
on 3 micron CMOS. CMOS has high noise margins to begin
with, as logic goes, and large geometry means that
a cosmic ray, at most, should upset one bit kind of thing,
not blow the bloody doors off it.

So something energetic happened, at a guess.

Maybe a cosmic ray went through the watch sideways,
but even that would not flip everything.

The metal of the watch, should give some protection
from an RF attack (say, a microwave beam).

That's the reason I now have two wrist watches sitting
on the desk. Yet another experiment, to see if there is
ever a next time for one of these, both watches are
kicked around or not. Other devices in the room, were
not affected. It's not like all sorts of stuff dropped
at the same time.

*******

With regard to the OPs device, it has 50 bytes of error
syndrome to protect 512 bytes of data. Even if hit with
a chain saw, it should be able to correct the error
and reconstitute the data. It should still be able to boot.

If you subjected a flash device to an XRay source, you
might be able to erase part of it that way. But how many
portable XRay sources do you have around the house ?
I don't think I have anything here, to give me 40kV to work
with. The best I can do is around 15kV.

Paul

Sticky tape?
https://physicsworld.com/a/sticky-tape-takes-x-ray-images/
It needs a vacuum though, otherwise it just ionizes the air.

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: A noiseless patient Spider (3:633/280.2@fidonet)

Ed Cryer <ed@somewhere.in.the.uk> wrote:
[...]

Everything looks OK. The image file is slightly larger than the previous one; the log consists of one line - image successful; ...

You probably mean the message in the popup window of the current
backup, or in the 'Logs' *list*. But the actual log should be much
longer (not that I expect more relevant info there):

'Logs' tab -> double click the relevant entry in the list (or
right-click it -> View)

(this is for the last free version, 8.0.7783)

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: NOYB (3:633/280.2@fidonet)

Ed Cryer <ed@somewhere.in.the.uk> wrote:

Ed Cryer wrote:

Frank Slootweg wrote:

Ed Cryer <ed@somewhere.in.the.uk> wrote:
[...]

Everything looks OK. The image file is slightly larger than the previous >>> one; the log consists of one line - image successful; ...

ÿÿ You probably mean the message in the popup window of the current
backup, or in the 'Logs' *list*. But the actual log should be much
longer (not that I expect more relevant info there):

ÿÿ 'Logs' tab -> double click the relevant entry in the list (or
right-click it -> View)

(this isÿ for the last free version, 8.0.7783)

That's the stuff that rolls up the screen during runtime. And yes, all
OK. And no mention of the warning I saw (which, incidentally, had the option to "not show again"; as if it were purely informational).

I'm on version 8.0.7783 also. So it's not a recent update.
So then, why haven't I had that display previously? Strange occurrence.
^M
Ed

https://www.dropbox.com/scl/fi/s70y6pifzb0e69wtllpok/Macrium-msg.jpg?rlkey=75kmtcripzu018e8fq84nvn3p&e=1&dl=0
The Macrium message box.

[Re-typed text of JPG file:]

"Macrium Reflect

Backup may not include all required partitions

It appears as if you are trying to create a backup of the operating system.

The current selection may not include all partitions neccessary for this.

Preceed anyway?


[ ] Do not show this again [OK] [Cancel]"

Ah, this *is* a familiar one and a bit different from what you said in
your OP, which was "a warning that parts of the operating system
wouldn't be saved", while the actual warning talks about not including
all *partitions* (not "parts").

I've only seen this warning message when creating or modifying a
Backup Definition file (Do you use a Backup Definition File?), but
perhaps it can also occur during other situations.

Anyway, as you said "I produced an image of my C disk today", the
warning message is exactly telling you that: You didn't include all the partitions needed to create a backup of your operating system, because
you didn't include the other 'boot' recovery, etc. partitions.

So if you tell us how exactly your perform your image backup, i.e. the detailed steps from starting Macrium Reflect till the actual warning
message occurs, we might be able to determine why you get the message.

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: NOYB (3:633/280.2@fidonet)

On Mon, 12/2/2024 1:40 PM, Frank Slootweg wrote:

Ed Cryer <ed@somewhere.in.the.uk> wrote:

Ed Cryer wrote:

Frank Slootweg wrote:

Ed Cryer <ed@somewhere.in.the.uk> wrote:
[...]

Everything looks OK. The image file is slightly larger than the previous >>>>> one; the log consists of one line - image successful; ...

ÿÿ You probably mean the message in the popup window of the current
backup, or in the 'Logs' *list*. But the actual log should be much
longer (not that I expect more relevant info there):

ÿÿ 'Logs' tab -> double click the relevant entry in the list (or
right-click it -> View)

(this isÿ for the last free version, 8.0.7783)

That's the stuff that rolls up the screen during runtime. And yes, all
OK. And no mention of the warning I saw (which, incidentally, had the
option to "not show again"; as if it were purely informational).

I'm on version 8.0.7783 also. So it's not a recent update.
So then, why haven't I had that display previously? Strange occurrence.
^M
Ed

https://www.dropbox.com/scl/fi/s70y6pifzb0e69wtllpok/Macrium-msg.jpg?rlkey=75kmtcripzu018e8fq84nvn3p&e=1&dl=0
The Macrium message box.

[Re-typed text of JPG file:]

"Macrium Reflect

Backup may not include all required partitions

It appears as if you are trying to create a backup of the operating system.

The current selection may not include all partitions neccessary for this.

Preceed anyway?

[ ] Do not show this again [OK] [Cancel]"

Ah, this *is* a familiar one and a bit different from what you said in
your OP, which was "a warning that parts of the operating system
wouldn't be saved", while the actual warning talks about not including
all *partitions* (not "parts").

I've only seen this warning message when creating or modifying a
Backup Definition file (Do you use a Backup Definition File?), but
perhaps it can also occur during other situations.

Anyway, as you said "I produced an image of my C disk today", the
warning message is exactly telling you that: You didn't include all the partitions needed to create a backup of your operating system, because
you didn't include the other 'boot' recovery, etc. partitions.

So if you tell us how exactly your perform your image backup, i.e. the detailed steps from starting Macrium Reflect till the actual warning
message occurs, we might be able to determine why you get the message.

On a GPT partitioned disk

ESP EFI System Partition (FAT32) Has boot materials (Microsoft folder, Ubuntu folder, ...)

C: OS Partition (NTFS 4K cluster)

Recovery Partition Has WinRE.wim file, and is marked as Required (Required in a backup too)

The Required flag seems to correlate with GPT Attribute 0x0000000000000001 <=== trailing bit set

But not all the partitions are marked as Required. Using our architecture
hat, we guess that the ESP is defacto-required. We guess that C:
is defacto-required. This leaves Recovery partition to receive an explicit Required flag, so it gets picked up in a backup. Not all OS installations
have a System Reserved or a Recovery Partition. But if it exists, it will
be marked as Required.

*******

This is a Recovery Partition

DISKPART> select partition 4

Partition 4 is now the selected partition.

DISKPART> detail partition

Partition 4
Type : de94bba4-06d1-4d40-a16a-bfd50179d6ac
Hidden : Yes
Required: Yes <=== Thus, a good idea to include next to C: in a backup
Attrib : 0X0000000000000001
Offset in Bytes: 127604359168

Volume ### Ltr Label Fs Type Size Status Info
---------- --- ----------- ----- ---------- ------- --------- -------- * Volume 4 NTFS Partition 649 MB Healthy Hidden

*******

[Picture]

https://i.postimg.cc/N0hDy3Wc/partition-sets.gif

Paul

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: A noiseless patient Spider (3:633/280.2@fidonet)

Ed Cryer wrote on 12/2/24 9:43 AM:

https://www.dropbox.com/scl/fi/s70y6pifzb0e69wtllpok/Macrium-msg.jpg?rlkey=75kmtcripzu018e8fq84nvn3p&e=1&dl=0

The Macrium message box.

Ed

That's different than your original post query.

The only way to answer correctly, now, would be for you to clarify what
option you chose in MR(left pane options) and specifically what you
selected for the primary disk's partitions for 'backing up'.


--
....w¡ñ§±¤ñ

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: windowsunplugged.com (3:633/280.2@fidonet)

Paul wrote on 12/2/24 7:20 PM:

https://i.postimg.cc/N0hDy3Wc/partition-sets.gif

Paul

With Macrium I also choose the MSR(locsted after System, prior to Windows
and WinRE partitions.

Macrium when MSR is present auto selects MSR and Recovery when choosing
the image selected disks option. When choosing the option to image the partitions required to backup and restore Windows - the MSR and WinRE
require user selection.

--
....w¡ñ§±¤ñ

--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: windowsunplugged.com (3:633/280.2@fidonet)

On Tue, 12/3/2024 1:16 PM, ...w¡ñ§±¤ñ wrote:

Paul wrote on 12/2/24 7:20 PM:

ÿÿÿÿ https://i.postimg.cc/N0hDy3Wc/partition-sets.gif

ÿÿÿ Paul

With Macrium I also choose the MSR(locsted after System, prior to Windows and WinRE partitions.

Macrium when MSR is present auto selects MSR and Recovery when choosing the image selected disks option. When choosing the option to image the partitions required to backup and restore Windows - the MSR and WinRE require user selection.

[Picture]

https://i.postimg.cc/3Rqjth4W/partition-sets-and-Macrium.gif

It's pretty hard to write a cohesive "story" about the OS, when
the MSR partition is

1) Not shown in Disk Management. <=== Why? Show it! This upsets the partition numbering FFS.
2) Does not have a File System.
3) Does not have a 1:1 relationship with installed OSes.
If you have two copies of Windows, there is no MSR in front of the second OS.

The only thing left, is the MSR has some cryptographic purpose.

And, I can tell you, that the MSR causes no end of problems in Linux gparted. When you want to move things in gparted, gparted is "careful to do a post-move verification of *file system*, to detect damage caused by GParted". It finds no file system. It then freaks out and now anything to the left of the MSR,
can't be moved. Windows doesn't move partitions. Linux can't move an MSR. Oh, joy.

*******

DISKPART> list partition

Partition ### Type Size Offset
------------- ---------------- ------- -------
Partition 1 System 100 MB 1024 KB
Partition 2 Reserved 16 MB 101 MB <=== the bastard partition
Partition 3 Primary 118 GB 117 MB
Partition 4 Recovery 649 MB 118 GB
Partition 5 Primary 128 GB 119 GB
Partition 6 Recovery 1025 MB 248 GB
Partition 7 Primary 682 GB 249 GB

DISKPART> select partition 2

Partition 2 is now the selected partition.

DISKPART> detail partition

Partition 2
Type : e3c9e316-0b5c-4db8-817d-f92df00215ae
Hidden : Yes
Required: No <=== For OS purpose No, for disk maintenance Yes Attrib : 0000000000000000
Offset in Bytes: 105906176

There is no volume associated with this partition. <===

DISKPART>

*******

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

"Formerly, on disks formatted using the master boot record (MBR) partition layout,
certain software components used hidden sectors of the disk for data storage purposes.
For example, the Logical Disk Manager (LDM), on dynamic disks, stores metadata
in a 1 MB area at the end of the disk which is not allocated to any partition.[3]

The UEFI specification does not allow hidden sectors on GPT-formatted disks.

Microsoft reserves a chunk of disk space using this MSR partition type,
to provide an alternative data storage space for such software components
which previously may have used hidden sectors on MBR formatted disks.
Small, special-purpose partitions can be allocated from a portion of
the space reserved in the MSR partition."

"The Microsoft-recommended size of MSR:

Windows 7 128 MB[4]
Windows 8 & 8.1 128 MB[5]
Windows 10 16 MB[6] "

*******

On the one hand, we have the claim it is "The Last Cylinder".

Yet the table of values, what does that pattern mean ?
Half a story, is no story at all.

128MB of space is not a last cylinder. And the GPT partition table
primary and secondary are not 128MB either.

But at least only having one MSR per disk drive, makes a little more sense now. It seems to be a disk level structure and not an OS level structure. Maybe.
Or maybe not.

And as a mystery meat, we'll be backing it up I guess.

*******

Windows 7 backup does not appear to back that up. Hmmm. The System Image
it made, it ignored the 16MB space. And none of the XML files, contain
anything of interest. The 6MB file is a declaration of system files
at the file level. Nothing in the "noise section" of the backup, equates
to an encoded MSR content.

Directory of D:\WindowsImageBackup\WALLACE\Backup 2024-12-03 222612 A Smart Copy, only Busy clusters recorded...

Tue, 12/03/2024 05:29 PM 60,817,408 Esp.vhdx ESP
Tue, 12/03/2024 05:29 PM 70,290,243,584 0bd6166a-0836-4041-891c-792df2c72abd.vhdx C:
Tue, 12/03/2024 05:29 PM 616,562,688 c3bc5ab1-c5f0-4dae-838c-751ef868e237.vhdx Recovery Partition (WinRE sized)

Tue, 12/03/2024 05:29 PM 776 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_AdditionalFilesc3b9f3c7-5e52-4d5e-8b20-19adc95a34c7.xml
Tue, 12/03/2024 05:29 PM 18,948 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Components.xml
Tue, 12/03/2024 05:29 PM 6,884 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_RegistryExcludes.xml
Tue, 12/03/2024 05:29 PM 10,728 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Writer4dc3bdd4-ab48-4d07-adb0-3bee2926fd7f.xml
Tue, 12/03/2024 05:29 PM 1,488 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Writer542da469-d3e1-473c-9f4f-7847f01fc64f.xml
Tue, 12/03/2024 05:29 PM 1,484 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Writera6ad56c2-b509-4e6c-bb19-49d8f43532f0.xml
Tue, 12/03/2024 05:29 PM 2,940 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Writerafbab4a2-367d-4d15-a586-71dbb18f8485.xml
Tue, 12/03/2024 05:29 PM 7,962 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Writerbe000cbe-11fe-4426-9c58-531aa6355fc4.xml
Tue, 12/03/2024 05:29 PM 5,402 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Writercd3f2362-8bef-46c7-9181-d62844cdc0b2.xml
Tue, 12/03/2024 05:29 PM 6,589,944 7a51ce53-7e37-4e53-a0e3-a8f4f1a54b4b_Writere8132975-6f93-4464-a53e-1050253ae220.xml
Tue, 12/03/2024 05:29 PM 1,528 BackupSpecs.xml

14 File(s) 70,974,271,764 bytes
2 Dir(s) 22,100,594,688 bytes free

An inconclusive story. OK to back it up. But don't know why.

Sure, when the Windows 7 (W11 Edition) backup is restored, the restoration process
could create an all-zeros 16MB area if it wants. That's possible,
but is not a "preservation" of whatever the content was before.
If there was a Veritas Dynamic Disks declaration, that would be missing.

Paul




--- MBSE BBS v1.0.8.4 (Linux-x86_64)
* Origin: A noiseless patient Spider (3:633/280.2@fidonet)