can one create an image of the hard disk that windows is running at the moment? I have a disk with 3 partitions and two of the three partitions have a different OS to boot with (Windows 7 and Windows 8). So the image of this disk will have the 3 partitions on it ?

CloneDisk
#401
Posted 25 March 2014 - 09:24 PM
#402
Posted 28 March 2014 - 04:49 PM
can one create an image of the hard disk that windows is running at the moment? I have a disk with 3 partitions and two of the three partitions have a different OS to boot with (Windows 7 and Windows 8). So the image of this disk will have the 3 partitions on it ?
You can backup a partition or a full disk.
Ideally you should make your (system) backup offline (i.e from a winpe for instance).
Nethertheless, you could use MS volume shadows to perform an online (system) logical drive backup.
See here.
#403
Posted 28 March 2014 - 04:56 PM
Thank you Wonko! but:
DriverInjection is not updated since 2010 and sometimes it does not work;
MSSTMake is a command line tool. I prefer a GUI tool;
If you give me the registry keys and a brief description of what it to be achieved, I can have a look.
- David Lynch likes this
#405
Posted 28 March 2014 - 05:52 PM
If you give me the registry keys and a brief description of what it to be achieved, I can have a look.
Sorry erwan.l, but I do not have the knowledge for that.
What I can tell you is that OfflineSysPrep is the tool that I used most of the times for this task.
Here is a LiveXP Script.
Here at reboot.pro we have a thread too.
Disk2vhd from Sysinternals has also a similar feature, which changes HAL to match VirtualPC. OfflineSysPrep does a convenient HAL auto detection, matching destination system.
So, to successfully boot a restored image that is hardware independent we have to auto detect HAL and disk (SATA) controller, and change them on the offline registry.
Edited by David Lynch, 28 March 2014 - 05:54 PM.
#406
Posted 29 March 2014 - 03:44 PM
if I read correctly, Windows XP can use 3 different HAL/Kernel :


- David Lynch likes this
#407
Posted 09 April 2014 - 08:58 PM
Hello erwan.l,
Can CloneDisk change the volume serial number?
#409
Posted 10 April 2014 - 05:52 PM
A word of warning (in passing by).
VOL does not reflect "immediately" a changed VolumeID, and changing a volume ID may cause issues, see:
http://technet.micro...s/bb897436.aspx
And (shameless plug ) remember that NTFS serial is longer than what appears:
http://reboot.pro/to...ed-drive-image/
Wonko
#410
Posted 10 April 2014 - 08:17 PM
Yes, sometimes we really need to change this serial.
#411
Posted 11 April 2014 - 10:03 AM
Yes, sometimes we really need to change this serial.
Latest CloneDisk version (2.1.2) can now patch the serial number within the bootsector for NTFS/FAT32/FAT.
Regarding NTFS, the serial number is stored in 8 bytes whereas the system seems to use only the first 4.
See post from Wonko here.
CloneDisk will therefore only read and write the 4 bytes as well.
Double click in the serial number boot sector field to modify it.
The serial number change is seen by the system only after a reboot or after the volume has been remounted.
#412
Posted 11 April 2014 - 10:39 AM
CloneDisk will therefore only read and write the 4 bytes as well.
Wonko disapproves of this .
The general idea should be of having more (or better) features than existing programs.
Sysinternals VolumeID can already change the serial (but only partially. i.e. the "last" 4 bytes).
IMHO at least as an option Clonedisk should allow to change the "full" serial.
Please also note how there is a "queer" concidence or "pattern" on the "hidden" part of the serial on NTFS volumes, see:
http://thestarman.pc.../mbr/NTFSBR.htm
It is interesting to note how the (3rd and 4th) and (6th and 7th) bytes repeat here!
Do you have a Serial Number where these two sets of bytes are not the same?
Though the reason for the "repetition" (as seen in a hex editor):
27 21 A6 C0 32 A6 C0 CE
is unknown, it is a pattern consistently found in NTFS bootsectors, and I personally would like to have it happen also on the "new" serial, possibly as a further option.
OT, but not much, and JFYI, the way the NTFS serial is generated (and more generally how NT based systems, also on other filesystems do that) is one of the fields that were never explored fully, whilst the DOS way to generate the "random" serial was based on date/time.
And, still as a JFYI, and as yet another shameless plug:
http://www.forensicf...ewtopic/t=2134/
http://www.msfn.org/...mages/?p=987748 <- (I am particularly proud of this completely unuseful spreadsheet
)
http://www.msfn.org/...mages/?p=980297
Wonko
#413
Posted 11 April 2014 - 11:23 AM
Hi Wonko,
I knew, or rather I was hoping you would step in
As indeed this "full vs half" serial number is puzzling me.
It seems that all tools outhere (volumeid.exe for example) change only 4 bytes whatever the file system is (ntfs/fa32/fat).
Hence me mimicking this but I am pretty sure these 8 bytes are not there for nothing in NTFS boot records and this especially when indeed it seems that there is a pattern in the 4 pseudo useless bytes.
But it is a fact that windows systems seem to care only about 4 bytes for NTFS (vol.exe reads 4 bytes, always).
What would you advise?
Give the user a choice for NTFS boot records to patch either 4 or 8 bytes?
Actually for me it could be as simple as write any user input bytes between 1 and 8 ...
Regards,
Erwan
#414
Posted 11 April 2014 - 12:37 PM
What would you advise?
Give the user a choice for NTFS boot records to patch either 4 or 8 bytes?
Actually for me it could be as simple as write any user input bytes between 1 and 8 ...
I personally would like a 3 (three) step approach:
- Normal (only the "visible 4 bytes) i.e. replicating exactly what VolumeId does (which I believe is "wrong")
- Advanced (the whole 8 bytes BUT with the user actually inputting ONLY 6 bytes, and have the program automatically replicate 2nd and 3rd to 5th and 6th - if seen as "serial" or 6th and 7th to 3rd and 4th if seen as "RAW" bytes)
- Reckless
(the whole 8 bytes, "freestyle")
Wonko
#415
Posted 11 April 2014 - 01:42 PM
I personally would like a 3 (three) step approach:
- Normal (only the "visible 4 bytes) i.e. replicating exactly what VolumeId does (which I believe is "wrong")
- Advanced (the whole 8 bytes BUT with the user actually inputting ONLY 6 bytes, and have the program automatically replicate 2nd and 3rd to 5th and 6th - if seen as "serial" or 6th and 7th to 3rd and 4th if seen as "RAW" bytes)
- Reckless
![]()
(the whole 8 bytes, "freestyle")
Wonko
Ok, here is what I have done (NTFS bootsector / offset $48):
-case 1 : user can enter 8 bytes ("freestyle" mode)
-case 2 : user can enter 4 bytes ("normal" mode)
In such case, I however make sure that offset $4d=offset $4a and that offset $4e=$4b
#416
Posted 11 April 2014 - 03:32 PM
A button to generate a random serial would be nice too. Call me lazy, yes, you can tell the truth
#417
Posted 11 April 2014 - 03:48 PM
http://xkcd.com/221/A button to generate a random serial would be nice too. Call me lazy, yes, you can tell the truth

Wonko
#418
Posted 11 April 2014 - 04:34 PM
A button to generate a random serial would be nice too. Call me lazy, yes, you can tell the truth
Done : each time you will double click in the serial number to change it, the inputbox will propose a random 4 bytes number.
- David Lynch likes this
#420
Posted 11 April 2014 - 04:54 PM


http://xkcd.com/1277/

Wonko
#421
Posted 13 April 2014 - 08:35 PM
Version 2.1.4 out.
I needed a simple partition editor in WINPE environement where I did not have the MMC.
Latest changelog :
- misty likes this
#422
Posted 14 April 2014 - 06:54 AM
Wow!!! I have barely scratched the surface and I'm already impressed with Clonedisk.
I did a quick test this morning in WinFE (based on WinPE 5.0) with a SANPolicy value set as 4 and NoAutoMount enabled.
On booting the system I ran a few diskpart commands -
Microsoft DiskPart version 6.3.9600 Copyright (C) 1999-2013 Microsoft Corporation. On computer: MINWINPC DISKPART> sel disk 0 Disk 0 is now the selected disk. DISKPART> detail disk HITACHI HTS723216L9SA60 Disk ID: 2344A9FF Type : SATA Status : Offline (Policy) Path : 0 Target : 0 LUN ID : 0 Location Path : PCIROOT(0)#PCI(1F02)#ATA(C00T00L00) Current Read-only State : Yes Read-only : Yes Boot Disk : No Pagefile Disk : No Hibernation File Disk : No Crashdump Disk : No Clustered Disk : No There are no volumes. DISKPART> attrib disk clear readonly DiskPart failed to clear disk attributes.Note that I could not clear the readonly disk attribute. I then used the new version of Clonedisk and cleared the readonly flag. Did a quick test in Diskpart -
Microsoft DiskPart version 6.3.9600 Copyright (C) 1999-2013 Microsoft Corporation. On computer: MINWINPC DISKPART> sel disk 0 Disk 0 is now the selected disk. DISKPART> detail disk HITACHI HTS723216L9SA60 Disk ID: 2344A9FF Type : SATA Status : Online Path : 0 Target : 0 LUN ID : 0 Location Path : PCIROOT(0)#PCI(1F02)#ATA(C00T00L00) Current Read-only State : No Read-only : No Boot Disk : No Pagefile Disk : No Hibernation File Disk : No Crashdump Disk : No Clustered Disk : No Volume ### Ltr Label Fs Type Size Status Info ---------- --- ----------- ----- ---------- ------- --------- -------- Volume 0 Partition 30 GB Healthy Offline Volume 1 Partition 118 GB Healthy Offline DISKPART>I had to assign drive letters manually afterwards. Also noted that running this command may have also cleared the readonly flag on a different disk - I'll verify this later as I'm about to start work.
It's worth noting that (IMNSHO) Clonedisk is intuitive to use and the UI does not appear overly cluttered with features. I look forward to playing around with it - fantastic work!
Regards,
Misty
#423
Posted 14 April 2014 - 06:57 AM
Just noticed that the about button refers to the new version as 2.1.3. Also changelog needs updating.
Regards,
Misty
#424
Posted 14 April 2014 - 07:05 AM
@erwan.l
Just noticed that the about button refers to the new version as 2.1.3. Also changelog needs updating.
Regards,
Misty
Hi Misty,
Thanks the positive feedback.
I'll take care of the about box and changelog.
Regards,
Erwan
#425
Posted 14 April 2014 - 10:31 AM
If you give me the registry keys and a brief description of what it to be achieved, I can have a look.
and @David Lynch
I have looked a bit at the thingy.
It seems to me like the easiest "manual" way is to use the AEK inftoreg tool (and NO other tool but that specific one):
http://www.mdgx.com/files/INFTOREG.ZIP
to convert the "F6 floppy" Mass storage driver .INF to a .REG and then merge this .REG to the "offline" Registry.
This converter - though not the easiest to use and seemingly completely UNlike documented - creates from the .INF file not only the "normal" driver entries but also the CDDB (Critical Devices DataBase) ones.
A minimal amount of "manual interaction" is needed to do the conversion, and it is possible that the reason why DriverInjectionGUI may sometime fail is because of it's "automagic" nature, that is tricked by some "overcomplex" "F6 floppy" .inf's
Probably a good idea (not only restricted to this specific topic) for erwan.l (time and will permitting of course) would be to either modify the offline registry tool:
http://reboot.pro/to...fline-registry/
to be able to parse a .REG file or create a converter from .REG to offline Registry tool commands.
@David Lynch
Can you test the above approach on *something* on which DriverInjectionGUI fails?
Wonko
- David Lynch likes this
3 user(s) are reading this topic
0 members, 3 guests, 0 anonymous users