• Re: Recovery Partition Resize

    From =?UTF-8?B?Li4ud8Khw7HCp8KxwqTDsQ==?=@winstonmvp@gmail.com to alt.comp.os.windows-11 on Fri Apr 24 12:38:47 2026
    From Newsgroup: alt.comp.os.windows-11

    On 4/24/2026 12:12 PM, Lars Poulsen wrote:
    On 2026-04-23 22:17, ...w¡ñ§±¤ñ wrote:
    On 4/23/2026 1:25 PM, Paul wrote:
    On Thu, 4/23/2026 2:52 PM, ...w¡ñ§±¤ñ wrote:
    On 4/22/2026 2:49 PM, Paul wrote:
    On Wed, 4/22/2026 5:11 PM, Lars Poulsen wrote:
    On 2026-04-22 09:42, ...w¡ñ§±¤ñ wrote:
    200 MB free on the Recovery is a concern.

    Did you perform the recommended Powershell command suggestion for >>>>>>> free space on the disk.
         GET-VOLUME
    If so, posting information on that results would be a starting
    point.
    Run the command, copy and paste the results in a reply.

    Additionally, for the recovery partition it would be helpful to know >>>>>>> your disk # and recovery partition #

    Powershell admin command. Copy the below command, paste or type into >>>>>>> Powershell, run the command, then copy and paste the results in the >>>>>>> same reply.

    reagentc /info

    Okay, Here goes:

    PS C:\Windows\System32> get-volume

    DriveLetter FriendlyName FileSystemType DriveType HealthStatus
    OperationalStatus SizeRemaining      Size
    ----------- ------------ -------------- --------- ------------
    ----------------- -------------      ----
                  Recovery     NTFS           Fixed     Healthy
    OK              199.92 MB   1.17 GB
                  SYSTEM       FAT32          Fixed     Healthy
    OK                31.3 MB     96 MB
    C           Windows      NTFS           Fixed     Healthy      OK
                309.88 GB 475.65 GB


    PS C:\Windows\System32> reagentc /info
    Windows Recovery Environment (Windows RE) and system reset
    configuration
    Information:

          Windows RE status:         Enabled
          Windows RE location: \\?
    \GLOBALROOT\device\harddisk0\partition4\Recovery\WindowsRE
          Boot Configuration Data (BCD) identifier:
    d05140b8-47c2-11f0- b542-acabeae3b1dc
          Recovery image location:
          Recovery image index:      0
          Custom image location:
          Custom image index:        0
          Windows RE Version:        10.0.26100.8235

    REAGENTC.EXE: Operation Successful.

    PS C:\Windows\System32>

    Disk management shows from left to right:

    [Disk 0   ]             [  Windows C:  ][        ]
    [  Basic  ][           ][475.65 GB NTFS][1.17 GB ]
    [476.92 GB][  100 MB   ][Healthy (Boot,][Healthy ]
    [  Online ][Healthy EFI][ PageF,Crash,B][Recovery]

    Does this convey any new information?

    That's perfectly normal looking, with the 100MB FAT32 ESP on the
    left as the first partition.

    +-----------+ - - - - - - - - - - - - - - - - - - - -+
    |  Disk 0   |            |  Windows C:    |          |
    |  Basic    |            | 475.65 GB NTFS |  1.17 GB | >>>>> | 476.92 GB |  100 MB    | Healthy (Boot, | Healthy  |
    |  Online   |Healthy EFI | PageF,Crash,Ba | Recovery |
    +-----------+ - - - - - - - - - - - - - - - - - - - -+

    That's a third party shrink of C:
    A third party move of Recovery, to the left.
    A third party resize of Recovery, using the freed-up space to the
    right of Recovery.

         Paul

    ?
    Looks like the Recovery partition is on the right(and where it
    should be)...
    ...and sometime in the past(b/c I suspect the original oobe Recovery
    partition was 1 GB) and one or more times Windows Update(when the
    recovery had 250 or more free space) shrunk C and enlarged the
    recovery to its current 1.17 GB.

    Based on the information Lars provided, it does not necessarily
    indicate that the Recovery partition was moved from any other
    location, nor being re-sized by 3rd party application.


    I was describing how you would go about fixing the situation.

    Currently the OP reports that there is a "blocker" that the
    Windows usage of the Defrag API, cannot move.

    Expected. Most likely protected files.
    Also, most wouldn't or need to defrag an SSD.


    Third party tools can move it. Some of them will require a
    reboot, into some media that comes with the tool, to do the
    operation.

    Afaics, moving via defrag or 3rd party tool is not the necessary or
    desired path.

    I use Linux gparted for this, and have done this a number of
    times now, as I experimented with various sizes of Recovery Partitions.

    I've already indicated, that in one failure case, I found *two*
    WinRE.wim files in the Recovery Partition. But I've only recorded
    that situation one time, in all the work I've done on Recovery
    Partitions.
    That would be 2*750MB plus a couple hundred more of slack space,
    so maybe 2GB for the Recovery Partition would cover all the
    failure scenarios possible.
    Yes, I recall that condition, it does not appear to be representative
    of Lar's issue.

    By doing it this way (Partition Manager), I don't have to lean on PBR
    resets as often.

    A quick test with AOMEI, and a fresh install of Win10, finds no
    blocker half way out on the disk. It was a Win10 21H2 (as part
    of another experiment). I checked with JKDEFRAG and there was
    nothing at the half-way point. As a result, there was no opportunity
    to test AOMEI on "blockers". I did manage to move the Recovery
    Partition OK and resize it. That part worked.

    Good,but afaics, nothing needs to be 'moved' - the partitions(System,
    MSR, Windows, Recovery), partition 1, 2, 3, 4. are in the correct order.

    Powershell Get-Volume verifed the partitions size and free space.
    Reagentc /info verifed the Recovery partition as disk 0, partition 4
    Get-Volume or Disk Management does not show MSR partition #2(its not
    formatted or labeled) but it is certainly present since Window is left
    adjacent(thus partition #3) to partition #4(Recovery).

    I am still stuck without a simple way to create the needed 400 MB of
    free space in the recovery partition. As I understand it, the path Paul
    has suggested is to create a bootable USB stick with a 3rd-party partitioning tool, and use that to move the partition boundary.
    That sounds a bit cumbersome and scary. And based on recent threads
    here, this is becoming a problem waiting to happen for many Windows-11 users. Every Windows-11 machine I have looked at recently has less than
    250 MB of free space in their recovery partition.

    I wouldn't do it that way.

    In my earlier reply, I asked for more details on the current recovery partition version and build. Here's the request again.

    Open a Powershell or command.com admin prompt window.
    Cut and paste the following line into that window.

    DISM /Get-ImageInfo /ImageFile:\\?\GLOBALROOT\device\harddisk0\partition4\Recovery\WindowsRE\winre.wim
    /index:1

    Report the results(or cut and paste them) in a reply.
    We're specifically looking for the ServicePack Build number, and the
    last created and modified date.

    Once verified, you can use the diskpart method to resize Windows and the Recovery Partition without the need for 3rd party or bootable usb or any
    other method.
    --
    ...w¡ñ§±¤ñ
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Paul@nospam@needed.invalid to alt.comp.os.windows-11 on Fri Apr 24 17:06:04 2026
    From Newsgroup: alt.comp.os.windows-11

    On Fri, 4/24/2026 3:12 PM, Lars Poulsen wrote:


    I am still stuck without a simple way to create the needed 400 MB of free space in the recovery partition. As I understand it, the path Paul has suggested is to create a bootable USB stick with a 3rd-party partitioning tool, and use that to move the partition boundary.
    That sounds a bit cumbersome and scary. And based on recent threads here, this is becoming a problem waiting to happen for many Windows-11 users. Every Windows-11 machine I have looked at recently has less than 250 MB of free space in their recovery partition.

    When I tested, there was no structure at all at 50%, so I failed
    to reproduce the situation. I know there are definitely blockers
    at 50%, because I've had them before and the Shrink Volume would
    not go below 50%. When I tried the AOMEI, it had no problem shrinking
    C: a lot more than I needed (and it could do that "online" as
    there were no blockers). I could also move the Recovery Partition
    to the left then, followed by resizing it as desired. The only problem
    with AOMEI, is having a numeric display for precise movement of
    the sizes I want. In some cases, the GUI is a tiny bit lacking.
    Some dialogs have dimensions, and some are just graphical.

    [Picture] aomei-first-test.gif (could only test Recovery Partition move/resize)

    https://postimg.cc/bZ4xPYgj

    I tried a bunch more archived VHDs I have on the NAS, and none of them have blockers
    either. (I use JKDefrag just for the "block view" so I can see where items are located on disk. I am not making any changes at all with JKDefrag. I use it
    for surveillance.) CoPilot suggested a list of items that can function as blockers, so I'm going to try that next (after I come back).

    https://www.diskpart.com/articles/shrink-volume-beyond-half-its-size-4348.html

    pagefile.sys \
    hiberfil.sys \____ For sufficiently large C: to begin with, these will
    $MFT / be below the middle
    $Extend\$UsnJrnl <=== Possible, more research needed
    System Restore / Shadow Copies <=== Possible, more research needed
    Bad clusters <=== Highly highly unlikely.

    You can see in the "style department", AOMEI will be using a reboot
    and running from a WinPE, while it does the resize in that case.

    I can't easily test those. About the only one easy to run, is
    artificially making a Restore Point is easy to do.

    *******

    I don't know about cumbersome. AOMEI just has to hijack something
    to use as a WinPE. It could build its own WinPE. It could use
    the WinRE.wim in the Recovery partition (Macrium Rescue CD builder
    does that, in the more modern versions of Macrium). It should not
    take long to reboot and do the job. Paragon Partition Manager 14 Free
    also has a lightweight reboot mechanism (about as lightweight as some
    earlier tools that did things like that, the OS item it used came
    with the package). Loading WinPE usually takes a bit longer than the
    old ways of doing it, but the WinPE is a lot more sophisticated in
    terms of subsystems and runtime capability.

    This shouldn't take long.

    1) Download the free version from the link on the diskpart page.
    AOMEI uses the diskpart web domain for advertising articles.
    This is what I got while testing yesterday (downloaded from some AOMEI page).

    Name: PAssist_Std_20260424.22679053.exe
    Size: 86,896,568 bytes (82 MiB)
    SHA256: 0C244FF57E35174E9FA017DF06CA54B7EDF5927D164E2ABD22AD44B8D7BBDE2C

    2) Install the program. Resize C: with Move/Resize.
    The program will prompt for a reboot. Since you are only
    making C: slightly smaller, not a lot of movement of files on C:
    will be required. It should only take "moments" for this operation
    to complete. You're not swinging from the rafters here, and squeezing
    all the air space out of C: . Basically, you're just moving the blocker(s)
    out of the way.

    3) After that process reboots, enter AOMEI again, move Recovery Partition to
    the left as your first operation. When it completes, use Move/Resize again
    to make the partition as wide as you see fit.

    Paul
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Paul@nospam@needed.invalid to alt.comp.os.windows-11 on Sat Apr 25 09:30:16 2026
    From Newsgroup: alt.comp.os.windows-11

    On Fri, 4/24/2026 5:06 PM, Paul wrote:
    On Fri, 4/24/2026 3:12 PM, Lars Poulsen wrote:


    I am still stuck without a simple way to create the needed 400 MB of free space in the recovery partition. As I understand it, the path Paul has suggested is to create a bootable USB stick with a 3rd-party partitioning tool, and use that to move the partition boundary.
    That sounds a bit cumbersome and scary. And based on recent threads here, this is becoming a problem waiting to happen for many Windows-11 users. Every Windows-11 machine I have looked at recently has less than 250 MB of free space in their recovery partition.

    When I tested, there was no structure at all at 50%, so I failed
    to reproduce the situation. I know there are definitely blockers
    at 50%, because I've had them before and the Shrink Volume would
    not go below 50%. When I tried the AOMEI, it had no problem shrinking
    C: a lot more than I needed (and it could do that "online" as
    there were no blockers). I could also move the Recovery Partition
    to the left then, followed by resizing it as desired. The only problem
    with AOMEI, is having a numeric display for precise movement of
    the sizes I want. In some cases, the GUI is a tiny bit lacking.
    Some dialogs have dimensions, and some are just graphical.

    [Picture] aomei-first-test.gif (could only test Recovery Partition move/resize)

    https://postimg.cc/bZ4xPYgj

    I tried a bunch more archived VHDs I have on the NAS, and none of them have blockers
    either. (I use JKDefrag just for the "block view" so I can see where items are
    located on disk. I am not making any changes at all with JKDefrag. I use it for surveillance.) CoPilot suggested a list of items that can function as blockers, so I'm going to try that next (after I come back).

    https://www.diskpart.com/articles/shrink-volume-beyond-half-its-size-4348.html

    pagefile.sys \
    hiberfil.sys \____ For sufficiently large C: to begin with, these will
    $MFT / be below the middle
    $Extend\$UsnJrnl <=== Possible, more research needed
    System Restore / Shadow Copies <=== Possible, more research needed
    Bad clusters <=== Highly highly unlikely.

    You can see in the "style department", AOMEI will be using a reboot
    and running from a WinPE, while it does the resize in that case.

    This is just some addendum, so you understand a bit about the blocker,
    and what my attempts to look at it are like...

    Looking through my notes, when you attempt a Shrink in Disk Management,
    it created an event in Event Viewer. I looked through my notes file, for
    the word "shrink", and to my surprise saw this text. This could be a shadow copy
    for example.

    Diagnostic details:
    - The last unmovable file appears to be: \System Volume Information\{ab59bee2-c38a-11e2-89db-0003ff781424}{3808876b-c176-4e48-b7ae-04046e6cc752}::$DATA
    - The last cluster of the file is: 0x813fff
    - Shrink potential target (LCN address): 0x1b883f
    - The NTFS file flags are: ---AD
    - Shrink phase: <analysis>

    So then I had to go look and see if I could reproduce this on something
    that has a blocker.

    *******

    Looking on the other machine, JKFrag (block-viewing mode, not modifying the disk)
    tells me there is a blocker half way, on Win10 F: partition.
    Notice that the red line is rather long. This particular partition has
    been "round the block" and on more than one disk drive, so it's history
    is "rich, and forgotten". This is the Win10 on my NAS-like disk set.

    [Picture] Win10-F-Test-Machine-blocker.gif

    https://postimg.cc/RWJL7wDb

    The length of that red line, implies there is also
    something bigger up there which could be a blocker.
    For JKDefrag, that's not a "blocker", it is merely something that
    JKDefrag cannot move at the moment, as JKDefrag doesn't shrink anything.
    But the red color is "indicative of trouble".

    When I indicate I would like to shrink F: , this is what Disk Management tells me.

    -------------------------------------------------------------------------------------------------
    eventvwr.msc
    Windows Logs
    Application
    Defrag Event 259

    A volume shrink analysis was initiated on volume WIN10 (F:).
    This event log entry details information about the last unmovable file
    that could limit the maximum number of reclaimable bytes.

    Diagnostic details:
    - The last unmovable file appears to be: \$MftMirr::$DATA
    - The last cluster of the file is: 0x2452d95
    - Shrink potential target (LCN address): 0x111237f
    - The NTFS file flags are: -S--D
    - Shrink phase: <analysis>

    To find more details about this file please use the
    "fsutil volume querycluster \\?\Volume{c65576d3-4316-4f15-b869-2cd2b3919d60} 0x2452d95" command.
    -------------------------------------------------------------------------------------------------

    If I use nfi.exe (microsoft utility, available via archive.org now...) to analyze F: then
    an awk script, the max sector is

    max value is 0X0012296CAF # Divide this sector number by 8 to match Event 259 cluster notation
    # and that gives 0x2452d95 cluster number.

    Listed by decreasing address (of about 300,000 files), these are at the top of the list.

    max value is 0X0012296CAF,
    0X0012296CAF,Master File Table Mirror ($MftMirr) <=== the blocker
    0X000B1BC9BF,\PROGRA~3\MICROS~1\Search\Data\APPLIC~1\Windows\Windows.edb <=== the next "item" is rather far away
    0X000B1B89BF,\PROGRA~3\MICROS~1\Search\Data\APPLIC~1\Windows\Windows.edb so the length of the red line is
    0X000B03C127,\PROGRA~3\MICROS~1\Search\Data\APPLIC~1\Windows\Windows.edb coming from some "mystery source"

    File 1
    Master File Table Mirror ($MftMirr)
    $STANDARD_INFORMATION (resident)
    $FILE_NAME (resident)
    $DATA (nonresident)
    logical sectors 304704680-304704687 (0x12296ca8-0x12296caf) <=== 8 sectors long, or one cluster
    That can't be my long red line.

    nfi.exe does not appear to list whatever that red line is.

    This is how you filter the output of nfi.exe . This is using the GNUWIN32 version of gawk.exe (which is version 3 of GAWK).

    ********* topfile.awk ************
    BEGIN {
    #
    # topfile.awk processes the output of nfi.exe
    # Output format suited to ingestion by spreadsheet, then Data:Sort
    #
    #**************************
    #File 0
    #Master File Table ($Mft)
    # $STANDARD_INFORMATION (resident)
    # $ATTRIBUTE_LIST (nonresident)
    # logical sectors 50425824-50426335 (0x3016fe0-0x30171df)
    # $FILE_NAME (resident)
    # $DATA (nonresident)
    # logical sectors 6291456-7340543 (0x600000-0x7001ff)
    # logical sectors 93369304-93479895 (0x590b3d8-0x59263d7)
    # logical sectors 99118840-99942135 (0x5e86ef8-0x5f4fef7)
    # $BITMAP (nonresident)
    # logical sectors 64892072-64892319 (0x3de2ca8-0x3de2d9f) #**************************
    #
    # This strips the punctuation by doing some ORed terms.
    # A repeat factor seems to help the OR work. This separator
    # is crafted to help with the "logical sectors" line.

    FS="[(]*|[)]*|-"
    max = 0
    filename = ""
    filenameisnext = 0

    # gawk.exe --non-decimal-data -f topfile.awk nfi-f-out.txt
    #
    #########################################################
    }

    filenameisnext == 1 {
    filenameisnext = 0
    # Need to process arcane line format of nfi, has extra carriage return
    #gsub(/[[:punct:]]/, "")
    gsub(/\r/,"")
    filename = $0
    }

    /^File / { filenameisnext = 1 }

    /logical sectors/ {
    cur = 0 + $4
    printf("\"0X%010X\",\"%s\"\n", cur, filename)
    if (cur > max) max = cur
    # exit
    }

    END {
    printf("max value is 0X%010X\n", max)
    }
    ********* topfile.awk ************

    Paul
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Andy Burns@usenet@andyburns.uk to alt.comp.os.windows-11 on Sat Apr 25 14:36:35 2026
    From Newsgroup: alt.comp.os.windows-11

    Lars Poulsen wrote:

    I am still stuck without a simple way to create the needed 400 MB of
    free space in the recovery partition.

    Have you considered just deleting the recovery partition, expand the C:
    drive and living without? Prepare a bootable USB in case of needing to perform maintenance on your Windows install ...

    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Paul@nospam@needed.invalid to alt.comp.os.windows-11 on Mon Apr 27 14:39:29 2026
    From Newsgroup: alt.comp.os.windows-11

    On Sat, 4/25/2026 9:30 AM, Paul wrote:
    On Fri, 4/24/2026 5:06 PM, Paul wrote:
    On Fri, 4/24/2026 3:12 PM, Lars Poulsen wrote:


    I am still stuck without a simple way to create the needed 400 MB of free space in the recovery partition. As I understand it, the path Paul has suggested is to create a bootable USB stick with a 3rd-party partitioning tool, and use that to move the partition boundary.
    That sounds a bit cumbersome and scary. And based on recent threads here, this is becoming a problem waiting to happen for many Windows-11 users. Every Windows-11 machine I have looked at recently has less than 250 MB of free space in their recovery partition.

    When I tested, there was no structure at all at 50%, so I failed
    to reproduce the situation. I know there are definitely blockers
    at 50%, because I've had them before and the Shrink Volume would
    not go below 50%. When I tried the AOMEI, it had no problem shrinking
    C: a lot more than I needed (and it could do that "online" as
    there were no blockers). I could also move the Recovery Partition
    to the left then, followed by resizing it as desired. The only problem
    with AOMEI, is having a numeric display for precise movement of
    the sizes I want. In some cases, the GUI is a tiny bit lacking.
    Some dialogs have dimensions, and some are just graphical.

    [Picture] aomei-first-test.gif (could only test Recovery Partition move/resize)

    https://postimg.cc/bZ4xPYgj

    I tried a bunch more archived VHDs I have on the NAS, and none of them have blockers
    either. (I use JKDefrag just for the "block view" so I can see where items are
    located on disk. I am not making any changes at all with JKDefrag. I use it >> for surveillance.) CoPilot suggested a list of items that can function as >> blockers, so I'm going to try that next (after I come back).

    https://www.diskpart.com/articles/shrink-volume-beyond-half-its-size-4348.html

    pagefile.sys \
    hiberfil.sys \____ For sufficiently large C: to begin with, these will
    $MFT / be below the middle
    $Extend\$UsnJrnl <=== Possible, more research needed >> System Restore / Shadow Copies <=== Possible, more research needed >> Bad clusters <=== Highly highly unlikely.

    You can see in the "style department", AOMEI will be using a reboot
    and running from a WinPE, while it does the resize in that case.

    An update on using Partition Managers.

    1) A second attempt to use AOMEI, resulted in a refusal of it
    to do a Move/Resize. A "free" product with a nervous disposition.

    2) Paragon Partition Manager 14 Free, continues to function (for as long
    as I've had it, it will do a Move/Resize). It reboots, and it took
    half an hour to shrink a 500GB partition to 250GB (when the files are
    in the lower 80GB of the partition). It puts text on the screen after
    a reboot, and the text, I didn't know where (what subsystem) prefers
    it that way.

    3) I installed Minitool Partition Wizard 13.6 or so.

    https://www.partitionwizard.com/offline-download.html

    Name: pw-free-offline--Minitool-Partition-Wizard-contains-PUP.exe
    Size: 77384904 bytes (73 MiB)
    SHA256: 8FEC38F06199D468D6F361C8742283C21DECDBB35227B914A36C2BE76E683F13

    It would download some AV where the name begins with an "A" if given
    a chance, which is why the testing was done with the Ethernet cable unplugged.

    The test disk had two Windows partitions. While booted into Partition 2,
    I had it shrink Partition 4 and it did that while Windows 10 continued to run.
    It moved the MFTMIRR out of the way (put it much lower on the partition).

    The interesting case, was when I was booted on Partition 2 and I asked it to
    shrink Partition 2. It wants to reboot. The same kind of text Paragon puts up,
    the Minitool tried to put up. In particular, it tried to "lock some memory" on
    my 4930K and this was failing to work. The text mentioned
    /filesystem/ntfsresize , so it is using a Linux utility (likely the same one
    that Linux GParted uses). That means the reboot environment it is using, is
    some pared-down Linux setup.

    This means, for Lars, the Minitool stands a chance of failing during the
    reboot session, after which it returns to Windows again.

    That means, so far, Paragon Partition Manager 14 (PM14) appears to be the winner on this round. But if Paragon is also using ntfsresize, you might
    as well just boot a Linux LiveDVD/ISO (via Rufus.ie if necessary) and just
    do it there with GParted, which would also use ntfsresize and friends.

    https://www.majorgeeks.com/files/details/paragon_partition_manager_express_free_edition.html

    Whereas this is the one I have on disk. This is an installshield, the Majorgeeks
    one is an MSI file.

    Name: pm14free_x64_eng.exe <=== the one I'm still using
    Size: 53091632 bytes (50 MiB)
    SHA256: 7F48097AC70AE6B27FCC5C40A10BBE7E42516DC7D73D0C48ABFFE1D50B2145CF

    Name: pm_2014_free.msi <=== the Majorgeeks one
    Size: 43603456 bytes (41 MiB)
    SHA256: 37B11D8DD6BF53F62BF8197F824A11B3E51835D8A92F0CF10D8058BFB969128A

    Paul


    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Lars Poulsen@lars@beagle-ears.com to alt.comp.os.windows-11 on Mon Apr 27 16:20:59 2026
    From Newsgroup: alt.comp.os.windows-11

    On 2026-04-27 11:39, Paul wrote:
    On Sat, 4/25/2026 9:30 AM, Paul wrote:
    On Fri, 4/24/2026 5:06 PM, Paul wrote:
    On Fri, 4/24/2026 3:12 PM, Lars Poulsen wrote:


    I am still stuck without a simple way to create the needed 400 MB of free space in the recovery partition. As I understand it, the path Paul has suggested is to create a bootable USB stick with a 3rd-party partitioning tool, and use that to move the partition boundary.
    That sounds a bit cumbersome and scary. And based on recent threads here, this is becoming a problem waiting to happen for many Windows-11 users. Every Windows-11 machine I have looked at recently has less than 250 MB of free space in their recovery partition.

    When I tested, there was no structure at all at 50%, so I failed
    to reproduce the situation. I know there are definitely blockers
    at 50%, because I've had them before and the Shrink Volume would
    not go below 50%. When I tried the AOMEI, it had no problem shrinking
    C: a lot more than I needed (and it could do that "online" as
    there were no blockers). I could also move the Recovery Partition
    to the left then, followed by resizing it as desired. The only problem
    with AOMEI, is having a numeric display for precise movement of
    the sizes I want. In some cases, the GUI is a tiny bit lacking.
    Some dialogs have dimensions, and some are just graphical.

    [Picture] aomei-first-test.gif (could only test Recovery Partition move/resize)

    https://postimg.cc/bZ4xPYgj

    I tried a bunch more archived VHDs I have on the NAS, and none of them have blockers
    either. (I use JKDefrag just for the "block view" so I can see where items are
    located on disk. I am not making any changes at all with JKDefrag. I use it >>> for surveillance.) CoPilot suggested a list of items that can function as >>> blockers, so I'm going to try that next (after I come back).

    https://www.diskpart.com/articles/shrink-volume-beyond-half-its-size-4348.html

    pagefile.sys \
    hiberfil.sys \____ For sufficiently large C: to begin with, these will
    $MFT / be below the middle
    $Extend\$UsnJrnl <=== Possible, more research needed
    System Restore / Shadow Copies <=== Possible, more research needed
    Bad clusters <=== Highly highly unlikely.

    You can see in the "style department", AOMEI will be using a reboot
    and running from a WinPE, while it does the resize in that case.

    An update on using Partition Managers.

    1) A second attempt to use AOMEI, resulted in a refusal of it
    to do a Move/Resize. A "free" product with a nervous disposition.

    2) Paragon Partition Manager 14 Free, continues to function (for as long
    as I've had it, it will do a Move/Resize). It reboots, and it took
    half an hour to shrink a 500GB partition to 250GB (when the files are
    in the lower 80GB of the partition). It puts text on the screen after
    a reboot, and the text, I didn't know where (what subsystem) prefers
    it that way.

    3) I installed Minitool Partition Wizard 13.6 or so.

    https://www.partitionwizard.com/offline-download.html

    Name: pw-free-offline--Minitool-Partition-Wizard-contains-PUP.exe
    Size: 77384904 bytes (73 MiB)
    SHA256: 8FEC38F06199D468D6F361C8742283C21DECDBB35227B914A36C2BE76E683F13

    It would download some AV where the name begins with an "A" if given
    a chance, which is why the testing was done with the Ethernet cable unplugged.

    The test disk had two Windows partitions. While booted into Partition 2,
    I had it shrink Partition 4 and it did that while Windows 10 continued to run.
    It moved the MFTMIRR out of the way (put it much lower on the partition).

    The interesting case, was when I was booted on Partition 2 and I asked it to
    shrink Partition 2. It wants to reboot. The same kind of text Paragon puts up,
    the Minitool tried to put up. In particular, it tried to "lock some memory" on
    my 4930K and this was failing to work. The text mentioned
    /filesystem/ntfsresize , so it is using a Linux utility (likely the same one
    that Linux GParted uses). That means the reboot environment it is using, is
    some pared-down Linux setup.

    This means, for Lars, the Minitool stands a chance of failing during the
    reboot session, after which it returns to Windows again.

    That means, so far, Paragon Partition Manager 14 (PM14) appears to be the winner on this round. But if Paragon is also using ntfsresize, you might
    as well just boot a Linux LiveDVD/ISO (via Rufus.ie if necessary) and just
    do it there with GParted, which would also use ntfsresize and friends.

    https://www.majorgeeks.com/files/details/paragon_partition_manager_express_free_edition.html

    Whereas this is the one I have on disk. This is an installshield, the Majorgeeks
    one is an MSI file.

    Name: pm14free_x64_eng.exe <=== the one I'm still using
    Size: 53091632 bytes (50 MiB)
    SHA256: 7F48097AC70AE6B27FCC5C40A10BBE7E42516DC7D73D0C48ABFFE1D50B2145CF

    Name: pm_2014_free.msi <=== the Majorgeeks one
    Size: 43603456 bytes (41 MiB)
    SHA256: 37B11D8DD6BF53F62BF8197F824A11B3E51835D8A92F0CF10D8058BFB969128A

    Paul

    I need to confess that I gave up on getting Windows-11 to work at this
    time, because my most urgent priority was to get Fedora Linux F43
    running again, so I reinstalled Fedora in "take the whole machine" mode.
    I suspect that it was a cascading series of problems, beginning
    (conjecture) that during the almost 4 months I had not booted into
    Windows, the laptop finally became eligible for the 25H2 upgrade, and
    that installing the feature upgrade also tried to upgrade the recovery partition, which failed because it ran out of space. And (still
    conjecture) it also tried to fix the secure boot key in the EFI
    partition, and breaking that too, because that (small EFI partition
    served double duty as Linux /boot/efi. Because Linux was installed after Windows, it had been set to boot into Grub, from where I was booting
    Windows by selecting "Windows Boot Manager" from the Grub boot menu.
    And the Windows update script was not set up to handle this kind of system.

    So for now, the little ASUS laptop, purchased for $150 at an
    after-Christmas sale in 2022, is a pure Linux machine. Bringing it up
    after this re-install has its own issues, but they do not belong in this newsgroup.

    Thank you for the detailed responses. I have learned a lot about how the system administration issues in Windows and Linux are both similar, but
    also different.
    --
    Lars Poulsen - an old geek in Santa Barbara, California
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Lars Poulsen@lars@beagle-ears.com to alt.comp.os.windows-11 on Mon Apr 27 16:22:59 2026
    From Newsgroup: alt.comp.os.windows-11

    On 2026-04-25 06:36, Andy Burns wrote:
    Lars Poulsen wrote:

    I am still stuck without a simple way to create the needed 400 MB of
    free space in the recovery partition.

    Have you considered just deleting the recovery partition, expand the C: drive and living without?  Prepare a bootable USB in case of needing to perform maintenance on your Windows install ...

    As described in my response to Paul, I decided to just live without ... Windows on that machine for now.
    --
    Lars Poulsen - an old geek in Santa Barbara, California
    --- Synchronet 3.21d-Linux NewsLink 1.2
  • From Paul@nospam@needed.invalid to alt.comp.os.windows-11 on Tue Apr 28 13:12:00 2026
    From Newsgroup: alt.comp.os.windows-11

    On Mon, 4/27/2026 7:20 PM, Lars Poulsen wrote:

    I need to confess that I gave up on getting Windows-11 to work at this time, because my most urgent priority was to get Fedora Linux F43 running again, so I reinstalled Fedora in "take the whole machine" mode.
    I suspect that it was a cascading series of problems, beginning (conjecture) that during the almost 4 months I had not booted into Windows, the laptop finally became eligible for the 25H2 upgrade, and that installing the feature upgrade also tried to upgrade the recovery partition, which failed because it ran out of space. And (still conjecture) it also tried to fix the secure boot key in the EFI partition, and breaking that too, because that (small EFI partition served double duty as Linux /boot/efi. Because Linux was installed after Windows, it had been set to boot into Grub, from where I was booting Windows by selecting "Windows Boot Manager" from the Grub boot menu.
    And the Windows update script was not set up to handle this kind of system.

    So for now, the little ASUS laptop, purchased for $150 at an after-Christmas sale in 2022, is a pure Linux machine. Bringing it up after this re-install has its own issues, but they do not belong in this newsgroup.

    Thank you for the detailed responses. I have learned a lot about how the system administration issues in Windows and Linux are both similar, but also different.

    I still don't know how to Boot Repair Fedora, so don't
    break the boot on that. The Yannbuntu Boot Repair disc
    doesn't seem to work with Fedora. You'd think GRUB2 is
    GRUB2, but for some reason that does not seem to be the case.

    After my experience with Paragon PM14 (being for such software,
    relatively well-behaved), I thought testing the others would not
    reveal such much crap-ware behavior. Boy, was I wrong.

    When I tried to run Raxco PerfectDisk (bankrupt and out of business)
    in trial mode, and I ticked the box "Prepare for shrink", the
    tool responded "this is only available in the Pro version",
    and as you can imagine, I set a new record for utility removed
    in least amount of time :-) The idea of Prepare for shrink, is
    they would move MFTMIRR out of the way.

    It would seem, it's going to require a dis-mount of a partition
    at a minimum, to do some of these things. A reboot neatly solves
    the problem. And the rather cheesy interface on the reboot, means
    you're hoping any problems it encounters, it prints a message
    on the screen. When it told me it was "ntfsresize", this is my
    shocked face :-)

    *******

    As for Linux installs, I *always* install them in Custom mode.
    As that's when I discover some of them don't have partition
    preparation capabilities in their Custom section. On one
    distro, it does not even inform you what kind of partitioning
    it would like. Which is beyond funny. I don't mind using my
    available tools here, to prep a disk suited to their liking,
    but if you don't tell me what you want, I'm not Kreskin. In
    some cases, not even the forum for the product, has a detailed
    discussion on that part. I guess "that's too simple for us"
    is their attitude.

    This is definitely the Year of the Linux Desktop.

    Paul


    --- Synchronet 3.21d-Linux NewsLink 1.2