Virtual machines deployed on VSAN without selected with THICK disk storage policy

Hello

We had an admin build 50 VMs on our data VSAN store but did not choose the VSAN storage policy while he got the default which allows you to have the THICKNESS of the disks provisioned

I have chnaged any of the virtual machines to use the VSAN storage policy (which uses a thin) but the virtual machine is still using all the space

Is there a way to make virtual machines recover white space, now they have a VSAN storage strategy?

Yes - I have to agree with Duncan. I don't think that there is a way to return/reduce the size.

An idea is, if you have additional storage, Storage vMotion VM to NFS or VMFS, converting to thin and then back to VSANS with a policy that doesn't have a "thickness".

Try it with a VM - see if it works.

A bit of a drag, I know, but I can't think of any other casual way.

HTH

Cormac

Tags: VMware

Similar Questions

  • Download the virtual machine in a store of data with powercli

    Hello

    How to list the files of virtual machine on a store of data with PowerCLI

    I have a virtual machine and I want to see all the files in a data store (such as hard,.-flst.vmdk, .log, .nvram) using powercli. Is this possible?

    TKS.

    Try the following, it uses a function of my position of friendly units .

    Function Get-FriendlyUnit {}

    <>

    . Numbers converted small binary multiples SYNOPSIS

    . DESCRIPTION the function accepts a value and him will convert

    in the largest available binary unit.

    . Author NOTES: Luc relaxing

    . Value of the PARAMETER

    The value to convert.

    This number must be positive.

    . IEC PARAMETER

    A switch to indicate whether the function should return the CIS

    Unit, or the names of the units most commonly used names.

    The default is to use the commonly used unit names.

    . EXAMPLE OF

    PS > Get-FriendlyUnit-value 123456

    . EXAMPLE OF

    PS > 123456 | Get-FriendlyUnit - CIS

    . EXAMPLE OF

    PS > Get-FriendlyUnit-value 123456,789123, 45678

    #>

    (param

    [CmdletBinding()]

    [parameter (mandatory = $true, ValueFromPipeline = $true)]

    [double []] $Value,.

    [switch] $IEC

    )

    Begin {}

    $OldUnits = "B", "KB", "MB", "GB", "TB", "PB", "EB", "ZB", "YB".

    $IecUnits = 'B', 'KiB', 'MiB', "GiB", "TiB", "PIF", "BANK", "ZiB", "YiB".

    If ($IEC) {$units = $IecUnits} else {$units = $OldUnits}

    }

    {in process

    $Value | %{

    If ($_ - lt 0) {}

    Write-Error "Numbers must be positive."

    breaking

    }

    If ($value - gt 0) {}

    $modifier = [math]: Floor([Math]::Log($_,1KB)))

    }

    else {}

    $modifier = 0

    }

    New-object PSObject-property @ {}

    Value = $_ / [math]: Pow(1KB,$modifier)

    Unit = & {if ($modifier - lt $units. {Count) {$units [$modifier]} else {'1KO E {0}' f $modifier}}

    }

    }

    }

    }

    foreach ($vm in (Get-View - ViewType VirtualMachine - property Name, 'LayoutEx.File')) {}

    $vm. LayoutEx.File |

    Select @{N = "VM"; E = {$vm. Name}},

    @{N = 'Name'; E={$_. Name.Split(' ') [1]. "Split('/') [1]}},"

    @{N = "Path"; E={$_. Name}},

    @{N = 'FileType'; E={$_. Type}},

    @{N = "Datastore"; E={$_. Name.Split(']') [0]. TrimStart('[')}},

    @{N = "size"; E = {}

    $val = get-FriendlyUnit-value of $_. Size

    "{0:n2} {1}" f $val. Value, $val. Unit

    }}

    }

  • We can recover a virtual machine of flat files without committing snapshots?

    I have a virtual machine that has died due to having excessive snapshots. "Too many levels of redo logs.

    Can I get out this virtual machine to the top without committing the delta files, IE just use original flat files?

    I tried going through the KB1004545 to solve the problem of the snapshot, and it fails

    Thanks in advance!
    Brendan

    It's really kind of unusual, but we will try to fix things.

    1.) create multiple subdirectories which you will pass some files during this procedure
    -not used
    -backup

    -fact
    -temp

    (2.) do a cleaning of unused files
    -move all the files to "...-ctk.vmdk" for the "unused" subdirectory
    -move the "000002" files in the subdirectory "unused" (not in the chain of snapshot)
    -move the '000239' files in the subdirectory "unused" (not in the chain of snapshot)
    -move the "000258' - '000287' files in the 'unused' subdirectory (contain no user data)

    3.) backs up the descriptor files

    -Copy all the hard of descriptor/header files in the subdirectory "backup."

    4.) merging the snapshots in the basic disk using the vmkfstools utility
    Given that problems with two virtual disks are the same, I'll explain the procedure for the first disc of "nzaklapp11" only. Once you are done with the first virtual disk, repeat steps to 'nzaklapp11_1 '.

    First round:

    • run: vmkfstools-i nzaklapp11 - 000030.vmdk temp/nzaklapp11.vmdk
    • the files '000030' moving 'nzaklapp11.vmdk', "nzaklapp11 - flat hard" as well as the "000001" - to the subdirectory 'done '.
    • Move hard files created in sub-repertoire 'temp' in the virtual machine file
    • Edit "nzaklapp11 - 000031.vmdk".
      parentFileNameHint = "nzaklapp11.vmdk".
      parentCID =

    Repeatedly executes:

    • run: vmkfstools-i nzaklapp11 - 000060.vmdk temp/nzaklapp11.vmdk
    • move the "000031' -"000060"files in the subdirectory 'done '.
    • Remove "nzaklapp11.vmdk", "nzaklapp11 - flat hard."
    • Move hard files created in sub-repertoire 'temp' in the virtual machine file
    • Edit "nzaklapp11 - 000061.vmdk".
      parentFileNameHint = "nzaklapp11.vmdk".
      parentCID =

    Repeat the "works on several occasions" by increasing the number of 30 until you are finished with all snapshots.

    Once finished with the two discs, and you confirmed that the new disks are ok, you can remove the subdirectories created in step 1.

    If anything is unclear, please feel free to ask.

    André

  • Deployment of a virtual machine from a store of data with less space, but enough for the virtual machine

    I received this delicate task, and I can understand not just how to replace all parts.

    I need to create a script that will be smartly decided what data store to deploy a virtual computer.  We do not want to deploy in a data store that has the most space, we want to deploy to the data store that has the least space but can still account for the space for the (vm + 5%) and still leave 50 GB free on the lun after the move.

    Thus, for example, if the virtual machine is 40 GB, we want the script to select the data store that has a close to 90 GB available without being under.

    So my thought for this approach is:

    • Create the query to get all the relevant LUNS.  This excludes all the LUNS with the 'local' name in it and excludes all LUN owners (who have a slightly different naming convention then our general shared storage LUNS)
      • This piece, which I partially understood
        • Get-datastore. WHERE-object {($_.)} Name: corresponds to "PAR0 [1-4] _ [edp] * disk *")- and ($_.) "." Name - notmatch 'local')}
      • Now, I need to get all of their total size and free space.  Perhaps export this list to a CSV, however if I have to.
      • Create a variable that contains: the size of the total virtual machine to the virtual machine that is deployed before its deployment.
      • Deduct vm size against each data store size and pull in some way that that also close to 50 GB free on the data store and still facilitates the deployment of the vm.

    I have a few other scripts, I scrounged on the internet that I tried to restore... but I just don't calm not having all the pieces...

    • Get-Datastore. Where-Object {$_.} ParentFolder-match 'Internal'} ' | Select-Object - property data center, FreeSpaceMB, CapacityMB name, ' | Tri-objet-property FreeSpaceMB
    • Select-Object Name,@{n="CapacityGB";e={[math]::round (($_.)) {{(CapacityMB/1024))}}, @{n = "FreeSpaceGB"; e = {[math]: round (($_.))}} {{(FreeSpaceMB/1024))}}, @{n = "FreeSpacePercent"; e = {[math]: round (($_.))}} FreeSpaceMB / $_. {{(CapacityMB*100))}} | Sort-Object FreeSpaceGB

    Any help would be greatly appreciated!

    -Knotz

    Try something like this

    # Get all data warehouses

    $ds = get-Datastore. Where-Object {($_.)} Name: corresponds to "PAR0 [1-4] _ [edp] * disk *")- and ($_.) "." Name - notmatch 'local')} |

    Select-Object Name,

    @{n = "CapacityGB"; e = {[math]: round (($_.))}} {{(CapacityMB/1024))}}.

    @{n = "FreeSpaceGB"; e = {[math]: round (($_.))}} {{(FreeSpaceMB/1024))}}.

    @{n = "FreeSpacePercent"; e = {[math]: round (($_.))}} FreeSpaceMB / $_. {{(CapacityMB*100))}}

    {foreach ($vm in Get - VM)

    # Find possible candidates

    $candidates = $ds | where {($_.)} FreeSpaceGB - $vm. (UsedSpaceGB-50) - gt 0}

    # Find the best candidate

    $target = $candidates | Tri-objet-property FreeSpaceGB-descending | Select - 1 first

    Write-Output "VM $($vm.). Name) can go to $($target.) (Name) ".

    }

  • Customization of the comments for the virtual machines deployed from model has stopped working

    Hi all

    (Environment: vCenter and ESXi 10 5.1 hosts running Build #3070626)

    We have a model 2012R2 and a file specification of customization that allows us to deploy the VMs 2012R2 server. It worked well for a while. After the Christmas break, it seems more work correctly and we havn't knowingly changed anything to cause this. We have fixed our ESXi hosts Build #3070626 October 6, but have deployed VMs of model since, without problem.

    We have all staged for the VM in advance (IE AD entry of objects and DNS). And then powered the deployed virtual computer, sysprep runs, but does not seem to apply settings in the customization file (IE rename the virtual computer, set the Admin password, join the domain etc.). If feed us the cycle of the VM, sysprep is running again, but even once, does nothing.

    I have attached the contents of my file guestcust.log of a computer deployed virtual (see below), but I do not see the clues about the problem. I thought initially the error on the first line was important, until I've found it is also present on the virtual machines that have been successfully deployed from this template in the past, so it's maybe a Kipper.

    Does anyone have advice for troubleshooting this?

    TIA

    Mark

    ==========

    guestcust.log

    ==========

    Could not query value reported in the registry (the system cannot find the file specified.)

    Sysprep from Windows Respecialize operating detected.

    Copy the files from C:\Windows\TEMP\vmw7DEB.tmp to C:\Windows\TEMP\vmw8A3E.tmp.

    Rescheduling of customization of the guest operating system.

    "Executing command C:\Windows\system32\schtasks.exe / create /sc ONSTART $ more HIGH/np/z v1 /ru SYSTEM /tr" C:\Windows\TEMP\vmw8A3E.tmp\guestcustutil.exe customize - sealparam \"/quiet/generalize/oobe /reboot\ ' NIC - 4000 - schedulenativeunobfusc ' /tn"VMware Guest OS task of customization.

    At the request of guest OS customization task invoked.

    Remove scheduled task of customization of the guest operating system.

    Execution of command C:\Windows\system32\schtasks.exe, /f/delete /tn "VMware Guest OS task of customization"

    Personalization current 2016 to 1 - Feb - 11 18:32:42

    RPCI: Send request = 'deployPkg.update.state 4 C:\Windows\TEMP\vmware-imc\guestcust.log 0'

    RPCI: Envoy = 'deployPkg.update.state 4 C:\Windows\TEMP\vmware-imc\guestcust.log 0' request, response = ", len = 0, status = 1

    Key successfully open SYSTEM\CurrentControlSet\Control\Session Manager\

    Size of 27 reg_multi_sz.

    Read the multi_sz registry autocheck autochk/q/v value *, size 27.

    value chain of registry autocheck autochk/q/v *.

    1 items back

    Obtained the session mgr BootExecute.

    Native App sysprepDecrypter.exe, arguments "

    Copied file sysprepDecrypter.exe to C:\Windows\system32\sysprepDecrypter.exe

    Update boot run value.

    Created / opened key \Guest Customization\ SOFTWARE\VMware, Inc. successfully.

    New shoe, run:

    autocheck autochk/q/v *.

    sysprepDecrypter.exe

    System drive: C:

    Move the folder "sysprep" to "C:".

    Select * from win32_networkadapter where manufacturer! "Microsoft" and ServiceName =! = "VMnetAdapter" and the manufacturer is not null and MACAddress is not null

    1 objects found. Pointer f049a8. return code 0(0x0)

    "" "0 items found." " Pointer to 0. return code 1(0x1)

    00:50:56:B0:70:46 the value for the property return system

    Setting dhcp for nic # 0

    Return value \\VMT-W2K12R2\ROOT\CIMV2:Win32_NetworkAdapter.DeviceID="10' for the system property

    ASSOCIATORS OF {\\VMT-W2K12R2\ROOT\CIMV2:Win32_NetworkAdapter.DeviceID="10"} where ResultClass = Win32_NetworkAdapterConfiguration

    1 objects found. Pointer eee290. return code 0(0x0)

    "" "0 items found." " Pointer to 0. return code 1(0x1)

    Clear the gateway ip addresses.

    Enabling DHCP on the computer

    Returns the value for the system property \\VMT-W2K12R2\ROOT\CIMV2:Win32_NetworkAdapterConfiguration.Index=10

    DNS server IP configuration

    Execution of command C:\windows\system32\sysprep\sysprep.exe/quiet/generalize/oobe/reboot /unattend:C:\sysprep\sysprep.xml

    Successfully executed the command C:\windows\system32\sysprep\sysprep.exe/quiet/generalize/oobe/reboot /unattend:C:\sysprep\sysprep.xml

    Try to connect network interfaces, try 1

    RPCI: Send request = 'deployPkg.update.state 4 103 C:\Windows\TEMP\vmware-imc\guestcust.log@4000'

    RPCI: Envoy = 'deployPkg.update.state 4 103 C:\Windows\TEMP\vmware-imc\guestcust.log@4000' request, response is "queryNicsSupported", len = 18, State = 1

    Got answer VMX 'queryNicsSupported '.

    RPCI: Send request = 'deployPkg.update.state 4 104 C:\Windows\TEMP\vmware-imc\guestcust.log@4000'

    RPCI: Envoy = 'deployPkg.update.state 4 104 C:\Windows\TEMP\vmware-imc\guestcust.log@4000' request, response = 'connected', len = 9, State = 1

    Got VMX response "connected."

    Network interfaces are connected to 0 seconds

    =============================================

    OK, I myself fixed the end.

    I created a new customization file with exactly the same settings as the old file, and it worked. I can only assume that the old customization file has been altered somehow.

  • Virtual machines Windows display "reboot and select the proper boot device or insert boot media in selected boot device.

    I downloaded the VM Windows and created a machine. The machine should start with ' Microsoft Corporation PXE Boot Agent v2.00 (built-in BIOS) (C) Copyright 2008 Microsoft Corporation all rights reserved.»

    "CLIENT MAC ADDR: 00 03 FF 8 1 C 2 C GUID: 2FA ' and more digits"DHCP. "... "with a spinny beside the points thing. After awhile, the... won't have a message "no boot file received" or "any offer DHCP or proxyDHCP was received" with a "reboot and device or insert boot media select start suitable for the selected boot device". What should I do? It also appears on other virtual machines. I tried pressing tools and clicking on install the built-in components, but nothing happens!

    Hello

    The question you posted would be better suited in the TechNet Forums. I would recommend posting your query in the TechNet Forums.

    http://social.technet.Microsoft.com/forums/Windows/en-us/home?category=w7itpro

  • Remove objects from the AD when a virtual machine is removed from the vCenter with vCO

    All,

    I am new to vCenter Orchestrator and have bad getting started with him.  I'm trying to figure out how to automatically remove the Active Directory from the computer object when it is deleted in vCenter.  My idea is when the virtual machine is removed, it starts a workflow that deletes the object.  I don't know how to call that or if it is still possible.

    The problem I am trying to solve, is that our developers are self provisioning of servers Windows and Linux through vCAC in the realm of 50 + per day.  They also remove most of them in 5 or 6 days causing much left on objects cluttering up AD.   So I try to find a way to have the deleted object when they destroy their vCAC provisioned machines.  My original idea was to make based on the GUID and UUID but I not figure out how to actually call with vRO.  I'm open to other ideas that can adapt to this task better, no need to force this idea of the solution to the problem if someone already knows a way to do that.

    Little background information is that I'm under vCenter 5.5 Update 2, 6.1 and vCO 5.5.2 vCAC.  The objects being deleted are Windows 2008 R2, Windows 2012 R2 and CentOS 6.5 servers.

    Thank you in advance, I think that once I get more of an idea on how it works, I'll be able to deploy more ideas.

    This can be achieved by simply adding following some custom properties in your Machine Blueprint or group of companies in VCAC.

    Plugin.AdMachineCleanup.UserName-

    Plugin.AdMachineCleanup.Password-

    Plugin.AdMachineCleanup.Execute

    Plugin.AdMachineCleanup.Delete

    Otherwise, create a profile to build using above set of custom properties, this way it will be easy for you to add in several plans in your environment vCAC.

  • "Failure of the application. "HR = 0 x 80040101: failed to initialize the virtual machine ' on Workstation TechPreview 07/2015 with Windows 10 x 64 final

    Hello

    I just did a new install of Windows 10 (x 64) on my Surface Pro. Then I installed Workstation 11.1.2. Installation went well. But when starting, I get the error message of the headline.

    Same thing with Workstation TechPreview 07/2015, vSphere Client 6.0 and Remote Console. Seems to be a general problem, access the console of the virtual machines remotely.

    No work around known yet? I can't connect to any virtual computer right now.

    See you soon,.

    -Carsten

    Hi Carsten,

    I have the same problem when I connected with an AD azure account. If I have connection locally without problem.

    What is the scenerio you face?

    Kind regards.

  • create the virtual machine from windows NT 4 server with adaptec raid 5

    I have an old Dell (1998) server running windows NT 4.0 sp6a.  He had a raid 5 configuration using the controller raid adaptec SCSI 2940.  2 discs failed who broke the raid set.  I had a disk recovery company to recover the data and they have scratched off readers and gave me a SATA with all volumes.  I found a procedure to create a virtual hard disk (.vhd) and then converted to a hard and used to create a virtual machine in VMware workstation 7.  When I turn on the virtual machine, it begins the process of starting windows, shows the boot menu, select normal or mode VGA, goes through the process of NTdetect, display processor and memory but then gives a BSOD with the message inaccessible_boot_device.

    I changed the hard to affect the buslogic ddb.adaptertype, but it gave me the same problem.

    is there a different driver that I need for an adaptec raid controller or if the typical SCSI drivers should work?  If the typical SCSI drivers should work, so my problem might be in my Boot.ini because it shows that he is supposed to start on the partition (4) and this sata drive may not have scores exactly as they were on the raid configuration.  I tried to mount the volumes not in read-only mode, but the partition that I need to change the boot.ini file is partition (2) what mapping VMware won't let me open mode not readonly.  whenever I have, select it and it opens the partition (1) mode not readonly.

    I know this sounds complicated then I hope I described with precision.

    Hello
    I live in Germany, but my internal clock seems to be adapted to the New York Times.
    Very useful, because most repairs or recovery to the VMware is done the evening or at night.

  • Model of centOS virtual machine deployment

    Hi all, I am met with the following problem when I tried to deploy the virtual machine from a template of centOS. "Customization of the guest operating system"centosGuest"is not supported in this configuration". I do turn on ESXi 4.1 and ensure that the vmware-tools is installed. I tested on a machine model virtual ubuntu which I am able to change on reviews customization then deploy the virtual machine. I would like to know if there is a solution to this? Thanks in advance.

    It seems that the new feature of type "CentOS" 4/5 as guest of OS has not evaluated correctly.  I changed my 'Red Hat Enterprise Linux 5' models as they were before the update.  CentOS is a recompilation of Red Hat anyway, so it makes sense to keep.

  • Need a solution to customize virtual machines deployed from models

    Here's the problem:

    When I deploy virtual machines from a template, the guest of the new virtual machine operating system takes the same size of hard drive than the model between what was created.  It is very fast to increase the size of the virtual hard disk after deployment.  But the size of the partition in the guest operating system is unchangeable.  This means that, for each virtual machine, I have to start the virtual machine and either extend the guest operating system by using an operating system disk management tool or a third-party for the partition management application.

    It takes a lot of time to do during the deployment of a large number of virtual machines.  Another option is to create a template for each size of hard drive - but the problem with this is that all of these models for each operating system must be maintained, that takes a lot of time.

    Does anyone have ideas for another or better solutions?  For example, is there such a thing as a tool that integrates with VMware to enable not only the size of the virtual hard disk, but also the size of the partition of the guest operating system to customize when you create a virtual model of computer?

    Any contribution is appreciated-

    Thank you

    It resizes the disk partition inside the virtual machine to the guest OS uses the complete virtual disk capacity.

  • How to migrate from vCenter as virtual machine to another host without vMotion running

    We organize ESX 3.5 hosts. I've already updated our vCenter VM to 4.0U1. We have not the vMotion Installer yet (but are licensed for her). What is the appropriate way for migrating the vCenter VM to another host without vMotion, so I can upgrade the host the vCenter VM is running on? I have examined all the documents and messages, I could find, but don't have not seen this addressed.

    You must turn off the virtual machine.

    Copy the files (if are not on a shared data store).

    Save the vmx file in the new ESX (with browser data store).

    Turn on the virtual machine.

    Answering the question "I_moved" (appears only the first time).

    André

  • Model of virtual machine deployment process.

    I'm trying to understand the model way VM deployment works to design a distributed environment of VirtualCenter with effective model commissioning. My scenario and the question is;

    • I have a NFS location to store my VM models.

    • Mount the NFS on one of my (ESX01) ESX hosts using a VMKernel port.

    • I save the VM model on this host ESX (ESX01).

    • I have a second host ESX (ESX02) who doesn't have the mounted via NFS.

    • Deploy a VM from this model to ESX02.

    Question: How is the copy of the model deployed/copied to the ESX host?

    • Is copied on the VirtualCenter server, then to ESX02 VM?

    • Is the copied virtual machine on ESX01 to ESX02 via the network of the Service Console?

    • There is another way, that the template is created?

    If anyone has a detailed understanding of the model deployment process, it would be much appreciated.

    Thanks for your time,

    Dan.

    Given that the source is mounting NFS using vmkernel, should be out of this interface in all cases.  If you have NFS mounted on both ESX hosts, then it will use vmkernel port on the computer virtual source where the template has been saved.  Because you can not have a template with the same name, you will have to re-register the template under a different name, so that it looks like ESX a completely different model.

    -KjB

    VMware vExpert

  • How to change the network when the virtual machine is connected, in Vmware ESXi with Perl in Linux?

    Hello

    If someone can answer that the way to change the network of a VM with Perl? I would periodically change the network where the virtual machine network adapter is connected.

    Thank you

    Take a look at the vGhetto repository of scripts, where you can find a lot of useful scripts, including updateVMPortgroup.pl , which may contain what you are looking for.

    André

  • Can I add a virtual machine to a dvswitch without using vcentre?

    Afternoon,

    Our vcentre is not set up to handle the dvswitches on the host - bit of a long story.

    Until he does, can we use the command line to add a virtual machine to a dvswitch?

    If so, can someone help me with this?

    Thank you

    Martin

    vCenter is required to set up and manage a distributed switch.

Maybe you are looking for