VMware

VMware, Windows 10 Customization Specification Not Completing

Here are details of my setup – in fact, I started over from scratch to document my steps.
This seems to be a new problem occurring with Windows 10, version 1709

 

Install of new VMware guest for the purpose of being a template

  1. Create new VM, named it windows_10_enterprise_version_1703
  2. Remove floppy drive
  3. Uncheck networking (for install)
  4. Mount ISO and enable, click OK to save settings

Install of OS on the guest virtual machine

  1. Boot newly created VM (windows_10_enterprise_version_1703)
  2. Step through the installation until completion
  3. Complete language settings to arrive at Windows desktop
  4. Enable the ‘Administrator’ account as its disabled by default
  5. Log out of installation user (the account I named ‘install’)
  6. Log into the ‘Administrator’ account
  7. Enter control panel, user and delete the ‘install’ user account.
  8. Install VMWare tools, Reboot once
  9. Shutdown

 

Prepare VM to be a template

  1. Edit the VM settings
  2. Connect networking
  3. Disconnect CD Drive
  4. Click OK
  5. Convert VM to a Template

 

Customization Specification Setup

  1. Create new specification, (I named mine Windows Desktop – DHCP)
  2. Applied registration information
  3. Computer Name – Use the virtual machine name
  4. Windows License – Left this blank (unchecked include server licensing information)
  5. Administrator Password – Set password, choose the option to automatically login as Administrator
  6. Time Zone – Set my desired time zone
  7. Run Once – Left this blank (blank for now, later intend on applying KMS details)
  8. Network – Use standard network setting (DHCP)
  9. Workgroup or Domain – For now just Workgroup and left workgroup name as ‘WORKGROUP’
  10. Operating System Options – Generate New Security ID (SID)
  11. Ready to complete – Clicked OK

Deploying template

  1. Right-click on the template – New VM from the template
  2. Gave a simple name – TEST01
  3. Selected Datacenter, Selected Cluster
  4. Selected Storage
  5. Selected Options (Customize VM) and (Power on the virtual machine after creation)
  6. Selected ‘Windows Desktop – DHCP’ from customized guest OS options
  7. Clicked Next, then Finish and wait …

 

Where things get stuck

  1. After the first boot, the guest gets an IP address from the network
  2. Customization starts in the background and system reboots
  3. When the system resumes I arrive at the following screen
  4. The system customization never completes, and I find my VM’s stuck at the “Let’s start with region …” screen

And, I’ll update this as soon as I find a solution, but for now…  I’m stuck

Log Shared via Pastebin:  https://pastebin.com/ETpuLX3U

 

Update: March 18, 2018

I’ve had others also test this using the latest ISO for Windows 10 – en_windows_10_enterprise_version_1703_updated_march_2017_x86_dvd, they too now encounter this same issue.  So the problem seems to be with the build of Windows 10 that was released.

I am going to download another build from MSDN and see if there is a change.

 

Update: March 19, 2018

I was able to get this working by way of reinstalling Windows 10 using the following ISO build: en_windows_10_multi-edition_vl_version_1709_updated_sept_2017_x64_dvd

Was even able to apply Windows updates and redeploy without error.

 

Tech News: VMware PowerCLI 10.0.0! Released

VMware just released PowerCLI 10.0.0. and before you ask; I thought they were just on version 6?  I wondered the same and here is the answer: The decision to move to version ten was a marketing choice as the PowerCLI project recently celebrated its 10th birthday.

Let’s get into the how to install or update to the latest

Requirments:

The only pre-requisite is to have PowerShell Core 6.0 installed. This adds support for Mac OS and Linux.

Installation Steps:

  1. Get yourself to a powershell prompt with administrative privileges
    In my case, I am on Windows 10 and prefer to use PowerShellISE
  2. Enter the following: Install-Module -Name VMware.PowerCLI -Scope CurrentUser
    This will initiate the install of the latest PowerCLI modules.

    If you receive a warning, use the -Force comamnd:

    “WARNING: Version ‘6.5.1.5377412’ of module ‘VMware.PowerCLI’ is already installed at ‘C:\Users\sysadmin\Documents\WindowsPowerShell\Modules\VMware.PowerCLI\6.5.1.5377412’. To install version ‘10.0.0.7895300’, run Install-Module and add the -Force parameter, this command will install version ‘10.0.0.7895300’ in side-by-side with version ‘6.5.1.5377412’.”

  3. Next enter: Set-PowerCLIConfiguration -InvalidCertificateAction Ignore

    This version of PowerCLI changes the way certificates are handled when connecting to a vCenter server or ESXi host with the Connect-VIServer cmdlet. If your connection endpoint is using an invalid certificate (self-signed or otherwise), PowerCLI would previously return back a warning. The handling has been updated to be more secure and now return back an error.If you are using an invalid certificate, you can correct the error with the ‘Set-PowerCLIConfiguration’ cmdlet. The parameter needing to be configured is ‘InvalidCertificateAction’ and the available settings are Fail, Warn, Ignore, Prompt, and Unset.

For more info ref: https://blogs.vmware.com/PowerCLI

 

 

Install VMware Tools Windows Server 2016 Core

I just completed my install of Windows Server 2016 Core as a guest in my VMware Lab. Now that this has been completed the next step is for me to install the VMware tools so that I can take advantage of various features; specifically, template deployment with customization options

About:VMware:Tools: VMware software tools enhance the performance of the guest operating system and improve the management of the virtual machine guests operating systems.

How to install:

  1. Select your VM from vCenter and select ‘Guest OS > Install VMware Tools
    This mounts the VMware CD Image containing the installation files
  2. Inside the guest machine type ‘powershell’
    This will drop you from the command shell to powershell prompt
  3. Next type the command Get-PSDrive
    This will return the drives attached to the system
  4. Change to the drive that the VMware tools are currently mounted
    In my case, this was drive letter “D”
  5. Issue the command .\setup64.exe to start the install process

    Note: issuing just setup.exe or setup64.exe will end in an error as Windows poweshell does not load commands such as this by default 
  6. Follow the steps of the VMware tools installer and restart when completed.

 

Ref: http://jermsmit.com/howto-install-vmware-tools-on-windows-server-2102-r2-server-core/

vSphere 6.5: OVF Import – The provided manifest file is invalid

Importing a template from vSphere 5.5 and importing to vSphere 6.5 the following error was encountered: The provided manifest file is invalidInvalid OVF checksum algorithm: SHA1

To get fix this error the following steps were taken:

Step 1 – is to extract your ova template (after all its only a zip)

You will notice 3 files once extracted

*.vmdk – is your disk containing all your data

*.ovf – is the configuration (also the file that we will edit)

*.mf – is a manifest containing a reference to the vmdk and ovf, also holding a SHA1 hash which ESXi will check for validation. This file needs to be deleted as we are making a change to the ovf and this will surely break that hash.

Example of what the contents of the .mf file looks like:

SHA1(template.ovf)= 908e804f140ffa58083b8bd154dace330b440c78
SHA1(template-disk1.vmdk)= 29c2d44d908d0207005360dabb58967f01a1

Step 2 – Delete the file with the *.mf extension. If this exists ESXi will attempt to validate and throw an error about the templates integrity being invalid. Once this has been deleted you can deploy your OVF Template.

Ref: http://jermsmit.com/unmount-local-iso-before-making-it-an-ovf-template/

Happy Importing

vSphere Customization Specification and Ubuntu 17.10

Has anyone noticed issues with #vSphere Customization Specification and #Ubuntu 17.10 (GNU/Linux 4.13.0-25-generic x86_64) – Now that the network settings are located in: /etc/netplan/01-netcfg.yaml, I’m back to manually updating IP information (Server Naming still works).

Although the traditional /etc/network/interfaces reflects the information provided by the customization process, the host still acquires a DHCP issued address. After some digging I was able to find out that this information was no longer being referenced and is now being pulled from:/etc/netplan/01-netcfg.yaml

Here are the IP configuration details

/etc/network/interfaces:

/etc/netplan/01-netcfg.yaml: