MDT – Windows 10 deployment watch

With our MDT environment up and running we’ve been refining our Windows 10 build over the past couple of months, sending out pilot builds to specific areas so we’re confident in the process when it comes to large-scale deployment over summer.#

This post focuses on a few Windows 10-specific tweaks that we’ve made to the Task Sequence that may be of interest…

Thin image approach

In the past I was a fan of what could be called a Hybrid image model in as much that I’d create a “Base” Reference image in a VM, usually comprised of Windows + Office + Updates. That would get captured and become the WIM file that goes into the Task Sequence.

However with Windows 10 I’ve decided to go down the completely thin approach that’s best represented as either a sandwich or hamburger depending on your culinary preference (!) Effectively the deployment gets built from its component parts, starting from an unaltered source Windows 10 WIM file extracted from its parent ISO image.

In our case we’ve settled on Education 1709 x64 as the build to deploy, due to some useful features such as OneDrive Files on Demand and Windows Defender Exploit Prevention. Along the way we’ve also used the 1607 and 1703 builds. The advantage of using the Thin image method is that we can swap the OS out at will with two clicks, rather than having to go through a Capture process that seems to have the potential for error.

Secure Boot validation

Windows 10 1709 brought in some new security features which benefit from machines being converted to UEFI rather than BIOS mode and in some cases (Windows Defender Credential Guard) needs Secure Boot too. Seeing as we need to update the BIOS > UEFI on older machines anyway it made sense to enable Secure Boot at the same time.

Ref: https://docs.microsoft.com/en-us/windows/whats-new/whats-new-windows-10-version-1709

The question was how to ensure that a machine is correctly configured before starting the imaging process (as converting later on is far from ideal).

The answer is to run cmd.exe to send a non-zero return code if specific requirements are met:

  1. Task Sequence variable isUEFI is false and \ or
  2. UEFISecureBootEnabled registry key is 0

If the machine is configured incorrectly the Task Sequence will fail before it even starts to pull down the image. To ensure you catch it early enough add the step here:

Putting the two together looks like this:

  

Removing the cruft

Sadly despite Microsoft giving Education our very own specific build of Windows they didn’t extend the effort into cleaning up the junk that gets pushed down with a standard Windows 10 installation. Seriously who wants Candy Crush on their business machines?!

Fortunately scripts exist to assist with cleaning up the junk shipped with the OS so it’s suitable for deployment. Now we can do this with DISM at image level but again my aim is to avoid tinkering with the Microsoft media if possible so I prefer the following PowerShell method…

PowerShell: Removing UWP apps from Windows 10 1607/1703/1709

Disable Refresh \ Reset

Another Windows 10-specific tweak is to disable the Refresh \ Reset menu that users can access either by using the Settings app or by holding shift while a machine reboots. In our case we don’t want users to wipe their machine clean of provisioned applications and it appears that this functionality will work even without local admin rights (!)

The solution to this one came via the EduGeek forums courtesy of ErVaDy using bcdedit commands:

Ref: http://www.edugeek.net/forums/windows-10/164236-preventing-shift-restart-into-recovery-mode-5.html

Place the commands below into a batch file and run as an Application or Task Sequence step:

reagentc /disable
bcdedit /deletevalue {current} recoverysequence
bcdedit /set {bootmgr} bootems off
bcdedit /set {bootmgr} advancedoptions off
bcdedit /set {bootmgr} optionsedit off
bcdedit /set {bootmgr} recoveryenabled off
bcdedit /set {current} bootems off
bcdedit /set {current} advancedoptions off
bcdedit /set {current} optionsedit off
bcdedit /set {current} bootstatuspolicy IgnoreAllFailures
bcdedit /set {current} recoveryenabled off

Updating OneDrive Files on Demand Client

In a way that only Microsoft can Windows 1709 shipped with an old version of the OneDrive client that doesn’t work with the much-anticipated Files on Demand feature straight out the box 😦

Although the client does auto-update we didn’t want any automatic sync starting without the placeholder functionality being in place so I’ve scripted an Application in the MDT Task Sequence to take ownership of the file on the newly deployed image, copy the latest version of the client over and then set everything back as it was.

For more details and the script itself please see my previous post OneDrive Files on Demand – update!

Pre-staging printer drivers

During our Windows 10 deployment we’re also migrating to a new set of Windows Print Servers, along with new GPOs to map them. However in initial testing I noted the first user to log in had a long wait whilst drivers were copied down from the server and installed.

Although subsequent logins won’t get this issue it doesn’t give a good first impression to the initial user so I wanted to find a way around it.

Step forward the very useful printui.dll 🙂

Ref: https://larslohmann.blogspot.co.uk/2013/12/install-printer-driver.html

Because we’ve rationalised our print fleet over the past few years in a move towards MFDs I only have 3 drivers to cover the entire range of hardware. By using a script method I can then pre-stage the drivers onto the machine at image time and speed up that first logon significantly!

Again paste this into a batch file and call as an Application (use an Application step instead of  Run Command Line as you want the driver files copied into the Deployment Share)

cscript "prndrvr.vbs" -a -m "HP Universal Printing PCL 6" -i "%CD%\HP Universal Print Driver\pcl6-x64-6.4.x.xxxxx\hpcu196u.inf"

Note the use of %CD% to ensure the path to the driver file is resolved correctly!

WSUS resources

Although there’s nothing special about running Windows Updates in MDT (use the built-in Task Sequence steps) we noticed that our WSUS server was struggling and sometimes hung the “Install Updates” step of the Sequence. The WSUS console then become unresponsive on the server end too.

After further research it turns out our increasing number of machines needs more resource than the default WSUS limit of 2GB  in the IIS Application Pool to handle the connections. Upon making the change below it’s back to being stable again.

Ref: https://sysadminplus.blogspot.co.uk/2016/11/wsus-console-crashed-after-running-some.html

Ref: https://www.saotn.org/wsuspool-keeps-crashing-stops

Ref: https://blogs.technet.microsoft.com/configurationmgr/2017/08/18/high-cpuhigh-memory-in-wsus-following-update-tuesdays

Run WinSAT

An oldie-but-goodie; running the WinSAT assessment tool at the end of setup will make sure your machine is properly benchmarked and appropriate performance tuning is performed by Windows. It doesn’t take long so I thought it worth continuing with:

Ref: https://deploymentresearch.com/Research/Post/624/Why-adding-WinSAT-formal-to-your-task-sequence-can-be-a-shiny-thing-to-do

Just add a Run Command Line step with the following in the box:

winsat.exe formal
Advertisements

OneDrive Files on Demand – update!

OneDrive logo

After our initial post getting the new Windows 10 1709 OneDrive client up and running with Files on Demand we had one or two little snags left to fix. Both of which are now resolved so thought I’d make a quick ICYMI post to cover the final pieces of the puzzle to getting everything up and running perfectly 🙂

Outdated client on the image

In true MS fashion the 1709 ISO ships with the old OneDrive client (epic fail) which means users have an annoying wait while it updates. There’s also the possibility to start off with the wrong client and therefore syncing files down by mistake.

I was trying out an updater script that would copy over the new client but didn’t have much success in MDT. After looking more closely at the logs with CMTrace I could see it failing on the copy operation so I added a Suspend action and tried each step manually. That flagged up an access denied error.

I then realised that MDT runs its scripts as the local Administrator user rather than SYSTEM as SCCM would, therefore the script’s permissions need tweaking for MDT use:

%SYSTEMROOT%\system32\takeown /f %SYSTEMROOT%\SysWOW64\OneDriveSetup.exe >> %SYSTEMROOT%\logs\Onedrive.log
%SYSTEMROOT%\system32\icacls %SYSTEMROOT%\SysWOW64\OneDriveSetup.exe /Grant Administrator:(F) >> %SYSTEMROOT%\logs\Onedrive.log
Copy OneDriveSetup.exe %SYSTEMROOT%\SysWOW64\OneDriveSetup.exe >> %SYSTEMROOT%\logs\Onedrive.log /Y
%SYSTEMROOT%\system32\icacls %SYSTEMROOT%\SysWOW64\OneDriveSetup.exe /Remove Administrator:(F) >> %SYSTEMROOT%\logs\Onedrive.log

This works like a charm! The updated client is installed during the Task Sequence and the first run as a user now begins with the 2017 client.

I’m also thinking of setting up a scheduled task on the MDT server to pull down the latest OneDrive client at regular intervals so the Task Sequence always deploys the latest version. That should do the trick until Microsoft see sense and push it out properly via WSUS.

Silently configure OneDrive using the primary Windows account

The final piece of the puzzle is to make the client log in via SSO so users have a fully configured OneDrive without any additional login prompts. I was puzzled by this not working initially as the GPO looks straightforward but it didn’t seem to do anything.

I’d read that the SSO relies on ADAL (aka modern authentication) so I initially wondered if our SSO provider hadn’t implemented that yet. That didn’t seem to make much sense as ADAL has been out for a while now so I hit Google a bit more deeply to try and find some further detail.

Soon came to this page, which I’m sure I’ve seen before:

Ref: https://support.office.com/en-gb/article/Use-Group-Policy-to-control-OneDrive-sync-client-settings-0ecb2cf5-8882-42b3-a6e9-be6bda30899c#silentconfig

The key (pun not intended, honest!) is the EnableADAL.reg file that’s squirrelled away at the bottom of the page. Deploy that via GPP et voila, one perfect blue OneDrive icon without any user interaction 🙂

What next?

Having got Files on Demand working how we want with minimal cache, SSO and the latest client we can now move onto piloting it with our users. I’ve been tweaking Windows 10 GPOs today for some of the newer features such as Windows Defender Security Center, Exploit Protection etc. so the configuration is looking good enough for some early adoption!

OneDrive Files on Demand – first steps

OneDrive logo

After much anticipation and playing with Windows Insider previews OneDrive Files on Demand finally hit general release alongside Windows 10 1709 (Fall Creators Update) the other week. I’ve been giving it a test drive over the past week or two along with fellow Network tech Matt Stevens – here’s a few of our observations so far along with workarounds for a couple of teething issues.

Windows 10 build

There is one pretty important requirement to bear in mind with the new Files on Demand feature; it’s only available in build 1709 and above. That means you need to be on the semi-annual (aka CB) branch rather than the LTSB route that some people have taken.

Ref: https://blog.juriba.com/windows-10-branching-timeline

It’s new features like Files on Demand that make the additional work of staying up-to-date worthwhile; so far we have a couple of hundred laptops running 1703 without too much fuss so 1709 should slot in fairly smoothly as we build our images layer-by-layer now using only the pure Microsoft WIM as a starting point.

We tamed (nuked) the built-in apps via a very handy Powershell script we found online (also see alternative version here) that runs during MDT deployment and the Start Menu default tiles are cleaned up via a GPO layout file. Configure your Windows Store for Business (or Education as case would have it), tweak a few more policies for Cortana, Telemetry etc. and Windows 10 becomes much more manageable even on the latest build.

Why Files on Demand?

If you don’t know what all the fuss is about check out the initial Insider announcement:

Ref: https://blogs.windows.com/windowsexperience/2017/06/13/onedrive-files-demand-now-available-windows-insiders/#kwLbqguOTefId6pv.97

Ref: https://blogs.office.com/en-us/2017/05/11/introducing-onedrive-files-on-demand-and-additional-features-making-it-easier-to-access-and-share-files/?eu=true

What it basically means is that we can finally integrate (huge amounts of) cloud storage with our on-premise desktops in a much tighter fashion and dispense with (unsupported) scripts or (expensive) third party tools to access OneDrive on a Windows desktop using File Explorer. It also means not having to deal with WebDAV, which always felt a horribly dated and clunky protocol to use for accessing cloud storage.

As soon as the 1709 ISO hit VLSC I grabbed it from Microsoft, slotted the new WIM into one of my MDT Task Sequences and deployed a VM to give the production version a try. It shows much promise but as always there’s some gotchas that mean nothing is ever quite straightforward.

Client version

Microsoft being Microsoft always have one shoot-self-in-foot moment whenever a new product comes out and this release was no exception. Despite having the freshly downloaded 1709 ISO I noticed that on first launch the client was showing up as 2016 and not the latest 2017 (17.3.7076.1026) that brings in Files on Demand

https://support.office.com/en-gb/article/New-OneDrive-sync-client-release-notes-845dcf18-f921-435e-bf28-4e24b95e5fc0


that’s the one that you want…

There’s a useful summary of the client install \ update process below. It does strike me as odd that the client self-updates and installs from appdata rather than being managed by WSUS.

Ref: http://deploynovellas.com/2016/05/25/install-onedrive-ngsc-update-windows-10-osd

Similarly it also takes a while to update when deployed on a clean 1709 build due to the initial client being out-of-date. This also means if a user is a bit too quick off the mark they can end up with an old-school full sync rather than Files on Demand.

I’ve been trying to replace the client during the deployment Task Sequence but more testing is required as my initial attempt failed with “Application Microsoft OneDrive 17.3.7073.1013 returned an unexpected return code: 1”.

Ref: http://model-technology.com/next-gen-onedrive-deployment-during-sccm-osd

I’ve added a Suspend action to the Task Sequence and will examine the logs to see what’s going on as the script tries to run…

Group Policy

To get more control over how the client is used grab the updated Group Policy templates from the local installation folder %localappdata%\Microsoft\OneDrive\BuildNumber\adm\

Ref: https://support.office.com/en-gb/article/Use-Group-Policy-to-control-OneDrive-sync-client-settings-0ecb2cf5-8882-42b3-a6e9-be6bda30899c

We force Files on Demand to be enabled as we don’t want sync cache eating up drive space on machines. We also configure our tenant ID (found via the Azure AD portal) so only Office 365 accounts can be used.

Configure these under Computer Settings > Administrative Templates > OneDrive

  • Allow syncing OneDrive accounts for only specific organizations > Enabled (using Tenant ID)
  • Enable OneDrive Files On-Demand > Enabled
  • Silently configure OneDrive using the primary Windows account > Enabled

I need to check if our third-party identity provider supports ADAL to make sure that last GPO setting works correctly. In the future we may well move to Azure AD Connect Passthrough authentication instead.

Clearing local cache (Free up space)

One important thing to remember about using Files on Demand is that when a file is either downloaded from the cloud, or freshly uploaded to it a cached copy will be kept on the local machine.

Over time (or with a large upload) this cache could grow and cause similar issues to what we were trying to avoid, especially with a shared machine and large volumes of users (pretty much the case for all our classroom machines)

At present it seems that no policies exist to force the “Free up space” option that removes the cached copies of files. However the article below suggests that using the new file attributes that have been brought in with 1709 can automate the process.

“Attrib.exe enables 2 core scenarios.  “attrib -U +P /s”, makes a set of files or folders always available and “attrib +U -P /s”, makes a set of files or folders online only.”

https://techcommunity.microsoft.com/t5/OneDrive-Blog/OneDrive-Files-On-Demand-For-The-Enterprise/ba-p/117234

We tried a script that runs on the root OneDrive folder and sure enough it resets all files back to Online only and reduces the space used down to a megabyte or so 🙂

cd "%userprofile%\Onedrive - Name of your Organisation"
attrib +U -P /s

Running this script on Logoff should in theory keep the cache files down to the bare minimum.

Disclaimer: we only just figured this one out today so again caveat emptor if you go and run this in production without testing it first!!!

MDT imaging megapost – part 1 (our first server)

The great thing about working in the tech field is that it keeps moving on, ever changing, always evolving. That means sometimes you have to let go of systems that were once the bright shining light of progress once it becomes apparent something better has taken its place. Now is that time for my trusty ZCM 11 custom imaging system; built back in 2013 and star of a 6-part thread series I look back on now and think “wow, I actually did that”.

Until I moved imaging onto a Satellite the stats say the original Primary server pushed out over 5000 images. Given the length time the Satellite has been in place, plus the stats from our other sites that figure can easily be doubled and over the course of 4 years around 10,000 image cycles have been completed.

Compared to the previous process that was in place a huge amount of time was saved and allowed us to complete a large-scale Windows 7 migration with relative ease. Add to that a 4-year saving on ENGL license costs and my motley crew of Bash and PowerShell scripts can retire with a satisfied feeling of a job well done 🙂

The future calls, and it’s shaped like the number 10…

However we need to move on, funny enough it’s another OS migration knocking on the door that prompted the change along with a shift in hardware and environment that meant the Linux-based PXE environment was starting to hold us back.

Windows 10 support from ZCM seemed patchy at best, as was timely support for new hardware such as Surfaces and their ilk. Reading the forums and email groups didn’t inspire much confidence either so we decided to start looking elsewhere.

SCCM was the natural direction of travel but having made a substantial investment of time creating ZCM Bundles we weren’t necessarily ready to move all that just yet. Similarly ZCM Patch Management works pretty well these days for covering our 3rd-party apps. With that in mind the Microsoft Deployment Toolkit was the obvious choice.

A nice GUI-based managed scripting environment with Windows PE as the underlying OS ticked all the boxes. Oh and did I mention it’s free!

It’s time for my own MDT… Massive Deployment Thread!

What originally started as a small side-project to push Windows 10 out to a couple of trial tablets has now expanded into a core system that’s been at the heart of our summer works. With that in mind it’s time to write up the journey and the numerous tips, tricks and tools used along the way.

Many of those ideas come from some of the best deployment pros in the business such as Johan Arwidmark, Michael Niehaus and Mikael Nystrom so a big shout out for all the knowledge they share. Hopefully this post will give an idea of how we put those pieces together in a live environment.

The beginning, our first server

Initially we started out deploying MDT for the sole purpose of imaging up a batch of demo Surface 3 devices so the first thing was to spool up a new VM with all the required software and roles installed. Links can be found below to save you some time:

Early fixes and customisations

After getting the basic Deployment Share running we hit a few minor issues that need resolving, which are worth bearing in mind:

Multiple DNS namespaces

We have two domains that are in use internally, one of which usually gets appended as part of the domain join process and the other via DHCP.

In the PE environment the machine isn’t domain joined and as such the default setting in Bootstrap.ini wouldn’t connect to the deployment share as it didn’t know the correct DNS suffix to append.

Ref: https://scottisageek.wordpress.com/2011/12/22/mdt-2010-and-multiple-dns-namespaces/

…we found it quicker in our case to change the DeployRoot setting to the MDT server’s FQDN rather than short name… problem solved 🙂

Share permissions

The default permissions applied to the Deployment Share by the installation wizard weren’t set up as we liked. Can’t remember the exact reason now but looking back documentation on other sites I think the share needed locking down to prevent users viewing the Deployment Share content or (even worse) making unauthorised changes to it (!)

We now have specific AD groups and a service account set up so nominated MDT Administrators can read \ write to the share to upload Application install files etc. but the imaging account (more on that later) can only read and all other users are denied access by virtue of having no rights.

Set UK Locale

A quick an easy tweak sets up the keyboard settings for UK users in Bootstrap.ini

Ref: http://kabri.uk/2010/01/20/sample-bootstrap-ini-for-uk-deployments/

Similarly set them also in CustomSettings.ini

Ref: https://scriptimus.wordpress.com/2011/06/23/mdt-2010-sample-customsettings-ini-for-fully-automated-deployments/

There are quite a few other settings you’ll want to add in CustomSettings.ini but more detail on those will follow in relevant posts so keep your eyes peeled!

Update the Deployment Share

This is one action you’ll soon need to get into the habit of! If you make changes to the settings in any of the .ini files or add drivers that you’ll need in the PE environment (basically network and storage) then you need to update the Deployment Share.

This recompiles the Boot Images to include your changes, otherwise you’ll find all those nice new additions above make no difference whatsoever!

Think of this as step 1 / 2 to completely updating the Boot Images though. If the MDT wizard says that the Boot Images have changed you also need to copy the new WIMs over to WDS so PXE boot is using the latest images.

In WDS browse your server select Boot Images then right click as per screenshot above and click Replace Image. Browse to your Deployment Share’s Boot folder and select the correct image for each architecture.

Windows Deployment Services service won’t start

At an early point in our testing WDS decided it didn’t want to start after a server reboot and was spewing error code 0x906. We weren’t sure why and were on the verge of reinstalling from scratch when I spotted this:

Ref: https://social.technet.microsoft.com/Forums/windows/en-US/265b4b53-63ac-491f-817c-6030daa39b81/cant-start-windows-deployment-services-service?forum=itprovistadeployment

As per Aaron Tyler’s advice in the link above run the wdsutil commands to uninitialize then reinitialize the server manually pointing to the RemoteInstall folder WDS creates.

wdsutil /uninitialize-server
wdsutil /initialize-server /reminst:[PATH_TO_REMOTEINSTALL_DIRECTORY]

Next time…

That should be enough to get your first server up and running. For the second post in the series we’ll look at the MDT Database and how it turns MDT from a good imaging solution into a great one 🙂

Build your own Thin-ish client with Windows 10 LTSB

After some positive user feedback from the launch of our new Server 2016-powered RDS setup I started wondering if it could have a wider use that just the remote access concept we initially wanted to address. One thought in mind was making use of old \ low-spec devices that would be a bit too clunky for running a modern OS but where the physical hardware itself was in good condition.

Chrome-OS esque distributions such as CloudReady sound nice but come at cost so I set up a little side-project to see if there’s anything that could be done with what we have on our licensing agreement or anything in the open-source space.

Looking around there do seem to be various thin-client “converter” products but again they all seem to be commercial e.g. https://www.igel.com/desktop-converter-udc/

The only other option I found was ThinStation which may also be worth a look when I have more time as it seems a bit more involved to get set up and I wanted to stick to the Microsoft RDP client for now for maximum compatibility.

Windows options

Going back some time I remember Microsoft released cut-down versions of Windows for RDS-type scenarios; going back to the XP days it was called Windows Fundamentals for Legacy PCs and morphed into Windows 7 Thin PC in its next incarnation. Effectively all I want the OS to do is boot up, log in quickly then pass the credentials to a pre-configured RDP file using the standard mstsc.exe application.

However building any solutions on a Windows 7 base going forward seems to be a false economy so I decided to have a look around to see what was available on the Windows 10 codebase – the results were interesting…

IoT is name of the day

Going forward it seems Microsoft have changed the branding for this kind of cut-down devices to Windows IoT. In fact there’s a free edition which sounds ideal but it only runs on certain devices and isn’t really geared for UI use:

Ref: https://www.theregister.co.uk/2015/05/21/first_look_windows_10_iot_core_on_raspberry_pi_2/
Ref: http://blogs.perficient.com/microsoft/2016/01/windows-10-iot-editions-explained/

Reading a bit further it appears Microsoft license an edition called Windows 10 IoT Enterprise for new thin client devices. Now it gets interesting… it seems that the OS itself is Windows 10 Enterprise LTSB but with some special OEM licensing. It just so happens the edu customers get Enterprise LTSB on EES licensing so it’s time to take a closer look!

What this does mean is that Windows 10 Enterprise LTSB gets features from the old Windows Embedded products such as Unified Write Filter, perfect for a locked down device that shouldn’t need to experience configuration changes to the base OS.

Ref: https://msdn.microsoft.com/en-us/windows/hardware/commercialize/customize/enterprise/unified-write-filter

All these features are available in Enterprise LTSB simply by going into Add \ Remove Windows Features window, look for the Device Lockdown section and add whichever ones meet your needs (more on this later).

Image & GPOs

After downloading the latest ISO the LTSB 2016 WIM was imported into MDT. I made a quick task sequence to get it up and running and deployed the OS to a Hyper-V VM.

Boot and logon speeds are very quick given the lack of any Modern Apps which usually need to be provisioned at each new login. The performance gain explains why quite a few people within education have used LTSB for their desktop builds against MS’ wishes; however they’ll miss out on new features such as the much-needed OneDrive Files on Demand that will only be provided to the Current Branch release.

In theory setting up a Mandatory Profile could speed up login even further but haven’t got round to trying that yet.

RDS domain SSO

Upon logging in with domain credentials the next aim is to seamlessly drop users into the RDS farm without any further prompts. After doing a bit of research this can be achieved by setting a couple of GPOs:

  • allow credential delegation
  • trust SHA1 signature of signed RDP file

The need to allow delegation of credentials is fairly commonly mentioned but a lot of the articles are old and don’t mention where this needs to be set in a 2016 farm. In fact you only need to allow the delegation on the FQDN of the Connection Broker based on the results of my testing so far.

Computer Configuration > Administrative Templates > System > Credentials Delegation

To avoid any unwanted prompts about trusting the signature of a signed RDP file populate the GPO mentioned above and copy \ paste the signature from the RDP file that is provided by RDWeb for whatever RDS Collection you want to connect to.

User Configuration > Administrative Templates > Windows Components > Remote Desktop Services > Remote Desktop Connection Client > Specify SHA1 thumbprints of certificates representing trusted .rdp Publishers

Custom shell

Now with the credentials side sorted out the final piece of the puzzle was to cleanly launch the session and (here’s the tricky bit) made a seamless logout once the RDS connection is closed. Now there’s a few ways to achieve the first part:

  • use the IoT Embedded Shell Launcher feature \ enable Kiosk Mode via System Image Manager
  • use the Custom User Interface User GPO

Ref: https://social.technet.microsoft.com/Forums/en-US/b4552957-45c2-4cc4-a13d-6397f06ee62e/windows-10-kiosk-build-embedded-shell-launcher-vs-custom-user-interface?forum=win10itprosetup

Ref: https://docs.microsoft.com/en-us/windows/configuration/set-up-a-kiosk-for-windows-10-for-desktop-editions

One thing to bear in mind with Shell Launcher is what happens when the shell i.e. mstsc.exe closes, you only have the choice of

  • Restart the shell.
  • Restart the device.
  • Shut down the device.
  • Do nothing

For the sake of speed logging off would be better so I decided to go with the Custom User Interface GPO – seeing as the Windows 10 device would be domain-joined anyway it also seemed a quicker more efficient way to configure multiple clients too.

Seeing as the Custom User Interface is a User GPO it goes without saying that Loopback Policy Processing needs to be enabled for the OU where the client resides. That also comes in handy for a few additional personalisation settings later on too.

The User GPO settings are summarised in the screenshot below, you can add more lock-down policies as you see fit:

Auto log-out on disconnect

Seeing as I wanted to automate the process as much as possible and all the devices would be domain managed anyway the GPO method seems to be the quickest way to achieve what I want. Also avoids needing to do an Add \ Remove Features step for each endpoint device.

Another important point is that the Shell Launcher method only provides options to relaunch the program, shut down or restart the machine. For speed I was aiming to log off the “client” when the RDS session is done so definitely going down the GPO route as a result.

In the GPO settings I initially tried the standard string you’d expect to launch a Remote Desktop session i.e. mstsc.exe C:\Default.rdp but noticed some strange behaviour:

  • Windows logs in
  • RDP file launched
  • connection starts
  • before the green bar completes i.e. handshake still in progress
  • host session logs out

This seemed like a behaviour I’ve seen with some other programs in the past where they appear to terminate mid-way through actions actually occurring. To check I tried manually with the “start” command with the same result. It appears mstsc.exe doesn’t play nicely so we need another way…

Plan b) was to monitor the mstsc.exe process then log out from the client once RDS disconnected and therefore the process was no longer running. After looking around and trying a few scripts out I settled on one I found here:

Ref: https://www.experts-exchange.com/questions/24218998/Check-if-a-process-is-running-in-vbs.html

Just add the logout command as the action to run when the desired process terminates and we have the desired behaviour. It takes a second or two to react to the process closing but there doesn’t seem to be a way to speed that up as far as I can see.

Final steps

Now just some finishing touches required to give the solution a bit of polish 🙂

  • set logon and desktop wallpaper
  • disable Task Manager and related lockdown setings

When the machine boots users see this login screen, easily customised via GPO…

After login connection to RDS is pretty much immediate and no further credential \ security prompts appear…

UWF

The final piece of the puzzle is tidying up after the client has been in use for a while. That’s where the Unified Write Filter from earlier comes in handy:

Enable-WindowsOptionalFeature -Online -FeatureName Client-UnifiedWriteFilter

Then enable the filter;

uwfmgr.exe filter enable

Ref: https://docs.microsoft.com/en-us/windows-hardware/customize/enterprise/unified-write-filter
Ref: https://developer.microsoft.com/en-us/windows/iot/docs/uwf
Ref: https://deploymentresearch.com/Research/Post/632/Using-the-Unified-Write-Filter-UWF-feature-in-Windows-10

And there you have it, a locked down RDS client that will run on older hardware (Windows 10 works on pretty much anything from the last 10 years) which can be managed through your standard AD infrastructure, all using stuff you already have access to via your Campus agreement… enjoy!