Tag: Microsoft

Duplicate SRV records are cousing domain join workflows to fail

Have you ever had problems with duplicate SRV records in your environment? This is a quite common phenomenon when you google it without any real solution to it (not at least what I could find). Some environments would not be affected by this, but I got into a specific situation recently where some workflows in Nutanix would fail because of duplicate SRV records.


  • Duplicate SRV records, one in lower-case – one in upper-case, are causing some workflows in Nutanix to fail.
  • When deleting the oldest record the duplicate is just recreated after some period of time (like 30 minutes or so).

So whats cousing this? In this specific case we managed (together with Microsoft support) to isolate the issue and found out that there were two main things that were related to this behaviour listed below.


  • Some Domain Controllers names were in lower-case, others in upper-case.
  • When you have a mixture of DNS servers running Windows Server 2012 and 2016 the way that machine names are registered differs between those Windows versions.

So how do we solve this? The preferred solution from Microsoft was to rename all domain controllers to lowercase, but since all Domain Controllers except one, in this case, was in uppercase we tried to rename that specific DC to uppcase instead. The following steps were performed on the server:

    1. Demote DC
    2. Rename to uppercase
    3. Promote DC
    4. Delete all duplicated SRV records in DNS
    5. (If  the issue is still happening):
      1.  Stop netlogon service
      2. Delete C:\Windows\System32\config\netlogon.dnb
      3. start netlogon service

After doing this the duplicate SRV records stopped being recreated in the environment.


  • The preferred way to solve the issue is to rename all domain controllers to lowercase (or uppercase which works too).

If you have any questions, feel free to email me at tobias.sandberg@xenit.se or comment down below. I will try to answer you as soon as possible.

Azure AD Connect and .NET Framework 4.7.2


Last week a discussion erupted on Microsoft forums regarding Azure AD Connect due to it’s Monitoring Agent using all free resources of CPU on the servers. These issues were caused by a .NET Framework update and a lot of administrators spent time uninstalling and blocking these patches to resolve the CPU usage issues on their servers. On Saturday Microsoft released an update (KB4340558) which contains a collection of several patches where one of the earlier mentioned .NET Framework updates were included. For more information, see this link.

Microsoft has recently published an article regarding this issue. In addition, Microsoft also published a new version of the health agent where they state that the issue is resolved, it can be downloaded from here. The new health agent version is set to be included in the next version of Azure AD Connect, which will be published for Automatic Upgrade (Auto Upgrade). The following patches have been identified with issues causing Azure AD Connect’s monitoring agent using huge amounts of CPU:

Auto Upgrade

In version of Azure AD Connect, Microsoft introduced Auto Upgrade. Although, not all updates are published for Automatic Upgrade. Whether a version is eligible for automatic download and installation will be announced on Microsofts version-history website for Azure AD Connect.

You can verify whether your Azure AD Connect installation have Auto Upgrade enabled by either using Powershell or viewing your configuration in It’s GUI.

Graphical User Interface of Azure AD Connect
PowerShell-command for determining whether Auto Upgrade is enabled or not.

This command will return either Enabled, Disabled or Suspended, where as the Suspended state only can be set by the system itself. Newer installations of Azure AD Connect enables Auto Upgrade by default, in case your installation applies to Microsoft’s recommendations. For more information, see this link.

Enabling Auto Upgrade

In case you have an installation of Azure AD Connect older than (February 2016), Auto Upgrade will be disabled, if you’ve not enabled it manually. Enabling this function can be done with below PowerShell-command if so wanted.

If you have any questions, feel free to email me at robert.skyllberg@xenit.se

Flickering Desktop Icons and re-directed folders

This blog post will only cover a scenario with Microsoft Windows Server 2016 Remote Desktop Services (RDS) and re-directed folders where flickering icons appear. Other solutions may apply to different scenarios.
Since the release of Windows 10 / Server 2016 and their different releases 1607, 1703, 1709 and 1803 there has been several issues regarding flickering icons on the Start-menu, in File Explorer and taskbar.


During the deployment of Citrix Virtual Apps and Desktops 7.15 on Windows Server 2016 with published Desktops and re-directed Desktop folder, users could experience that the desktop icons kept flickering continuously. The more shortcuts, folders or files on the Desktop the more prevalent the issue was. Constantly blinking icons on the desktop looked like refreshing the desktop with F5 or Ctrl+R and would also flash when browsing network shares.

My first thought was to activate ”Always show icons, never thumbnails” in Folder Options since there seemed to be a constant query to network shares where the re-directed Desktop folder resided.

File Explorer - Options

File Explorer – Options

File Explorer - Always show icons

File Explorer – Always show icons


The moment I clicked on View in Folder Options the desktop icons ceased flashing in my session. Dwelling deeper with Procmon investigating what actually happens when opening View tab in Folder Options I found out that explorer.exe queries a registry key in the users HKEY_CURRENT_USER registry. If the registry entry does not exist it will be created.

  • HKEY_CURRENT_USER\SOFTWARE\Classes\CLSID\{031E4825-7B94-4dc3-B131-E946B44C8DD5}
Explorer query and creation of registry key

ProcMon – Explorer.exe query and creation of registry key


With the knowledge that the registry key was missing and creating they key would stop the icons from flashing for users on Windows Server 2016 RDS, the appropriate solution was to use Group Policy Preferences (GPP) that created the registry key for users during logon (run in logged-on users’s security context) and apply it to Windows 2016 RDS servers.
Gorup Policy Preferences - User Configuration - Registry

Gorup Policy Preferences – User Configuration – Registry

Apply to Current User

Apply to HKEY_CURRENT_USER and set Key Path

Run in logged-on users security context

Run in logged-on users security context

Step 1: Create a USER GPP that will be applied to affected targets

Step 2: Create a Registry Item

Step 3: Add registry key

  • Key Path: SOFTWARE\Classes\CLSID\{031E4825-7B94-4dc3-B131-E946B44C8DD5}
  • Tab Common: [v] Run in logged-on user’s security context (user policy option)

If you have any questions regarding above solution, or ideas on how to handle above in a better way, please contact me at viktor.glinski@xenit.se or post a comment below.

Sending CSS formatted tables in Outlook

If you’ve ever used Powershell to send HTML tables in Outlook containing CSS you’ve probably been disappointed of the outcome.
There is some archived documentation for Outlook 2007 that is still viable for Outlook 365 (https://msdn.microsoft.com/en-us/library/aa338201(v=office.12).aspx).

Basically the function accepts a csv and css file, hardcodes the css into the table and outputs a formatted HTML table that is compatible with Outlook.

Example table sent using the function and send-mailmessage
The css has odd/even for readability, bolded column 1/4 and red text for column 3.
This is by default impossible to achieve using just css in outlook.


HTML output


Since the CSS does not work perfectly the style.css file imported needs some specific configuration..

  • classes has some specific name structure”
    • columns are named .coln
      • n is the number of the column starting with 1 to infinity. .col1 .col2 and so on
    • one whitespace is required between class name and the curlybrackets.
      • Curlybrackets must be on the same row as class name
      • Ending curlybrackets must be on a separate line
    • Data must be on separate rows
  • Odd/even css is the only tr handled code.
    • Must be named exactly
      • tbody tr:nth-child(odd) {
      • tbody tr:nth-child(even) {

Example style.CSS



Nyheter på väg till RDS 2016

Microsoft presenterade tidigare i höstas nyheter som är på väg till Remote Desktop Services (RDS) 2016. Det är några stora förändringar på gång som är viktiga att känna till, och detta inlägg sammanfattar några av de nyheter som ska komma inom kort.


I en traditionell RDS infrastruktur måste alla servrar i uppsättningen vara med i domänen. Det innebär att RD Gateway och Webaccess servrarna både är med i domänen och har direkt kontakt mot internet, vilket gör dem sårbara för attack.

Med den nya infrastruktur design som Microsoft presenterar så är Gateway, Webaccess och de övriga rollerna ej längre med i domänen. Kontakten från domänen till infrastrukturen görs endast genom utgående trafik på port 443. Förutom att detta ökar säkerheten, så möjliggör det för organisationer att drifta flera olika miljöer med samma RDS infrastruktur. Inte längre behövs den en RDS miljö för varje domän, utan nu kan infrastrukturen sättas upp en gång för att drifta flera olika miljöer och låta användare ansluta till deras respektive domän och Sessionhosts.

Microsoft presenterar även en ny roll inom Remote Desktop Services; Diagnostics, vilket har som uppgift att samla in information om uppsättningen och kan användas för att felsöka anslutningsproblem.


Integration med Azure Active Directory (AAD) är snart här. Med hjälp av AAD så kan Multi-Factor Authentication, Intelligent Security Graph och övriga Azure tjänster nyttjas i RDS miljön. Azure AD är något som många organisationer redan nyttjar, om de använder sig av Office 365 tjänster.


Om RDS miljön sätts upp i Azure så kan organisationer installera RDS rollerna som Platform as a Service (Paas) tjänster. Det innebär att det inte längre krävs ett VM för varje roll i infrastrukturen, Administratörer slipper alltså managera varje VM individuellt, samt de får tillgång till den smidiga skalbarheten som Azure erbjuder. Denna uppsättning stödjer även hybrid-lösningar, Sessionhosts kan alltså ligga on-premise och resten av infrastrukturen i Azure.

Det finns fortfarande ingen ETA på när dessa nyheter görs tillgängliga. För mer information och demo på några av dessa funktioner, se inlägget från Microsoft.

Outlook Search index med FSLogix

Något som upptäckts snabbt efter uppsättningen av sin ”FSlogix Office 365 Containers”-lösning i en fleranvändarmiljö är att sök-indexeringen för Outlook i vissa miljöer görs om vid varje ny inloggning, det gäller miljöer där man har flera Session Hostar användarna kan logga in på.

Sök-funktionen i Outlook använder sig av ”Windows Search” vilket är en databas över indexeringarna på hela Operativsystemet, det är alltså inget som lagras för varje enskild användare. Det innebär t.ex.  att en Citrix miljö med flera servrar kommer en användares Outlook indexera om hela Outlook vid varje ny server man loggar in på. Detta medför en långsam sökning (tills indexeringen är klar) och en onödigt belastning på CPU som i sin tur kan påverka hela miljön negativt. Det kan bli ännu värre i de fall man använder Citrix Provisioning Services (PVS) då den uppdaterade indexeringen försvinner vid varje omstart av servern.

FSLogix to the rescue

För att komma runt detta problem finns en funktion i FSLogix som tar med din Outlook indexering i VHD-filen, på så vis har du alltid din uppdaterade indexeringsdata med dig på vilken server du än hamnar på. Du behöver ändra på två stycken registervärden för att aktivera detta, jag själv föredrar att skapa/editera en GPO för detta.

Följande två registervärden ska justeras:


Type:                      DWORD

Value Name:          RoamSearch

Value Data:            2



Type:                      DWORD

Value Name:          RoamSearch

Value Data:            2


Hör gärna av er om ni skulle vara intresserade av eller vill veta mer om produkter från FSLogix, se gärna våra tidigare blogginlägg om FSLogix nedan:

FSLogix Profile Containers – Enkel och snabb Profilhantering

Office365 med FSLogix i en fleranvändarmiljö

OneDrive with simulated Single Sign-On



Azure Automation – Running scripts locally on VM through runbooks

I was tasked to create a powershell script to run on a schedule on a Azure VM. Normally this would be running as a scheduled task on the VM but seeing as we’re working with AzureVM and schedule tasks are legacy I wanted to explore the possibilities of running the schedule and script in Azure to keep the VM clean and the configuration scalable.

After some research the best option would be running the powershell script as a CustomScriptExtension on the VM, and the schedule would be handled by a Process Automation Runbook (using Automation Accounts).

What I ended up with is the script below. It’s fairly easy to configure and contains almost all the required configuration in the parameters.

Just enough Administration & RDS

The Problem?

Microsoft RDS has limitations when delegating access, in fact there is no built-in access delegation whatsoever!

The solution? Powershell!

A Just enough Administration (JEA) endpoint, also known as a Constrained Powershell Endpoint.

I’ve created a powershell app to list and logoff users I also created a simple tool in powershell to connect to connection brokers and search users. All without delegating users access to any RDS servers.

  1. Connection broker/Endpoint connection
  2. Collection field, used to search and filter. Multiple collections can be selected at once
  3. Wildcard search field, can be blank (list all users)
  4. User information & selection are
  5. Press to logoff all selected users

How does this work? Constrained Powershell Endpoint.

The endpoint is restricted to only access List collection, list users and subsequently force logoff users. The endpoint configuration to only accept users from a certain AD group.

To configure endpoint, run the following code on all connection brokers. The script is somewhat generalized for easy adaptability. The main parts are in the parameters where we configure who can run, using which account and also what cmdlet are available in the constrained session.

The frontend application code is not posted in the blog.

Print drivers and Microsoft Update KB3170455

Typically users get their printers mapped by Group Policies or Group Policy Preferences. Especially in Citrix environments, users should not have the right to add their own printers or drivers that are not approved for multi-user environments. On July 12th 2016, Microsoft released a security update (KB3170455) to safeguard Man-in-the-Middle (MITM) attacks for clients and print servers. Then an updated version was released again September 12th 2017.

Users could encounter the dialog boxes below if the driver did not meet the requirements of Microsoft where the driver would be packaged and signed with a certificate:

Scenario 1

For non-package-aware v3 printer drivers, the following warning message may be displayed when users try to connect to point-and-print printers:


Do you trust this printer?

Scenario 2

Package-aware drivers must be signed with a trusted certificate. The verification process checks whether all the files that are included in the driver are hashed in the catalog. If that verification fails, the driver is deemed untrustworthy. In this situation, driver installation is blocked, and the following warning message is displayed:


Connect to Printer

Even if you enabled Point and Print restrictions in GPO and specified which server’s clients could get drivers from, users could encounter an installation prompt and request administrator privileges to install.

For most printers this is not an issue if there is an up-to-date driver which is compliant. Some manufacturers do not always provide printers drivers that is both packaged and signed. The first thing you should do is update the driver to one that both is signed and packaged. Usually the drivers from the manufacturer are signed according to Microsoft Windows Hardware Quality Labs (WHQL) but may not be packaged correctly and the users get prompted for administrator credentials when the printer is being added to the client computer or in the remote desktop session.

Since KB3170455 we need to enable point and print restrictions and specify our print servers in the GPO. For most printers there is no issues, however a couple of printers will not be pushed out by Group Policy Preferences since the update. Even though the print server was listed in the point and print GPO. Browsing the print share and trying to connect the printer manually would result in the ”Do you trust this printer” pop up which will then prompt for administrator credentials to install the driver. Looking at Print Management on the server in question shows that the problem printer drivers have a ”Packaged” status of false.


If you are pushing out printers via Group Policy or Group Policy Preferences and they are of Non-Packaged type you will always get a prompt to install, ignoring the point and print GPO, which will cause the install to fail. A workaround to this is a registry edit on the print server – test and verify this first before putting it into production:

  • HKLM\System\CurrentControlSet\Control\Print\Enviroments\Windowsx64\Drivers\<…>\<Driver name>\PrinterDriverAttributes

Change the value from 0 to 1 and reboot the printspool service or/and server. The value for other print drivers may not be 1, but to make this work the value needs to be set to an odd number. For example, if the value is 4 change it to 5. Only do these changes if you have no other means of getting a valid driver or printer swapped. In RDS/Citrix environments you could pre-install the printer driver on the host if viable and you only have a few session-hosts.

Back in Print Management you will see the Packaged status is now changed to true, and the printer should deploy. If you can find packaged print drivers then use those, but some manufacturers have not bothered supplying them.


PrintManagement – Packaged True

Source: https://support.microsoft.com/en-us/help/3170005/ms16-087-security-update-for-windows-print-spooler-components-july-12

New modern management features

On Microsoft Ignite this year I learned that we are going from a classic workplace to a modern workplace. Microsoft refer to this as the digital transformation and that require us to change the way we manage our environments today. The characteristics of traditional PC management is on-premise infrastructure, high control with Configuration Manager/GPOs and business-owned devices while the characteristics of mobile device management is cloud services, a simpler IT process and BYOD.

Microsoft suggest that a new organization should go cloud-first, while existing companies should go for one of the other methods, like the Big Switch Transition or the Group by Group Transition. The paths to modern management is illustrated by this picture:

One of the most common ways for companies to follow this path will be with a new management feature called Co-management. This feature uses a combination of Configuration Manager and Intune so the transition to the modern workplace can be done by an iterative way instead of switching to a cloud based management right away, like the Big Switch Transition illustrated above.

Another reason for using Co-Management is because Intune doesn’t have all the existing ConfigMgr policies and settings available yet, something that Microsoft is currently working on to bring more features and functionality into Intune. A good way to check which policies and settings that can be replaced with Intune in your existing environment is to use a tool called MMAT. The tool can be downloaded here. Like the picture below illustrates you will get a report stating which policies and settings you can replace with Intune. Note that the report doesn’t care about the order or precedence of the policies and settings.

A computer managed by Co-management can be managed both by on-premise ConfigMgr and cloud based Intune. Like I mentioned before, Microsoft is currently working towards bringing more settings and policies into Intune to extend the MDM capabilities for the cloud based management. Some of those new features are shown below.

To get more information about the transition to the modern management, please read Microsofts documentation here.