Share via

Exchange SE Management Shell & Toolbox HTTPS Connectivity Challenges (Post-Upgrade to Server 2025)

Michael Lovett 5 Reputation points
2025-07-21T20:28:14.7866667+00:00

šŸ“ Question:

I’m investigating an Exchange Management Shell and Toolbox connectivity issue following an upgrade to Exchange Server Subscription Edition (SE).

šŸ”§ Environment Details:

  • Original hybrid deployment: Exchange 2016 running on Windows Server 2016
  • Recently added: Exchange 2019 CU15 on Windows Server 2025
  • Performed in-place upgrade on the new server → now running Exchange Server SE
  • All remoting connections use Kerberos authentication

āœ… Observed Behavior:

  • I can remotely connect to the new Exchange SE server from the legacy Exchange 2016 server (or other machines with Exchange SE management tools installed) using Exchange Management Shell and Exchange Toolbox only if SSL is not enforced on the /PowerShell virtual directory.
  • When I enable RequireSSL, remote remoting fails with:
WinRM cannot determine the content type of the HTTP response

—often resulting in a 403.4 Forbidden response, indicating IIS is blocking the HTTP request expected by the client.

  • Locally, on the Exchange SE server, if I manually launch PowerShell using:
New-PSSession -ConnectionUri "https://exchangeserver.company.org/powershell/" `
  -Authentication Kerberos `
  -AllowRedirection `
  -SessionOption (New-PSSessionOption -SkipCACheck -SkipCNCheck -SkipRevocationCheck)

…the IIS logs show successful 200 OK responses over port 443, but the session hangs and never returns a usable prompt.

  • Initially I thought the Exchange Toolbox was connecting locally, but turns out it fails too. It throws the following error:
FX:{714FA079-DC14-470f-851C-B7EAAA4177C1}
Type is not resolved for member 'MonadDataAdapterInvocationException...
System.Runtime.Serialization.SerializationException

This suggests a failed deserialization attempt related to the MonadDataProvider and possibly a version mismatch after the upgrade.


🧪 Troubleshooting Done So Far:

  • Verified:
  • WinRM HTTPS listener on port 5986
  • Certificate binding via netsh http show sslcert
    • GPO settings → confirmed AllowUnencrypted = false, TrustedHosts = *.company.org
    • Ran EMTshooter.ps1, which showed:
      • HTTP fails as expected when SSL is enforced
        • HTTPS request gets redirected to /PowerShell-LiveID
          • Recommended enabling AllowRedirection, which I confirmed helps—but only if sessions are manually scripted
          • Manually tested the Exchange PowerShell snap-in: loads fine when added directly
          • Confirmed IIS /PowerShell VDir is healthy and responding
          • Checked RPS.Proxy health set → all monitors report Healthy
  • Observed multiple successful HTTPS POST /powershell requests in IIS logs, but no prompt appears in shell
  • Followed various steps from here:
    https://learn.microsoft.com/en-us/troubleshoot/exchange/administration/connecting-remote-server-failed

Can't Run the: https://microsoft.github.io/CSS-Exchange/Diagnostics/HealthChecker/


šŸ“‰ Monitoring Impact:

In addition to the shell/toolbox behavior, SCOM is generating alerts against the new Exchange SE server due to RPS health probes failing. These probes attempt to use PowerShell remoting to validate availability, but they seem to default to HTTP, which is now blocked if SSL is required.


ā“ Questions:

  1. Why does the Exchange Toolbox fail locally while attempting remote management via HTTPS? Is this related to .NET serialization or assembly mismatches from the CU15 → SE upgrade?
  2. Is there a supported or recommended way to force the Exchange Management Shell to use HTTPS with Kerberos, especially if editing RemoteExchange.ps1 isn’t viable?
  3. Has anyone seen similar behavior with Exchange SE running on Windows Server 2025 after performing an in-place upgrade?
  4. What’s the best way to enforce SSL on the /PowerShell VDir without breaking:
    • Remote Exchange Shell access
      • Local shell behavior
        • Exchange Toolbox
          • Monitoring probes like those used by SCOM RPS

Any guidance, insights, or experience with these SSL + remoting edge cases in Exchange SE would be hugely appreciated. Happy to share more config or logs if helpful!

Exchange | Exchange Server | Management
Exchange | Exchange Server | Management

The administration and maintenance of Microsoft Exchange Server to ensure secure, reliable, and efficient email and collaboration services across an organization.

0 comments No comments

5 answers

Sort by: Most helpful
  1. Simon Burbery 731 Reputation points
    2025-10-16T23:53:53.6633333+00:00

    Thanks for the fix guys! For anyone coming across this post - the issue is solved but is buried in the comments under the previous post (which we initially missed):

    Step 1: View Current WinRM Configuration

    A. Open PowerShell as Administrator

    B. Run the following command to see the full WinRM client configuration:

    winrm get winrm/config/client

    Example Output:

    Client

    NetworkDelayMs = 5000
    
    URLPrefix = wsman
    
    AllowUnencrypted = false
    
    Auth
    
        Basic = false
    
        Digest = true
    
        Kerberos = true
    
        Negotiate = true
    
        Certificate = false
    
    DefaultPorts
    
        HTTP = 5985
    
        HTTPS = 5986
    
    TrustedHosts = *
    

    šŸ”“ If TrustedHosts = *, this is causing the issue!

    Step 2: Remove TrustedHosts Setting from Group Policy (or use local policy gpedit.msc or PowerShell at #3 below)

    A. Open Group Policy Management Console (GPMC) on a domain controller

    B. Navigate to:

    Computer Configuration → Administrative Templates → Windows Components → Windows Remote Management (WinRM) → WinRM Client

    C. Locate and edit the Trusted Hosts policy:

    Double-click Trusted Hosts

    Set it to Enabled with Blank value (not *)

    Click OK

    D. Force Group Policy Update on the Exchange server:

    gpupdate /force

    Step 3: If not using GPO to set this, you can change TrustedHosts Locally on the Exchange Server

    A. Clear the TrustedHosts Setting:

    Set-Item WSMan:\localhost\Client\TrustedHosts -Value ""

    As soon as you set this the Exchange Shell works - thanks Microsoft! :)

    2 people found this answer helpful.
    0 comments No comments

  2. Wilson, Bob 15 Reputation points
    2025-09-19T15:56:31.0466667+00:00

    So, we figured out what was causing the issue. We have a GPO that sets the TrustedHosts="*" for the WinRM client. Removing this allowed EMS and Toolbox to load without issue. You would think this value would not be a problem as it is essentially telling the client that it can connect to any WinRM server but who knows. Will be working with our Domain Admins to get this fixed for all our new Exchange servers.

    2 people found this answer helpful.
    0 comments No comments

  3. Seiun 1 Reputation point
    2025-09-13T09:14:14.69+00:00

    Hi

    I've the same problem.

    In our envronment we have Exchange 2016 on Windows Server 2016.

    We added Exchange 2019 with CU15 on Windows Server 2025 with full update.

    Exchnage Management Shell and Exchange Toolbox don't work. Very frustrated.

    And soon we'll have to migrate to Exchange SE. I can already see the problems. I don't understand why Microsoft is rushing this when there are so many bugs and imperfections.

    To get local access to Exchange 2019 from powershell you have to use Exchange PowerShell snap-in. Only that is working.


  4. Michael Lovett 5 Reputation points
    2025-07-25T15:50:41.1933333+00:00

    We've worked through the checklist and even used the Upgrade option with Exchange SE to reinstall and verify that the binaries and files are intact. We suspect the issue lies with the Kerberos configuration.

    Ultimately, our goal is to retire the current Exchange 2016 Hybrid Server and transition to the new Exchange SE Hybrid Server. This new server will serve as the sole system for:

    • Account management for newly created Office 365 users
    • SMTP mail relay for copiers, notification systems, Microsoft Operations Manager, and other monitoring tools that rely on SMTP

    Right now, we're trying to determine the correct permissions and delegation settings, specifically:

    • What permissions need to be configured on the PowerShell Virtual Directory (VDir)
    • What delegation settings are required on:
      • The Exchange computer account
        • The service account used to run the Exchange Health Check PowerShell script, which generates an HTML report on server health

    Do you have detailed guidance on how Kerberos rights, delegation, and permissions should be configured for the relevant virtual directories and accounts?

    I have looked at this document, but it doesn't really match our simple single server configuration we are trying to do:

    Here are some detailed guides that walk you through setting up Kerberos authentication for Exchange Server:

    šŸ” Microsoft Official Documentation

    šŸ›  Step-by-Step Community Guides

    Configure Kerberos Authentication for Exchange Server – Ali Tajran A comprehensive walkthrough including ASA creation, SPN association, PowerShell commands, and verification steps.

    Enable Kerberos Authentication in Exchange – AventisTech Focused on Exchange 2016 and 2019, this guide includes script usage and configuration of authentication methods for Outlook and MAPI.

    How to Enable Kerberos Authentication for Accessing Exchange in a Resource Forest – Microsoft Tech Community Useful if you're working in a multi-forest or resource forest topology.

    All of these seem to indicate this ASA account creation, and I don't see or understand the need to do so for such a simplified 1 server environment when all is said and done. Can you highlight the parts of any of these that we should follow?


  5. Hin-V 14,175 Reputation points Microsoft External Staff Moderator
    2025-07-22T07:49:45.6333333+00:00

    Hi @Michael Lovett

    Thank you very much for the detailed information you’ve shared, it’s greatly appreciated and helps us better understand the situation.Ā 

    At this time, we are actively conducting a series of internal research to further investigate the issue. As this process involves multiple layers of configuration and synchronization between services, it may take a bit of time to ensure we’re covering all possible angles. Our goal is to be as thorough as possible so that we can provide you with accurate and effective support.Ā 

    We completely understand how important this is for your environment, and we truly appreciate your patience while we work through it. Please rest assured that as soon as we have any findings or updates, I will get back to you immediately, you’ll be the first to know.Ā 

    Thanks again for your cooperation and understanding. If there’s anything else you’d like to share in the meantime, feel free to reach out.Ā 


Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.