Hi

From time to time we have BSOD on Session Host servers. Users which had sessions on these server can't reconect to another Session Host server, because Connection Broker redirects them to inaccessible server. It lasts about 30 minutes.

Is some way to "inform" Connection Broker that Session Host is unavailable now? Or reduce time out?

Hi all,

I wonder if you could help me. I have deployed a new RDS server and everything seem to be in place. Its based on Windows Server 2016, and all licences applied, and certificate added.

Internally I can access it through the browser fine, but not externally (outside our LAN). Also everytime i try to add the server to Access RemoteApp and desktops, internally or externally, i get the message "Your credentials did not work".

This is a big project that i am working on and i find my self stuck.

So far i have tried:

• GPOs to add TERMSRV/* and thumbprint
• Bypass RD Gateway credentials for local addresses
• SSL Certificate for RD Connection Broker Enable SSO, RD connection broker Publishing, RD Web Access and RD Gateway is the same.
• Checked IIS for redirections and SSL (all off, tried on too)...all i could think of, but no luck.
• Authentication policy is set via GPO as: allow delegating default credentials with NTLM-only server authentication and Allow delegating saved credentials with NTLM-only server authentication. 

I am struggling to find a reason why it's not working.

Any help is much appreciated.

Hi,

have a hard time right now. My setup is a Windows Server 2016 on a HP DL380p Gen8 Server with a Nvidia Quadro K4000 GPU. My goal was to create a VDI with user based Windows 10 Enterprise desktops my users can connect to with RemoteFX enabled.

Unfortunately I cant get it to work. As soon as I enable the RemoteFX GPU within the Hyper-V settings of any of the VM's I cant connect via RDP from another Windows 10 with MSTSC anymore.

- I get the user/password promt and enter both. (I now the authentication works because if I enter wrong credentials I get an password error)

- I see the "Remote Connection is being secured..." message and a second later the error message: "Your Remote Session has ended, possibly for one of these reasons:. The Administrator ended, the session Network Error...

BUT:

If I RDP into the Hyper-V Server (RDVSH) and open the Hyper-V Manager I can click on the same machines console and login - with RemoteFX enabled (as I can see within device manager "Microsoft RemoteFX Graphics Adapter WDDM...")

If I remove the GPU using Hyper-V Manager I can connect from the Windows 10 client with MSTSC to my VM Windows 10 just fine.

Other things I checked:

- Using Windows Server 2019 (same effect)

- clean non-domain windows 10 machine (same effect)

- MSTSC with the 10 MBit LAN option and not automatic detection (same effect)

We have 2 connection broker servers and few RDS session servers in the environment managed with VHD. whenever we try to remote in,its asking to login several times even though we enter the correct password. After the third time,users are getting authenticated and was able to login with the same user name and password.

Any solutions for this?

Configuration:

Hyper-V - Server 2016 Desktop Virtualization Host will all available updates as of 09/12/2019
     -nVidia Tesla M10 with newest Graphics Drivers available

VDI Desktop - Windows 10 1903 with all available updates installed as of 09/12/2019
     -RemoteFX Adapter added to VM with 2x monitors with 512MB

RDP Client Endpoint - Win10 1803, Win10 1809, Win10 1903 all with all available updates as of 09/12/2019

Scenario 1: RDP from client to the VDI through RDCB or directly: Fails to connect with administrator message

Scenario 2: Connect to VM through Hyper-V Console with RemoteFX enabled - Works flawlessly

Scenario 3: Enable the RDP XDDM graphics policy - Connection works over RDP but it appears RemoteFX isn't functioning fully as graphics are sluggish. This should not be the fix as the nVidia RemoteFX drivers should be fully WDDM compatible.

Scenario 4: Remove RemoteFX Adapter - Connection works over RDP but is sluggish due to no RemoteFX

This person is having the same problem but the fixes are not working for me and since he did not respond, I'm not sure the fixes are actually working - https://social.technet.microsoft.com/Forums/office/en-US/25fa7511-c21d-44e7-9c5f-b77ffdd4d2ac/remotefx-on-hyperv-console-but-not-through-rdp?forum=winserverTS

Hi All,

I am hoping someone is able to help or point me in the right direction. My google foo powers have let me down.

I am receiving the error:
RD Licensing Manager
Unable to perform operation: Access denied.

The overview is:
I have two domains "Adomain.local" & "Bdomain.local" (I know .local historic and not yet migrated)
Both domains have been joined and AD trust relationships configured and working.

Adomain.local has our terminal server licences installed. We have purchaed additional licences and added these to the rds server at Adomain.

When accessing RD Licensing Manager on Bdomain and attempting to add the license server "rds.Adomain.local" I receive the above error.

I have also tried this the other way around. ie adding rds.Bdomain.local to the Live license server rds.Adomain.local

Just incase it has a bearing:
rds.Adomain.local is Server 2016
rds.Bdomain.local is Server 2012 r2

Thanks for any help

Hello,

I haven't find any information about and I have just answered myself, but need to be sure about the answer :)

I have a RDS Farm with a Broker that hosts also UPD on dedicated virtual disk with Microsoft Dedupe Enable.

As I have migrated the Broker on a deduplicated storage, the Microsoft Dedupe is not needed anymore, so I want to disable it.

So the question is: is it possible to disable it (Windows 2012) without service disruption (considering that this is a 24/7 production server)?

I'm having trouble with this on Windows 2019.

I have thee services installed and the RDWeb/Pages/en-US/password.aspx page loads OK.

When i enter the credentials for a user and try to change the password, no matter the complexity I set for the new password, I get the following message:

"Your new password does not meet the length, complexity, or history requirements of your domain. Try choosing a different new password."

I've disabled password complexity in the GPO to be sure that's not impacting the issue, but the problem persists.

I have seen an error on the RDWeb server hosting this page, in the servers application error log, and am unsure what to do with it:

Event code: 3005
Event message: An unhandled exception has occurred.
Event time: 01/09/2019 18:02:51
Event time (UTC): 01/09/2019 17:02:51
Event ID: 9fd930d7635347de8ce161b010f23485
Event sequence: 26
Event occurrence: 2
Event detail code: 0

Application information:
    Application domain: /LM/W3SVC/1/ROOT/RDWeb/Pages-3-132118305533185037
    Trust level: Full
    Application Virtual Path: /RDWeb/Pages
    Application Path: C:\Windows\Web\RDWeb\Pages\
    Machine name: PWADMIN

Process information:
    Process ID: 5576
    Process name: w3wp.exe
    Account name: IIS APPPOOL\RDWebAccess

Exception information:
    Exception type: HttpException
    Exception message: Cannot redirect after HTTP headers have been sent.
   at System.Web.HttpResponse.Redirect(String url, Boolean endResponse, Boolean permanent)
   at Microsoft.TerminalServices.Publishing.Portal.FormAuthentication.TSFormsAuthentication.RedirectBackWithErrorValue(HttpContext objHttpContext, String strDestPage, String strErrorMessage)
   at Microsoft.TerminalServices.Publishing.Portal.FormAuthentication.TSFormsAuthentication.<OnAuthenticateRequestAsync>d__71.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.TaskAsyncHelper.EndTask(IAsyncResult ar)
   at System.Web.HttpApplication.AsyncEventExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute()
   at System.Web.HttpApplication.ExecuteStepImpl(IExecutionStep step)
   at System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously)



Request information:
    Request URL: https://<myserver address>:443/RDWeb/Pages/en-US/password.aspx
    Request path: /RDWeb/Pages/en-US/password.aspx
    User host address: 10.2.0.38
    User:  
    Is authenticated: False
    Authentication Type:  
    Thread account name: IIS APPPOOL\RDWebAccess

Thread information:
    Thread ID: 27
    Thread account name: IIS APPPOOL\RDWebAccess
    Is impersonating: False
    Stack trace:    at System.Web.HttpResponse.Redirect(String url, Boolean endResponse, Boolean permanent)
   at Microsoft.TerminalServices.Publishing.Portal.FormAuthentication.TSFormsAuthentication.RedirectBackWithErrorValue(HttpContext objHttpContext, String strDestPage, String strErrorMessage)
   at Microsoft.TerminalServices.Publishing.Portal.FormAuthentication.TSFormsAuthentication.<OnAuthenticateRequestAsync>d__71.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.ExceptionServices.ExceptionDispatchInfo.Throw()
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.TaskAsyncHelper.EndTask(IAsyncResult ar)
   at System.Web.HttpApplication.AsyncEventExecutionStep.System.Web.HttpApplication.IExecutionStep.Execute()
   at System.Web.HttpApplication.ExecuteStepImpl(IExecutionStep step)
   at System.Web.HttpApplication.ExecuteStep(IExecutionStep step, Boolean& completedSynchronously)





Can anyone assist?

Hi All,

Quick question (I think).

I'm trying to find the best way to monitor our Per-User RDS CAL usage on Server 2016 v1607.  Is there any automated way to find out how many CALs are in use so I can write a monitor to warn me when we're at, or approaching, our licensed limit?

Any help to ensure we maintain compliance is greatly appreciated.

Thanks!

Justin

Hi,

We have set up ( with GPO) session timeout for RDS users :
- IDLE session : 1 hour
- disconnected session : 1 hour
- close session when timeouts are reached

But sessions never hit timeout because each time a new user logs in or an IDLE user reconnects, all IDLE session counter are reset to 0.

I am not able to figure out what is happening...

Does anyone have an idea ?

Regards,

Arnaud

Hello

Im managing a customer infrastructure and I've been asked if it is possible to hide the RDP Connection on the RD Web Portal.

So i am searching for a solution to showcase the option "download the RDP Connection" AFTER a successful Login on the web portal. Is that possible?

Hi,

I'm kicking off an extensive Excel macro on a network machine via Remote Desktop Connection that takes roughly 50 hours to complete. The macro VB calls a DLL file to execute a chunk of code in C++ (I did not build it and am not well versed).  Running the macro in an active session works fine, but if I disconnect and return a good time later, I can tell the macro has made zero progress since when I disconnected from the session--I can tell from status bar communication in Excel.  Once I reconnect my session, the macro continues running like I'd never left and no error messages or code execution breaks messages to be found.

It gets weirder--I have 4 options of VMs to choose from, all of which have identical specs (Windows 10, 64-bit Office, 64-bit OS).  The macro will continue to run when I'm disconnected on only ONE machine. The only difference I can see on the one machine that's successfully running the macro, is that it's the least recent to have been rebooted (and therefore updated) of the lot.

I do not have local admin rights on any of the machines. Any incite would be greatly appreciated. 

I'll caveat that I'm kind of a layman in tech speak; I apologize if I chose the wrong categories to post on.

Hi,

I have recently set up a number of new Server 2016 Remote Desktop Servers. I have a timed script which runs at 6am to reboot them all, as we have some Excel add-ins which have tendency to crash and require a reboot, and a scheduled reboot means the servers don't have to be checked regularly for this.

I've always done this with the previous Remote Desktop Server (2012 R2), and never had any issues. With the new ones, most of them are fine but on two of them there is an intermittent problem (at least 1 in every 2 days) of them being inaccessible by RDP after the reboot. When I go into thm via VMWare, I can see that the Remote Desktop Connection Broker (Tssdis) service is stopped. If I manually start the service, or reboot the server again, it's then fine.

I can't replicate the problem manually - if I copy the line from the batch file which reboots all the Remote Desktop Servers and use it to reboot one of the problem ones from a command prompt, they are fine.

I've checked the event log, and on both of them there's a information item saying that the service has started after the reboot, as would be expected - but then another one saying it has stopped a few seconds later.

Anyone able to suggest anything? At the moment I'm checking these two servers before most of our staff start work, and starting the service if it's stopped - but this isn't a long-term solution!

Thanks

David

I've read a lot of documentation regarding VDI and RDS, yet I am still a little confused on some aspects.

Consider my company's scenario:

- We have a Windows Server 2019 Datacenter Edition running on a physical machine with 2 8-core processors (total of 16 cores)

- All employees have Software Assurance included with their Windows OS license on their laptops

We want:

- To host Client-based VMs (Win10 mainly) on our physical WS2019 and allow employees to remotely access their dedicated personal virtual machines

- Very rarely, we would occasionally like to host Server-based VMs (WS2012RS and WS2016 mainly) on our physical WS2019 and allow employees to access their personal server virtual machines. Our employees will not need to use this scenario often, only in unique circumstances.

- None of these VMs would be in use long-term. At max, each VM would be used for a month or two. More often, VMs will probably be needed on a week-long basis, then destroyed.

To my understanding:

- We need to apply a core-based license to our physical WS2019 Datacenter, covering all 16 cores, for the hardware to be in compliance.

- We can deploy as many Windows Server-based VMs as we would like, and can properly handle licensing VM OS's using AVMA

- Employees can connect to Server-based VMs only if they have a CAL

- We can deploy Client-based VMs, and employees can concurrently access up to 4 client virtual machines at any given time because they have SA

My questions:

- Do I have any of the information above wrong? Please correct me if I do

- Do we need to worry about licensing each individual Client-based VM? It seems ridiculous to need to provide a separate product key/activation for each VM instance when they're only needed for a short time.

- Can we deploy both Server-based and Client-based VMs like this in the same VDI environment?

- Do we need to still have the core-based license for our physical WS2019 if we only deploy Client-based VMs? 

- How do we go about purchasing and applying a core-based license for our physical WS2019?

Please let me know if there is any other critical information I may need that I have missed.

Thank you in advanced for helpful insights.

Hello,

I am attempting to setup an RDS farm with 2 servers:

2 Server 2012 R2 RDS servers hosting Broker services in HA mode (setup and seemingly working)

both servers should act as session hosts as well, but 1 fails to become part of the group.1 server generates the following error when I attempt to add it as a session host in the collection:

the property sessiondirectorylocation is configured by using group policy settings

I can add and remove the session host role at will, no issues.  and the second server can become a session host without issue in the collection.  They are identical otherwise.  I have tried unconfiguring all group policy objects and then configuring various objects to no avail.  Any help would be appreciated.

Hi,

All our win 10 1903 clients, immediately after been updated from 1803 to 1903 start repeating keystrokes in remote apps. Servers are 2012 R2. RemoteFX cannot be disabled on session hosts servers since that solution causes problems to some of the apps running. Any one facing the same problem ? Any solution?

after configuring and installing eh webclient on my deployment. when I try to connect to one of the resources I get the error..



when I check for the error in the console i get the following errors...

Can someone help me...

what to do next...

rsamayoa

Hello,

I have one RDGW and three RDSH server (rdsh1, rdsh2, and rdsh3) in a same domain and we need to purchase some license/CALs. But I have some questions about the licensing:

1. Do I need to purchase the per device CALs for the RDGW/RDSH? We already activated the Windows Server 2016 std using the open license.
2. If I purchase 50 per user CALs, can rdsh1, rdsh2, and rdsh3 share those CALs? or I need to purchase 50 CALs for each RDSH server separately? which means we need to purchase 150 per user CALs in total for 3 of RDSH servers?

Thanks.