This is a technical limitation that Android application cannot obtain keyboard event for ALT+TAB. To push a window to foreground, one way is to use the unityTouch sidebar, open the sidebar, and in the "running apps" section, clicking on an item would push the corresponding remote window to foreground.
Perhaps Horizon Client for Chrome from Chrome Webstore could work for this ALT+TAB requirement. It is worse a try if ALT+TAB matters for you.
Problem machines is still available:
Hi experts. I'm battling this issue for a day now and maybe you can point me to the right direction to fix it.
We are deploying a new vdi image in our VMware Horizon Environment and i'm having issue when connecting via PCoIP, it connects for a few seconds and just freeze, and goes black screen. No issues when using Blast and RDP on the new image. Its not a port issue as we have a number of existing pools using PCoIP as default protocol, it only happens on this new images.
Our version of Horizon is 7.2 ..yes its old as i just inherited this environment. Upgrade will be done for sure. The images are Windows 10 Pro.
Thanks in advance for those who will help.
@AaronWei Thanks, trying the Chrome client. How do you set the resolution using the Chrome client?
There is no scaled resolution or customized resolution in Horizon Client for Chrome. Change the resolution/scale in the Chrome OS display settings may help. ChromeOS settings -> Devices -> Display.
We are using the Chromebook in a "managed guest session" mode administering it from the Google Admin Console. In managed guest sessions, the resolutions resets each time.
Managed public session is the special usage of Chromebook, if the display setting of ChromeOS would be reset, I'm afraid any kind of application data, like previous connected broker address, would also be reset. It seems to be expected behavior for managed public session. I think the way to avoid settings being reset, is to use an account to login the chromebook.
I was able to fix the problem. This is what, I did
1. Uninstalled Replica Connection Server
2. Restarted the Server
3. Open Control Panel - Program and Features
4. Uninstalled few leftover from Replica Connection Server
5. Reinstalled Replica Connection Server
6. Restarted the Server
7. Opened Replica Connection Server using URL
8. Check Event - No more issue with certificate
9. Both Standard and Replica Connection Server working fine. Installed Composer as well and added vCenter. So far everything is working fine.
Next project is to start with desktop clone.
Thanks
Ram
Here is screen shot
VMware supports setting the broker address etc using JSON settings https://docs.vmware.com/en/VMware-Horizon-Client-for-Chrome/5.4/chrome-client-installation/GUID-DEA8515C-4949-4E57-B6DE-DB95A235B68E.html Also supports setting the resolution, but I would have to have separate JSON settings for each resolution from what I see. https://docs.vmware.com/en/VMware-Horizon-Client-for-Chrome/5.4/chrome-client-installation/GUID-B8683360-93A7-4C3A-9D0E-FA0DE74CB060.html
Hi, before sealed up the master image did you run the OS Optimization Tool? Did you review the VMware Horizon view agent logs? The below links may help in further diagnosing the issue.
Ports may be open, but what about Protocol? are both TCP and UDP Allowed? The newer clients automatically switch between TCP/UDP I believe. I think you can force it on the Agent side, and maybe it's set that way on the old image?
Came across this thread after having same question. Looks like the links no longer work though.
They are all referenced in this document and the scripts are all on code.vmware.com
Modernizing VDI for a New Horizon | VMware
That particular script is here - it has been totally re-written: PersonaManagement-to-FSLogixProfileContainer-Migration - Samples - VMware {code}
Howdy all,
Facing a very odd problem with Macs, 4K screens and the Horizon Client (either 5.4.3 or 2006, I have tried both).The problem I am facing seems to be with Mac clients and scaling. As I run 4K monitors, I have the scaling on Mac set to emulate 'Looks like 3008 x 1692'. However, when there is any scaling in place on the Mac (other than Native Resolution of 3840 x 2160 I get blurry text on the screen.
Example Below:
In Microsoft RDP, the text is crisp and clear, in Horizon PCOIP and Blast, it is not (nor funnily enough - Horizon RDP). I have tried the latest client 2006 to no avail. I have also deployed the Group Policy for DPI Scaling (again to no avail).
If I remove scaling from the Mac, at native resolution 3840 x 2160, everything is crisp and clear via Horizon and scales perfectly, even if using the Apps. The moment there is any scaling set on the Mac at all (in System Preferences), I get the blocky text as though Horizon is selecting the scaled resolution of 3008 x 1692 (not, the equivalent hi-dpi).
I have followed the guides at https://docs.vmware.com/en/VMware-Horizon-Client-for-Mac/5.4/horizon-client-mac-installation/GUID-B821ECF5-D1F5-4D00-8F0C-0EDCBEC6A924.html
and
Using a High-Resolution Mac with Retina Display
However, I get no such option for 'Connection > Resolution > Full Resolution' when connecting to the Desktop via Horizon (running sessions on a Windows 2019 standard server). I have ensured I have selected 'Automatic Resolution' as well (as a condition of getting the 'Connection --> Resolution' menu, again, it does not show up in the Mac Horizon Client.
Has anyone experienced this, or know what might be the cause? I have run into little speed humps here and there, all of which I have been able to overcome (and enjoyed, hey, tis a learning process after all) but this one has me more than a little stumped and pulling my hair out.
Thank you all!
Stuart
Forgot to mention, I have since found in 2006 a possibly related bug. If you have two monitors plugged in, despite selecting single monitor, you will never get the referenced 'Connection --> Resolution' menu.
If you unplug your second monitor, it will show up, however, even with that I was still unable to get proper scaling working, still looked blocky.
#soFrustrating #hopingSomeoneCanHelp
Thanks!
Hi StuartCRyan,
Currently 'Connection --> Resolution' menu only exists when single monitor is selected. We'll improve it to support multi monitor case in the coming release.
And in current released version, high resolution only works in single monitor with retina display.
> I have the scaling on Mac set to emulate 'Looks like 3008 x 1692
How did you set this scaling?
Have you tried to restart Agent after you connected to it?
Please help to provide the steps of this issue and we'll try to reproduce and see what happend.
Thanks.
Has anyone successfully hybrid joined with Azure AD with Instant Clones?
Azure AD is not supported. Only AD DS is supported, you can still have an Azure AD connect that links AD DS with Azure AD but the AD objects need to be in the AD DS not Azure AD.
iOS 14 release today, but there's no update. Killing me
I hope the release will come today, before the GA of IOS14.
The Beta of IOS14 is out for a longer time. Can't believe that there is nor Horizon for IOS14 release so far. At least Fusion 12 was release before the GA of Big Sur.
Hopefully it will come today.
I have struggled with this for the last couple of months as well. vBritinUSA is right when they say that it isn't supported. "Technically" not supported, rather. I do have it working in my environment. The problem is consistency with the instant clone VM's being registered. Sometimes when a user logs into a floating pool, they might be assigned a VM that didn't successfully join AzureAD. This obviously causes a ton of problems with Outlook due to the fact that a proper license token cannot be acquired if the machine is not AzureAD joined. In our case, it also causes issues with Teams. By policy, we don't allow non company assets to log into anything 365 related. So if the instant clone VM is not AzureAD joined, Teams will fail to login and tell the user that "Personal devices are not permitted". Only way I have found to remedy the issue is have the user log out and back in in hopes of getting a VM that registered successfully when vCenter provisioned it.
In terms of "how" I got it to work, the key is to make sure your Golden Image VM is not already AzureAD Joined. I have always published my Golden Image snapshots to a pool while they were domain joined. As a byproduct of that, the Golden Image was also AzureAD joined. This led me to start experimenting with leaving my Golden Image non domain joined when provisioning to a pool. I found collectively, there are less problems (in my case) in general doing it this way. I know people will debate back and forth whether a non-domain joined vs domain joined Golden Image is the "right way", but in this specific instance where AzureAD comes into play, it seems to be a requirement that your Golden Image VM is non-domain joined and not already AzureAD joined.
It's also important to note that I also had to do the following:
- Created a startup task that runs "dsregcmd /join" on the Golden Image VM
- Created a logon task in DEM that runs the same command at logon of the VM (Not sure if this is actually required but I have it for good measure)
This guide directly from Microsoft regarding this topic was helpful:
Device identity and desktop virtualization - Azure Active Directory | Microsoft Docs
As previously stated, with all of this in place, I seem to get about 70% consistency with my instant clone VM's being AzureAD joined. The other 30% of the time end users call claiming "Outlook says it isn't activated" or "Teams says it can't sign in". And of course every time the culprit is always the VM not being AzureAD joined. Like most people, I search these forums all the time and have seen several instances of people claiming they've got it working perfectly in their environment. Not sure how, but I hope Microsoft comes up with a more reliable solution in the near future, as this does get tedious and tends to annoy the end users with the issues it causes...
So I'm not sure if anyone is still looking but here is what I am using in our production environment and don't have any issues. When I googled searched I landed her and a few other pages and came up with an alternative solution. I am already using a simliar method with Get-Printer/Add-Printer for persisting IP mapped printers so that's why I went this way. I am open to any feedback or questions. Don't be afraid to reach out. Ok, so here we go...
I used Dynamic Environment Manager (formerly known as UEM) to create a config file to persist the CSV file by creating a custom config file from the Config File Creation Wizard:
[IncludeFiles]
<LocalAppData>\Microsoft\Windows\defaultprinter.csv
## You can put file wherever you want to, that's just where I put it.
Then I created 2 .ps1 files
defaultprinterexport.ps1
defaultprinterimport.ps1
I put them into a directory that I created that hast other scripts I use - C:\ProgramData\vdiscripts\
I created a Logon Task with the following:
powershell.exe -ExecutionPolicy Bypass C:\ProgramData\vdiscripts\defaultprinterimport.ps1
I created a Logoff Task with the following:
powershell.exe -ExecutionPolicy Bypass C:\ProgramData\vdiscripts\defaultprinterexport.ps1
Here is what was in each powershell script:
defaultprinterexport.ps1
------------------------
Get-WmiObject -Query " SELECT * FROM Win32_Printer" | Where-Object { $_.Default -like 'True' } | Select-Object Name | Export-CSV "$env:localappdata\Microsoft\Windows\defaultprinter.csv" -NoTypeInformation
defaultprinterimport.ps1
-----------------------
$printer = Import-CSV "$env:localappdata\Microsoft\Windows\defaultprinter.csv"
(New-Object -ComObject WScript.Network).SetDefaultPrinter($printer.Name)
Hi All,
We have a challenge to join Horizon view Windows 7 Non-Persistent VDI's to Hybrid Azure AD domain.
I got to know that joining the Windows 7 Non-Persistent VDI's to Hybrid Azure AD domain will leave stale computer objects in Azure AD when the user logs off from the VDI.
Where as it is not the case in On-premises where the VDI where there are no stale computer objects are not created.
I got the information from the below link.
https://azureera.com/configure-hybrid-azure-ad-joined-with-non-persistent-vdi/
Please help me if someone has more information on my questions.
Regards,
Ajay
There is an 8.0.1 version that is in the store for iOS 14 support now.
Updated the client to 8.0.1 but still having the exact same issue (app crashes when trying to actually connect to the pool). We are running version 7.8 Horizon components server-side. Anyone else having success or continued issues with the app update?
We have exactly the same, updated the app to version 8.0.1 and crash when connecting to the pool. On the server runs the current version 2006.
Same. Just updated to v8.0.1. It just crashes when launching App Pool. Also, once you open Horizon Client and switch to Recent tab, it crashes.
I had reported the issues during beta. From what I can tell, VMware messed up and release the beta to public. They need to get their act together. This is holding up iOS 14 updates for our enterprise until its fixed.
It's fun in a BYOD environment, we have a large number of users having this issue right now.
We need to wait for the 8.02 version which is hopefully coming out soon, 8.01 is confirmed not working.
8.0.2 is live now and working in our environment!
I have this working in our environment currently but it is an absolute joke and I am looking to leverage any other solution to get around this.
The way I was able to get it 100% of the time is by using a pool of desktops that is provisioned completely from the get go, not using machines on demand.
I then force a sync or wait for a sync to occur into AAD, for us we get syncs at 00 or .30. I check AAD to ensure the devices are there and then force or wait for another sync to happen (cant figure out the science behind this second sync, but its needed in our environment for some reason)
Now I restart the machines from the horizon portal and when they come back up they will be hybrid joined in AAD and a user will receive their PRT on log in so they get auto logged into office etc etc. I have noticed that on user log out and vm refresh it doesnt lose the hybrid join status which is a positive.
My golden image is set the same as your by the sounds of it.
I really don't like this solution though, to do maintenance is a chore of a task if you want to hit that 100% joined status.
Hello:
I'm trying to find this information in the documentation and so far cannot find the answer. We are setting up a new Horizon environment and linking it to our old environment at another site using Cloud Pod Architecture. Everything is working good internally. Now we are going to set up external access using UAGs. What I'm trying to find out is if the UAGs at the new site need to communicate directly to the Connection Server at the old site. To better phrase the question:
We have Site A, and Site B, both linked with CPA.
I have UAGs in Site B in a DMZ. We want to point all users to the UAGs in Site B, for VMs running in Site A.
Do the UAGs in Site B, need to be able to reach the Connection Servers in Site A, for authentication? Or will they only need to reach Connection Servers at their site (Site B)? Currently the UAGs are pointed at the Connection Servers at Site B.
Hopefully this is a simple answer. We are just trying to get firewall rules in place for our DMZ network.
Thank you.
Hi guys,
Any development on this issue?
I'm having the same error msg...
UAG 3.9/CS 7.12
Regards
Moderator: Thread moved to the Horizon > View area.
Because both sites are in the same CPA, you can point the UAGs to either site’s Connection Servers as long as the pools are behind Global Entitlements, and your GE scope is set to all pods.
Are there desktop pools in both sites? Home sites and dedicated assignments can be used to manipulate which site the user actually connects to behind the Global Entitlement.
Thank you for the reply. At the moment, we only have desktop pools at the old Site A. At some point we will need to repoint our access URL to the Site B UAGs for users to access. Based on my understanding, we should be able to update our access URL at any time, and the users can traverse through Site B's UAGs and Connection Servers and access everything at Site A (to start with). We will be building new pools at Site B and will eventually cut them over. The overall plan is to migrate all users from the old Site A, all to Site B, pool by pool.
Also, our VMware engineer also shared this video with us which helped also, for those that may be looking at doing the same thing.
https://techzone.vmware.com/vmware?share=video1232
Thank you.
So where do I go from here? How do I open the VMWare Horizon Client GUI?
Hello averyone
I have a problem
When installing Vmware Horizon Agent on Windows 10 and wWindows Server 2019
"Register with Horizon Connection Server" dialog box Not displayed And the installation process ends without registering the "Horizon Connection Server" address.
If I do this with the command line, it will still not be registered on the Horizon Connection Serevr !!
Do you have a solution to this problem?
Are you trying to install the view agent on a physical machine? If so and the information is not requested, try installing by command this way:
VMware-viewagent-xxx-yyyyyy.exe /s /v"/qn VDM_VC_MANAGED_AGENT=0 VDM_SERVER_NAME=connection_server_name VDM_SERVER_USERNAME=admin user name VDM_SERVER_PASSWORD=admin_password"
A VMware Horizon Client on ARM setup is explicitly listed in the 2006 releasenotes in deployment instructions ("CPU instruction set architecture, i386, x64, or ARM"). The tar with the 2006-8.0.0-16522670.armhf binaries must be processed accordingly to the description. Horizon Client 64bit is supported on 64bit Ubuntu 20.04. In releasenotes in 'building dependendies' there are additional hints for building AND running horizon client. Haven't tinkered around with it yet, but VMware ARM apps have more support than some months ago.
Had a couple pools running out of space on the Local C drive so i redid the master VM. The question i have is concerning the overhead that there is when creating VM's in the pool. Currently on my new master i have a 50GB partition and there is 29GB of free space. However when i compose the desktops in the pools the desktop now shows only 5GB of free space. This is going to be an issue and would like to find out what is going on. This master is Optimized and i believe configured correctly. Any help or info is appreciated. Thanks
Hi there
Is it allowed to connect to the connection server admin console via UAG?
I have a unified URL whether I am remote or on orem. If remote it the requests goes to the UAG loadbalanced VIP and then hits the internal connection servers.
On-prem everything goes directly to connection servers. There is no issue using the admin console.
When I am off-site I can connect fine to the portal client via UAG but not admin console.
I am trying to figure out where issue lies and I wonder if the UAG block access the horizon connection admin console somehow? I could understand if they did from a security point of view but just trying to understand and cannot find anything on this
Cheers
Hey AllBlack,
So basically UAG has another admin console and is not the same one as the Connection Server and the purpose of the UAG is to proxy the connections and made them secure when connecting to the VDI.
Also if you are connecting over the WAN you are probably doing a DNAT to the private IP of the UAG and not to to the Connection Server IP, however the Connection Server is for administering the platform so this should never be published to the internet.
There is no issue here, everything is working as it is supposed to do.
But apparently, this is not how you would like it to work.
;-)
The UAG is an appliance built to allow for secure remote connections to (among others) VMware Horizon, within the secure part this also means to not allow access to /admin of the Connection Server(s).
If this is something you would like to achieve, which is something I would not recommend, you could add a proxy pattern on your UAG’s.
You can add one under Horizon Settings > Proxy Pattern. On my fresh version 2006 UAG this looks like:
(/|/view-client(.*)|/portal(.*)|/appblast(.*))
But if you want the admin portal enabled externally you could change it to something like:
(/|/view-client(.*)|/portal(.*)|/appblast(.*)|/admin(.*))
Again, not recommending this!
We´re currently running horizon client under Dell Optiplex 3XXX series, the headset in use is Jabra Biz 2300 USB and there more than 900 users using this technology.
The context is, users are making calls on this environment and during the day they are experiencing issues with the microphone that is not working, in order to fix this, we either restart the drivers in Windows within Dizzion or reconnect to the horizon client itself. The instructions that users are following is selecting the prefered microphone driver and while connected to the client are enabling the options Automatically Connect at Startup and Automatically Connect when Inserted.
Please let me know if we need to follow up any other recommended steps in order to fix it.
Regards
The 8.0.5 iOS Horizon Client update ostensibly adds support for ”any external mouse” for iPadOS 13.4, which I presume also means the Magic Keyboard Trackpad. However, while you can click on items on a desktop, the cursor disappears on the screen.
It seems like the mouse support is not functioning correctly.
Well, I figured out how you had to turn off Windows Native Touch Gestures. Not super intuitive, but works great once it is off.
Thanks for contacting VMware. We'd like to know a few more details to help diagnose the issue. Would you please kindly provide the following information?
1) Horizon client OS - windows or linux?
2) Horizon client & agent & server versions
3) Are you using USB redirection for these headset? Because if you select Automatically Connect at Startup and Automatically Connect when Inserted options, the USB headsets will be auto-redirected into remote desktop via USB redirection solution which is not the recommended way for headsets. We recommend you to NOT select the Automatically Connect at Startup and Automatically Connect when Inserted options and NOT redirect the headsets devices. Please use the RTAV solutions for headsets devices. You can refer to this document for details.
Configuring Real-Time Audio-Video
Real-Time Audio-Video (RTAV) for Horizon View | VMware End-User Computing Blog
We are using the 2-Factor Authentication from Google in Horizon View for all our external Horizon users. Therefore we have setup a CentOS VM with the Google Authenticator package installed. The Guide for this can be found here: https://blogs.vmware.com/consulting/files/2015/02/VMW_15Q1_TD_Horizon-View-Google-Authenticator_021715_FINAL_EMonjoin.pdf
Because Sysadmins are a bit lazy and we also do not want to generate every QR-code by hand for all our Users, we have come up with this fully automated Linux script.
A cronjob runs this script every 5 minutes. If a new user and HomeDir is created, the script automaticly creates the QR-code for this user and places a textfile with the URL for the QR-code in the Users Homedir. It works like a charm and saves us a lot of time.
Feel free to use this script in your own environment, but please leave the header info as it is
#!/bin/bash # This script is created by John at RTV-Noord. You may use and distribute # it as long as you keep this header as it is. # # THE USAGE OF THIS SCRIPT IS AT YOUR OWN RISK! # # There are some prerequisites to make this script work: # - machine you run it from is a Active Directory Domain member # - the share with AD homedirs is mounted on /mnt/home # - the google-authenticator package is installed # - replace domainname.com with your AD domainname ls /mnt/home > all_users MAKEQR=$(comm -3 all_users users_with_qr | wc -l) if [ $MAKEQR -gt 0 ]; then echo "Found users without QR." comm -3 all_users users_with_qr > users_without_qr for user in $(cat users_without_qr); do echo " $(date) make qr for $user..." su -l "$user@domainname.com" -c "google-authenticator -tdf -r 3 -R 30 -w 17 -Q UTF8 | grep http > google-authenticator.txt" mv /home/$user@domainname.com/google-authenticator.txt /mnt/home/$user/ done cat all_users > users_with_qr rm users_without_qr echo echo "$MAKEQR QR-code has been created" fi
