The one time I believe I saw something like this we did a full restart of horizon,
I would start by looking at the connection server logs though, the recompose request shows up in the debug log I believe.
The one time I believe I saw something like this we did a full restart of horizon,
I would start by looking at the connection server logs though, the recompose request shows up in the debug log I believe.
I have a view 6 environment that has been operating for 3 years now (1 of which has been under my administration). Just last week, users could connect remotely (to security gateway), see their desktop pool and select it (static assignments/persistent disks). It begins to load the desktop, then immediately says server connection was lost. This environment has been stable for a long time and the certs have not expired. I have no idea why this happened and the logs on all servers involved don't show anything of value.
Firewall rules have not changed on edge FW or servers
No updates have been applied to any of the servers (MSFT or Vmware)
Servers are green in view admin page (Sec servers are red but that's due to self signed certs)
I know a lot of people point first to FW rules, I've triple checked the rules and the Windows FW's are disabled at the moment, yet the problem persists.
I get the same outcome if I try:
External >> Sec Gateway >> Connection Server
Internal >> Sec Gateway >> Connection Server
Internal >> Connection Server
I'm not extremely familiar with the view/horizon environment, but the KB's and discussions have not yielded any insight to my issue.
Is your View environment set to connect using TLS 1.0? VMWare horizon view 4.10 was released a couple weeks ago and removed support for TLS 1.0.
I copied this over from a different thread.
Re: Horizon View not working on mobile devices
What version of the Horizon Client are your endpoints using? 4.10.0 was released earlier this month which disabled support for TLS 1.0. You need to be on Horizon 6.2.1 or newer where TLS 1.0 is also disabled for the newer clients to work.
You should start prioritizing an upgrade or migration to Horizon 7.x as Horizon View 6.x goes end of general support in just over 6 months on 2019/06/19.
There is a VMware supplied workaround available.
Re: Version 4.10.0 not working on ipad
There is a per-device workaround that allows Horizon Client for iOS 4.10 to continue working with TLS 1.0:
TLSv10 = enable
- Create a new file named config.txt
- Add a new line to this file:
- Connect the iOS device to a computer that has iTunes installed
- Open iTunes and select File Sharing on connected device
- Select Horizon in Apps
- Click Add… button to add config.txt into Horizon App
- Relaunch Horizon App
The latest Horizon Clients have turned off support for TLS 1.0, an older encryption framework with multiple security vulnerabilities. If your company is on an older version of the server software, for instance Horizon View 5.X, it won’t support the newer encryption connections. There is a work-around for mobile clients that I have been sharing with companies when they reach out to me. Otherwise, you can stay on the Horizon Clients 4.9 until you can update your environment.
Thanks, Kristina
Product Line Manager
several of the users have complained that when using VMware to log in from home they cannot use the keyboard while using an IBM System i (AS400) 5250 terminal emulator.
The users are using VMware 4.10. They can connect to the network and the keyboard does work there, but when they start an IBM terminal emulator, they lose the keyboard in that window. This would be very similar to the user's question who was having trouble using PUtty and losing the keyboard.
One of our users downloaded VMware 4.3, and reports he has the functionality of his keyboard restored.
I, myself, cannot reproduce this problem from home, I am strictly a Linux users at home. On my 2 home computers with Ubuntu 16.4 LTS, using VMware 4.10, I do not have those issues with the keyboard while running IBM terminal emulation.
Help would be appreciated. Is there a configuration item in 4.10? Is there a work-around? I don't want to tell my uses that they must all downgrade.
even on windows 8.1 we faced the same issue and its randomly
Having same issue. Tried 4.3.1 and no joy. Starting to make me crazy. Even the VMware engineers are having trouble.
Did you ever resolve this? I am having a similar issue. Thanks.
what do you mean by 4.3.1 ? Horizon version ?
Hi,
I am new to Published Apps and would like to know how to hide the server C drive from view whenever I try to save a file from my published app.
Can this possibly be done? I have already used the group policy on the server to hide the c: drive and d: drive from My Computer. But it's still visible when I click the save option in the app.
Appreciate to hear from anyone who have done this before. Thank you.
Regards,
Yazid-
So I've figured out the issue, it seems to come down to how Firefox is handling the URL in the first place. This has been tested with the Horizon 4.8.x and 4.10.x clients and Firefox v64.0. Both are 64bit versions, running on Ubuntu 18.04.1
I owe a workmate lunch now
Sorry. That was a typo. I meant 7.3.1
For clarity of the issue are you explaining, is this issue with the PCoIP protocol or the BLAST protocol, or both?
Same results using both BLAST or PCoIP
On my initial connection using BLAST or PCoIP it shows USB Unavailable. If I disconnect from the desktop (not log off) and reconnect using BLAST or PCoIP again, the USB becomes available.
If I log off from the desktop and reconnect using BLAST or PCoIP it again is USB unavailable until I disconnect and reconnect
I found this regedit helped with the issue in 7.3:
reg.exe add "HKLM\Software\VMware, Inc.\VMware VDM\Agent\USB" /t REG_DWORD /v UemTimeouts /d 120 /f
I am assuming you are adding this to the master golden image and recomposing or are you using a script to add it to the desktops themselves?
I put this in the master golden image.
Going to try now without UEM and see if that is the issue first. then I will try the Regedit. Will update shortly. Takes me 20 minutes to fully recompose each time. Thanks for the advice.
Alternatively, set the pool to neither refresh nor delete on logoff. Then you can modify the existing test VM and reboot it without losing your changes. This would save you the potential time "waste" of recomposing from a snapshotted template. :-)
But I would not get the same machine each time, correct?