Björn Schläfli
-
Posts
273 -
Joined
-
Last visited
-
Days Won
15
Content Type
Forums
Articles
Labs
Videos
TechZone
Citrix Community Articles
Events
Profiles
Posts posted by Björn Schläfli
-
-
I've installed 2203 cu2 https://support.citrix.com/article/CTX338807/users-may-fail-to-reconnect-or-start-a-new-session-on-virtual-apps-and-shared-desktop which should fix the reconnect issues, but it doesn't. On server 2022 os the issue still exists.
-
I've had this issue this morning on my new set up server 2022 multi-session vda with vda 2203 cu2, where EnforceUserPolicyEvaluationSuccess, FastReconnect and DisableGPCalculation are set (including vda reboot). This issue CTX338807 is not fixed.
-
Thank you for your input.
FastReconnect isn't valid for server 2016.
EnforceUserPolicyEvaluationSuccess should be fixed with 2203 CU1 (Citrix fixed issues in edocs). I use vda 2203 CU2.
This issue existed before with 2103 and 1912. Golden image has been completely new set up.
-
Virtual Apps 2203
VDA multi-session 2203 CU2
WEM 2203
W2k16
Every now and then we have an issue with some of our vda servers. Users are unable to log off completely. Server logs a lot of 7011 events in event log (Services not responding for ....ms). Citrix Director and Studio shows the user as connected. In task-manager the affected users on this server are not visibile. Searching for this userid in process explorer shows some left handles under system context like WEM service, lsass. CPU and RAM have low load. Log off these users in task-manager, Studio or Director has no effect. Restarting WEM services, UPM service has no effect.
I'm able to enable maintenance mode and change the hidden attribute of these users to true, so they are able to logon to another server again.
The only way at the moment to fix this is to reset the virtual machine.
-
All kind of webcams didn't work with my own user profile because of the reg key HKCU:\SOFTWARE\Citrix\HdxRealTime\Policies. Webcams are recognized but not available for video meeting software or webcam test web sites. As soon as I delete the key Policies all webcams are working again. In a new user profile the reg key Policies does not exist. But it exists in several user profiles as I've checked. I cannot find any information about this key.
Generic or optimized doesn't matter.
Does anybody know what this key and it's DWORDs (Name of vda machine with value 1) are for?
-
Hi James,
thank you.
I'm currently testing fslogix 2210, as I've seen the appx improvement in their release notes but it's not helping. I think as long as Microsoft will write user reg keys into hklm it will not be possible to optimize logon duration when using pooled desktops.
-
We use Citrix upm roaming profiles. These are network printers mapped by gpo in session, no client printers.
I was able to fix it by running a logon ps script. Remove-Printer -Name ...
-
We use FSLogix as profile solution. Control Ups logon duration analyzer shows 15 - 25 seconds at logon at the phase 'load appx packages'. It happens at every logon, not just the first one.
DelayedDesktopSwitchTimeout is set to 0 (also tested 1)
We use Citrix Optimizer
We removed all possible appx packages while staging windows 10 22h2
The cause is probably that the appx packages are stored in the regkey HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\Appx\AppxAllUserStore\UserSID for each user. Of course, this SID key is missing on non-persistent virtual desktops. Therefore, the AppReadiness service thinks it's the first login by that user and installs the AppX packages each time. But how should it be possible to create this UserSID entry beforehand, since you don't know with which desktop gets assigned to which user with pooled desktops. HKLM for user keys...bad idea.
Does anybody have any helpful tips how I'm able to reduce this time? Is it possible to bypass this appx load package phase?
Disabling the service appreadiness kills the load of appx but that's no solution as you can't open start menu afterwards and a lot more issues.
-
Hi
Our infrastructure team, which is responsible for printer mapping uses group policy preferences with mode update to map network printers. That works beside every mapping logs a warning in event log saying it doesn't work but it does. Anyway. They use gpp's also to remove printers, if user is no longer a member of the security group for this printer. This is not working on Citrix. Printers are not removed, also after a few logoff and logon cycles. The only way to remove these printers is to remove them manually or in registry and relogon. On fat clients locally it works with the gpp way.
How do we have to configure the removal of network printers in Citrix sessions?
Another issue:
They moved printers to a new printer server. While the removal of the old didn't work because of the first issue, these user's had their printers twice and weren't able to print anymore. Deleting the printers manually and relogon didn't work if the user was brokered to a server vda where the wrong queues already had been mapped. I guess because of the registry entrys on the server in HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Print\Providers\Client Side Rendering Print Provider\.... I've had to run a powershell script, which removed all the registry keys from hkcu and hklm of these printers, log users off and restart the servers. This is pretty bad.
Anyone has a helpful idea for this issue?
-
Hi Ryan
try this one:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Citrix\wfshell\TWI
Create the following DWORD value ApplicationLaunchWaitTimeoutMS.
Default is 60000 (decimal / value is in ms) which is 1 minute.
Set it to 120000 and try again. You can use group policy preferences. You need to open a new session (restart isn't necessary).
more information CTX Article waittimeout
-
Hello Gian,
Carl already sent you the helpful link to my post.
Citrix has no fix until now.
Workarounds are:
Back to vda 2112 or older one
disable html5 access in StoreFront
private fix (but it's necessary to allow testsigned drivers with bcdedit)
-
Hello sjoerd,
thank you very much. the activated logging setting was the cause. Thank you.
-
We are in the process of evaluating SentinelOne Agent. I've installed this agent with parameter /vdi in the master image of my Citrix Virtual Desktops. After around 1 hour most of the time the vdiskdif.vhdx is quickly filling up. I use 4 GB RAM cache with overflow and 12 GB RAM per machine. Procmon shows a lot of writes to vdisk C:\, but I cannot find any large new files created. vdiskdif.vhdx is filled up to 15 GB after 1 day.
Does anyone have best practices for configuring Sentinel Agent for Citrix PVS? I was unable to find some deep dive information about it.
-
I've set Install locally in receiver for web settings. That's it. No bsod since then. Strange it's working for me and not for you.
-
Citrix sent me a private fix (new tdica.sys) but it's necessary to allow testsigned drivers with bcdedit. So it's unuseable for me in production.
If you need the private fix my Citrix case number is 81386211.
Workarounds are:
Back to vda 2112 and older
disable html5 access in StoreFront
-
I still use vda 2203 cu1, because it fixes the reconnect issue, where users are unable to launch apps after they disconnected their session (wfshell.exe stops unexpectedly).
-> My workaround is to disable html5 access on StoreFront.
ctx463756 is restricted for me too.
-
Citrix support engineer confirmed bsod tdica.sys is an ongoing issue in vda 2203.
-
Hi Anthony,
I've disabled HTML5 in StoreFront. The crashes stopped immediately. I assume Citrix has changed something related to html5. I was unable to reproduce it over html5 so far.
Have you html5 enabled? Try to disable it if possible.
-
Hi Andy,
thank you for your information.
I now have a guess what could be the trigger. In one site the crashes started this Monday. On Monday I allowed HTML5 access in StoreFront. In the other site html5 was already enabled but used very rarely.
I've disabled html5 a few minutes ago and hope the crashes stop now.
Do you have html5 access enabled?
-
What kind of AV software do you use Andy?
-
updated to VMware Tools 12.0.5. First server crashed 10 minutes ago. Now we have only two differences left between Andy's and my site. OS and pvs.
I have a dump now and will open a case.
-
I was able to catch the crash and it seems to be Citrix related. Blue screen with tdica.sys system thread exception.
Found another person with this issue. https://www.reddit.com/r/Citrix/comments/u4jovd/issues_with_2203_ltsr/
Since Andy has no crashes with VDA 2203 CU1 (as in my site), the difference between his environment and mine is the os (we use w2k16) and VMware Tools (we use 12.0 at the moment).
-
Hi Andy (once again ? ),
thank you for your answer.
I use PVS 2203 since june. The issue with the crashes startet two weeks ago. So it should not be related. We use VMware Tools 12.0.0. Maybe an update can help.
Do you have auto recovery (vmware HA options for the cluster) enabled? It's disabled for our cluster but the VM HA recovery does take effekt anyway.
Have a good day!
-
Our vmware support team has upgraded our esx to 7.0.3. Since then we have noticed crashes on our vdas. "the guest operating system has failed" event on esx.
PVS 2203 LTSR
Virtual Apps & Desktops 2203 LTSR
VDA 2203 LTSR CU1 (W2k16)
The affected servers are rebooted automatically.
Sometimes there is a bnistack 158 error logged in event log of the vda server.
Anyone else with these issues?
vda server 2016 users unable to log off
in XenApp 7.x
Posted
found this CTX477137 right now.