Hi everyone, I've found a bug in the current Workspace 26.3.10.96 x64 that I can reproduce in our CVAD 2507Cu1 environment. We work with full desktops, which also have Workspace installed—currently Workspace 25.5. Today, I installed the latest 64-bit 26.3 Workspace on our system, and now I’m having an issue where keyboard inputs no longer work when connecting to another desktop via a session in session, Keyboard input lost in nested ICA session. It works fine with version 25.5 or earlier. Can anyone confirm or verify this, or does anyone have a resource I can check? I'm assuming it's a bug at the moment. Operating system: OS 2022 Server, CVAD 2507 Cu1, PVS infrastructure. Thanks to everyone in the community, Mark

Is anyone able to access the login page for citrix.com right now (for downloads)? The redirect appears to be broken and I am unable to actually get to the login page at all. DaaS login page works fine, though.

2607 LTSR and 2507 LTSR CU2 are now available for download at Citrix. Docs are also live.

What's everyone's option of HDX direct? Today, we push everyone, external or internal, through a NS GW. we don't leverage workspace service or GW cloud service stuff, we have everything on-prem. Looking at HDX direct for internal users is somewhat appealing if we take the NS's out of the mix. what is everyone's experience with it?

2607 LTSR is live: citrix.com/downloads/citrix-virtual-apps-and-desktops/product-software/citrix-virtual-apps-and-desktops-alleditions…

Well found my first bug. App Assistant isn't work. Looks like it is adding a comma after the session number preventing it from loading.

Does anyone use Palo Alto Cortex XDR within their non-persistent images? If so how do you handle the impact on the write cache? I have tried creating a junction for the C:\ProgramData\Cyvera folder to the persisitent disk but that seems to eventually get removed whenever the agent performs certain actions like for example pausing protection then resuming.

So working on getting all the new 2607 items setup in the lab. Does anyone have info on what needs to be done to get Endpoint security setup? docs.citrix.com/en-us/citrix-virtual-apps-desktops/director/endpoint-security.html

I have an issue where one VDA is not accepting HDX connections over EDT, while another VDA in the same environment is working as expected. The strange thing is that the UDP listener on port 2598 is missing on the affected VDA. Has anyone seen this behavior before or knows what could prevent the EDT listener from being created? Working vs Nonworking Screenshot:

What's your experience with VDA 2507 CU2 in terms of stability?

Is there any way to configure the LAS service to send notifications when the activation expiration drops below XX days? I want to avoid being caught off guard by an expired license just because the uploads haven't been executed for 30 days. Essentially, we are looking for a way to monitor the LAS service status

Hey friends have a strange question. How do you alert on hosting / hypervisor issues like password or secret expiring or just connection being down? Within DaaS when I go to monitor and alerts I do not see anything hypervisor related at all. Any way to alert on these issues natively?

Heads-up: Citrix DB migration per CTX280675 — don't forget the VDA Update Service (VuS) When moving the Citrix databases to a new SQL server, every service (Connection String) has to be repointed individually. The VDA Update Service (VdaUpdateService, cmdlet prefix Vus) is easy to miss because CTX280675 doesn't list it and was never updated afterwards— the article covers 16 Set-*DBConnection commands, and Set-VusDBConnection isn't one of them. Symptom: Studio (both Web Studio and the MMC console) fails to start on all Delivery Controllers with an error along the lines of "Could not find a working 'VDA Update' service." Running sessions and VDA registrations are unaffected. Fix: Run Set-VusDBConnection on every DDC. How we found it: After completing all CTX280675 tasks, neither Studio would open, while desktop logons and active sessions kept working. We asked the DBA team to bring the old SQL server back up — Studio worked again within seconds. On the old server we could see both DDCs still writing to the site DB:

sql

SELECT spid, loginame, hostname, program_name, DB_NAME(dbid) AS DatabaseName

FROM  sys.sysprocesses

WHERE program_name LIKE 'Citrix%';

There was a successor article, CTX558936, but it has been unavailable for some time as well. We still have the PowerShell script Citrix shipped with it — VuS isn't in there either, so that article would have had the same gap. Two gaps to watch for if you follow CTX280675: 1. Missing service — add Set-VusDBConnection as described above. 2. Wrong order — old SQL server shut down too early. Step 3 says to shut down the old SQL server, but the connection strings are only cleared in step 7. That asks the services to release a connection to a server that no longer exists. Set-*DBConnection ... -DBConnection $null -Force then fails with Citrix.XDPowerShell.Status.DatabaseError, and Set-LogSite -State Disabled throws a NullReferenceException because the logging service can no longer read its site object. Correct order: keep the old SQL server online until all connections on all DDCs have been cleared and repointed to the new server. Only then shut it down.

Anyone faced the following error when opening Office365/OneDrive in the past and got a permanent fix by any chance: 58tm1 or 4nsvw

anyone ever come across a citrix published app where in seamless mode it has a black border round it when you minimise it and bring it to front it just happens on this single app. When you maximise the app it is fine this is on a single monitor / native 1080p resolution / 100% dpi scaling / high dpi on CWA set to use native res and also tried let the OS scale the resolution with no joy. Also tried a different use video codec for compression policy in studio to test and was same. Added the process for the app to uviexcludes for api hooking and no joy either. breaking out of seamlessless mode shift + f2 doesnt show much and i cant tell if i am getting the black border or not. Tried with and without graphics accelleration enabled for CWA. Via Published desktop using ICA and RDP it's fine so issue specific to seamless

CWA client detection hanging on StoreFront (internal-only store, works fine on all other farms) Symptom:

GetDetectionStatus

polls forever, detection never resolves, users fall back to ICA file download instead of protocol handler launch. Confirmed via Task Manager:

WebHelper.exe

never spawns on this farm when clicking through detection, while it briefly appears on every other working farm (same client machine, same account, tested in both Edge and Chrome, identical result). So the failure is happening client-side, before `WebHelper.exe`/CWA even spawns, and it seems to be browser-agnostic. Anyone run into

receiver://

silently not launching on a specific store while others work fine?

Did somebody already successfully setup Modern Entra SSO on-prem without any issues? I know it is not officially supported for Server OS 2022, but did somebody get this to work technically? I have seen a blog post from the tech preview, where Julian Jakob got it to work with OS22: Citrix DaaS - Entra ID SSO with PRT and without FAS Haven't read anywhere that SAML + Entra SSO is not supported, but I think it wasn't in the past... do we necessarily need OAuth or does SAML also work with this combination?

Defender Onboarding / Write Cache Just a question: When you use Onboard-NonPersistentMachine.ps1 on you non-persistent VDIs, do you also get about 2GB Write Cache after onboarding is done? Before onboarding, Cachefile stays at 4MB, while onboarding it grows to approx. 2GB in a few seconds. Procmon only shows me writes to C:\programdata\Microsoft\Windows Defender\Scans

Looks like there is as issue with Windows 11 September patches and Explorer.exe crashing with Citrix and FSLogix container profiles. Multiple people reporting the issue on the Citrix sub-reddit. Edit: Added an "and" between Citrix and FSLogix.

I see numerous mentions in different forums that the September patches can cause issues with Remote Desktop Services on Win2016, 2019, 2022 and 2025. Bleepingcomputer.com has an article on it, bleepingcomputer.com/news/microsoft/september-windows-server-updates-break-remote-desktop-services. So far I have not seen anyone confirm that this also affects Citrix servers, but I would assume it does. Does anyone know for sure?

Does anyone have a solid document that explains to a technical and non technical audience why Citrix VDA session hosts need to be restarted regularly? Looking for something I can put in front of management rather than another forum thread.

Just another Writecache Thread.. i modified @Chris Twiest Control Up script a bit, to compare the actual findings with the mcsdif.vhdx. It filters out only writes on C:\ and also includes only data since the last reboot. I cannot find what causes my cache to grow, as the scripts data finds about 2943 MB, the cache size is 7707 MB. And yes, the cache initially starts after reboot with 4096kb. The sum itself is based on all entries, not only the top 200. I modified my mcs template with every tweak i found (and also did it without any tweaks). So Disabled NTFS Last access timestamps, disabled system optimize / defrag tasks etc., Updates, etc. When sth is misconfigured here it would also show up as "visible" data to c:\, so my issue is the "invisible" data. This image is running on sophos central, also this is optimized for VDI, no scheduled scans, etc.

I am doning Citrix for some years now but I never had to do a Windows OS migration of the Management Servers (DDC, SF, PVS) - there is a first time for everything! 🙂 We have two nodes per role DDC1/2, SF1/2, and PVS1/2 and I would like to keep the IP and Names of the servers. My idea was to remove the "2" servers from each service, rebuild the server with the new OS using the same Hostname and IP, reinstall the Citrix Components and rejoin it to the existing site, farm, or deployment. After some tests that they are stable enough, I would repeat that with the "1" Servers from each service. The Only "issue" I found is that SF does not support different OS Versions in the same Server-Group, but then will just export the config and import it and create a new Server-Group. Obviously there is a risk because only one server is available for some time and the rollback would probably be more complicated. Has anyone used this approach? Any recommendations? Is it valid and are there any issues or recommendations I should consider?

I see a lot of admins manually updating teams and O365 on master images, making changes to the registry and so on. They put in hours to basically re-install teams and office each month. That opens up for mistakes and changes that can damage the master image. It's 2026 and I am surprised that people are working like this instead of centralizing the configuration/installation. Do you guys also do this each month or are you scripting this / using 3rd party tools? I'm thinking SCCM, BigFix or other tools which can check the current versions of the apps on the master and update teams and office automatically for example after I boot up the master. This feels more like 2026 to me but what do I know :D

Does anybody know if Citrix is working on Powershell Modules for PS7? Or somebody get a good workaround (-UseWindowsPowershell isn't sufficient. )

Soooo how's slimcore Teams going for everyone?

Anyone seen issues with Workspace for Mac not saving MS Project mpp files to client drive? It looks like its saving but file never actually saves. The strange thing is if you save as XML it works. This is not an issue on Windows though.

Good morning all, what is the recommendation for software metering/VDI migration planning tools today? Its been a while since I've had to go through migration planning so want to be sure I'm using the right tools. Thanks!

I am looking for information on all issues and bugs in VDA 2507 CU2 multi-session.

Got a site, where we got 2 machine catalogs with different zones in a single delivery group. Basically Primary Zone is newer hardware, Secondary Zone older hardware. We want the user to first use all VMs on new hardware, before using older hardware. This works, except with power-on rules. When i set a percentage or count on the delviery group, it spans these settings over all machine catalog. So for example i got the case that the user wants to access a desktop, 2 machines are unused in primary zone but powered off, because "ready" machines reside in secondary zone. When starting the desktop he has to wait for power on and register VM. Any way to get rid of this (without setting power-on to 100%)?