World of EUC

I remeber that i have seen something from <@ULJM9DQF2> on twitter regarding this :wink:

If memory serves me correctly this can be done via the WFAPI SDK using WFQuerySessionInformation

Also the Citrix Powershell snapins can be installed on the Server VDA machine and you can query the broker for the information too but this isnt a local query

Try my recently updated script function Get-WTSSessionInformation that calls the WTS API which is based on the good old WFAPI and let me know how you get on.
<https://github.com/guyrleech/Microsoft/blob/master/WTSApi.ps1>

I knew you can offer something Guy :grin:

Lots of cool information, which will help me someday for sure, but unfortunately the function doesn’t provide any information about the Citrix client type.  I’m starting to think that this is not possible…

How does director do it? Get’s it from the DDC?

To be honest, I’m not sure Director even knows the type of client. 

I think it does. A few years ago I got the connected Receiver/CWA data at a customer via an ODATA connection to the Director.

It knows the version number, you’ll need your own logic to derive the client from that 

It would be a lot more useful if it did it for you though 

Yes, that’s what I mean. It only knows the version number. 

There’s a controlup script that does it, let me see what it’s doing 

<https://www.citrix.com/mobilitysdk/docs/clientdetection.html|https://www.citrix.com/mobilitysdk/docs/clientdetection.html>

Isn’t that what Guy is doing with the wfapi though?

I think you would need to build that code into Guys script that takes the client product ID and converts it to the Platform 

YES! With Get-WTSSessionInformation I get the ClientProductID!

<https://www.controlup.com/script-library-posts/show-citrix-receiver-os-platform/|https://www.controlup.com/script-library-posts/show-citrix-receiver-os-platform/>

That should do it, thanks a lot for finding this!

The logic is there at the end of the script, you could reuse that

No problems it’s something that has been bugging me too

I figured someone must’ve worked it out 

Make sure to fork guy’s script or get him to add the code to GitHub once you get it all working 

<@ULJM9DQF2>:  I was wrong! Your script gives me exactly the information I need. It’s the ClientProductId. With the link Mike posted I can translate it or use the registry key. Great, so I can go on with my script, maybe I will publish it later. It’s a request from a customer.

I have 20H2 in production for some time already on AHV - no issues so far. Backend is CVAD 2009, FSLogix Profile Containers, no CC.

Ah and AOS 5.19 but this shouldn’t matter I guess

Are your clients internal, external, or both? Do you have MRA already up and running? Do you currently use Jabber?

I do not believe they have MRA set up. But external and they are currently using jabber.

If your users will be external, and using Jabber from within Citrix, then you'll need MRA. Those have specific ports required by Cisco to be open on the firewall. Outside of that, we did not open any ports on or to our VDAs. We use it for about 100 agents across various departments. Seems to work really well!

Also using Jabber and it seems to be working well. I know <@ULNE983N3> did some magic for a remote call center by using EDT for UDP to improve call quality. It made sense for his environment for where I work it has been fine. We are running it on virtual machines access via Thin Clients and even some connecting into RemotePCs. Have used it no issue from my own devices when connecting in too. The VDI client works pretty well as long as it's there and running.

<@ULK5ZS52L> are your users internal or external? If they're external, i assume you already had MRA enabled as Tim mentioned. I did ask, and they do have MRA setup and running so we're good on that front. Just need to get the Wyse terminals updated and JVDI Agent/Client standardized.

Hi Gents, on acall right now, but SIP/RTP sources from the end-point, so those ports do need to be open from the end-point

I think so <@ULNE983N3>. At least the MRA ports for sure to the endpoint/terminal.

Yup - We wen't down the road of configuring this for a large Customer but they ended up backing out, and decided to just do HDX real-time /w EDT + UDP Audio via gateway, got pretty solid results. Of course no offloading so higher CPU on VDAs. I wish they stuck with the offload engine

We do the offloading, and our users love it. IGEL TCs as endpoints, just click the box to enable, and away we go! We had CDW config the Cisco side, they did us right on that regard, and I cannot recommend them enough. Highly satisfied with our results.

Internal and external. Mostly external now. We also deploy IGEL

Do you have FSLogix in the environment? You can redirect with that...

We had FSLogix in the past. Suffice it to say that it's not coming back. We had to give clients too many credits because of the outages we suffered with it before it was taken over by Microsoft. CEO and CTO won't allow it back.

I'm not convinced that FSLogix would roam them any better anyway, since they can't be written to VHDX files, unless FSLogix doesn't something differently than when we had it in house previously.

I really need to try to figure out what that error means and how to let UPM roam the logs. (I would prefer to just not roam the logs myself, but I have been told that is not an option either as the vendors require the logs. I disagree, since they can get the log information when the users are logged in..it just looks odd when there is very old information from the users first login and then the current session login. That part isn't technical, it's more of a management issue and the vendors kind of dictating what data we keep for them to maintain support. It's rather frustrating.)

I'm not saying to necessarily use it for profiles - you can use it for the app masking part. Might be worth a look.

Are they just looking to keep the logs for historical troubleshooting purposes or for active use?  Could you do something with WEM at logoff to copy the files to a share for reference?

Yeah, don't use the VHD part, just use the Redirection engine as Tim says

Interesting thought. I will look into it since I'm not aware of that piece of FSLogix, obviously. Appreciate the tips.

As far as the logs, one vendor is being kind of weird about them having missing data, but I think that's just because they don't understand the issue with roaming the files that I discovered when I was brought in to assist with the issues. They are looking for errors to troubleshoot the issues, but because we are losing data at logoff, gathering the needed information has been difficult and they were pointing at the logs being part of the issue. I *may* have moved them past that part of it now. Waiting for the vendor front line support to hear back on my feedback from their developers.

what part of localappdata are they writing to? maybe you can set up a variable in your system path to have them write somewhere else? If it’s writing to temp for example you can just change the path out of localappdata

Remove the machines from the delivery group, add to the new delivery group, and publish it.

Sorry, I didnt explain the requirement properly. DG1 and DG2 stays, also the machines in them. Published Desktop from DG1 should be moved to DG2 without deleting and recreating. Like we point applications from one DG to another.

Moving a desktop from one DG to another isn't actually deleting or recreating the VM.... That should just be tied to the machine catalog.  The DG really is just an assignment which you use to give access to the users, assign policies, etc so you should be able to move those around with minimal work like <@UK6S90JDR> described. I may be still misunderstanding what you're meaning though

Yep just create a new published desktop on the new delivery group. Not sure if you can move without doing that.

we also had the problem and it was caused in the logoff phase of a user. We were able to isolate the problem and it was caused by the AV in our case.

Hrmmm it wouldn’t surprise me with AV, I’ll ask the guys to go digging and remove whatever is there - thanks for replying 

<@ULABCUGTV> nice procmon trace above.... <@U015G4SG5LY> Thank you sir :slightly_smiling_face:

I have the same issue at a customer site. Citrix Ticket is still open - in our case it is related to the Trendmicro AV. First attempt was to update to “private fix” VDA2009 wich did not resolve our issue. -&gt; Citrix is still developing this case. We’ve received a list of dll to exclude from scanning. Hoping this resolvs our issue. To be honest i’m not that optimistic…

We had a Server 2016 / 7.15 LTSR VDA exhibit similar gray screen issues yesterday - may not be the same but symptoms sound eerily similar even though it's a different OS and VDA version.

James we had the same issue when going to 1912. It was a CTX API hook that we had for a electron base applicaiton. Visual Code was the bad guy in our case.

We was manipulating the hooks registry key via group policy and it messed us up causing these same things.  In our case, it was for Edge Chromium.  Removed the hooks change and it all started working

Thanks fellas, no bad hooks here, tis first point of fun for grey screens. Trend micro is in the mix 

closing the loop on this one lads - it was trend AV causing it - thank you very much :slightly_smiling_face:

Thanks for the update. Trend has bitten me before. I know that feeling man.

If you remove UPM from the equation, does it still happen?

Logging on with UPM disabled or with a local account on the server populates the WinX\Group1 folder with the "Desktop" shortcut, but the Group2 folder is empty.  Right Click works but only has show desktop and log off.  It stays like that after multiple logons, it never populates the Group2 folder.  File explorer doesn't close on the first logon, (you click the X button to close the window, which fades slightly but doesn't close.  Clicking on the window activates it again).  On subsequent logons it works correctly.

In summary with UPM enabled right click on the Start Menu is broken on first logon but repairs itself on the next logon.  With UPM disabled, right click is semi-broken and never repairs.

image.png

I had this happen for a client just this week ! I ended up providing feedback to Citrix on the related KB of <https://support.citrix.com/article/CTX224551|https://support.citrix.com/article/CTX224551>, as it’s lacking some key details. The fix in the end , was to start the wizard to add a new connection , choosing to TRUST the new cert on the creation wizard , then cancel. The existing connection then magically started working 

when you do get-childitem within XDHyp:\Connections whats the sslthumbprint saying ?

<https://blog.misjageuskens.nl/2020/03/25/citrix-studio-cannot-connect-to-vcenter-server-due-to-certificate-error/>

Like Owen said, I had the best luck with adding a new hosting connection to the existing vCenter. No PS tinkering needed.

<@UMH935X7H> I tried, I even have a working connection pointing to the same URL, I still have the same error.
<@ULLAD12P2> I changed the SSLThumbprint, still have the following error (cert is in trusted publisher in local machine store)

Ok, really strange thing here. I removed the working connection, restarted both controlers, and while I was doing some screenshots for an article, I added the connection and clicked cancel button as suggested by <@UMH935X7H> because my "Screenshot tool of choice was not properly configured."
Then, "what if..." I tried to exit maintenance mode on the previously-non-working connection, it worked like a charm.
I restarted both Controler, issue is here again!

Hey Steven! That's odd, so your issue returned after rebooting the controllers? As per John's above post, what do you get when you run the following ?

*Get-Childitem xdhyp:\Connecitons | Select Name, SSLThumbprint*

I got this :
SSLThumbprint matches the certificate installed in local machine store

Can you try turning off one of the controllers and running the tests again? Maybe the issue is isolated to one or the other? The above command just shows what's the SQL DB for the site, but the certs are technically per machine, so maybe one of them is having an issue? Otherwise, you'll probably want to open a case to Citrix

if even one of the machines has the cert issue you’ll have problems

is your backend 1912 now? I had this issue with pre cu3 7.15

I have installed the certificate on both :disappointed:
Backend is 1912cu2...

I raised an incident to Citrix Support, we’ll see. I keep you updated!

We finally found the issue. When I changed the SSLThumbprint property of the connection, I copied/pasted the value from the certificate and it was in lower case. Citrix requires upper case!
I missed it in <https://support.citrix.com/article/CTX224551> "** *Note*: The Certificate Thumbprint has to be written in Capital Letters"
Fun fact, my working connection in 7.15 does not have any value under SSLThumbprint property...

Ah! Yes, I actually wrote to Citrix to update the above KB, one of my feedback requests was related to their thumbprint examples 

Yes it does! Used it in the past defining CPU, memory and concurrent logons thresholds

This work out of the box. You only have to set the load index policy. If resource metrics are even of above the maximum counter the load index will be 10000 and sessions will be redirected to the server with the lowest load index.

Are they having issues during the session it disconnects them?

Not sure if i got you right...active sessions just get randomly disconnected. There is no other indication like specific application usage or issues.

I mean it seems to hits all kind of user groups. Not related to specific use cases or workloads.

Are all the clients get disconnected from the same VDA or at the same time?

We're seeing a similar problem, although not as widespread. Our users reconnect fairly quickly. Do you have EDT enabled? We disabled that and it seemed to help quite a bit.

EDT was disabled already in early migration stages. Too much pain in this environment!
Reconnect completely fails for us, if SR kicks in. Not 100% sure if that happens in every case, but regarding the high numbers of event log messages it seems to happen quite often.
We discussed a few ideas already. One is to set a higher value for SR timeout, because it’s way below the default setting. We want to check if it helps to re-establish the session on a higher rate. But that’s surely not the root cause here...:face_with_raised_eyebrow:

Just searched my log files. Holy crap! We've had over 300 in the last 24 hours O.O

My users aren't complaining that much, though. Odd - normally they are quite vocal (healthcare).

This message shows up at the same time: The Citrix ICA Transport Driver connection from 172.31.5.132:60492 has been closed

Not sure - will have to see if it occurs before we upgraded. Going to take a few minutes to figure that out...

I can't say. I modified my logging at the beginning of the year, after we did the update. I'm not seeing the logs either before or directly after the CU2 update.

<@UL5TD2QM9> : thank you for participating, hope you’re good! :wink: I‘ve to admit I’m not sure right now. We‘re still on recap. Which takes some time...
It’s difficult to say, because the workload was quite low before we rolled out 1912 CU2 and entered the first heavy production migration stages.
Let’s say there were 300 session with 1912 CU1 state.
Now we’re on about 1000 with 1912CU2, and much more to come! 
It’s a long term (healthcare) project where we originally started with 1903. My guess is it’s a mixture of issues where 1912CU2 takes its part. There are tons of possible issues right now, because of the long project time frame.
They‘re rolling out new NAP-Configurations on access switches, new IGEL OS versions, we got vSphere upgrades and rubbish like Palo Alto Software. We‘re definitely considering rolling back to CU1 if reported support cases increase.



<@ULHA2PS2H> What is your count for ID12 / source RPM on your VDAs?

For what time frame? I have 32 VDAs, and over the last 1 day I have 288.

that’s a lot of changes at the same time <@ULD72Q44R>. I have seen session disconnects with IGEL lately but they fixed it with a private build. But since you see it with different clients it’s obviously not your issue. I‘d give it a try and roll back to CU1 or just to test, see how 2012 compares.

<@ULHA2PS2H> Ok, thanks! 
My guess is these log records do not represent real issues. Right?

As far as I can tell, that is correct. I have not heard any complaints from these users. That's not to say there aren't problems!

<@UL5TD2QM9> „lot‘s of of changes at the same time“ is relative in a project time frame of 12 month right now :wink:
We will definitely bring that rollback in! Going forward in version will be a hard discussion. I hope Citrix client logging will bring some news next week.
Just for a therapeutic measure regarding IGEL versions ...we are forced to run OS5, OS10 and OS11 in this environment. That alone is a nightmare on it’s one. But hey! It’s the hard things you learn from, right?

<@UL5TD2QM9>: Which firmware resolved the disconnect issues? I vaguely recall seeing posts about it in IGEL’s Slack, but had not seen which firmware fixed it.

<@ULC9J5XPB> 11.04.264/11.04.270 should have the fix

Update: Discussed that issue with Patrick Matula via Twitter chat. He confirmed and already opened a citrix support call. No feedback so far. Will test and check a few things next week, and involve citrix support.
For your reference:
<https://discussions.citrix.com/topic/412245-vda-1912-cu2-random-disconnects-on-server-2016/|https://discussions.citrix.com/topic/412245-vda-1912-cu2-random-disconnects-on-server-2016/>

Can someone confirm if the "3Dconnexion Spacemouse Enterprise Kit (3DX-700058)" is working with Citrix Native USB Redirection (Win10)? I tested this about 18 months ago and the mapping did not with with the full feature set. Display was broken and only some buttons worked. It is still not listed as a supported device (CTX123569)

Maybe something to look into: <https://support.citrix.com/article/CTX296639|https://support.citrix.com/article/CTX296639>

Still get flickers, background gets lighter and darker, delays in audio and video

Not that I'm aware of no, we are partners but I believe my employer still coughs up for the licences to enable us to try stuff out.

Okay.. thanks for reply...but there should be some options for partners trail or test labs of Citrix Cloud

On my Partner account CVAD Trail is expired ... So unable to do any testing...

Not sure Citrix want anyone to trial anything easily any more partner or not. :slightly_smiling_face:

You can get 100 user "Demo/Partner Use" license of Workspace Premium Plus - but it's about $4K / year.

:smile::stuck_out_tongue_winking_eye: Partners have to to pay for DEMO/Partner Use License...

You definitely want to reach out to your partner account manager to see what's available - they've got SEs as well that should be able to help.

<@ULPS6HFAT> I tried to reach them first but no proper response but only opportunity talks...

You should have personal demo account. It's very limited, but you can show to the customer the UX and play with some published applications.
If the customer is ready for real tests including the admin interface - try to reach Citrix representative on customer behalf. I had my demo tenant deployed at the same day.

I have check on personal demo accounts which is free for certified partners but, we cannot do any changes in it...

Yes, as I told it's very limited. If the customer is happy with the UX and performance - time to schedule PoC with defined success criteria and request test tenant for it.

Hmmmm....but I want to test MFA authentication with Citrix Cloud

There are two kind: Citrix own TOTP (just works) and MFA provided by IdP (can be tested without Citrix).
No BYO RADIUS or something like that.

Ok thanks did you tested any one of them with Citrix Cloud

Yep. Citrix one and Azure MFA. In first case Citrix asks you to download any authenticator app and enroll. After entering password it will ask for passcode which you can see in your app. Azure MFA fully controlled by Azure, you'll be just redirected to Azure authentication page.

RSA cannot be integrated directly to Citrix Cloud. You need to integrate it first with your IdP (e.g. Azure, Okta or onprem Citrix Gateway) and use it as part of IdP.
BYO RADIUS tech preview was cancelled

Citrix SSO App and Microsoft Authenticator App both are similar

Often you already have some authenticator app so no need to download second one.

Are you sure BYO Radius is cancelled? Bcz I saw one tech video from Product Manager in 2020 it's mentioned in roadmap

It was cancelled this January (hope its not under NDA :grin:)

Ohhh ....that is the reason I was looking to do some testing before confirming the customer.

Yeah £3k for 100 user workspace premium plus license for partners. Seriously sticks in the throat.....

<https://docs.citrix.com/en-us/citrix-cloud/citrix-cloud-management/identity-access-management/saml-identity.html>

If I remember correctly you can't have delivery controllers on different versions within the same site.

Out of curiosity, why are you still on 1906?

I see. Honestly the only reason is that we haven't gotten to upgrade it yet - we're planning to upgrade to 1912 LTSR CU2 soon though. So we have no intention of staying on 1906 :slightly_smiling_face:

Now is probably a good time to move to 1912 CU2 then?

Yeah - we have other sites running 1912 CU2 so probably time to look at it and find the time to do an upgrade.

We have quite a lot of PVS based VDA's, i just need to find a good script to automate the VDA/Target device upgrade process - manually updating 45 images is quite time consuming, so i'm not doing that again untill i've automated it :sweat_smile:

Well, I can recommend a scripted install via MDT, <@ULJ3D468M> can probably hook you up with something :)

<http://xenapptraining.com|xenapptraining.com> with the Automation Framework, there's also a free Community Edition <https://courses.euctraining.com/>

45 vDisks without an automated procedure??? You have to change that! Updating a controller to 1912 is no big deal btw. You don’t have to update the VDA immediately. 
Maybe also think about FSLogix AppMasking to reduce the number of vDisks. 

any reason it needs to be a script? Can't use something like PDQ deploy (Even free should do it) or ConfigMgr etc to drive that?

No real reason other than I don't have those currently. I guess free version of pdq deploy might work. I was just going through the InstallVDA.bat file included with the installation to see if I could make it work.

ah - yeah i dont have a lot of faith in those scripts - lack of feedback scares me

PDQ is amazing and cheap as chips for what it gives you at scale (even free is awesome)

PDQ for the win, its only $500. Use the VDA clean up, run it a couple of times with reboot in between. Do you have the screenshot I shared <@URVNXB96C> 

You’ll still need to come up with the command line install regardless of the tool used <https://docs.citrix.com/en-us/citrix-virtual-apps-desktops/2009/install-configure/install-command.html#command-line-options-for-installing-a-vda|https://docs.citrix.com/en-us/citrix-virtual-apps-desktops/2009/install-configure/install-command.html#command-line-options-for-installing-a-vda>

Yeah this is what I've come up with:

XenDesktopVDASetup.exe /controllers MyControllers /quiet /enable_remote_assistance /virtualmachine /optimize /enable_real_time_transport /enable_hdx_ports /enable_hdx_udp_ports /includeadditional "Citrix Supportability Tools","Citrix User Profile Manager","Citrix User Profile Manager WMI Plugin" /exclude "User Personalization layer","AppDisks VDA Plug-in","Citrix Files for Outlook","Citrix Files for Windows","Citrix Telemetry Service","Citrix Personalization for App-V - VDA","Personal vDisk" /components vda

my test environment only  has a couple VDAs so I'd like to get it ironed out there. My prod is  not huge, but i don't want to run the install manually on each machine.

<https://github.com/haavarstein/Applications/blob/master/Citrix/XenDesktop/Install_VDA.ps1>

I’m lost on that script. I thought I would just be able to push the XenDesktopVDASetup.exe with all desired parameters to the server and it would update the VDA. Am I missing something

The script by itself won’t - that’s where you’d need something like PDQ Deploy

Ok yeah I downloaded the free version and tried it but must have something wrong. 

You need to run the VDA Cleanup multiple times : VDACleanupUtility.exe /unattended

The free version won’t let you sequence unfortunately, so you would need to see why your install fails and then work around that - pdq is ridiculously good for the price 

Tags are site elements for mainly brokering. Quiet sure they are not populated or cached locally on the VDA.

Sorry, that’s wrong. There is a GUID on the VDA’s registry in HKEY_LOCAL_MACHINE\SOFTWARE\CitrixVirtualDesktopAgent\State....


You can see an example here: <https://docs.citrix.com/en-us/storefront/1912-ltsr/manage-citrix-receiver-for-web-site/unified-receiver-experience.html#unified-user-experience>

I think it's supposed to be a restoration of the older "folder" view.

Hadn't realized that had gone public - but it just goes to show though, sometimes they will listen

seems i misread the original comment.  It’s not for customers.  At least its not back for us.

<https://blogs.nvidia.com/blog/2020/03/24/expanding-free-virtualization-licenses/>

does that link not do anything now? I got some that way last year

thank you so much. on the normal way there is an link with an contact ...

I think <@ULLH5JGFR> and/or <@ULC9J5XPB> did something, not AppLocker, to prevent this..

WEM can be used, but it's basically just applocker in itself. Why can't you use applocker?

<https://docs.citrix.com/en-us/workspace-environment-management/current-release/user-interface-description/security.html>

You can use FSLogix app masking to block it

Yep, I used App Masking to create a hiding rule blocking the local folder, which was preventing the user based install from running:

But didn't that result in the Teams install ended up i C:\ProgramData instead?

<@ULC9J5XPB> do you have the same for OneDrive?

Not in my testing, no. The install would actually throw an error message.

I only hide the native OneDrive shortcut, nothing else.

All OneDrive startup tasks, etc. are removed via Citrix Optimizer.

OK because we are not ready to deploy machine installer of OneDrive and users are installing the user path way, so i thought this would also be possible to hide

we are using server 2019 and the user are downloading onedrive client from <http://office.com|office.com>

Oh, I'm sure you could hide the OneDrive executables. I just don't do it because the shortcuts are all hidden from the users, and it doesn't auto launch. We have the latest OD installed in the build.

Yes OneDrive is not installed on Server 2019 by default

I've never tried this method with that form of OneDrive, but I bet it could be done. I'd launch ProcMon and see what it's writing to initially. If you hide it and it forces the install to fail, there you go.

But it should be similar to Teams i think? (hiding rule)

Maybe somebody else here will chime in later this morning when they wake up. There may be a cleaner way to handle it.

If it were me, I'd try to hide the path that OneDrive uses in the per-user installer to see if that will prevent it from actually installing.

The per-user OneDrive is installed in AppData\Local\Microsoft\OneDrive

That folder also contains user based configuration, which the per-machine OneDrive is using

Yeah just create a hiding rule for “C:\users\*\appdata\local\microsoft\onedrive”. That stops the OneDriveSetup.exe launching. Though remember to remove the rule when you’re ready to deploy the system wide installer 

<@ULL6F7DRA> In my testing it would go in program data as well. We blocked both with app masking.

I would create the user folder in the Program data\microsoft\teams\%username%

Yeah I see this. I send you later a registry key for that. I’m not online in the moment. *family time* :sweat_smile::the_horns::skin-tone-4:

Yeah it crashes when it tries to enumerate the Citrix hdx webcam and the client isn’t running the Webex vdi agent 

If they have the agent installed it talks to the webcam directly and doesn’t crash 

Something changed with webcam pass through in 2012 and Webex is not happy about it 

I wonder if it’s because Webex is 32bit?

Doesn’t seem to matter what version of workspace is passing through the webcam 

Maybe not a complete similar issue, but I have one User where Citrix Desktop on Server 2016 TS is freezing completely when the camera is being passed through. When opening Teams on the TS and check camera Desktop is also freezing. I'm having a ticket open at Citrix since this morning. We are on 1912 CU1 with latest Workspace App

This only happens with webex and only the webex app crashes, Teams works ok. It’s weird. Does it on Windows 10 too.

so it’s definitely as citrix issue, just looked at the webex crash logs and it’s  “C:\Program Files (x86)\Citrix\HDX\bin\HostMMTransport.dll” that causes the crash.

Rolling back to 1912CU2 fixed it, going to work with Citrix to try and work out what it is, suspect it has something to do with being on a 7.15 backend

<@ULLH5JGFR> may be running it on S16. I've got it in W10.

It ain’t supported on server OS to start with so that’s a bucket of giggles 

It's supported on W10, but the product is still a bucket of ~giggles~ shit, IMO.

Yea, its doest work on Server OS. I made it work, and I have video issues.

<@UKJTJD4KA> Do you have the VDI installers? It will install on 2016 OS, Just really wonky for video? You may have better luck.

Here is what they gave me.
please install the following BlueJeans app within the Citrix desktop environment:

•	<https://swdl.bluejeans.com/desktop-app/VDI/Tech%20Preview/BlueJeans.2.17.791m.msi>
	

On the thin client or windows client, install the following BlueJeans VDI Connector:

•	<https://swdl.bluejeans.com/desktop-app/VDI/Tech%20Preview/BlueJeans.VDI.Client.msi>
	

With the above two pieces of software installed, you can verify that the correct software is installed and that the VDI Connector is functioning properly by checking Menu &gt; About BlueJeans, attached is a screenshot of what you should expect to see.

FWIW, assuming that the clients are the same, the app that goes in the VDA is up to v2.17.1486m

We tried it in Server 2016, and ended up telling our staff to use it locally (even on their thin clients). Much better performance. We spent *months* trying to get it to work, and eventually just gave up.

yeah i am thinking i am potentiall chasing ghosts here - we have no idea what it is supposed to look like or how its supposed to interact - we just get a screen that looks like this when the customer launches the app (we didnt in stall)

That looks like the initial screen when it loads, and it should definitely shift to showing the meeting info. Are you blocking or restricting access to *AppData\Local\BlueJeans*? The BJ client will throw data in there when you launch the app. If you get to the point where you're actually in the meeting, go to Help and then About. To confirm that you're optimized (client has the VDI plugin installed and is functional), you should see the following:

What if you aren’t joining a meeting and just opening the app? What should that show?? Thanks by the way 

2021-02-23_6-35-09.gif

This is what a first launch looks like, and the user logs in as guest:

If they have already logged into BJ, they see this screen at launch: