World of EUC

InTune? Ivanti would need a whole load of supporting infrastructure

Hmm, haven't thought about Intune. Seems a valid way. Thanks!

Really like this tool.   Hoping they continue it.  Very handing in troubleshooting scenarios.

image.png

I have seen this when the ca is way to old and needed to be upgraded to s more modern version. 

Way to old on OS Layer? It's indeed a 2012 R2 one, not that modern :smile:

This is a tough one. Satellite connections have fairly high latency, so there will definitely be a delay in responsiveness and interactivity in the session. There are many variables in this scenario that can influence how successful this solution would be:
What is the latency of the connection?
1. What’s the bandwidth available?
2. What type of work/activities will the users be doing within the session?
3. The users’ tolerance for delay can vary from one to another, either because of their work requirements or personal preference
4. The users’ tolerance for visual/graphics quality can vary from one to another, either because of their work requirements or personal preference
5. Graphics settings. Check out this article for some guidance on graphics settings for various scenarios: <https://docs.citrix.com/en-us/tech-zone/design/design-decisions/hdx-graphics.html> 
6. Redirection/optimization features such as BCR, multimedia redirection, Teams optimization, etc.
7. OS-level optimizations such as cursor blink rate (the more blinks, the more screen updates are sent), disabling menu animations, etc. The <https://docs.citrix.com/en-us/workspace-environment-management/current-release/user-interface-description/policies-and-profiles/environmental-settings.html#sbchvd-tuning|SBC tuning section> of the WEM doc lists they key settings to look at. <https://support.citrix.com/article/CTX224676|Citrix Optimizer> also covers many Windows optimizations that are also important.
8. The user device platform. (Not all CWA platforms have the same capabilities, to mileage may vary. Windows would be recommended)
9. VDA 2112 or 2203 might be better here due to enhancements to Adaptive Throughput


With that being said, I think the only way to know for sure whether this will work for you is to set it up and test it so you can assess the performance and experience, and then you can determine if it works for your use case

and dont forget to tweak policies to adjust for WAN optimizations, there is a template for this in the policies section.

might need to experiment with a few of those policies

I had a user on Hughesnet Gen 5 and they had 12-15mpbs. I followed a lot of Miguel advice, as well as the WAN Template. I had to tune things to get it working. The average latency was 200ms+.

I will say the Thin Wire stuff can be hard to wrap your head around at times. Especially for vGPU. I have always had to do a good bit of research before I start tuning these settings. That is a really good graphics doc by tech zone.

If you take the time to read all this stuff Miguel Posted. I think it's amazing how much work Citrix has put into their products to make the thing that were impossible 10+ or so ago. Possible now. Seems like the HDX team really has their stuff together.

I used to have users on 64 and 128kb links working fine in Citrix. But with modern workloads - well let's just say I wouldn't expect Teams to be a good experience if using audio and video. As everyone says though - suck it and see

CU4 for 1912 LTSR: anyone can see WebSockeyService crash with exception code: 0x0000005 ?

The NVv4 series doesn't have an GPU Encoder, so we can't utilize it.
I think if you check Taks Manager -&gt; Performance and select the GPU you wont see Video Encode.
It's also noted here: <https://docs.microsoft.com/en-us/azure/virtual-desktop/configure-vm-gpu#configure-gpu-accelerated-frame-encoding>

This is not a Citrix limitation but a VM limitation. :slightly_smiling_face:

yeah fingers crossed for the new nVidia VM. Once they're available in Canada, I'll start testing them.

Yea this has been the case for the amd based gpus. You still get a gpu but specifically for encoding offload to GPU that feature is a no go

yeah i used ms sql always on for my XD site in my lab and used it in production enviroments many times. Even got a Youtube vid on the setup

I am doing it for the first time with AlwaysON

<https://www.youtube.com/watch?v=OLFkiLu5nO8&amp;list=PLLboZpnlvjGq7rlgzcp6zIuZ8jkgAjoDP&amp;index=1>

I saw ur blogpost - Configure a file share witness is mandatory?

it depends what your trying to do i only used a file share witness as i wasnt using shared storage. If you got shared storage and want to use failover cluster instances you have to install the clustered install of SQL

clustering especially with VMware brings a whole another bunch of issues such as clustering in a box  if using SCSI bus sharing (both nodes must run on same host hence cluster in a box). You can use RDM's to get round that limitation but then RDM's bring there own issues

Let me check YT video and if i have any query will ping here

Thanks for sharing your inputs...the video was helpful

<@ULCBDDJCR> 
<https://patrickvandenborn.blogspot.com/2018/01/microsoft-sql-and-microsoft-sql.html?m=1|https://patrickvandenborn.blogspot.com/2018/01/microsoft-sql-and-microsoft-sql.html?m=1>

We're using SQL AlwaysOn Availability Groups for both CVAD and PVS and using the file share witness. Pair it with group managed service accounts so you don't have to mess with changing the service account password periodically.

I use the Cloud Witness feature, but I wouldn't need to at this point. Originally put in place when I built the cluster across data centers, but that ended up being less that ideal in the long run and split them into two clusters and use Distributed AGs for DR across Data Centers.

FileShare Witness I can try <@U01258VQW56>

I know a bit late on this thread.  I do not use a file share witness.  I use a 3 node cluster in majority node mode.  It will give multiple copies of the DBs and tends to be a little more redundant for an always on situation.  Key to that is to never have more than sql box shutdown at a time.  SQL is installed independently on each server and not in a shared cluster.  Each has its own local storage as well

Clustering is only relied on for the listener name that way.  SQL is stand alone on each node.

Hi,
The published application (Visio) including FTA are assigned to EndPoint (Windows 10) and calling Visio files also works. the customer wants to edit the files from MS Teams with published application.
Unfortunately, MS Teams does not recognize the endpoint's FTA.
Does anyone have an idea or tip for me?

bcr.PNG

i have seen this in my environment also, but i think this is a new way the extension is working. had no time to check it really

i don't think so, cause on a different farm with the same chrome version and same addon version, the url is shown normaly.

Hi guys,

Looking to update my current EoL VDA version for Multi-user as Citrix have requested I do so to attempt to fix an issue im seeing with Linux endpoints disconnecting on Logon (event id 1049).

Looking at using a VDA north of 1912. whats everyone's take on latest LTSR VDA in terms of issues and stability?  I imagine its OK to use with 1912 CVAD.

thanks

Yes, works fine. But I suggest to configure some default settings (reg values) for the user via WEM or your preferred user environment manager. 

I would have to check again but I think I disabled the automatic update check and the transmission of anonymous data. I also enabled the radio button “Don't show any notifications”

Another thing would be to allow or prevent the automatic start of SnagIT. 

You cant use a UNC location for the snagit picture library.  So if you are planning on using a redirected location, that won't work.  I also have a problem where wifica32.exe just consumes 10% on my local endpoint when i sporadically 'draw a region for selection' with snagit inside the citrix session.  Pretty sure it's environmental though.

Did i miss this?  anyone else mention? <https://www.citrix.com/blogs/2022/04/01/modernizing-citrix-digital-workspace-solutions/>

Appflow enabled? <https://support.citrix.com/article/CTX285094|https://support.citrix.com/article/CTX285094>

Solid suggestions! I don't believe appflow is enabled on their SDX ADCs, will verify tomorrow. Thanks, <@UK6S90JDR>

App flow is not enabled, confirmed! Client will do some troubleshooting with their network team to ID any dropped connections on 2598

I usually run through this to check FW

Get-NetTCPConnection

<https://support.citrix.com/article/CTX214599>

netstat -a -n
netstat -ab

program's name that has a specific port open
netstat -aon

other tools
TCPView, Nirsoft CurrPorts, and PortQry

what we encountered and have an active CTX incident is the following.

logon to virtuall Windows10 with VDA version 2112 and 2203,
user logon stops with (error sessionstart, policies could not be loaded. Disconnect) and session gets unloaded.
Same Windows Version, only VDA is different, VM's are totally fresh staged.
As soon as, we uninstall VDA and install version earlier then 2112, logon runs through.
internal access.

RDP connection works good, with admin and "normal" user rights.

best regards
Thorsten

Yep ! That's our exact issue, <@ULK2FDTAQ>

i have a answer from Citrix, why our session gets unloaded after error pops up --&gt; Error at session start, policie could not be loaded.

Answer from Support:

• *In latest releases we break the session launch when group policy fails to process*. The error is found on 1912 traces as well but back then we did not break the session, which explain why it works.
hope this helps for further troubleshooting.
regards
Thorsten

Interesting, however, define "group policy breaking"

<@UMH935X7H>

we have still an active CTX Case, but we are getting closer, and it looks like, some one else has the same problem.

inside CDF trades, those errors are visible:
_*GpoListLoaderWorkItemDomain::GetGpoListInternal: GetGPOList result = 0 13",""*_
_CGpoListLoaderWorkItemDomain::AddComputerGpos,9,Error,"*GetGpoListInternal (Computer) FAILED - Error = 13",""*_
_CGpoListLoaderWorkItemDomain::ExecuteInternal,9,Error,"*AddComputerGpos FAILED - Error = 0x8007000D",""*_
_2022/04/21_

and it goes in direction "loopback setting" as soon as we disable loopback, logon works. after enabling loopback, logon fails.
So, at the moment it is again at CTX to figure our, to define "group policy breaking",
because this setup runs now for years without problems.

i would post the result.
regards
Thorsten

<@UMH935X7H> still under investigation, they can reproduce the issue.
if we disable "loopback" logon works again, but this can not be the solution :man-shrugging:

very curious what the cause is :question:

a Citrix implementation without loopback? yikes! :stuck_out_tongue:

Hi all

we received a working private fix, a few files and creating a new registriy setting.
logons to single user session with 2203 VDA works now.

our case number = 81100761

regards
Thorsten

Do you have to crrect video codec installed? If I remember correctly Windows Media PLayer does not support the mkv format out of the box

I dont think thats the issue, from the ICA session if i RDP to another session host and run the same file it works no issue.....im thinking its some strange HDX Setting which might be blocking the video

Ah, yes if it works via RDP then it's not a codec issue

Single or dual monitor on the client side?
Have you tried with VLC?

is it trying to do redirection because it’s windows media player? Try it in another program like VLC and see if it works

check if this is enabled: <https://docs.citrix.com/en-us/citrix-virtual-apps-desktops/multimedia/windows-media-redirection.html>

“When you’re using Windows Media Player and Remote Audio &amp; Video Extensions (RAVE) enabled inside a session, a black screen might appear.”

Ill have to tinker with the settings around Windows Media redirection, i am sure it will be something simple like that - ill confirm on RAVE as well...

Circling back to this one-  the HDX Policy which resolved this issue is the "Prohibit" Windows Media Redirection

dedicated Win10 VDI's on Azure with Citrix SF &amp; DDC onPrem. Is it possible to get faster boot times? VM's are Gen2(V2), comparing D and F series, but no difference.

Nope, I have not seen that. I have Studio running on a couple of WS2022 servers here in my lab

Are you logged in as a domain user? Was it installed while logged in as a domain user?

Haven't seen that one. Looks like domain issues.

Yes <@UL8ACTGSF> I have installed wih domain user and logged in with domain user.

Identified the issue late night, customer has applied hardening on DDC VM, enabled RC4_HMAC_SHA1. everything started working.

It would be nice if there would be an installer that does not put the term LTSR in every component. Some customers really don't like LTSR :smile:

If you are currently using a CR release, then you have "the right" to skip this release, and jump on the next CR release instead

I know what you mean though. I am not a big fan of LTSR either, from a technical perspective I haven't really had that many issue with the LTSR releases. However the current pace of new features in getting released the LTSR release is fairly easy an "old" product/release.

This is how it's always been with LTSR releases. It's a combined CR/LTSR release. Next release being another CR

My Citrix Presales tech was telling me there were two versions of 2203 on Tuesday. 

First I've seen registration groups limited to a site - I know where you are playing and I am 99% sure I have registration groups in use for some of their standalone app servers for dr failover between sites - look at the gpos, should be an example in there under preferences… I reserve that 1% coz it's been a while since I built it 

 You'dbe better off doing this with GPP and editing the reg keys for list of DDCs/ListOfSIDs

That's exactly what I did above :arrow_up: 

Easier: use dns aliases as DDCs, in case of a failure, update the dns records

You can even write a script to detect a site is down and update the record accordingly 

Thanks for the feedback lads.. got enough to go make summit fly now :) 

Screen Shot 2022-04-08 at 11.45.32 AM.png

Screen Shot 2022-04-08 at 1.07.12 PM.png

By default, profile folders in our environment creates two folders for each user, in the form of username.domain and when the user logs off both folders are removed.

When we enable local cache for profile container it changes to a more standard looking structure of a single folder, but that folder does not get removed on logoff.

hi Paul

yes, we have the same problem and are in contact with Citrix.
As you wrote, only container activated, everything looks good including startmenue, outlook and user search on Server 2019.
after enabling (local cache for container):
Folders below users look different, second logon startmenue is broken.

I have an open discussion below citrix-profilemgmt on slack.

what i see at documentation and is contradictory.
If you enable profile container, streaming is automatically disabled, at the end of the page, where "local caching is mentioned" you see that profile streaming can be activated to safe time?

best regards
Thorsten

How do you calculate the Page file and Write Cache for a vDisk?

We used Storefront API and I see in fixed issue of Storefront 2203 readme about API and incorrect parameters passed to the launch session.
I will try an update

you can also try the HKLM keys suggested by dannyboyayyo

VDA 2112 and we released some test 2019 servers. users would connect fine, after auto disconnect period (3 hours) they would try and reconnect but they start a new session

somtimes on the same server. you would see two same user names in task manager

i fixed it (i hope, hasnt happened since i applied this) by adding reg keys

Local machine - software - Citrix   created a new folder called "reconnect"

added dword key "DisableGPCalculation" with 1

Maybe your customer should upgrade XenDesktop to get away from an EOL version?

According to this article, you should implement a couple of registry fixes to fix some issues with 4K resolution in XenDesktop 7.6:
<https://support.citrix.com/article/CTX200257>

Is this option available in windows 10 as well? This article shows settings for win7 I guess...

A very old article from 2016 does indeed mention W10 support with 7.6 LTSR:
<https://www.citrix.com/blogs/2016/06/22/adding-windows-10-compatibility-to-xenapp-and-xendesktop-7-6-ltsr/>

And to answer your question, I don't know if the registry changes will work with Windows 10. I don't think I have ever seen a combination of 7.6 and W10, and on top of that I don't see 7.6 any more. 7.15 is the oldest I see at the moment, and even that is changing

Yep definitely upgrade to supported version. Start there.

Kids, you can't use your new toys unless you clean the mess first.