Jump to content
Welcome to our new Citrix community!
  • 2

Session States are not updated after resume


Sascha Rodszies

Question

We observe the problem that sessions do not update their status. It is easy for us to reproduce.

1) A user connects to a desktop for the first time.
   In Studio, the Session State is correctly displayed as "Active".
2) The user disconnects the session.
   Studio Session State is correctly "Disconnected".
3) The user resumes the session.
   In Studio, the Session State is obviously wrong displayed as "Disconnected"!
   The user works as normal.
4) The user disconnects the session again.
   The Session has disappeared in the Studio, but in fact it is disconnected.
5) The user would like to resume his work.
   A new session is brokered, see step 1.

After the second resume we see User Connection Failures in Citrix Director. Failure Type is "Client Connection Failure" and Failure Reason is "Other".
Users get a new session on a random RDSH, but the old sessions are still in use on other RDSH, albeit disconnected.
Using FSLogix for profile management, users can no longer log in (virtual disk locked), but fortunately we use Citrix Profile Management.

 

  • Windows Server 2022 RDSH with CVAD 2203 under Provisioning 2203 with Nvidia vGPU 14.0
  • Windows Server 2019 with CVAD 2109 shows similar behavior
  • CVAD database is on a Microsoft SQL Server 2019 Cluster
  • Latest Workspace app for Windows and latest HTML5 Receiver
Eventlog:
Level: Warning, Source: TdIca, Event ID: 1019
The Citrix TDICA Transport Driver connection from x:x to port 2598 received an invalid packet during its SSL handshake phase.

Communication between VDAs and Delivery Controllers looks good on the network side, but we can provide Wireshark dumps if desired.

 

Has anyone observed similar problems with CVAD 2203?
The suggestions in the discussions so far have unfortunately not produced any useful results. We are grateful for ideas.

Link to comment

5 answers to this question

Recommended Posts

  • 0

Neither the DDCs nor the VDAs caused our problem.

 

We have solved our issue that the session status, after a reconnect, was no longer updated (changed not back from disconnected to active as it should).
Our cause was a Group Policy.

User Configuration > Administrative Templates > System > Ctrl+Alt+Del Options
Policy: Remove Lock Computer
Setting: Enabled

We have set the GPO to 'Disabled' on the VDAs and now everything works again.

 

Maybe this will help one of you.

Our - successfully closed - Support Case was 81177826

Associated articel: CTX330692 (Reconnected session shows as disconnected on Cloud Studio and Monitor in multisession OS)

 

Greetings, Sascha

  • Like 1
Link to comment
  • 0

I have recently updated the VDA agent my Windows Server 2019 RDSH golden image from CVAD 1912LTSR to 2112 & later 2203 LTSR but the storefront+DDC server was only upgraded to 2112 and not yet 2203. For the windows clients this works fine in production as far as I can see using CWA 2112 and later (starting to roll out 2204.1 as we speak). For the MAC users however it's nothing but trouble but that's another story.

 

Next to that productional environment I have an exact copy of everything running but there (since this environment is QA) the Storefront+DDC server was upgraded to CVAD 2203 LTSR already for evaluation purposes and there I stared to see some weird stuff like copy-paste not working etc.

 

My first hunch goes towards problems in Delivery controller version 2203 for what it's worth. I highly recommend rerolling or setting up a complete QA environment like me to test out the differences of each component without productional impact.

 

Windows Server 2022 RDSH with CVAD 2203 under Provisioning 2203 with Nvidia vGPU 14.0 => I doubt PVS or vGPU affects or causes it. Even CVAD 2203 not as likely as DDC since DDC does the brokering and this sounds like a brokering issue to me.

 

Windows Server 2019 with CVAD 2109 shows similar behavior => make perfect sense since none of that affects the brokering part

CVAD database is on a Microsoft SQL Server 2019 Cluster => not relevant I think because of the same reason again

Latest Workspace app for Windows and latest HTML5 Receiver => I suppose that's 2204.1 which I'm about to roll out so if issues start happening then I'll know it .

 

The Citrix TDICA Transport Driver connection from x:x to port 2598 received an invalid packet during its SSL handshake phase. => I have thousands of these per day gathered in my monitoring suite. I doubt those are the cause since they appear with perfectly working sessions for me

Link to comment
  • 0

Hi Rodszies.

Seems very similar to my issue here https://discussions.citrix.com/topic/416346-2203-virtual-apps-disconnection-issues-and-questions/ like you said.

We are a very small IT company so not really able to do the mirror system to do extensive testing.

We do use FSlogix and it does allow the logon to complete.  The reason we know that the session is a second session is that Group policy does not run, so no mapped Drives for the user and also they cannot open Outlook.  We use Cached Exchange mode and it says the OST is already in use, which it is in the disconnected session.

Like you say it affects Apple Macs way more than PC's.  Although I have seen it on some PC's where the user is lazy and relies on Windows Hibernation/close the lid stuff so we are trying to educate them that that is not good for our situation.  The Macs are home devices and are configured as Mac intended so they go to sleep after 2 minutes, so we have asked the users to extend to over 15 minutes and we are having significantly less help desk calls to reset both sessions.

I look forward to some potential relief from Citrix on this issue

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...