Chris Marreel

Posted June 27, 2023 · June 27, 2023

do we have an example (powershell) script that queries the Monitoring database (= Citrix Director data) to list or count the failed user connections in the last (e.g) 30 minutes, and if there are more than 3 failed user connections to one specific Citrix VDA-worker, then that VDA-worker should be put into maintenance mode so a system administrator should get an e-mail or trigger and check that VDA-worker ?

Examples are welcome.

If we run that script every (e.g.) 15 minutes it can take an bad- behaving XenApp-worker 'offline' so it no longer accepts connections. All this to avoid the "black hole" issue of a bad-behaving VDA-worker that attracts all the new sessions as they don't succeed, but the session-count-number (used in the session balancing) isn't increased...

Greetings,

Chris Marreel

June 3, 2020

Hi,

I have a PVS 1912 setup that was running fine. No issues with hanging PVS-targets.

I exported my vDisk, and imported again after renaming the PVP and VHDX-files. And on this new vDisk I enabled the "Asynchronous I/O" option. That was the ONLY Change !

I assigned this new vDisk to my collection and restarted HALF of my PVS-targets.

After only 2 days I already had a few PVS-targets that were no longer responsive, and were not registered anymore. After a RESTART of these devices they become functional again.

And this morning I again had 3 of the 8 PVS-targets that were hanging...

I will switch back to my vDisk WITHOUT this option "Asynchronious I/O". I know there is a newer 1912 CU1 version of the PVS-software. There is no information mentioned in the release notes about issues fixed with that "Asynchronious I/O", but I will give it a new try...

Any other experiences ? At the moment I think this new options is not production ready yet...

April 10, 2020

Hi Maurits and others,

I had EXACTLY the same issue, and exactly the same experience with Citrix Support.

I finally changed the NIC from VMxNet3 to E1000E and my issue is solved. Of course changing the NIC-type from VMxNet3 to E1000E I had to 'reverse image' my master-image and import it again in PVS after re-installing the PVS-target software when switching the NIC-type.

But I hope, once the real underlaying issue is found and solved, to return to the VMxNet3 NIC as this network-adapter is much performant and better than the legacy E1000E Nic.

So I'm more and more convinced it must be a combination between PVS-server/PVS-target/VMware ESXi-version and maybe the VMware Tools version as we only experienced it with the VMxNET3 NIC.

My citrix-support ticket : 79445951 (started on 20 jan 2020).

I hope this helps.

Greetings,

Chris

March 13, 2020

Hi,

I'm looking to provide our application development team information how they can detect inside the application if the application is running in a CONNECTED or DISCONNECTED session.

This is for an in-house developed Windows Application for patient records inside a hospital. The medical staff uses this application to keep track of medical information. They use IMPRIVATA to connect and disconnect their Virtual Apps-session. But if they DISCONNECT their session when they are in the edit-mode of one specific person, then a record lock exists inside the application. And at the moment that lock stays in place until that end-user RE-connects to the session and continues with the edit-action.

But this situation (a long lock on a database-record) can be very annoying.

So we were thinking, if the application can detect that it is running inside a DISCONNECTED session, then it could discard the edits after e.g. 5 minutes, and releasing the record-lock so it no longers causes issues when other medical staff wants to edit that record.

Sounds good, isn't it ?

But "how" can an application-developer detect the state of a session ? And I mean inside the Windows-OS, and not by interacting with the central SQL-database.

I'm not an application programmer, but I want to get some input so I can provide this to the application development team so they can pick that up and incorporate this in their application.

Can someone shine a light on this ?

Thanks and greetings,

Chris Marreel

February 7, 2020

Hi All,

cbenali69 described the issue as:

> Just to reiterate, for us the core issue symptoms were VDAs using Vdisks via PVS were experiencing a full lock up

> and becoming unregistered with the DDC with the error message in Studio stating that the VDA should be registered

> but isnt, only a restart of the VDA would bring it back up and the issue was completely intermittent affecting all VDAs

And I have 2 setups experiencing this PVS-issue. And this is EXACTLY the same as we are experiencing.

One environment is running “VMware ESXi, 6.7.0, 13004448” -> this corresponds with “ESXi 6.7 EP 07” with a release date “03/28/2019”.

The second environment is running “VMware ESXi, 6.7.0, 15160138 ” -> this corresponds with “ESXi 6.7 P01” with a release date “12/05/2019”.

BOTH Environments had VMware DRS (Distributed Resource Scheduler ) Active.

At this moment we have added an extra DRS-rule "VM Overrides" and DISABLED DRS for the VDA-workers/PVS-targets AND the PVS-servers.

And we are again evaluating this annoying issue.

Greetings,

Chris

February 6, 2020

Hi cbenali69, Hi Chris,

Thanks for your detailed input.

In one environment the ESX-level was already upgraded to the "ESXi 6.7 P01", which is build numer "15160138", and this is already NEWER than the "ESXi 6.7 Update 3" which is build number "14320388". And even with this higher build-number we experienced this issue.

So this is very strange. Can you have a look a which build-number your environment is running now ? Just to be sure...

Besides the ESX-level upgrade you also mentioned that you switched off "DRS". And that DRS is at the moment still switched OFF in your environment.

In other words, maybe switching OFF DRS is the solution for this annoying PVS-issue...

What do you think ?

Thanks for sharing your thoughts.

Greetings,

Chris Marreel

February 4, 2020

Hi all,

I'm experiencing EXACTLY the same as cbenali69 and Lewis Barclay are expiriencing.

On saturday morning we do a weekly reboot of 5 XenApp Workers using PVS. 1 or 2 days later a few of these 5 are hanging, and I only can REBOOT them to get them back on-line.

In the eventlog I see first :

- BNIStack error, Event ID 84 : [MIoWorkerThread] Too many retries Initiate reconnect.

- BNIStack information, Event ID 155 : [IosReconnectHA] HA Reconnect in progress.

- BNIStack information, Event ID 156 : [IosReconnectHA] Invalid socket error, trying another socket.

And this last eventlog message is repaeting thousands of times...

We are running the PVS Servers and the PVS targets on VMware. Hypervisor: VMware ESXi, 6.7.0, 15160138

I have no idea exactly when this began happening, as at first we simply RESTARTED the hanging VM's without investigating. Only after a while I started to follow-up on this and at the moment I don't see a pattern and I have idea how to trigger this.

All the components on the PVS Servers and the PVS Targets are up-to-date:

- PVS Target version : 1912.0.0 LTSR

- PVS Server version : 1912.0.0 LTSR

- VMware Tools on the PVS Target : 11.0.5

I have at the moment a citrix support ticket for this issue : 79445951 (started this ticket on 20-jan-2020).

Can others share their support ticket number so we can bundle all the gathered knowlege at Citrix Support ?

Can others, who are experiencing the same, share their versions of Hypervisor ?

We are a Citrix-partner, and I think I have some other customer-setups who are experiencing exactly the same, but I haven't dived into this deeper in these environments as I don't want it to happen more often.

Thanks,

Chris Marreel

January 6, 2020

Hi Warnox,

Thanks for pointing me in the right direction. As far as I understand Citrix Receiver already works a long time via TLS 1.2, so it is seems to be a good decision to disable TLS 1.0 and 1.1 on the Virtual Server running the Gateway.

Thanks and greetings,

Chris

January 4, 2020

Hi All, a quick question related to this topic.

If you disable TLS 1.0 and 1.1, on the Netscaler Gateway Virtual Server, you mentioned you loose the possibility to connect from some (older) browsers and probably allso some (older) Citrix Receivers.

Do you have an idea which older Receiver versions will have issues ?

Is this known or listed somewhere ?

I know e.g. that disabeling SSL 3.0 had impact on Citrix Receiver versions lower than version 4.5.

I noticed SSLLABS is making it harded to score an A+ rating from january 2020. If you still have TLS 1.0 and 1.1 enabled the rating will be downgraded to a B rating.

on the SSLLABS-test page you see at the moment -> "This server supports TLS 1.0 and TLS 1.1. Grade will be capped to B from January 2020."

Thanks for you thoughts,

Greetings,

Chris

December 9, 2019

9 hours ago, Alan Lantz said:

Something else I have ran into that is along this lines is something like DNS where I want VM's to never been

on the same hosts together. Assigning them to servers was the best I could do. It would be nice if WLB could

do inclusions/exclusions like this.

--Alan--

Hi Alan,

That is what I mean with "rules in WLB". With these 'rules' we should be able to specify which VM's should be on SEPARATE hosts, or even GROUPED together on the SAME host.

I hope the development of WLB can include these "rules", one can call them 'affinity-rules' and 'anti-affinity-rules'.

Hi, development team of WLB, this would be a very handy function to include in a next release of WLB. What do you think ?

Thanks and greetings,

Chris MARREEL

December 9, 2019

11 hours ago, Tobias Kreidl said:

One can up-vote it using the little blue arrow on the upper left. Just gave it a vote up. You probably cannot up-vote your own posting, I suspect, but have never tried that.

-=Tobias

thanks Tobias for the vote ;-)

December 8, 2019

1 hour ago, Tobias Kreidl said:

Chris,

No, that is not a currently available feature. Best to do, as suggested, is to add your request to the request and suggestion forum.

-=Tobias

Hi Tobias, My post is in the correct forum I assume.

I can understand I can't 'vote' for my own request, but it's totaly unclear to me how one could "vote" for a request in this forum...

As far as I can see all the forum-posts have at the moment 0 votes.

Or can you, by means of prove one can vote, try and do a test-vote on my request-post ?

Greetings,

Chris

December 8, 2019

Hi Alan and Thobias,

Thanks for you input.

But I assume that what I want, isn't yet available in the current WLB-setup. If I compare with the possibilities in VMware's DRS there it would be possible, so I hope the WLB-development-team can implement this in the next version of WLB.

Let's hope for some votes in my forum-post.

Greetings,

Chris Marreel

December 8, 2019

Hi all,

I'm at the moment looking at WLB, but I can't find a way to create 'custom rules'.

e.g. Is it posssible to keep some VM's TOGETHER on the SAME XenServer-host ?

The biggest reason is PVS-Accelerator. So I would love to have all the VM's that are using a specific PVS-image running on the SAME XS-Host, so they make optimal use of the PVS-Accelerator-cache. Ofcourse in the aspect of HA this is not optimal, but we must be honest, we don't have that much hardware-failures, so we are looking at optimal every-day use and maximum performance for the end-users.

And if not possible at the moment, wouldn't it be a great idea to add this in future WLB-versions ? How can I file this WLB feature-request ?

Thanks for sharing your toughts.

Greetings,

Chris Marreel

August 28, 2019

Hi,

Anyone ever had to look for a reclaim process of those TemporaryStorage.vhdx-files ? At the moment I DELETE the VDI's and re-create them, but there should be a more elegant way.

At the moment I have 40 VDI-stations, and each VDI-station has a TemporyStorage.VHDX-file of 15 - 19GB, which means 750GB of 'tempory storage'... (on my super-fast SAN = expensive SAN!)

Thanks for helping me further on this.

Greetings,

Chris Marreel

February 4, 2019

Hi Carl,

I was not yet aware of that 50.31 build for the 12.1 firmware.

I have tested this and I can confirm the "rewrite policies" are working as they should with this upgrade.

Thanks for making me attent of this maintenance-build 50.31 for the 12.1 firmware.

P.S. I noticed this was indeed mentioned in the Release Notes :

In a Citrix Gateway appliance, responder and rewrite policies bound to VPN virtual servers might not process the packets that matched the policy rules.

[# NSHELP-18311]

Greetings,

Chris Marreel

February 4, 2019

Hi,

I could solve my issue with the "Rewrite Policies" by downgrading to the 49.37 build.

I had a Citrix Support case (december 2018), and they confirmed the issue/bug with those Rewrite Policies, but they would not release a fix until the new build would come available in march 2019.

I suggested that support engineer to withdraw that 'bad' 50.28 build, but he mentioned that would not be the case. I have warned my college's to not use that 50.28 build, and stay at the 49.37 build.

I would suggest to do a downgrade to that 49.37 build.

Greetings,

Chris

December 29, 2018

Hi,

I experienced with this firmware-build an issue where the Rewrite Policies I create are not working. I tested with a downgrade to the 12.1 build 49.37 and my Rewrite Policies are functioning again. I have for this issue an open support-case running. More news in the near future once I have more news from the support-engineer.

In other words, is this 12.1 build 50.28 actually a good firmware-build ?

Any other experiences ?

Greetings,

Chris Marreel

July 10, 2018

Hello,

According to my experience, this issue is solved by doing the next 2 actions:

- upgrade your PVS Server AND PVS Target software to the 7.15 CU2 version (= 7.15.3)

- upgrade the SEP Management Server AND SEP Agent to the version 14.0 RU1 MP2 (= 14.0.3929.1200)

These were until recent the latest and most recent verions. And I have several setups running (on W2K16 VDA's) without problems.

Greetings,

Chris

March 2, 2018

Hello,

Thanks Peter for the feedback.

I can confirm that the tests with the 7.15.2 version of PVS, that is delivered in the form of Hotfixes for the 3 different PVS-components, seems to be the answer to these Target Device boot-issues.

I don't know why there is not more noise about this new PVS-version.

Keep on testing, and providing some feedback here.

Greetings,

Chris

February 28, 2018

Hello,

I'm following this thread now for a while.

I noticed that there is a hotfix available for PVS 7.15 :

https://support.citrix.com/article/CTX231112 : Provisioning Services 7.15.2 for Server x64
https://support.citrix.com/article/CTX231115 : Provisioning Services 7.15.2 for Console x64
https://support.citrix.com/article/CTX231111 : Provisioning Services 7.15.2 for Target Device x64

Are there already persons who have done this upgrade ? If yes, was this boot-issue solved with this hotfix ?

Thanks for sharing,

Greetings,

Chris Marreel

January 31, 2018

Hello,

The main reason why I'm not a fan of "Deleting" and "Adding" VM's to this machine catalog : all these machines have a DHCP-reservation on their MAC-adress, and recreating these VM's gives them new MAC-adresses on the hypervisor-level, and this involves then a lot of copy/paste actions to set these MAC-adresses correct again.

Shouldn't Citrix implement some kind of a "Reclaime diskspace"-button in the Citrix-Studio that re-initializes that Temporary Space VHDX-file ?

Share your thoughts, I think this is a valuable enhancement request

Greetings,

Chris Marreel

January 31, 2018

Hello,

We are using MCS with Hyper-V.

I have configured all the VDI's with 20GB of Temporary Space to hold all the write-actions. In normal circumstances I notice this only gets to 3 à 4GB in size. But I have a few of my VDI's who have a 20GB TemporaryStorage.vhdx and I'm looking for a procedure to RECLAIM this diskspace.

What can I do ? I assume deleting and recreating is one option, but that is to drastic, so I'm looking for an alternative way.

I already found that the 20GB "Disk Cache"-file is "not intialized" when the VDI's are started, so I assume the MCS is handling the content/filestructure inside that VHDX-file.

So I cannot mount this VHDX-file as it doesn't contain a valid filesystem I assume. But I still want to reclaim the unused diskspace from those VHDX-files.

Thanks for helping me out on this,

Greetings,

Chris Marreel

Forums

Articles

Labs

Videos

TechZone

Citrix Community Articles

Events

Profiles

Posts posted by Chris Marreel