This section provides details of the known issues in Horizon Management Pack.

Resources move to Old Data Receiving state when collection fails in VMware Horizon version 2312 with Manual Desktop Pools.

Collection fails due to empty networkLabel in the manual desktops of Manual Desktop Pools. This is only applicable for VMware Horizon version 2312.

In Brownfield, Connection Servers gets listed for UAG objects and UAGs gets listed for Connection Server objects.

Sometimes, the Connection Server related objects are seen in the API response for Gateway API from Horizon. Due to this, the Connection Server objects are visible in the search results for the Horizon Unified Access Gateway object in VMware Aria Operations.

Time to load profile metric goes missing inconsistently.

When FIPS is enabled on VMware Aria Operations, Telegraf installation fails for connection servers or UAG servers.

The installation fails only with openSSH version 3.0.9.

Workaound:
  • ssh to cloudproxy with root credentials.
  • Modify CSR generation digest from -sha1 to -sha256 in /ucp/ucp-config-scripts/ucp-firstboot.sh file (Line no : 1717).
  • Reset ucp-firstboot.sh file ownership to admin. Cmd : chown admin:admin ucp-firstboot.sh.
  • Retry Agent install again after performing the above steps.

Actions fail for desktops with lower versions of VMtools.

Actions fail for those desktops that have VMtools versions lower than version 11.0.1.

VMware Aria Operations requires the following VMware VMTools versions 11.0.1 for and above installed on the monitoring VMs.

While earlier versions may seem to work, there are no guarantees that they will behave correctly and produce reliable data. It is not recommended to install any other version than the ones explicitly stated above as they will not be supported.

Actions cannot be performed on logged off sessions.

Any Actions performed on logged off VDI Desktop Sessions will fail. There will be a delay of one collection cycle in VMware Aria Operations for the session objects.

Concurrent Actions on same session object causes failure.

When concurrent actions occur, you can see the following exceptions for different scripted actions:
  • com.sun.xml.ws.fault.ServerSOAPFaultException: Client received SOAP Fault from server: A general system error occurred: vix error codes = (1, 0).
  • com.vmware.vim25.InvalidStateFaultMsg: The operation is not allowed in the current state.
  • Exception occured while performing LoginTimeAnalysis
  • Adapter:onAction() returned a null ActionResult, dispatch mode of the action is UNKNOWN. Task status could go to UNKNOWN state

Unable to decrease the desktops on the VDI Pools or RDS Farms.

You can expand the VDI Pools or RDS Farms with more desktops but cannot descrease the desktops. This action results in failure.

Actions can be performed on Multiple client sessions but the output of all the sessions will be displayed under one of VM.

When you perform actions on multiple client sessions, all the outputs of those session will be displayed under one of the VMs in the Select VM to view Actions Results widget. Individually check every VM in this widget to get the VM with all the outputs.

Horizon adapter collection might stop if configured Horizon Server undergoes upgradation.

Impacted versions:
  • Any Horizon Server upgraded to Horizon Servers 8.12 version
  • Horizon Server 7.13 upgraded to any Horizon Servers 8.x
  • Horizon Servers 8.x upgraded to Horizon Servers 8.3 or higher
  • Horizon Server 8.3 upgraded to Horizon Servers 8.7 or higher

Workaround

Restart the Horizon adapter to continue collection.

VDI Pools capacity widget of the Horizon Day to Day Operations dashboard show incorrect data, pointing to session remaining.

Workaround

This widget can be edited to select VDI Pool / Capacity list instead of Capacity / RDS Farm distribution.

Number of sessions at Horizon Connection Server level shows incorrect value.

Issue originated from Horizon.

Logon Duration is shown negative.

The issue is originating from Horizon.

Helpdesk API fails to fetch the protocol metrics for sessions.

The issue is originating from Horizon.

Logon Metrics is missing if there is time sync issue in Event DB, Horizon Desktop & Connection Servers.

To check whether there is time sync issue or not, follow these steps:
  1. On the Connection Server, Navigate to Monitor > Session.
  2. Identify the user who is logging off.
  3. Go to Events tab and note down the Updated timestamp in the right corner of the table.
  4. Log off the same user session, note down the time stamp from the Time column of the table with Agent Module & Message User xxx has disconnected from machine yyy.

    If the timestamps are not matching then there is time sync issue between Event DB, Horizon Desktop & Connection Servers

    Workaround: Fix the time sync issue to resolve the missing logon metrics.

Data Missing in RDS Performance Dashboard.

To populate the data for the RDS Performance Dashboard, follow the steps given below:
  1. Click Edit Widget in Profile Performance widget.
  2. Under configuration, select Box columns value as 4 (from 2).
  3. Under Output Data, select the existing two metrics and click Remove selected metrics.
  4. Click Add New Metrics and search for Horizon RDS Desktop Session and select it.
  5. Select the metrics Session|Time taken to load Profile and Session|Time taken to Logon and click OK.
  6. Repeat 4 and 5 steps for Horizon RDS Application Session.
  7. Click SAVE.

Performance KPI metrics are not populated.

Performance KPI metrics are not populated in the Horizon Management Pack if::
  • The underlying vCenter infrastructure is hosted on AVS or GCVE
  • The ESXi version is less than 6.7u1
  • The VMware tool version is less than 10.3.5
  • The metrics are not enabled in the default policy of the vCenter adapter
Workaround:

Use VM Tools 10.3.0 on ESXi 6.7u1 and later.

Duplicated Objects and Relations found in the Traversal spec when a vCenter is mapped to 2 different Horizon Pods.

If a vCenter server is shared across multiple Horizon Pods:
  • The adapter does not support the configuration
  • The computed metrics are wrongly calculated due to shared relations
  • Related Objects are duplicated under both pods
  • Traversal spec duplicates child objects under both pods

Some of the sessions are having start time as NULL causing the incorrect number of sessions/users.

The issue is originating from Horizon.