VDA Fails to Register: Cannot Communicate With Delivery Controllers

To resolve this issue:

1. If the communication between VDA and Delivery Controllers were set using

a) Policy or Manually/Registry-based:

  • Verify the ListOfDDCs is not empty, and that the hostnames are correctly entered and can be resolved. To do this, you can ping each host name or use nslookup from the command prompt.
  • Value will be stored in:
HKLMSoftwarePoliciesCitrixVirtualDesktopAgent (ListOfDDCs)

or

HKLMSoftwareWow6432NodeCitrixVirtualDesktopAgent (ListOfDDCs)

*For more information, see CTX133384 – Best Practices for XenDesktop Registry-based DDC Registration

b) Active Directory OU-based discovery:

  • Value will be stored in:
32 Bit: HKEY_LOCAL_MACHINESoftwareCitrixVirtualDesktopAgentFarmGUID

64 Bit HKEY_LOCAL_MACHINESoftwareWow6432NodeCitrixVirtualDesktopAgentFarmGUID

*For more information, see Active Directory OU-based Controller Discovery

c) Provisioned by MCS

  • The MCS process creates C:Personality.ini, containing a list of contactable DDCs in following format:

[VdaData] ListOfDDCs=<FQDN of the Controller>

2. Verify the VDA’s DNS settings are correctly configured so the Delivery Controller’s FQDN can be resolved from the VDA.

3. Verify the network communication by pinging VDA from the Controller and vice versa.

4. Verify the VDA and the Delivery Controller can communicate on the same port.

5. Verify that any Delivery Controller host names in the Windows Hosts file are correctly entered and can be resolved. To do this, you can ping each host name or use nslookup from the command prompt.

Related:

Printers created from WEM are not being mapped to user sessions

To resolve this issue, configure the printer on WEM admin console using the hostname of the print server instead of the FQDN.

If you want to use the FQDN in the printers configuration, you have to fix the issue that causes that the agent machines fail to map the printers manually using the FQDN of the print server.

Related:

Running Add-PvsDeviceToDomain commandlet from PowerShell without any parameter causes unexpected behavior.

This problem has already been identified by Citrix and is currently working towards fixing the same in the future releases.

The only work around available for now is to use a Parameter along with the PowerShell Commandlet – “Add-PvsDevicesToDomain”, which are as below:

-Guid or DeviceId [<Guid[]>]

GUID of the Device to Add to the Domain.

-Name or DeviceName [<String[]>]

Name of the Device to Add to the Domain.

-DeviceMac [<PvsPhysicalAddress[]>]

MAC of the Device to Add to the Domain.

-CollectionId [<Guid[]>]

GUID of the Collection to Add all Devices to the Domain.

-SiteViewId [<Guid[]>]

GUID of the Site View to Add all Devices to the Domain.

-FarmViewId [<Guid[]>]

GUID of the Farm View to Add all Devices to the Domain.

-FarmViewName [<String[]>]

Name of the Farm View to Add all Devices to the Domain.

-CollectionName [<String[]>]

Name of the Collection to Add all Devices to the Domain.

-SiteViewName [<String[]>]

Name of the Site View to Add all Devices to the Domain.

-Domain [<String[]>]

Domain to add the Device(s) to. If not included, the first Domain Controller found on the Server is used.

-OrganizationUnit [<String[]>]

Organizational Unit to add the Device(s) to. This parameter is optional. If it is not specified, the device is

added to the built in Computers container. Child OU’s should be delimited with forward slashes, e.g.

“ParentOU/ChildOU”. Special characters in an OU name, such as ‘”‘, ‘#’, ‘+’, ‘,’, ‘;’, ‘>’, ‘=’, must be

escaped with a backslash. For example, an OU called “commaIn,TheMiddle” must be specified as

“commaIn,TheMiddle”. The old syntax of delimiting child OU’s with a comma is still supported, but deprecated.

Note that in this case, the child OU comes first, e.g. “ChildOU,ParentOU”.

-SiteId [<Guid[]>]

GUID of the Site.

-SiteName [<String[]>]

Name of the Site.

-Confirm [<SwitchParameter>]

The impact of this operation is “low”. If -Confirm is specified, the operation will be confirmed.

$ConfirmPreference can be set to “low” to have confirmation without the Confirm parameter.

-Object [<PvsObject[]>]

Object: will use the DeviceId, CollectionId, SiteViewId or FarmViewId.

Related:

Naming Windows 10 Computers and Joining to Domain

I do not need a solution (just sharing information)

Hello, I am fairly new to Norton Ghost. Here is what I would like to achieve with it. Right now we are transitioning to Windows 10 using Symantec Ghost 3.3, Build 2121 and we have a somewhat unusual location based naming convention for our dell desktops.

To give you an idea of what this hostname looks like it would be Location-DesktopClient-SerialNumber IE: SEA-DCL-S3R1AL.

I would like to see if it is at all possible to have Ghost name these computers like this. In a previous life I once used SCCM where there was a intranet webpage where we could enter the MAC address/Serial Number, Hostname, Domain, OU then it would submit this information into a database. Then it would automatically issue the Hostname, Join to the domain, and put the computer in the designated OU in corrolation with the MAC Address or Serial Number from the database. 

I am just looking into if this is all possible, or gathering ideas as to how I can automate this process, or make it very simple as we have alot of computers to do.

Thank you,

Jason

0

Related:

Sophos Anti-virus for Linux: Linux endpoint not reporting as registering to Central though the MCS.log file and config file show that it has registered.

A Linux endpoint is not reporting as registering to the Cloud though the MCS.log file and config file show that it has registered. The following error may be seen if the Linux machine is not registered in the DNS A records or hosts file so the lookup against itself fails:

subprocess.CalledProcessError: Command '['hostname', '-f']' returned non-zero exit status 1

This will probably be due to a name resolution issue when the Endpoint is trying to register itself to Cloud. During this process two DNS queries are performed from the EP, one to the AWS cloud server,the other is to the Linux machine itself

The lookup process is as follows:

  1. DNS lookup from EP for AWS cloud
  2. Once IP address is identified by DNS lookup, TLSv1 session to AWS cloud is made. (typically ‘Server Hello’ is communicated.)
  3. DNS lookup for the Linux machine itself.
  4. Once the lookup for itself is successful, the next TLSv1 session with AWS cloud is made. (typically ‘Client Hello’ is communicated.)

When this error is seen the Linux machine is not registered in the DNS A records or hosts file so the lookup against itself fails.

The following sections are covered:

Applies to the following Sophos products and versions

Sophos Anti-Virus for Linux

Once a record in the DNS server for the Linux machine has been specified the registration with Sophos Central should proceed. Alternatively, the hosts file can be updated by adding the machine name of the Linux machine itself.

If you’ve spotted an error or would like to provide feedback on this article, please use the section below to rate and comment on the article.

This is invaluable to us to ensure that we continually strive to give our customers the best information possible.

Related:

VDA Cannot Communicate With Delivery Controllers

To resolve this issue:

1. If the communication between VDA and Delivery Controllers were set using

a) Policy or Manually/Registry-based:

  • Verify the ListOfDDCs is not empty, and that the hostnames are correctly entered and can be resolved. To do this, you can ping each host name or use nslookup from the command prompt.
  • Value will be stored in:
HKLMSoftwarePoliciesCitrixVirtualDesktopAgent (ListOfDDCs)

or

HKLMSoftwareWow6432NodeCitrixVirtualDesktopAgent (ListOfDDCs)

*For more information, see CTX133384 – Best Practices for XenDesktop Registry-based DDC Registration

b) Active Directory OU-based discovery:

  • Value will be stored in:
32 Bit: HKEY_LOCAL_MACHINESoftwareCitrixVirtualDesktopAgentFarmGUID

64 Bit HKEY_LOCAL_MACHINESoftwareWow6432NodeCitrixVirtualDesktopAgentFarmGUID

*For more information, see Active Directory OU-based Controller Discovery

c) Provisioned by MCS

  • The MCS process creates C:Personality.ini, containing a list of contactable DDCs in following format:

[VdaData] ListOfDDCs=<FQDN of the Controller>

2. Verify the VDA’s DNS settings are correctly configured so the Delivery Controller’s FQDN can be resolved from the VDA.

3. Verify the network communication by pinging VDA from the Controller and vice versa.

4. Verify the VDA and the Delivery Controller can communicate on the same port.

5. Verify that any Delivery Controller host names in the Windows Hosts file are correctly entered and can be resolved. To do this, you can ping each host name or use nslookup from the command prompt.

Related: