"One of more transport node(s) realization failed while creating the transport node collection" error is displayed during VCF 9.0 Installation or workload domain creation

Products

• VMware NSX
• VMware Cloud Foundation

Issue/Introduction

• VCF Installer Interface shows the message:

One or more transport node(s) realization failed while creating the transport node collection with profile <uuid> on compute collection <uuid>:domain-### through NSX manager <fqdn>. Remediation: Check the error(s) on NSX Manager <fqdn> and resolve them before restarting the failed task. Reference token: ######.

• The NSX interface (System, Fabric, Hosts) does not show any attempt to configure the cluster or any ESX hosts:
  
  
• In the SDDC Manager log /var/log/vmware/vcf/domainmanager/domainmanager.log, the following exception is generated:
  
  Exception occurred while creating transport node collection using profile <uuid> on the cluster(with compute collection) <uuid>:domain-c10. com.vmware.vapi.std.errors.InvalidRequest: InvalidRequest (com.vmware.vapi.std.errors.invalid_request) (statusCode:400) +> { messages=[], data = => {error_message=# Hosts in Transport node collection <uuid>:domain-### are incompatible with ENS based HostSwitchMode config in transport node profile /infra/host-transport-node-profile/<uuid>. Sample incompatible host and Pnic info <ipaddr>=[vmnic2],<ipaddr=[vmnic2],<ipadd>r=[vmnic2],<ipaddr>=[vmnic2]. Remove ENS incompatible hosts from the cluster or use STANDARD as HostSwitchMode in the TransportNodeProfile., httpStatus=BAD_REQUEST, error_code=609679, module_name=policy}, errorType = INVALID_REQUEST}
  
  
• In the NSX Manager log /var/log/syslog, entries similar to the following are present for one or more vmnic's of the hosts in the cluster being prepared for NSX:
  
  <NSX Manger>.example.com NSX 82516 FABRIC [nsx@4413 comp="nsx-manager" errorCode="MP8287" level="ERROR" reqId="1ca438c0-####-####-####-0300f0d480b7" subcomp="manager" username="########"] Physical nic [vmnic5] on TransportNode [00000000-0000-0000-0000-000000000000] is not capable for ENS_INTERRUPT mode for given HostSwitch [## ## ## ## ## ## ## ##-99 9e cb db 57 e1 9d 27]
  
  
• Running the command esxcfg-nics -e on the ESX host shows ENS Capable as True or False and/or INTR Capable are False.
  

  Name    Driver      ENS Capable   ENS Driven    INTR Capable  INTR Enabled  MAC Address       Description
vmnic0  nenic       True          False         False         False         00:25:b5:a1:55:04 Cisco Systems Inc Cisco VIC Ethernet NIC
vmnic1  nenic       True          False         False         False         00:25:b5:b1:55:04 Cisco Systems Inc Cisco VIC Ethernet NIC
vmnic2  nenic       True          False         False         False         00:25:b5:a1:55:0d Cisco Systems Inc Cisco VIC Ethernet NIC
vmnic3  nenic       True          False         False         False         00:25:b5:b1:55:0d Cisco Systems Inc Cisco VIC Ethernet NIC
vmnic4  nenic       True          False         False         False         00:25:b5:a1:55:15 Cisco Systems Inc Cisco VIC Ethernet NIC
vmnic5  nenic       True          False         False         False         00:25:b5:b1:55:15 Cisco Systems Inc Cisco VIC Ethernet NIC

Note: When ENS Capable is True, it means the driver can support ENS (Enhanced Network Stack), which is the former name of EDP (Enhanced Datapath), if it is False, it means the installed driver does not support EDP.

Environment

• VMware Cloud Foundation 9.x
• VMware NSX 9.x

Cause

This is caused by the physical NICs on the ESX hosts being incompatible with, or having an incompatible configuration for, an NSX preparation using the Enhanced Data Path feature.

This may be a full incompatibility, a NIC driver installed that does not support EDP, or a feature that needs to be enabled/disabled to make it compatible with the EDP Enabled Transport Node Profile.

In the above case, the host is using nenic drivers, these drivers do not support EDP, there is a Cisco driver called nenic_ens which does support EDP.

Resolution

Multiple resolutions are possible, depending on environmental needs.

1. Update the physical NICs in the ESX host(s) to meet the requirements of the EDP feature. This resolution will be specific to your chosen hardware vendor. If in doubt, please consult with the nic vendor.
   
   OR
   
   
2. Install the appropriate driver for the NICs to be used for EDP. Check the Compatibility Guide to identify the correct NIC driver.
   
   OR
   
   
3. Disable EDP in the assigned Transport Node Profile per Adding a Transport Node Profile documentation. This will require modifying the Transport Node Profile and selecting Standard in the Advanced Configuration Section.
   
   
   1. Navigate to System → Fabric → Hosts → Transport Node Profiles
   2. Select the profile in question and click the blue number beneath the Host Switch header
      
      
      
      
   3. In the dialog that appears, click the blue Edit link to enable editing:
      
      
      
      
   4. Click the vertical ellipsis to the left of the Transport Node Profile name and click Edit:
      
      
      
      
   5. In the Advanced Configuration Section of the Host Switch, select Standard
      
      
      
      
   6. Click Add.
   7. Click Apply.
   8. Click Save.

Additional Information

Enhanced Data Path
Enabling Enhanced Data Path Standard in New Environment