Partial Success, Failed to add VTEP errors when configuring ESXi as transport node in NSX-T

Products

• VMware NSX

Issue/Introduction

• Configuring ESXi as a transport node in NSX-T may result in partial failures, with error messages like "Partial Success" or "Failed to add VTEP."
• Observed from logs:

 

nsxapi.log:

<YYYY-MM-DDTHH:MM:SS> ERROR L2HostConfigTaskExecutor4 TransportNodeAsyncServiceImpl 14543 FABRIC [nsx@6876 comp="nsx-manager" errorCode="MP8700" level="ERROR" subcomp="manager"] DefaultGateway [xx.xx.xx.xx] failed to be applied: Failed to update vxlan default gateway: {'fault': 'InvalidArgument', 'msg': 'A specified parameter was not correct: ', 'faultMessage': []}; VTEP [vmk10,xx.xx.xx.xx] failed to be applied: Failed to add VTEP: {'fault': 'InvalidArgument', 'msg': 'A specified parameter was not correct: ', 'faultMessage': []};
com.vmware.nsx.management.switching.common.exceptions.SwitchingException: null
        at com.vmware.nsx.management.switching.sync.host.TransportNodeAsyncUtils.getErrorFromHostReply(TransportNodeAsyncUtils.java:945) ~[libnsx-switching.jar:?]
        at com.vmware.nsx.management.switching.sync.host.TransportNodeAsyncServiceImpl.stageHandlingHostConfigReplyMessage(TransportNodeAsyncServiceImpl.java:1170) ~[libnsx-switching.jar:?]
        at com.vmware.nsx.management.switching.sync.host.TransportNodeAsyncServiceImpl.syncTransportNode(TransportNodeAsyncServiceImpl.java:2571) ~[libnsx-switching.jar:?]
        at com.vmware.nsx.management.switching.sync.host.TransportNodeAsyncRequestHandler$TransportNodeConfigTask.run(TransportNodeAsyncRequestHandler.java:652) ~[libnsx-switching.jar:?]
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) ~[?:1.8.0_281]
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) ~[?:1.8.0_281]
        at java.lang.Thread.run(Thread.java:748) [?:1.8.0_281]

 

hostd.log:

<YYYY-MM-DDTHH:MM:SS>info hostd[2103876] [Originator@6876 sub=Hostsvc opID=b8d80a88 user=nsx-user] Invalid shaping policy: shaper is enabled, but at least one of the avgbps, peakbps or burst is not set.
<YYYY-MM-DDTHH:MM:SS> info hostd[2103876] [Originator@6876 sub=Hostsvc opID=b8d80a88 user=nsx-user] Invalid network policy: there are invalid properties in network traffic shaping policy.
<YYYY-MM-DDTHH:MM:SS> info hostd[2103876] [Originator@6876 sub=Hostsvc.NetworkVmkSimulator opID=b8d80a88 user=nsx-user] Cannot add portgroup [vMotion Network]: invalid network policy specified:(vim.host.NetworkPolicy)

Environment

VMware NSX-T Data Center 3.x
VMware NSX

Cause

The problem arises due to traffic shaping policies applied to the vMotion VMkernel. When traffic shaping is enabled without proper configuration parameters (e.g., missing average or peak bandwidth), NSX-T transport node preparation is blocked or partially applied. This results in the inability to fully configure the transport node, preventing proper network setup and creation of essential VMkernel interfaces like vmk10.

Resolution

To address this, follow these steps to temporarily adjust the configuration and enable successful transport node preparation:

1. Remove the vMotion VMkernel Interface: Temporarily delete the vMotion VMkernel from the ESXi host configuration to avoid policy conflicts.

2. Update the Transport Node Profile:

   • Modify the IP Pool settings associated with the Transport Node Profile or update the ESXi node configuration directly.
   • This will reapply network configurations without interference from traffic shaping policies on the vMotion interface.

3. Re-add the vMotion VMkernel: Once the transport node updates are successfully applied, reconfigure and add back the vMotion VMkernel.

4. Prepare the ESXi host once again with NSX.