When performing a new NSX-T deployment on a cluster through the NSX-T Plugin in vSphere the following error is seen
which reads as
'Cluster is not homogenous in terms of VDS connections or physical adaptors. Please select another cluster or attach a common VDS(7.0 or higher) to all hosts within this cluster.'
this can happen when the VDS version is 7.0 and higher and when cluster configuration seems to be same on all hosts
Upon going to the NSX Manager UI to perform installation you can se a more detailed error that appears as
which reads as
'Standby uplink list must be empty for LOADBALANCE_SRCID teaming policy associated with dvportgroup-xx. Please modify the teaming policy from VSphere and retry the Quick Install workflow. Alternatively, please use the advanced host switch configuration workflow. (Error code:610506)
both of which will make it so you cannot select the VDS on the cluster and continue the installation
this is due to a requirement that NSX-T has in which there cannot be any dvportgroups assigned to the cluster with a standby uplink
To fix this error, followed these steps:
Log in to the vSphere Client.
Navigate to Networking and locate Distributed Virtual Portgroups with Route based on originating virtual port configured
Right-click the portgroups and select "Edit Settings."
In the Edit Settings window, go to the "Teaming and failover" section.
Under "Load balancing," ensure that "Route based on originating virtual port" (LOADBALANCE_SRCID) is selected.
Clear the "Standby uplinks" list. For the LOADBALANCE_SRCID teaming policy, there should be no standby uplinks defined.
Click "OK" to save your changes.
Retry the NSX-T Quick Install workflow or the advanced host switch configuration workflow, as suggested in the error message and if that is the only issue the install should continue as expected