Disclaimer! I am learning NSX-T, part of my learning is to deploy in my lab - if I contradict the official docs then go with the docs!
This NSX-T lab environment is built as a nested lab on my physical hosts. There are four physical ESXi hosts, onto which I will deploy three ESXi VMs, a vCenter Server Appliance, NSX Manager, an NSX Controller cluster, and two NSX Edge Nodes.
I will follow the deployment plan from the NSX-T 2.0 documentation:
When this post series is complete, the network topology should be something like this, with two hostswitches configured. The ESXi Hosts will have a Tunnel Endpoint IP address, as will the Edge. The Edge will also have an interface configured for a VLAN uplink.
In this post I will walk through configuring the Transport Zone, Transport Nodes, Edge Cluster and other configuration required to support the deployment.
With ESXi hosts you can prepare them for NSX by using the “Compute Manager” construct to add a vCenter server, and then prepare the hosts automatically, or you can manually add the hosts. Since this is a learning exercise, I am going to manually install the VIBs because I think it gives a good understanding of how the automatic process works in the background.
To manually install the NSX VIBs, download the nsx-lcp* package from My VMware and copy it to the /tmp folder on each ESXi host. The VIB installation is exactly the same as any other VIB installation
In the NSX Manager web interface, you can see the host appears in the Fabric section. Here you can see I’ve prepared my three nested hosts, you can see NSX is installed, it’s at version 2.0 and connectivity with the Manager is up. As we’re not yet participating in a transport zone, there is no active connection to the Controller cluster and Transport node is not configured.