VCF / VLC Holodeck Lab – Networking Multiple Hosts

In this article, I’d like to talk about some advanced networking use-cases when deploying a VLC Holodeck environment into your lab. The supplied documentation only covers deployment onto a single host – this can be restrictive due to the amount of compute resources required to spin up a nested VCF environment.

By following this guide, you will be able to split the Holodeck/VCF VMs across multiple ESX hosts and save yourself from purchasing a ton of high density memory modules!

I won’t cover the basics of Holodeck deployment as William Lam has already written a fantastic article regarding this – I’d consider it mandatory reading before following this guide.

Quick Overview
A Holodeck VCF Environment will look like the below network diagram. This can either be created as a single VCF (Site-A) or you can also deploy a second site (Site-A2) with the pre-built configs provided out of the box.

The physical specs required for this is pretty hefty, so you might find it’s more cost effective for your lab to scale out, rather than scale up.

ESX Network Configuration

One of the installation pre-requisites is to create 2x Standard vSwitches and 2x Port Groups on a standalone ESX host. These vSwitches have no uplinks assigned which will allow the VMs connected to the switch to communicate with each other, as long as they reside on the same physical host.

To allow these VMs to communicate across multiple hosts, we need to add a physical NIC uplink to the vSwitch. My lab consists of 2 hosts, so I have created the same PG on both and linked them together with a crossover network cable. I have these configured as separate vSwitches as I had plenty of spare ports to separate VLC-A and VLC-A2 but you could put both Port Groups on the same vSwitch.

If you have more than 2 hosts they can be uplinked to a switch, but you would need to make sure the appropriate VLANs are configured on your network, and that these are not overlapping with anything else on your LAN (VLAN 10 for Site-A and VLAN 20 for Site-A2).

Now that this has been completed on both hosts, and DRS enabled on the cluster in vCenter we can see that the Holodeck VMs have been distributed across both physical hosts and they can still communicate with each other!

The VLC deployment script might complain during pre-checks when deploying a fresh environment into a cluster that has been configured like this. First, you must set DRS to Manual as VLC expects all the VMs to be running on a single host and vMotioning a VLC VM might drop its connectivity. The pre-checks also ensure you have set the MTU to 9000 and vSwitch/PG security settings as-per the documentation to ensure that the nested networking will function correctly.

Stay tuned for the next article where I will demonstrate the deployment of VMware HCX between the two Holodeck VCF sites.