Table of Contents
Overview
I run a home lab built on older Intel Xeon CPUs, which are no longer supported by modern versions of VMware Cloud Foundation (VCF). Since this is a non-production lab environment, I’m using nested virtualization to continue my testing and learning. In this post, I’ll walk through how I’m bypassing the CPU-compatibility checks so I can deploy VCF on this older hardware.
Note: Did you catch that? This is the ESXi installer for ESX…and this is even on a 9.0.1 installer! Both ESX and ESXi terms will be used interchangeably in this post.
RTFM
Virtual Hardware Configuration
These are the settings per nested host. There will be four nested hosts for a consolidated VCF Management Domain.
- vCPU: 16 (Hardware virtualization enabled)
- RAM: 96 GB
- OS Disk: 128 GB (thin provisioned)
- vSAN Cache Disk: 64 GB (thin provisioned)
- Attribute:
scsi0:1.virtualSSD - Value: 1
- Attribute:
- vSAN Data Disk: two x 300 GB (thin provisioned)
- Network adapter: two x VMXNET 3 with 10 Gbps fiber backing
- Secure boot enabled
Installation – Edit Boot Options
When loading the ISO during an interactive installation, notice the option to Edit boot options with SHIFT + O. (Big “Oh”, not zero!)
Add the following to the runweasel cdromBoot line.
runweasal cdromBoot allowLegacyCPU=true
Press the ENTER key to apply the options and boot.
Installation
Now the installation should proceed as normal.
And the moment of truth, now this is a CPU_SUPPORT WARNING.
Preparing the ESX Host
Press F2 to customize the system.
Log in with the password configured during the installation.
Configure the Network
Select Configure Management Network.
Configure settings as needed.
The following image is for informational purposes, only. For actual use, my lab has two vmnic adapters, with only one configured since this will be for VCF.
When configuring DNS, always use the fully qualified domain name and ensure the case is lowercase.
Same thing with custom DNS suffixes. Always add in lowercase suffixes as these will be searched in the order given if only a short name is provided.
When finished configuring all the settings, press Esc to exit. When prompted to apply changes, press Y.
Even though the Hostname: does not show the fully qualified domain name here in the Direct Console User Interface (DCUI), the hosts files on the host will display the fully qualified domain name. This is important for preparing the host for VMware Cloud Foundation (VCF) to get the local certificates regenerated correctly.
cat /etc/hosts # Do not modify this file directly, please use esxcli. 127.0.0.1 localhost.localdomain localhost ::1 localhost.localdomain localhost 172.16.2.51 vcf9esx-162-51.lab.aaronrombaut.com vcf9esx-162-51
Enable SSH
Enable SSH by selecting Troubleshooting Options.
Press Enter to enable ssh. When the option turns to Disable SSH, ssh is enabled.
One Script
The following is a quick script that can be added to an SSH session to quickly configure the host. More detailed instructions are below and of course, there are the official Broadcom Techdocs in the RTFM section.
Update the VLAN ID, NTP server IP 1, and optionally, NTP server IP 2.
# Configure the Virtual Machine Network Port Group esxcli network vswitch standard portgroup set --portgroup-name='VM Network' --vlan-id=<vlan id> # Configure NTP esxcli system ntp set --enabled=true --server=<ntp-server-ip-1> --server=<ntp-server-ip-2> # Regenerate the Self-Signed Certificate /sbin/generate-certificates # Reboot the host reboot
Configure the Virtual Machine Network Port Group
First, check the default VLAN ID.
esxcli network vswitch standard portgroup listesxcli network vswitch standard portgroup list Name Virtual Switch Active Clients VLAN ID ------------------ -------------- -------------- ------- Management Network vSwitch0 1 162 VM Network vSwitch0 0 0
Modify the VM Network port group to be the same as the Management Network. When the command is run, there will not be any output.
esxcli network vswitch standard portgroup set --portgroup-name='VM Network' --vlan-id=<vlan id>esxcli network vswitch standard portgroup set --portgroup-name='VM Network' --vlan-id=162
To verify the network change is successful, return to the first command.
esxcli network vswitch standard portgroup listesxcli network vswitch standard portgroup list Name Virtual Switch Active Clients VLAN ID ------------------ -------------- -------------- ------- Management Network vSwitch0 1 162 VM Network vSwitch0 0 162
Configure NTP
First, check the current NTP settings.
esxcli system ntp getesxcli system ntp get Enabled: false Loglevel: warning PID: 0 Runtime Seconds: 0 Servers: Service Providing Kernel Time: Time Service Enabled: false Time Synchronized: false
Now configure up to two NTP servers. Note that running this command on the command line will not update the web GUI NTP service startup policy to Start and stop with host until after the host is rebooted.
The command for adding one NTP server.
esxcli system ntp set --enabled=true --server=<ntp-server-ip>esxcli system ntp set --enabled=true --server=172.17.0.10
The command for adding two NTP servers.
esxcli system ntp set --enabled=true --server=<ntp-server-ip-1> --server=<ntp-server-ip-2>esxcli system ntp set --enabled=true --server=172.17.0.10 --server=172.17.0.11
To verify the NTP configuration is successful, return to the first command.
esxcli system ntp getesxcli system ntp get Enabled: true Loglevel: warning PID: 2100600 Runtime Seconds: 156 Servers: 172.17.0.10, 172.17.0.11 Service Providing Kernel Time: Network Time Protocol Time Service Enabled: true Time Synchronized: false
It’s also a good idea to test the configuration. Keep running the test until the Timeinsync reads true. This should not take more than 10 minutes!
esxcli system ntp testesxcli system ntp test
Comments:
Service analysis started on host: vcf9esx-162-51.lab.aaronrombaut.com,
Test started at: 2025-12-09T19:38:30Z,
Time Service is administratively enabled.,
Verifying NTP service.,
NTP server: 172.17.0.10 resolves IPv4: 172.17.0.10,
Virtual NIC vmk0 : Admin: Up,
IP Interface: vmk0 IPv4 Address: STATIC 172.16.2.51,
IP Interface: vmk0 connected to Management Network on vSwitch0,
IP Network Stack: defaultTcpipStack,
Physical NIC vmnic0 : Admin: Up Oper: Up,
Firewall Rule: ntpClient allows traffic on port: 123,
Firewall Rule: ntpClient allows all IPs,
ntpd is running, PID: 2100865,
Kernel clock type: ntp,
NTP is in sync ,
Peering with: 172.17.0.10,
Stratum: 5,
Accuracy to within: 521.534000 msecs,
Polling every: 64 secs,
Network delay round trip: 50.828000 msecs,
Difference from remote clock: 0.170933 msecs,
Service analysis completed.
Timeinsync: true
Regenerate the Self-Signed Certificate
When the host is first installed, the hostname will reflect localhost.localdomain. Even after the host is configured prior to regenerating the certificate, the certificate will be self-signed and reflect this name from a browser.
Regenerate the self-signed certificate on the ESX host. There will be no output after the command.
/sbin/generate-certificatesReboot the host.
rebootIf you are (or were) connected to the ESXi Host Client from a browser, refresh the page and accept the new self-signed certificate.
Leave a Reply to Tom Cancel reply