Holodeck 9.0.2 Deployment Notes – Single-Host R740 Lab (VCF 9.0.2)

Author: Brian | Date: March 2026 | Hardware: Dell PowerEdge R740 (2x Xeon Gold 6242, 336 GB RAM, 4tb NVMe on PCIe card)

This guide documents the exact steps and workarounds I used to deploy a nested VMware Cloud Foundation 9.0.2 lab using Holodeck 9.0.2 on a single physical ESXi host managed by vCenter. Includes common pitfalls and fixes for Supervisor activation and All-Apps Org creation.

1. Pre-Deployment Checklist

• I used Offline depot already populated. I scp’d all the binaries to a Temp folder on the depot VM. Then I had Grok make a bash script to create the expected folder structure and move all the files to the correct places. If you can create a Download key its much easier but I didn’t have that option.
• ESXi ISO and SDDC Manager OVA need to be places on the Holorouter VM in /holodeck-runtime/bin/9.0.2.0/
• Use a hardwired jumpbox VM or Run inside ‘screen -S holodeck‘ to survive SSH drops from personal machine going to sleep.

2. Create Cluster in Lab (non-nested) vCenter (Critical Step)

You usually have to recreate cluster HolodeckCluster with defaults and put esx3 into it. Even if it existed, sometimes needs to be deleted and recreated to get past ‘VI Container not found’ error.

1. Log into your vCenter
2. Right-click Datacenter → New Cluster
3. Name: HolodeckCluster (exact spelling, case-sensitive)
4. Take defaults, leave DRS/HA off
5. Drag your physical ESXi host into the cluster (can be ESXi 8.x, ESXi 9.x runs fine nested in a 8.x host)

3. Create & Load Holodeck Config (vCenter Target)

New-HoloDeckConfig `
  -Description "R740 Lab" `
  -TargetHost "<Your vCenter>" `
  -Username "administrator@vsphere.local" `
  -Password "VMware1!"

Take note of the new Config ID (shown in output).

Get-HoloDeckConfig   # confirm the ID

Import-HoloDeckConfig -ConfigId <NEW-ID>

4. Run Deployment (takes 12+ hours on R740)

New-HoloDeckInstance `
  -Version "9.0.2.0" `
  -InstanceID "MyLab" `
  -ManagementOnly `
  -CIDR "10.10.0.0/20" `
  -vSANMode "OSA" `
  -NsxEdgeClusterMgmtDomain `
  -DeployVcfAutomation `
  -DeploySupervisorMgmtDomain `
  -DepotType "Offline" `
  -LogLevel "INFO" `
  -Site "a" `
  -DeveloperMode

Install may fail — just rerun the same command. Holodeck is idempotent and will resume where it left off.

5. Supervisor Activation (common failure point)

The last part, Supervisor, always fails initially because the cluster is out of compliance due to alerts.

Fix:

1. In nested vCenter → cluster → Monitor tab → clear all alarms
2. Go to vSAN → Skyline → SILENCE all alerts
3. Go to Update tab → Remediate each host in REVERSE ORDER (esx-04a → esx-03a → esx-02a → esx-01a)
4. Wait until “All hosts in this cluster are compliant” appears
5. Supervisor activation should automatically finish

If still stuck after remediation:

• In nested vCenter → cluster → Configure → vSphere with Kubernetes → Deactivate
• Rerun the deploy command (it will retry Supervisor only)

6. Deployment Success Confirmation

Look for this message in the pwsh output:

[SUCCESS] HoloDeck deployment completed successfully

7. Create All-Apps Org in VCFA (Day-2 Operation)

After Supervisor is complete, run:

Before running: Log into VCFA and NSX to wake them up and confirm they are responding.

Update-HoloDeckInstance `
  -Site a `
  -AddVcfAutomationAllAppsOrg `
  -VIDomain Management

8. Cleanup / Reset (if needed)

Remove-HoloDeckInstance -Site a -InstanceID MyLab -Force -Confirm:$false

Use this before any rebuild to avoid conflicts.

Access URLs (after success)

• SDDC Manager: https://sddcmanager-nslab.site-a.vcf.lab (admin / VMware1!)
• Nested vCenter: https://vc-mgmnt-a.site-a.vcf.lab (administrator@vsphere.local / VMware1!)
• VCFA: https://auto-a.site-a.vcf.lab or vcfa-a.site-a.vcf.lab
• Self-signed certs: Click “Show Details” → “Visit this website” in Safari

Etc Hosts and Routing

Holorouter has a Desktop with links to all the VMs. Its pretty cool but cut&paste doesn’t work so I prefer to access from my laptop browser.

Built in desktop: http://<holorouter IP>:30000 (there are no credentials)

How to access from your laptop:

1. Add static route (on a Mac):

sudo route -n add -net 10.10.0.0/20 <holorouter IP>

sudo nano /etc/hosts

<add these entries and save>
10.10.1.10   vc-mgmt-a.site-a.vcf.lab vc-nslab.site-a.vcf.lab
10.10.1.5     sddcmanager-a.site-a.vcf.lab
10.10.1.21    nsx-mgmt-01a.site-a.vcf.lab
10.10.1.30    ops-a.site-a.vcf.lab
10.10.1.41    opscollector-01a.site-a.vcf.lab
10.10.1.70    auto-a.site-a.vcf.lab vcfa-a.site-a.vcf.lab
10.10.1.36    opslcm-a.site-a.vcf.lab

Tips & Troubleshooting

• The first build I had weird issues where I couldn’t create a Namespace in VCFA, I never could fix it so I deleted the whole Holodeck deployment using ‘Remove-HoloDeckInstance’ and started from scratch and the issue went away. Sometimes you just have to rebuild.
• Use -DeveloperMode to bypass strict resource checks
• If Supervisor hangs after remediation: Deactivate in vCenter → rerun deploy script
• Before redeployment: Run Remove-HoloDeckInstance first

Good luck with your VCF 9.0.2 lab!