Since the brand new VCF 9 release (and the Beta before that) has been available, I have deployed it in a few different scenarios in our demo lab and right now I am in a few customer PoCs. Hereby I want to share my experience and gradually expand the blog.
Overall I really love the extraordinary smooth automated deployment of a complete SDDC with 13 Management VMs in about 3 hours. Like every other big GA release there are a (very) few minor issues, but you can all resolve them manually without much knowledge and without wasting much time.
1. Perform vSAN ESA Auto Disk Claim – Failure
Description
The issue initially occured during my first VVF deployment. We had just received 4 new Dell vSAN Ready Nodes R7615 (all NVMe) and I tested the VVF installation without doing any Firmware updates first. After the workaround and the continued deployment a Skyline Health warning showed up in vCenter regarding a vSAN hardware compatibility issue (the Ready Nodes came from teh factory with an NVMe FW Version 1.0.0, but vSAN ESA needs 1.1.1). So I updated all the FW using OpenManage Enterprise and tested a few other things in VVF (like a VCF Import). Then, I deleted the environment because I wanted to test the VCF deployment in High Availability (3 Nodes for NSX, VCF Operations and Automation) and ran into the same issue again. It seems like it is not Firmware related but that it could be a bug. We will see.
Workaround
Log into the newly deployed vCenter and manually claim all unused disks from the other nodes for te vSAN ESA datastore. After that just return to the VCF Installer UI and push the “Retry and Proceed with Deployment” button.
2. Configure NSX Fabric Compute Manager (vCenter) – Failed
Description
During the NSX deployment step I ran into this issue. I logged into the newly deployed NSX Manager VIP and recognized a registration failure – apparently because auf DNS. I double checked DNS and everything looked fine. It could be a timing / timeout issue and probably be related to our lab DNS servers.
Workaround
Log into the newly deployed NSX Manager and manually resolve the error (if DNS is actually fine). After that just return to the VCF Installer UI and push the “Retry and Proceed with Deployment” button.
3. Retrieve the status for VCF Operations with VCF Operations collector Deployment request – Failed
Description
During the VCF Operations deployment step I ran into this issue. It looks like a timing / timeout issue to me and could probably be related to our lab environment.
Workaround
Log into the newly deployed vCenter and VCF Operations Admin UI, check that all 5 nodes are deployed, the Ops HA cluster has been created, Fleet Management has been activated and that everything looks healthy. After that just return to the VCF Installer UI and push the “Retry and Proceed with Deployment” button.
4. Retrieve the status for VCF Automation Deployment request – Failed
Description
During the VCF Automation deployment step I ran into this issue. It looks like a timing / timeout issue to me and could probably be related to our lab environment.
Workaround
Log into the newly deployed vCenter and VCF Automation provider UI, check that all 3 nodes are deployed and that everything looks healthy. If that is the case just return to the VCF Installer UI and push the “Retry and Proceed with Deployment” button.
5. NFS Datastore Configuration – Validation Failed
Description
NFS Datastores have to be preconfigured on the ESX hosts to use them for the initial VCF Deployment. Unfortunately we always run into the same failure during the validation “Failed to validate NFS configuration : Invalid vswitch port group name NFS. Error was: Unable to get node: Not found: VSI node (5007:)”.
Workaround
After some trial and error and research of our customer, he found the following article about NFS deployment issues:
VCF 9 NFS Fails To Deploy
https://www.lab2prod.com.au/2025/10/vcf9-nfs-fails-deployment.html
It looks like it is necessary to configure the right MTU size out-of-the-box. The VMware documentation is very vague about the correct expected configuration of NFS. But with 9000 configured on the vSwitch and the VMKernel port the VCF deployment has been succesful!
I hope this post can be helpful to you. Feel free to share if you like…
// footnotes:
Date: 07. Nov 2025
Version: 1.1