Month: October 2025

VCF 9 Upgrade Paths Explained: From vSphere 8, VCF 5.x and VVF to VCF 9.x

How to get to VCF 9.x: upgrade from VCF 5.x, converge vSphere 8 or VVF with the VCF Installer, minimum versions, component order and upgrade traps.

Short answer: There are three ways into VCF 9.x. A VCF 5.x instance is upgraded: management components first, then SDDC Manager, NSX, vCenter and ESX, with the management domain done before any workload domain. A standalone vSphere 8 environment (with or without Aria Operations 8) is converged with the VCF Installer. A greenfield site is deployed new. vSphere 7 must first go to 8.

This page is the broad overview of upgrade paths into VCF 9.x. For the exact version matrix into 9.1, including which source builds qualify for a direct upgrade, see VCF 9.1 Supported Upgrade Paths.

Which VCF 9 upgrade path applies to you

Broadcom’s Paths to Building 9.0.x Environments groups everything by your starting state. The guidance covers only the management domain, because that is the minimum building block of a VCF Instance.

Starting pointPathTool
Nothing (greenfield)DeployVCF Installer, then VCF Operations
vSphere 8 (vCenter + ESX), optionally vSAN, NSX, Aria Operations/Automation 8Converge to VCF 9 or to vSphere Foundation (VVF) 9VCF Installer
VCF 5.xUpgradeManagement components first, then SDDC Manager
vSphere 7.xUpgrade to vSphere 8 firstStandard vCenter/ESXi upgrade

Upgrading from VCF 5.x to VCF 9.0

According to Upgrading Your VCF Management Domain to VCF 9.0, you can upgrade sequentially or skip-level to VCF 9.0 from VCF 5.0 or later. Anything older than 5.0 has to be brought to 5.0 or later first, for the management domain and all workload domains. The management domain upgrade is mandatory. Workload domains can be upgraded later as a Day-N task.

VCF 9 moves the Aria products into the platform. The order is:

  1. Management components: Aria Operations becomes VCF Operations, and a new VCF Operations fleet management appliance takes over from Aria Suite Lifecycle. Aria Suite Lifecycle has no upgrade path to 9.
  2. Core components through SDDC Manager: SDDC Manager, NSX, vCenter, ESX.
  3. Remaining components, such as VCF Operations collector and, optionally, VCF Automation.

Some products cannot be upgraded in place. VMware Identity Manager has no upgrade path, so you deploy VCF Identity Broker instead. Aria Operations for Logs has no upgrade path either, so you redeploy VCF Operations for logs 9.0. If you have Aria Automation 8.x, keep Aria Suite Lifecycle 8.x and VMware Identity Manager running, because Identity Manager is still an authentication source for VCF Automation 9.0.

vSphere 8 to VCF 9: converge instead of upgrade

A vSphere 8 environment that was never a VCF instance has no SDDC Manager to upgrade. Instead, the VCF Installer adopts the existing vCenter and hosts as the new management domain and deploys whatever is missing. Minimum versions for converging a vCenter instance and ESX hosts:

TargetvCenterESXExisting NSX
VCF 9.0.09.09.0Not converged; a new NSX 9.0 is deployed
VCF 9.0.1 / 9.0.28.0 Update 3 (9.0 if no NSX is registered)8.0 Update 14.2.1 or later, without Enhanced Linked Mode

Before converging, check the supported configurations. The cluster needs VDS 8.0 or later and shared storage accessible from all hosts. You need at least 3 hosts with vSAN, or 2 with external storage, for a simple deployment, and at least 4 hosts for high availability. Broadcom also lets you upgrade standalone vCenter, ESX and Aria Operations to 9.0 on their own first. That is enough to evaluate 9.0 for up to 90 days before licensing through VCF Operations.

VVF and vSphere 7: what changes

  • VVF 5.x / vSphere 8: converge to vSphere Foundation 9 in the documented convergence order. KB 437250 covers vCenter 8.x, Aria Operations 8.x, VVF 5.x and VCF 5.x as sources for VVF/VCF 9.0 and 9.1.
  • vSphere 7: vCenter 9.0 supports upgrades from vCenter 8.0 and later and cannot manage ESX 7.0 or earlier hosts. Plan a 7 → 8 → 9 sequence, and check VMFS5 datastores, which cannot be upgraded in place to VMFS6.
  • Licensing: 9.x uses subscription license files managed in VCF Operations. KB 437250 states that 8.x license keys cannot be upgraded to 9.x.

Rules that break upgrade plans

  • Forward-only: VCF only lets you upgrade to a target version released after your current version. A recently patched 5.2 or 8.0 U3 build may be newer than the 9.x release you are aiming for, so check release dates in the Interoperability Matrix.
  • Management domain first: workload domains stay on 5.x until you upgrade them as Day-N.
  • Prechecks on integrated products: an old Aria Operations for Networks (vRNI) registration can block the SDDC Manager upgrade. See VCF 9.x upgrade stuck on vRNI SSL thumbprint validation.
  • SDDC Manager UI deprecation: 9.0 still upgrades core components from the SDDC Manager UI. After that, lifecycle management moves to VCF Operations.

If you are planning a greenfield lab alongside the upgrade, see Deploying a Minimal VCF 9.0 Lab.

Further reading

Scaling ESX DCUI & Console for Compressed…

Scaling ESX DCUI & Console for Compressed Screen Resolution

Scaling ESX DCUI & Console for Compressed…

I recently shared my VMware Cloud Foundation (VCF) 9.0 Mini Rack build, which includes the use of a tiny 7″ LCD screen for direct console access to the ESX hosts. Since the LCD is optimizing for the 10″ inch rack, it is also using a non-standard resolution of 1280×400, which can make the output […]

Broadcom Social Media Advocacy

Quick Tip – Deploy OVF/OVA with Basic…

Quick Tip – Deploy OVF/OVA with Basic Authentication Endpoint

Quick Tip – Deploy OVF/OVA with Basic…

As part of rebuilding my VMware Cloud Foundation (VCF) 9.0.1 environment, a refresh of my VCF (Offline) Software Depot was needed to include the latest 9.0.1 binaries including various OVF/OVAs like Data Services Manager (DSM). My VCF Software Depot has been configured with basic authentication, which is a requirement for the VCF Installer. This meant […]

Broadcom Social Media Advocacy

VCF Operations Fleet Manager 9.0 Upgrade to…

VCF Operations Fleet Manager 9.0 Upgrade to…

With VCF 9.0.1 dropping earlier this week I thought I’d upgrade my lab, I’m started with VCF Operations Fleet Manager. Logging into VCF Operations – Fleet Management – Lifecycle – VCF Management, you’ll notice right away there is a new version available.From here I went to Binary Management […]

Broadcom Social Media Advocacy

Automating VCF Operations Active Directory over…

Automating VCF Operations Active Directory over…

When VMware Cloud Foundation (VCF) Single-Sign On (SSO) is configured with the Active Directory over LDAP Identity Provider, VCF Operations will periodically synchronize the users and groups from your identity source. By default, the sync frequency will default to once per week but it is user […]

Broadcom Social Media Advocacy

How Many VCF Instances Can a VCF Fleet Support? VCF 9.0 Fleet Sizing Explained

VCF 9.0 has no fixed instances-per-fleet maximum. The limit is VCF Operations capacity in objects and metrics. Here is how to size a VCF fleet.

Short answer: Broadcom does not publish a fixed maximum number of VCF instances per VCF fleet in VCF 9.0. A fleet is defined by one VCF Operations and one VCF Automation deployment, so the practical limit is how many objects and metrics that VCF Operations cluster can collect. The largest supported cluster, 12 Extra Large nodes, handles about 1,056,000 objects and 126 million metrics. Network latency and your availability requirements set the other limits.

VCF fleet vs VCF instance in VCF 9.0

VCF 9.0 adds a level above the instance. The hierarchy is VCF private cloud, then fleet, then instance, then domain, then cluster. The VCF fleet deployment models page in the design guide splits the components like this:

ScopeComponents
Per VCF fleet (one set)VCF Operations (monitoring, logs, licensing, fleet lifecycle, certificates, passwords, identity), VCF Automation
Per VCF instanceManagement domain with SDDC Manager, vCenter and NSX Manager, optional VI workload domains, the instance’s vSphere clusters, storage and networking

The VCF Installer either starts a new fleet, where the first instance also hosts the fleet management components, or adds a new VCF instance to an existing fleet. A fleet can also include standalone vCenter instances that VCF Operations manages.

Why there is no single “instances per fleet” maximum

Each instance is self-contained: its own SDDC Manager, vCenters and NSX. What all instances in a fleet share is the fleet layer. VCF Operations does not size itself by instances or hosts. It sizes by the number of objects (VMs, hosts, datastores, NSX objects and so on) and metrics it collects. Ten small instances and two very large ones can therefore put the same load on the fleet. That is why the limit is expressed in objects and metrics rather than as an instance count.

VCF Operations 9.0 sizing that bounds a fleet

Values from Broadcom KB 397782 – VCF Operations 9.0 Sizing Guidelines (standard, non-extended cluster):

Node sizevCPU / RAMMax nodesMax objects (cluster)Max metrics (cluster)
Extra Small2 / 8 GB1700140,000
Small4 / 16 GB212,0002,800,000
Medium8 / 32 GB8136,00032,000,000
Large16 / 48 GB16576,00081,600,000
Extra Large24 / 128 GB121,056,000126,000,000

Notes from the same KB:

  • All nodes in a multi-node cluster must be the same size.
  • Per-node object capacity in a multi-node cluster is lower than for a single node. For Extra Large it is 88,000 per node, compared with 100,000 for a single node.
  • An adapter instance (for example a large vCenter) that exceeds a node collector’s capacity can be pinned to a Cloud Proxy. Each node supports up to 200 Cloud Proxies on Large and Extra Large configurations.
  • Latency limits apply between analytics nodes and between nodes and Cloud Proxies. A fleet spread across distant regions has to stay within them.

How to plan the number of instances in a fleet

  1. Estimate objects and metrics per instance: hosts, VMs, datastores and NSX objects, plus any management packs you enable. Broadcom’s sizing worksheet referenced in the KB does this.
  2. Add up all planned instances and standalone vCenters and pick a VCF Operations size with headroom for growth.
  3. Check the latency and bandwidth from every site to the VCF Operations nodes and Cloud Proxies.
  4. Decide on fault domains. Because one fleet means one VCF Operations and one VCF Automation, compliance, tenancy or DR boundaries can be reasons to run several smaller fleets instead of one large one.

For a lab, one fleet with one instance is enough. See deploying a minimal VCF 9.0 lab. For licensing and edition differences, see the vSphere 9.0 product line comparison.

Further reading

Support for nConnect Feature Added in ESXi NFS Client

Use nConnect in ESXi 8.0 U1+ for several TCP connections per NFSv3 datastore: esxcli syntax, limits (4/8/256), runtime changes in 8.0 U2 and how to check.

Short answer: since ESXi 8.0 Update 1 the NFS client supports nConnect: several TCP connections per NFSv3 datastore. Mount with esxcli storage nfs add ... -c <connections> (default 1, up to 4 per datastore, or 8 with /NFS/MaxConnectionsPerDatastore, and 256 per host in total). Since 8.0 Update 2 you can change it on a mounted datastore with esxcli storage nfs param set -v <label> -c <n>. It’s NFSv3 only and CLI only.

With the release of VMware ESXi 8.0 Update 1, VMware quietly introduced nConnect support in the NFS client — a long-awaited enhancement for environments using NFSv3 datastores.

The nConnect feature allows administrators to establish multiple parallel TCP connections per NFS datastore, significantly improving throughput, resiliency, and traffic isolation.

This guide walks you through how to configure nConnect, supported versions, and key operational tips.

🆕 What’s New in ESXi 8.0U1

From ESXi 8.0U1 onwards, a new parameter -c (number of connections) is available in esxcli when mounting NFSv3 datastores.

esxcli storage nfs add \
  -H <host> \
  -v <volume-label> \
  -s <remote_share> \
  -c <number_of_connections>

Key points:

  • Supported for NFSv3 only (NFSv4.1 not yet supported).
  • Currently available only via CLI, not via vCenter UI.
  • Host Profiles don’t officially support it yet (though there’s a workaround, see below).
  • Default max connections per datastore: 4
  • Adjustable up to 8 connections via advanced option:
esxcfg-advcfg -s 8 /NFS/MaxConnectionsPerDatastore

⚠️ The total number of connections across all mounted NFSv3 datastores is capped at 256.

Runtime Scaling in ESXi 8.0U2

The nConnect feature got a useful upgrade in VMware ESXi 8.0 Update 2.
You can now increase or decrease the number of TCP connections on an existing NFSv3 mount without unmounting the datastore:

esxcli storage nfs param set \
  -v <volume-label> \
  -c <number_of_connections>

This flexibility helps admins fine-tune NFS performance on the fly — for example, ramping up connections during heavy backup windows or scaling down during low I/O periods.

Note: In both 8.0U1 and 8.0U2, the default number of connections remains 1.

How to Check Active NFS Connections

To view the number of RPC clients currently used by mounted NFSv3 datastores, run:

vsish -e get /vmkModules/nfsclient/info

Sample output:

NFS Connections Info {
mountedVolumes:10
totalConnections 40
}

This command is particularly handy for performance troubleshooting and capacity planning.

🧪 Experimental: Host Profile Integration

Although official Host Profile support for nConnect is not yet released, VMware added an experimental field named “Number of TCP connections” in:

Host Profile → Storage Configuration → NFS

What it does:
When you set a value here, ESXi will use that number of parallel TCP sessions when mounting the NFS datastore during profile remediation.

This is ideal for consistent deployment across multiple hosts in larger clusters.

Best Practices & Tips

  • Start with 4 connections and scale up as needed after performance testing.
  • Keep the 256 connection limit in mind when mounting multiple datastores.
  • Use consistent configuration across hosts to avoid mount mismatches.
  • Monitor performance metrics via VMware vSphere or CLI.
  • If you’re using Host Profiles, experiment carefully with the TCP connection parameter.

Summary

The nConnect feature in ESXi’s NFS client marks a major step forward in improving NFSv3 performance and scalability in enterprise environments.

Whether you’re running backup workloads, large VMs, or latency-sensitive applications, multiple TCP connections per NFS datastore can make a measurable difference.

If you’re upgrading to VMware ESXi 8.0U1 or later, nConnect is worth testing and adopting in your storage strategy.

Links

MS-A2 VCF 9.0 Lab: Deploying VMware Private AI…

MS-A2 VCF 9.0 Lab: Deploying VMware Private AI…

In this blog post, we will now deploy an instance of VMware Private AI Services (PAIS) that will use the Vector Database that was provisioned earlier and the OIDC Client Application that we had also setup earlier using the Authentik Identity Provider (IdP). Requirements: VCF Automation (VCFA) […]

Broadcom Social Media Advocacy

MS-A2 VCF 9.0 Lab: Configuring Authentik…

MS-A2 VCF 9.0 Lab: Configuring Authentik…

In this blog post, we will walk through the configuration of an Identity Provider (IdP) and create an OIDC Public Application Client, which will be used to grant access to a PAIS deployment. As of this blog post, the instructions for setting up the required IdP will be based on Authentik (which […]

Broadcom Social Media Advocacy