Lab: Create & Apply Cluster Image

VCF 9.0Intermediatevcp-foundation⏱ 120 min

Objectives

Lab: Create & Apply Cluster Image

Prerequisites

VCF lab environment deployed and operational

Lab Environment

Standard VCF lab environment for vSphere Foundation 9.0 Administrator

Tasks

Task 1 Lab: Create & Apply Cluster Image

Cluster image management via vLCM is the standard lifecycle approach in vSphere 9.0. Understanding image composition (base ESXi + vendor add-on + components) and the compliance check workflow is essential for Day-2 operations.

Step 1

Download ESXi 9.0 ISO from vmware.com

Step 2

Download OEM image: Dell iDRAC Lifecycle Controller → OMIMSWAC driver bundle

Step 3

vCenter UI → Clusters → Lifecycle → Images → Add image → Select ISO + OEM image

Step 4

Tag image with version (e.g., "ESXi-9.0-2024Q1-Dell")

Step 5

Set as desired state for cluster

Step 6

Run health check (Precheck): Identifies incompatibilities before remediation

Step 7

Remediate one host in preview mode (HA suspended host, updates, restores HA)

Step 8

Verify image version: SSH to ESXi → vmware -v (should show new build number)

Validation Gate

Check: After creating and applying a cluster image: verify all hosts show 'Compliant' status, no hosts are in maintenance mode, and vSAN health is green

Expected: All hosts compliant with the cluster image. No hosts pending reboot. vSAN health shows all checks green after rolling update completes.

Common Errors

Creating a cluster image without the vendor add-on (firmware/drivers)

Fix: A cluster image with only base ESXi lacks vendor-specific drivers and firmware. VMs may run but hardware features (RAID controller, NIC offloads, GPU passthrough) won't function correctly. Always include the vendor add-on (e.g., Dell OpenManage Integration, HPE Agentless Management) in the cluster image.

Applying an image update during production hours without pre-check

Fix: Image remediation reboots hosts sequentially (rolling update). If DRS cannot evacuate VMs from a host (insufficient capacity), remediation stalls. Always run: (1) compliance check first, (2) verify DRS can evacuate each host, (3) schedule during maintenance window. Pre-check: vLCM → Check Compliance → Review remediation impact.

Not validating the cluster image in a test cluster first

Fix: A new cluster image (especially with firmware updates) can cause boot failures on incompatible hardware. Always validate the image on a non-production cluster or single host first. In Holodeck, test on a nested cluster before applying to production-simulated domains.

Mixing image-managed and baseline-managed clusters

Fix: vSphere 9.0 removed baseline-based patching. All clusters use image-based management via vLCM. If upgrading from vSphere 7.x/8.x, convert baseline-managed clusters to image-managed during the upgrade. Mixed management modes are not supported.

Final Validation

Lab completed successfully

✓ All steps completed → No errors observed

Cleanup / Restore

• Revert to snapshot if needed

Design Reflection (VCDX)

Lifecycle management is a Day-2 operational requirement in VCDX designs. Panelists test whether you planned for patching cadence, image composition, and the operational impact of rolling updates on cluster availability.

Requirements

Create cluster images with base ESXi + vendor add-on + components
Validate images before production deployment
Schedule remediation during maintenance windows

Constraints

vSphere 9.0 requires image-based management only (no baselines)
Rolling updates require DRS capacity to evacuate each host
Firmware updates may require additional reboot cycles

Assumptions

Vendor add-on packages are available and compatible
DRS has sufficient headroom to evacuate VMs during rolling update

Risks

Boot failure from incompatible firmware in cluster image
Stalled remediation from insufficient DRS evacuation capacity

⚠ Known Pitfalls (from Community KB)

Applying cluster image updates without pre-checking DRS evacuation capacity — if a host can't be evacuated, the update stalls mid-process.

Forgetting vendor add-on in the cluster image — base ESXi without vendor drivers causes hardware feature degradation.

Mark complete