13 — Demo Day Guide

Pre-Demo Checklist

Complete this checklist at least 30 minutes before the session begins:

Environment Health

• [ ] RDP to hv-host01 (10.250.2.5) successfully
• [ ] All nested VMs are Running: Get-VM | Select-Object Name, State
• [ ] Cluster is healthy: Get-ClusterNode -Cluster hvlab-clus01 | Select-Object Name, State — all Up
• [ ] All CSVs Online: Get-ClusterSharedVolume -Cluster hvlab-clus01 | Select-Object Name, State
• [ ] Cloud Witness reachable: Get-ClusterQuorum -Cluster hvlab-clus01

WAC vMode (Session 1)

• [ ] https://10.250.2.6 loads without certificate error
• [ ] All 4 cluster nodes show as Connected in WAC vMode
• [ ] At least 2 demo VMs are running on different cluster nodes (verify node placement)
• [ ] PostgreSQL service running on hvwac01

SCVMM (Session 2)

• [ ] SCVMM console connects to hvscvmm01.azrl.mgmt:8100
• [ ] hvlab-clus01 shows OK in SCVMM Fabric view
• [ ] At least 1 VM template available in library
• [ ] SQL Server service running on hvscvmm01

Connectivity

• [ ] From on-premises machine: Test-Connection 10.250.2.6 -Count 2 succeeds
• [ ] From on-premises machine: Test-Connection 10.250.2.7 -Count 2 succeeds
• [ ] Azure Local cluster node can reach 10.250.2.6

---

Restoring the DEMO-READY Checkpoint

If anything is broken or the environment needs to be reset to a clean state:

Option 1 — GitHub Actions (Preferred)

1. Navigate to: GitHub → Repository → Actions → hvlab-07-demo-reset
2. Click Run workflow → select branch main
3. Click Run workflow
4. Wait approximately 20 minutes
5. Re-run the pre-demo checklist above

Option 2 — Manual Reset from Host VM

# On hv-host01 — restore all VMs to DEMO-READY checkpoint
$vms = Get-VM
foreach ($vm in $vms) {
    # Stop VM if running
    if ($vm.State -eq "Running") {
        Stop-VM -VM $vm -Force
    }
    # Restore checkpoint
    $checkpoint = Get-VMSnapshot -VM $vm -Name "DEMO-READY"
    if ($checkpoint) {
        Restore-VMSnapshot -VMSnapshot $checkpoint -Confirm:$false
        Start-VM -VM $vm
        Write-Host "Restored: $($vm.Name)"
    } else {
        Write-Host "⚠ No DEMO-READY checkpoint found for $($vm.Name)" -ForegroundColor Yellow
    }
}

Expected Post-Reset State

After restoration:

VM	State	Notes
hvdc01	Running	DC services started
hviscsi01	Running	iSCSI target serving 4 LUNs
hvnode01-04	Running	Cluster nodes Up, CSVs Online
hvwac01	Running	WAC vMode web UI accessible at 10.250.2.6
hvscvmm01	Running	SCVMM console accessible at 10.250.2.7:8100

---

Session 1 Walkthrough — WAC Virtualization Mode

Duration: ~25 minutes
URL: https://10.250.2.6

Talking Points Opening (2 min)

• WAC Virtualization Mode is not WAC Administration Mode — completely different product
• Purpose-built for Hyper-V fabric management: stateful agents, PostgreSQL backend
• WS2025 only during preview
• Today's demo: manage a 4-node Hyper-V cluster the same way an Azure Local operator would

Demo Step 1 — Cluster Overview (3 min)

1. Open https://10.250.2.6 from your demo machine
2. Log in as AZRL\<admin-account>
3. Navigate to Cluster → hvlab-clus01
4. Show the Overview tab: node health, CSV status, cluster events
5. Talking point: Real-time agent data, not polling — the health view updates without page refresh

Demo Step 2 — Host Health Dashboard (4 min)

1. Navigate to Hosts → click hvnode01
2. Show CPU utilization graph (real-time)
3. Show Memory pressure and NUMA topology
4. Show Network adapter throughput across all vNICs
5. Show Storage — iSCSI paths and MPIO status
6. Talking point: This is the same depth of host telemetry that Azure Monitor provides for Azure VMs — now available for on-prem Hyper-V through the WAC vMode agent

Demo Step 3 — Live Migration (8 min)

1. Navigate to Virtual Machines — show VMs distributed across nodes
2. Select a VM on hvnode01
3. Click Move → select hvnode03 as destination
4. Click Move and watch the status bar

Expected behavior:
  - VM disappears from hvnode01 VM list
  - VM appears on hvnode03 within ~30 seconds
  - No guest OS restart — live migration is near-zero downtime

5. After migration completes: navigate back to Hosts → hvnode01 to show reduced load
6. Talking point: Kerberos constrained delegation is what makes this work — no password prompts, the cluster nodes trust each other via AD delegation

Demo Step 4 — Create a VM (8 min)

1. Navigate to Virtual Machines → New VM
2. Configure:
   • Name: demo-workload-01
   • Host: hvnode02
   • vCPU: 2
   • RAM: 4 GB
   • Storage: CSV-Vol2 (show the CSV selection)
   • Network: Attach to vSwitch-Workload
3. Click Create
4. Watch VM appear in the VM list, boot up
5. Click Connect to open the in-browser console
6. Talking point: No VMConnect needed — browser-native console is part of WAC vMode

---

Session 2 Walkthrough — SCVMM 2025

Duration: ~25 minutes
Console: SCVMM Admin Console → hvscvmm01.azrl.mgmt

Talking Points Opening (2 min)

• SCVMM 2025 for enterprise-scale Hyper-V management
• Logical networks abstract physical switch configuration
• Integration with System Center Suite and Azure

Demo Step 1 — Fabric View (4 min)

1. Open SCVMM Admin Console → Fabric → Servers
2. Expand hvlab-clus01 — show all 4 nodes with OverallState = OK
3. Click each node to show vCPU, memory, and storage capacity
4. Navigate to Storage — show the 3 CSVs and their capacity/usage
5. Talking point: SCVMM aggregates fabric inventory — SCVMM knows every NIC, every vSwitch, every storage path across all nodes

Demo Step 2 — VM Deployment from Template (8 min)

1. Navigate to VMs and Services → Create Virtual Machine
2. Select an existing VM template (WS2022-Base or similar from the library)
3. Configure:
   • VM Name: scvmm-demo-vm-01
   • Destination: Cloud/Cluster hvlab-clus01
   • SCVMM will auto-select the least-loaded node
4. Show Intelligent Placement — SCVMM scores each node and picks the best
5. Click Deploy
6. Show deployment progress in the Jobs pane

Demo Step 3 — Maintenance Mode (6 min)

1. Navigate to Fabric → Servers → right-click hvnode04
2. Select Start Maintenance Mode
3. Choose Move all virtual machines to other nodes → click OK
4. Watch the Jobs pane: each VM live-migrates away from hvnode04
5. Show hvnode04 in Maintenance status
6. Right-click hvnode04 → Stop Maintenance Mode
7. Talking point: This is the same workflow used before patching in production — drain the node, patch, return to service

Demo Step 4 — Azure Integration (5 min)

1. Navigate to Settings → Azure Management
2. Show the subscription 00cd4357-ed45-4efb-bee0-10c467ff994b connected
3. Show on-premises VMs alongside Azure VM inventory
4. Talking point: Unified inventory — one pane for hybrid workloads. SCVMM 2025 extends this to Azure Local cluster integration for a single management plane across on-premises and Azure

---

What to Do If Something Breaks

WAC vMode Web UI Unreachable

# 1. Test from host VM
Test-NetConnection -ComputerName "172.16.10.30" -Port 443

# 2. RDP to hvwac01 via host VM (172.16.10.30)
# 3. Check services
Get-Service | Where-Object { $_.DisplayName -like "*Windows Admin Center*" -or
                              $_.Name -like "*postgres*" }
# 4. Start any stopped services
# 5. If TLS cert expired — see docs/09-wac-virtualization-mode.md troubleshooting

Fallback: Use WAC Administration Mode (installed on hv-host01 as backup) to at least show a Hyper-V management interface.

Cluster Node Not Responding

# From hv-host01
Restart-VM -Name "hvnode02" -Force
Start-Sleep 60
# Check it rejoined cluster
Get-ClusterNode -Cluster "hvlab-clus01" -Name "hvnode02"

Fallback: Continue demo with 3 nodes. Live migration and all scenarios still work.

iSCSI Disconnected (CSV Offline)

# Restart iSCSI initiator on affected node
Invoke-Command -ComputerName "hvnode02" -ScriptBlock {
    Restart-Service MSiSCSI -Force
    Start-Sleep 30
    Get-Disk | Where-Object BusType -eq "iSCSI" | Select-Object Number, HealthStatus
}
# If CSV comes back Online — continue demo
# If not — failover CSV ownership to another node
Move-ClusterSharedVolume -Name "CSV-Vol1" -Node "hvnode01"

SCVMM Console Can't Connect

# Test port from on-premises
Test-NetConnection -ComputerName "10.250.2.7" -Port 8100

# Restart SCVMM service on hvscvmm01
Invoke-Command -ComputerName "172.16.10.40" -ScriptBlock {
    Restart-Service vmmservice -Force
    Start-Sleep 30
    Get-Service vmmservice | Select-Object Name, Status
}

Fallback: Use PowerShell (Import-Module VirtualMachineManager) to demonstrate the same scenarios via script if the GUI is unavailable.

---

Backup Plan — Screenshots and Recording

If a complete environment failure occurs (host VM unreachable, networking down), use the pre-recorded materials:

Asset	Location	Contents
WAC vMode recording	scripts/demo/recordings/wac-vmode-full.mp4	Complete Session 1 walkthrough
SCVMM recording	scripts/demo/recordings/scvmm-full.mp4	Complete Session 2 walkthrough
Architecture screenshots	scripts/demo/screenshots/	Architecture diagrams, cluster health

Presenter note: If using the backup recording, narrate over it live. Focus the commentary on the architecture diagrams and talking points — the audience learns more from your explanation than from watching a perfect UI recording.

---

Post-Demo

After the demo session:

1. Debrief: Note any questions that came up for the FAQ / future sessions
2. Restore checkpoint (optional): Trigger hvlab-07-demo-reset.yml to clean up any VMs created during the demo
3. Deallocate the host VM (if not needed for more sessions):

az vm deallocate `
  --name "hv-host01" `
  --resource-group "rg-hvlab-mms26-eus-01"

Deallocation stops billing for compute (~$10/hour). Storage costs continue. Restart when needed — the DEMO-READY checkpoint will be preserved on the managed data disk.