Failover clustering: build a two-node cluster

Start DC01, SRV01 and NODE1. Work from DC01 as SAATLAB\adm.rahul in PowerShell as administrator unless a step says otherwise.

1Build NODE2 and add a witness LUN

Create NODE2 exactly like NODE1 in Lesson 36 (two NICs on LabNet and LabNet2), with addresses 10.0.0.42 and 10.0.1.42, join it to the domain in the Servers OU, install Multipath-IO and run Enable-MSDSMAutomaticClaim -BusType iSCSI.

On SRV01, allow NODE2 and add a 1 GB witness disk to the same target:

Set-IscsiServerTarget -TargetName NodeTarget -InitiatorIds 'IQN:iqn.1991-05.com.microsoft:node1.saatlab.internal', 'IQN:iqn.1991-05.com.microsoft:node2.saatlab.internal'
New-IscsiVirtualDisk -Path D:\iSCSI\Witness.vhdx -SizeBytes 1GB | Out-Null
Add-IscsiVirtualDiskTargetMapping -TargetName NodeTarget -Path D:\iSCSI\Witness.vhdx

On NODE2, connect both paths (and on NODE1, refresh so it sees the new LUN):

Set-Service MSiSCSI -StartupType Automatic; Start-Service MSiSCSI
New-IscsiTargetPortal -TargetPortalAddress 10.0.0.20 -InitiatorPortalAddress 10.0.0.42 | Out-Null
New-IscsiTargetPortal -TargetPortalAddress 10.0.1.20 -InitiatorPortalAddress 10.0.1.42 | Out-Null
$iqn = (Get-IscsiTarget).NodeAddress
Connect-IscsiTarget -NodeAddress $iqn -TargetPortalAddress 10.0.0.20 -InitiatorPortalAddress 10.0.0.42 -IsPersistent $true -IsMultipathEnabled $true | Out-Null
Connect-IscsiTarget -NodeAddress $iqn -TargetPortalAddress 10.0.1.20 -InitiatorPortalAddress 10.0.1.42 -IsPersistent $true -IsMultipathEnabled $true | Out-Null
Get-Disk | Where-Object BusType -eq 'iSCSI' | Select-Object Number, Size, OperationalStatus

You should see something like this on NODE2:

Number        Size OperationalStatus
------        ---- -----------------
     1 10737418240 Offline
     2  1073741824 Offline

The disks are offline on NODE2 because Windows doesn’t bring shared disks online automatically: two servers writing to the same NTFS volume outside a cluster would corrupt it. On NODE1, initialize the 1 GB disk as GPT with one NTFS partition, then take both iSCSI disks offline so the cluster can claim them.

✅ Checkpoint: Both nodes see the 10 GB and 1 GB iSCSI disks over two paths.

2Install the feature and validate

On DC01:

Install-WindowsFeature RSAT-Clustering-PowerShell, RSAT-Clustering-Mgmt | Out-Null
Invoke-Command -ComputerName NODE1, NODE2 -ScriptBlock { Install-WindowsFeature Failover-Clustering -IncludeManagementTools | Select-Object Success }
Test-Cluster -Node NODE1, NODE2 -ReportName C:\Temp\Cluster-Validation

You should see something like this:

WARNING: Network - Validate Network Communication: The test reported some warnings..
WARNING:
Test Result:
HadUnselectedTests, ClusterConditionallyApproved
Testing has completed for the tests you selected. You should review the warnings in the Report.  A cluster solution is
supported by Microsoft only if you run all cluster validation tests, and all tests succeed (with or without warnings).

Mode                 LastWriteTime         Length Name
----                 -------------         ------ ----
-a----        20/10/2026     00:41         702114 Cluster-Validation.htm

Open the HTML report. Warnings are common in labs (for example, the storage network has no gateway, or both NICs are on similar networks). Failures in the Storage or Network categories must be fixed before you create the cluster.

Microsoft supports a cluster only if validation passes, with or without warnings.

✅ Checkpoint: Validation finished with “ClusterConditionallyApproved” (warnings only) and a report in C:\Temp.

3Create the cluster and set quorum

On DC01:

New-Cluster -Name CLUSTER1 -Node NODE1, NODE2 -StaticAddress 10.0.0.45 -NoStorage | Out-Null
Get-ClusterAvailableDisk -Cluster CLUSTER1 | Add-ClusterDisk -Cluster CLUSTER1 | Select-Object Name, State, OwnerNode
Set-ClusterQuorum -Cluster CLUSTER1 -DiskWitness 'Cluster Disk 2' | Select-Object Cluster, QuorumResource
Get-ClusterNode -Cluster CLUSTER1 | Select-Object Name, State, NodeWeight

You should see something like this:

Name           State  OwnerNode
----           -----  ---------
Cluster Disk 1 Online NODE1
Cluster Disk 2 Online NODE1

Cluster  QuorumResource
-------  --------------
CLUSTER1 Cluster Disk 2

Name  State NodeWeight
----  ----- ----------
NODE1 Up             1
NODE2 Up             1

Check the 1 GB disk really is Cluster Disk 2 (look at its size in Failover Cluster Manager) before making it the witness. Two node votes plus the witness make three, so the cluster keeps running with any two of them. The cluster also created a computer object, CLUSTER1, next to the nodes in AD.

✅ Checkpoint: CLUSTER1 has two Up nodes, two cluster disks, and Cluster Disk 2 as its disk witness.

4Make a file server highly available

On DC01:

Add-ClusterFileServerRole -Cluster CLUSTER1 -Name FS-CLUSTER -Storage 'Cluster Disk 1' -StaticAddress 10.0.0.46 | Out-Null
$owner = (Get-ClusterGroup -Cluster CLUSTER1 -Name FS-CLUSTER).OwnerNode.Name
Invoke-Command -ComputerName $owner -ScriptBlock {
    New-Item -ItemType Directory S:\Projects -Force | Out-Null
    New-SmbShare -Name Projects -Path S:\Projects -ScopeName FS-CLUSTER -ContinuouslyAvailable $true -FullAccess 'SAATLAB\Domain Admins' -ChangeAccess 'SAATLAB\Domain Users' | Out-Null
}
Get-ClusterGroup -Cluster CLUSTER1 | Select-Object Name, OwnerNode, State

You should see something like this:

Name              OwnerNode State
----              --------- -----
Available Storage NODE1     Online
Cluster Group     NODE1     Online
FS-CLUSTER        NODE1     Online

The 10 GB disk kept the drive letter S: from Lesson 36 on whichever node owns it. Users and scripts connect to \\FS-CLUSTER\Projects, never to a node’s own name. The share belongs to the role, so it follows the role from node to node.

✅ Checkpoint: FS-CLUSTER is Online and the Projects share is reachable as \\FS-CLUSTER\Projects from CL01.

5Move the role between nodes

On DC01:

Measure-Command { Move-ClusterGroup -Cluster CLUSTER1 -Name FS-CLUSTER -Node NODE2 | Out-Null } | Select-Object TotalSeconds
Get-ClusterGroup -Cluster CLUSTER1 -Name FS-CLUSTER | Select-Object Name, OwnerNode, State

You should see something like this:

TotalSeconds
------------
      4.8732

Name       OwnerNode State
----       --------- -----
FS-CLUSTER NODE2     Online

This planned move is what Cluster-Aware Updating does for every role before it patches a node. Copy a large file to the share from CL01 while you move the role back to NODE1: with a continuously available share, the copy pauses briefly and carries on.

✅ Checkpoint: FS-CLUSTER moved to NODE2 in a few seconds and stayed Online.