ConfiguringWindowsServer2016andNanoServerFailoverClusters (3288 CD WS)

Dell Storage SC Series: Configuring Microsoft
Windows Server 2016 and Nano Server

Failover Clusters
Dell Storage Engineering
October, 2016
A Dell EMC Configuration Guide

Revisions
Date Description
10/27/2016 Initial release
Acknowledgements
This document was authored by Kris Piepho, a member of the Dell EMC Storage Applications Engineering
team.
The information in this publication is provided as is. Dell Inc. makes no representations or warranties of any kind with respect to the information in this
publication, and specifically disclaims implied warranties of merchantability or fitness for a particular purpose.
Use, copying and distribution of any software described in this publication requires an applicable software license.
Copyright 2016 Dell Inc. or its subsidiaries. All Rights Reserved. Dell, EMC and other trademarks are trademarks of Dell Inc. or its subsidiaries. Other
trademarks may be the property of their respective owners. Published in the USA [10/27/2016] [Configuration Guide] [3288-CD-WS].
Dell EMC believes the information in this document is accurate as of its publication date. The information is subject to change without notice.
2 Dell Storage SC Series: Configuring Microsoft Windows Server 2016 and Nano Server Failover Clusters | 3288-CD-WS
Table of contents
Revisions.............................................................................................................................................................................2
Acknowledgements .............................................................................................................................................................2
1 Introduction ...................................................................................................................................................................5
1.1 Audience .............................................................................................................................................................5
1.2 Prerequisites .......................................................................................................................................................5
2 Introduction to Microsoft failover clustering ..................................................................................................................6
2.1 Overview .............................................................................................................................................................6
2.2 Active/Active clusters ..........................................................................................................................................6
2.3 Active/Passive clusters .......................................................................................................................................6
2.4 Cluster quorum ...................................................................................................................................................6
2.4.1 Quorum configurations .......................................................................................................................................7
2.5 Cluster shared volumes ......................................................................................................................................7
2.5.1 Cluster shared volumes cache ...........................................................................................................................8
2.6 Using MPIO with Microsoft failover clusters .......................................................................................................8
2.7 High availability ...................................................................................................................................................8
2.8 Cluster validate ...................................................................................................................................................8
3 SC Series array configuration ....................................................................................................................................10
3.1 Creating a cluster server object in Dell Storage Manager ................................................................................10
3.2 Volume considerations .....................................................................................................................................11
3.3 Mapping SC Series Volumes to clustered server object in DSM .....................................................................11
4 Windows Server configuration....................................................................................................................................12
4.1 Network configuration .......................................................................................................................................12
4.1.1 Public interface .................................................................................................................................................12
4.1.2 Private network .................................................................................................................................................12
4.1.3 Cluster NIC adapter assignment and NIC teaming ..........................................................................................13
4.2 Installing the failover clustering feature ............................................................................................................13
4.2.1 Installing the failover clustering feature on Windows Server 2016 ...................................................................13
4.2.2 Installing the failover clustering feature on Nano Server ..................................................................................14
4.3 Preparing disks for a new cluster .....................................................................................................................16
4.3.1 Preparing disks on Windows Server 2016 .......................................................................................................16
4.3.2 Preparing disks on a Nano Server....................................................................................................................17
5 Create a failover cluster .............................................................................................................................................19
6 Adding disks to a cluster ............................................................................................................................................21
6.1 Creating a cluster shared volume (CSV) from available cluster storage..........................................................22
7 Failover clustering and Hyper-V .................................................................................................................................25
8 Administrative tasks ...................................................................................................................................................26
8.1 Testing failover .................................................................................................................................................26
8.1.1 Moving roles .....................................................................................................................................................26
8.2 Cluster-Aware updating ....................................................................................................................................27
8.3 Windows PowerShell ........................................................................................................................................28
8.3.1 Dell Storage PowerShell SDK command set ...................................................................................................28
A Additional resources ...................................................................................................................................................29
A.1 Technical support and resources .....................................................................................................................29
A.2 Related documentation .....................................................................................................................................29
1 Introduction
Dell Storage SC arrays are designed from the ground up with redundancies to avoid downtime for such
events as component failures, maintenance, upgrades and expansion. These redundancies benefit clustered
Windows Servers and their respective workloads. Windows Server also includes similar features. When
SC Series arrays are integrated with clustered Windows Servers, their feature sets complement each other.
There are several other means of presenting storage to a Windows Server 2016 or Nano Server cluster, such
as traditional on-board storage in physical servers. Another is Storage Spaces; which Microsoft introduced
with the initial release of Windows Server 2012. Storage Spaces Direct and Storage Replica were introduced
with Server 2016. While Storage Spaces, Storage Spaces Direct, and Storage Replica offer similar basic
features as traditional SAN arrays, the SC series SAN provides a much more powerful and complete set of
integrations, management and monitoring tools that are not available with Storage Spaces or Storage Replica
alone.
1.1 Audience
This document was written for technology professionals interested in learning more about configuring a
Microsoft Windows Server 2016 or Nano Server Failover Cluster with SC Series SAN arrays.
1.2 Prerequisites
Understanding the material in this document requires advanced working knowledge of the following:
Installation and configuration of Windows Server 2016 and Nano Server

Configuration and operation of the SC Series arrays
Operation of Dell Storage Manager (DSM) software
Note: Unless otherwise specified, the recommendations and examples in this document apply to both
Windows Server 2016 and Nano Server.
2 Introduction to Microsoft failover clustering
2.1 Overview
Microsoft Failover Clustering provides the capability to tie multiple Windows Server 2016 or Nano Server
servers together to offer high availability and scalability for business-critical applications such as Microsoft
Exchange, Hyper-V, Microsoft SQL Server and file servers. Clustering is designed to maintain data
integrity and provide failover support. Windows Server 2016 failover clustering can scale up 64 nodes in a
single cluster.
Improving on the previous release of failover clustering included with Windows Server 2012 R2, the 2016
version includes many new features and improvements. Some of these are cluster operating system rolling
upgrades, a new quorum witness hosted in Microsoft Azure, site-aware clusters, the ability to use every NIC
on a subnet for cluster traffic, and various improvements to Hyper-V Virtual Machines hosted on a cluster. In
addition, servers that function as cluster nodes can belong only to workgroups (and not joined to a domain),
or nodes in the same cluster can be joined to different domains. For a complete list of changes and new
features included in Windows Server 2016 failover clustering, please refer to Microsoft TechNet.
Failover clustering is included in both the Standard and Datacenter versions of Windows Server 2016 and
Nano Server.
2.2 Active/Active clusters

In active/active clusters, all nodes are active. In the event of a failover, the remaining active node takes on the
additional processing operations and cause a reduction in the overall performance of the cluster.
Active/passive cluster configurations are generally recommended over active/active configurations because
they often increase performance, availability and scalability. Microsoft Exchange 2016 and SQL Server 2016
support a configuration that falls into the realm of what would be considered an active/active cluster
configuration. These particular configurations will not be discussed as they are beyond the scope of this
document. For detailed information about Microsoft Exchange 2016 and SQL Server 2016 clustering, refer to
Dell TechCenter.
2.3 Active/Passive clusters

In active/passive clustering, the cluster includes active nodes and passive nodes. The passive nodes are only
used if an active node fails. Active/Passive clusters are commonly known as failover clusters. For example,
file and print environments use the active/passive cluster model since two nodes cannot own the same disk
resource at one time.
2.4 Cluster quorum

Put simply, cluster quorum is the collective status of all elements in the active cluster. Each element in the
cluster casts a single quorum vote which contains the element status. If the majority of quorum votes show
the cluster to be healthy, the cluster will start properly or continue running. Elements in the cluster include
server nodes, and can also include a quorum witness. A quorum witness is configured when the number of
nodes in the cluster cannot reach majority (such as an even number of cluster nodes) and works as the
determining vote if needed. The witness can be a shared disk or a file share that runs outside of the cluster.
2.4.1 Quorum configurations

In Windows Server 2016, four types of quorum configurations exist:
1. Node majority (no witness)

Only the cluster nodes have votes. Quorum is the majority of voting nodes.
2. Node majority with witness (disk or file share)

Cluster nodes as well as a witness have votes. Quorum is the majority of voting nodes plus the
witness vote.
3. No majority (disk witness only)

Only a disk witness has a vote. No nodes have votes. Quorum is determined by the state of the disk
witness. This is not a recommended configuration, as the witness disk becomes a single point of
failure for the cluster.
4. Cloud witness
Functions much like a node majority with witness quorum, except the file share witness is hosted in
Microsoft Azure.
Note: In clusters with an even number of nodes the default quorum configuration is node majority with
witness. In clusters with an odd number of nodes the default quorum configuration is node majority with no
witness.
In most cases, a disk witness should be used when a witness is required. A file share witness should only be
used in case shared storage between cluster nodes is not available.
Note: As mentioned above, Windows Server 2016 will automatically assign the proper cluster quorum
configuration based upon the number of nodes in the cluster. As a best practice, do not change the cluster
quorum configuration unless the number of nodes in the cluster has changed.
For more information about configuring and managing the quorum in a failover cluster, please refer to
Microsoft TechNet.
2.5 Cluster shared volumes

Originally Introduced in Windows Server 2008 R2 Failover Clustering, cluster shared volumes (CSVs) allow
all nodes in a cluster to simultaneously have read-write access to the same LUN that is formatted as an NTFS
or ReFS volume. Using CSVs, clustered roles can failover quickly from one node to another without requiring
a change in drive ownership, or dismounting and remounting a volume. In Windows Server 2008 R2 CSVs
were only allowed to be used with a Hyper-V virtual guest workload. Starting with Windows Server 2012 and
continuing on to Windows Server 2016, expanded CSV functionality allows CSVs to be utilized as file shares
with the scale-out file server role. Scale-out file shares can host application data, such as Microsoft SQL
Server and Hyper-V guests.
2.5.1 Cluster shared volumes cache
CSV cache is a failover clustering feature that uses system memory (RAM) as a write-through cache on
Cluster Shared Volumes. CSV cache is designed to be used with Hyper-V guest and scale-out file server
workloads. When properly configured, CSV cache can significantly improve performance of the workloads
hosted on the CSV.
Note: CSV cache is disabled by default on Windows Server 2016 and Nano Server clusters.
For more information about CSV cache, refer to TechNet.
2.6 Using MPIO with Microsoft failover clusters

Using MPIO with clustering is supported in both round-robin and failover-only configurations. In Windows
Server 2016, the MPIO load balance policy for each volume can be modified in Disk Management under the
MPIO tab of the volume properties. A default load balance policy for MPIO can also be set, and any new
volumes on the server will automatically be set to use the specified MPIO load balance policy. If the default
load balancing policy has not been set, Windows Server 2016 automatically defaults SC Series volumes to a
round robin MPIO configuration.
The use of MPIO with SC Series volumes is highly recommended. Refer to the Dell SC Series: Microsoft
Multipath I/O Best Practices Guide on Dell TechCenter for more information on configuring MPIO in Windows
Server 2016 and Nano Server.
2.7 High availability

In previous versions of Windows clustering, cluster groups were able to manage failover and determine how
failover was handled. In Windows Server 2008/R2 failover clustering, cluster groups were referred to as
services and applications. Starting with Windows Server 2012, highly available services and applications are
referred to as roles. Examples of a highly available cluster role might be a file server, DNS or a virtual
machine. It should be noted that in order to add a role to the cluster, that role or feature must already be
installed on each node of the cluster. For example, in order to add the File Server role to the cluster, the File
Server role must be installed from the Add Roles and Features Wizard on each node of the cluster.
2.8 Cluster validate

Historically, cluster configurations (Windows 2000 and Windows 2003) had to be certified through the
Windows Hardware Quality Lab (WHQL) in order to be approved and eligible for support by Microsoft.
Starting with Windows Server 2008, the Cluster Validate tool has been included with the Operating System to
test the functionality and compatibility of the servers and storage involved in a cluster. This tool verifies that
the storage meets the requirements (supports specific commands) to operate in a failover cluster. The end
user can run Cluster Validate and save the output as proof of supportability for the configuration.
As a best practice, perform Cluster Validate on any cluster configuration that uses SC Series arrays to ensure
the customer can receive Microsoft Premier Support should a cluster issue arise.
Cluster Validate is a wizard-driven tool in Failover Cluster Manager that can be run as part of configuring a
new cluster or at any time thereafter. The Cluster Validate tool assumes that the storage is attached and
accessible by all nodes participating in the test.
Note: Running Cluster Validate on a live Failover cluster with active resources may require a maintenance
window for all validation tests to run as some resources may temporarily go offline.
3 SC Series array configuration
Configuring an SC Series to present storage to a Microsoft failover cluster is a straightforward process.
Before continuing, be sure that all Windows Server 2016 or Nano Servers used as cluster nodes are properly
connected to the SC Series array and configured according to best practice guidelines detailed in the Dell SC
Series: Windows Server 2016 Best Practices Guide, located on TechCenter.
3.1 Creating a cluster server object in Dell Storage Manager

Microsoft failover clusters require that storage is accessible by all nodes in the cluster. This means that all
volumes that the cluster need to be mapped to each cluster server node individually. To make mapping
storage to a cluster easier, Dell Storage Manager (DSM) offers the ability to create a cluster server object.
The individual server objects used for cluster nodes are placed into a cluster server object. Any volumes that
are mapped to the cluster server object are automatically mapped to each server contained in the cluster
server object.
Note: Snapshot profiles can be applied to cluster server objects, and will apply to all volumes mapped to the
object.
To create a cluster server object in DSM, perform the following steps:
1. Connect to the SC Series array using the DSM Client.

2. Select the Storage tab.
3. Right-click on Servers and select Create Server Cluster.
4. Enter a name for the cluster object.
5. Select a folder to place the cluster server object.
6. From the Operating System drop-down, select Windows 2012 MPIO.
7. Click Add Server to Cluster.
8. Select a cluster server node to add to the cluster and click OK.
Repeat this process for all cluster server nodes.
9. Click OK.
To add a server to an existing cluster object, perform the following steps:
1. Right-click on the cluster object and select Add Server to Cluster.

2. Select a server to add to the cluster and click OK.
To remove a server from a cluster object, perform the following steps:
1. Expand the cluster server object.

2. Right-click on the server and select Remove Server from Cluster.
3. Click OK.
4. When prompted about removing volume mappings, click Yes.
3.2 Volume considerations
When considering how to size volumes for use in a failover cluster, it is important to size cluster data volumes
based on the workload that will be stored on them. For example, cluster data volumes that host Hyper-V
workloads need to be sized to accommodate the desired total number of virtual machines running on a single
data volume.
Cluster quorum volumes do not require more than 1GB of space and should be sized accordingly.
To increase the amount of storage available to cluster roles, additional volumes can easily be added to a
cluster. Cluster volumes can also be resized to use any space that may be added to the volume
3.3 Mapping SC Series Volumes to clustered server object in DSM

An SC Series volume can be mapped to a cluster server object individually, or as part of a group of volumes
that all get mapped to the object at the same time.
Note: To map multiple volumes to a cluster server object at the same time, the volumes will all need to be
contained in the same volume folder.
To map an individual volume to cluster server object, perform the following steps:
1. Connect to the SC Series array using the DSM Client.

2. Select the Storage tab.
3. Expand Volumes and any volume folders to locate the volume.
4. Right-click on the volume and select Map Volume to Server.
5. Select the cluster server object and click Next.
6. Click Finish.
To map multiple volumes to a server cluster object, perform the following steps:
1. Expand Volumes and select the volume folder containing the volumes.
2. In the volume folder summary, select the volumes to map.
3. Right-click on the highlighted volumes and select Map Volume to Server.
4. Select the cluster server object and click Next.
5. Click Finish.
4 Windows Server configuration
Before creating a failover cluster, be sure the Windows Servers (Server 2016 and Nano Server) that are to be
used as nodes in the cluster are configured using Dell EMC best practice guidelines as detailed in the Dell SC
Series: Windows Server 2016 Best Practices Guide, located on TechCenter.
Along with the Dell EMC recommended best practice settings, additional settings and best practices should
be applied to Windows Servers before creating a failover cluster. These settings are detailed in this section.
4.1 Network configuration

Networking is a critical component in any Windows Server Failover Cluster. In a failover cluster, network
connections can be specified as public (connections where the cluster communicates over the network), and
private (reserved for intra-cluster communications, such as heartbeat and Hyper-V Live Migration).
During the cluster creation process all available NICs on the system are assigned to the cluster. NIC adapters
on the same subnet are automatically grouped together and configured for cluster use. The adapters that can
access the network are configured as part of the public connection. Any adapters that can only communicate
with each other are configured as part of the private network.
After cluster creation, specific NIC adapters can be excluded from cluster communications by editing the
cluster network settings.
4.1.1 Public interface

The public interface contains the IP address of the server that is accessed over the network. It should contain
the specific subnet mask, default gateway and DNS server addresses for accessing the network.
4.1.2 Private network

The private network is reserved for intra-cluster communications and is commonly referred to as the cluster
heartbeat network. To create a private cluster network, a dedicated NIC port on each node must be
configured. In a two node cluster, a standard RJ-45 cable can be used to directly connect the first node to the
second node.
In a larger cluster configuration, or in configurations where high-bandwidth connections are required between
cluster nodes, dedicated hardware and a separate subnet may be required for these cluster communications.
Note: Although recommended, a private network is not required to create a failover cluster and will not
cause cluster validation to fail. Be advised that intra-cluster communications between failover clusters
configured without a private network will occur on the public network. In cluster configurations where large
amounts of data are transferred between cluster nodes (such as Hyper-V Live Migrations), overall network
and cluster performance can be negatively impacted as the internal cluster data is transferred on the public
network in addition to normal network and cluster traffic.
When configuring a private network adapter, the following guidelines should be followed:
Do not specify a default gateway or DNS servers.

On the DNS Settings tab, be sure to uncheck Register this connections addresses in DNS.
On the WINS Settings tab, uncheck Enable LMHOSTS lookup and select Disable NetBIOS over
TCP/IP.
4.1.3 Cluster NIC adapter assignment and NIC teaming

By default, the create cluster process will automatically create a public and private cluster network based on
the IP address assigned to the cluster. The NIC adapters assigned to each network are automatically
assigned based on the subnet assigned to each adapter.
In previous versions of Windows Server failover clustering, in cases where there were two or more NIC
adapters on each node that belonged to the same subnet, the cluster would only use one of the adapters on
each node to allow access from the public network. In order to use more than one adapter on the same
subnet in the cluster, the adapters needed to be teamed prior to cluster creation.
As mentioned previously, Windows Server 2016 automatically uses all available NIC adapters on the same
subnet for cluster communications through the Simplified SMB Multichannel and Multi-NIC Cluster Networks
feature. This capability eliminates the need to team multiple NICs in order to use them all in the cluster. For
more information about Simplified SMB Multichannel and Multi-NIC Cluster Networks, refer to TechNet.
Using teamed NICs in a cluster is still supported, although it is not necessary. Advantages to using teamed
NIC adapters in a cluster include increased failover capability and in some cases, increased bandwidth. In
Windows Server 2016, NIC adapters can be teamed via software from the NIC manufacturer (such as Intel or
Broadcom), or through the built-in load balancing and failover option (LBFO) within the operating system. For
more information on NIC teaming in Windows Server 2016, refer to Microsoft TechNet.
Note: Once a NIC has been added to a cluster it can no longer be modified. NIC teams must be configured
before cluster creation.
4.2 Installing the failover clustering feature

Failover clustering is an optional feature on both Windows Server 2016 and Nano Server, and needs to be
installed on all servers that will serve as nodes in the cluster.
4.2.1 Installing the failover clustering feature on Windows Server 2016

To install the Failover Clustering feature on Windows Server 2016, perform the following steps:
1. Click Start > Server Manager to open Server Manager.

2. On the Server Manager Dashboard, select Add roles and features and click Next.
3. Choose a role-based or feature-based installation and click Next.
4. On the Select destination server screen, select the local server and click Next.
5. Click Next on the Select Server Roles screen.
6. On the Select Features screen, check the box next to Failover Clustering and click Next.
7. When prompted to Add features that are required for failover clustering, click Add Features.
8. Click Next when returned to the Select Features screen.
9. Click Install to start the feature installation.
10. Click Close when the installation is finished.
11. Repeat this process on any other servers that are to be used as cluster nodes.
4.2.2 Installing the failover clustering feature on Nano Server

Ideally, the failover clustering feature is installed on Nano Server by specifying the -Clustering switch during
Nano Server image creation. The -Clustering switch installs the failover clustering package, which includes
the files required by the feature, as well as the associated PowerShell module to manage the feature. In
situations where a Nano Server image has already been built and deployed without the failover clustering
package installed, the package can be added when the Nano Server is online, or added to an existing image
when the Nano Server is offline. The example shown below details how to install the failover clustering
package to an online Nano Server using a package downloaded from an online repository.
For more information about installing Nano Server packages, refer to TechNet.
To install the failover clustering package on an online Nano Server, perform the following steps:
1. Connect remotely to the Nano Server using a PSSession.

2. Install the NanoServerPackage provider by typing:
Save-Module -Path $env:ProgramFiles\WindowsPowerShell\Modules `

-Name NanoServerPackage -MinimumVersion 1.0.0.0
3. If prompted to install the NuGet provider, click Yes.

4. Import the NanoServerPackage provider by typing:
Import-PackageProvider NanoServerPackage
5. Obtain a list of available packages by typing:
Find-NanoServerPackage
6. Install the failover cluster package by typing:
Install-NanoServerPackage -Name Microsoft-NanoServer-FailoverCluster-

Package
7. Verify the package install by typing:
Get-WindowsOptionalFeature -Online -FeatureName FailoverCluster-NanoServer
8. Repeat this process on any other Nano Server servers that are to be used as cluster nodes.
4.3 Preparing disks for a new cluster

Before any SC Series volumes can be used in a failover cluster, they need to be prepared on one of the
server nodes that will be used in the cluster. Once a LUN is mapped to the server, it needs to be brought
online, initialized, and a Windows volume needs to be created and formatted in NTFS or ReFS. Drive letters
do not need to be assigned to volumes that will be used in a cluster.
Note: Although it is not a requirement to create a cluster, it is recommended to prepare two SC Series
volumes to be presented to the cluster. One of the volumes will be used for the cluster quorum, while the
other volume will be used for cluster data.
4.3.1 Preparing disks on Windows Server 2016

To prepare SC Series volumes to be used in a failover cluster on a Windows Server 2016 server, perform the
following steps:
1. Open the Disk Management MMC by clicking Start > Run > diskmgmt.msc and then press [Enter].
In Disk Management, if the SC Series volume does not appear, click Action > Rescan Disks to
display the newly available volume as unknown and offline.
2. Right-click on the new disk and choose Online.

3. The disk will now show as Not Initialized. Right-click on the disk and choose Initialize Disk.
4. On the Initialize Disk screen, validate the correct disk is selected for initialization, and select a
partition style (MBR or GPT) for the disk. Click OK.
5. The new disk will now show as Online.
6. The disk must be formatted before it can be used by the cluster.
a. Right-click on the disk and choose New Simple Volume.
b. Click Next on the dialog screen.
c. Specify the volume size, and click Next.
d. Select Do not assign a drive letter or drive path, click Next.
e. Specify partition format options:
i. Select Format this volume with the following settings.
ii. Select NTFS of ReFS for the File System.
iii. Leave Allocation unit size set to default.
iv. Assign a volume label.
Note: To better manage cluster volumes, provide intuitive and descriptive volume labels that correspond to
the name assigned to the LUN on the SC Series array. This is particularly useful when multiple LUNs of the
same size are presented to the same cluster.
v. Leave the box selected to Perform a quick format.

vi. Make sure the box to Enable file and folder compression is not selected.
vii. Click Next.
viii. Verify settings and click Finish.
f. The SC Series volume is now available to be added to the cluster.
g. Repeat this process for all disks that will be added to the cluster. It does not matter in which order
the disks are added.
4.3.2 Preparing disks on a Nano Server

Nano Server does not offer the ability to manage disks with a GUI, or a remote management tool other than
PowerShell.
To prepare SC Series volumes to be used in a failover cluster on a Nano Server, perform the following steps:
1. Connect remotely to the Nano Server using a PSSession.

2. Obtain the list of disks attached to the system by typing:
Get-Disk
Note: If an SC Series volume is not visible, run the Update-HostStorageCache command to rescan disks.
3. Bring the SC Series disk online by typing:
Set-Disk -Number 1 -IsOffline:$false
4. Initialize the SC Series disk and create a GPT partition by typing:
Initialize-Disk -Number 1 -PartitionStyle GPT
5. Create a new Windows volume using all available space on the disk, format the volume in NTFS and
assign a label to the volume by typing:
New-Partition -DiskNumber 1 -UseMaximumSize | Format-Volume `

-FileSystem NTFS -NewFileSystemLabel "Cluster Disk 1" -Confirm:$false
6. At this point, the volume is ready to be added to the cluster. Repeat this process on any other
SC Series volumes that are to be used with the Nano Server cluster.
5 Create a failover cluster
The Failover Cluster Management MMC is used to create failover clusters, validate hardware for potential
failover clusters and perform configuration changes to failover clusters. Failover Cluster Manager can be used
to create clusters on Windows Server 2016 servers, as well as Nano Server servers.
Note: Failover clusters can also be created using PowerShell on both Windows Server 2016 and Nano
Servers. Both the Failover Cluster MMC and PowerShell offer the same functionality.
To create a new cluster, perform the following steps:
1. Start Failover Cluster Manager by clicking Start > Windows Administrative Tools > Failover
Cluster Manager.
2. In the actions pane, click Create Cluster.
3. Click Next on the Before You Begin screen.
4. Enter the Windows Server 2016 servers or Nano Server servers that are to be part of the cluster.
After a server name is entered, click Add. When all server names have been entered, click Next.
5. In the Validation Warning dialog of the Create Cluster Wizard, select to run the configuration
validation tests and then click Next.
Note: Microsoft requires performing cluster validation in order to obtain assistance from Premier Support.
The Cluster Validation tool will report a failure if both nodes of the cluster are not presented with (or are
unable to access) available storage from the SC Series array.
6. When the Before You Begin dialog appears, click Next.

7. On the Testing Options screen, verify Run all tests is selected, and click Next.
8. On the Confirmation screen click Next to begin cluster validation.
9. Review cluster validation results on the Summary dialog screen. Click Finish.
Note: Cluster validation can succeed even if warnings are reported about the configuration. It is highly
recommended to resolve any warnings before continuing with cluster setup.
10. Type the Cluster Name, and provide an IP address for the cluster access point. Click Next.
11. Verify the information, and then click Next.
Note: By enabling add all eligible storage to the cluster on the confirmation screen, all storage that is
presented to both nodes of the cluster and formatted with NTFS or ReFS will be added to the cluster
available storage pool. In a 2 Node cluster configuration if two (or more) drives are added to the cluster, the
smallest capacity disk will be configured automatically as the cluster quorum disk.
12. Once the cluster creation is completed, click Finish.
6 Adding disks to a cluster
Adding disks to the cluster is a simple process once the new volume has been created on the SC Series array
and mapped to the cluster nodes using the server cluster object in DSM. Complete the following steps from
either node to add a new disk (or disks) to the cluster:
Note: For instructions on how to present SC Series volumes to the cluster, refer to section 3.2 of this
document.
1. In Failover Cluster Manager, expand the cluster, expand Storage, and then click Disks.
2. In the Actions pane, click Add Disk.
3. Select the disk or disks to add and then click OK. The disk(s) will be mounted and brought online.
Note: When adding multiple disks to the cluster (especially if they are the same size), it is a best practice to
add each disk individually to keep track of the SC Series LUN assigned to each cluster disk.
4. The newly added disk will be shown in Failover Cluster Manager.
5. Optionally, rename the cluster disk.

a. Right-click on the cluster disk to rename and select Properties.
b. Change the name of the disk and click OK.
Note: For more efficient disk management, the cluster disk name should match the name that was set when
the volume was formatted.
c. The disk name change will be reflected in Failover Cluster Manager.
6.1 Creating a cluster shared volume (CSV) from available cluster

storage
Once a disk has been added to Available Storage in the cluster, it can be converted to a CSV. The process
for converting a disk to a CSV is:
1. Open Failover Cluster Manager, expand Storage, select Disks, right-click the disk to convert and
select Add to Cluster Shared Volumes.
2. The conversion takes place immediately. The disk will now show as a Cluster Shared Volume in the
Assigned To column.
3. The newly created CSV is presented as a mount point in the C:\ClusterStorage directory on each
node of the cluster.
Note: When CSV mount points are created they are named sequentially as Volume1, Volume2, etc
CSV mount points can be renamed for ease of management reasons. As a best practice, rename the CSV
mount point to match the associated cluster disk name.
Note: CSV mount points should be renamed before cluster resources are configured to the corresponding
CSV. Renaming a CSV mount point with associated active cluster resources will cause the resources to fail.
7 Failover clustering and Hyper-V
When clustered, Hyper-V hosts can greatly enhance host server hardware utilization (including processors,
memory, NICs, and power) by allowing many VMs to share these resources at the same time. When used
with Cluster Shared Volumes, Hyper-V guests can be assigned to highly-available cluster roles providing
seamless failover capability for the guests.
Like physical servers, failover clustering can be used to cluster Hyper-V guests together. Clustered Hyper-V
guests can share SC Series storage using in-guest iSCSI, virtual Fibre Channel, pass-through disks, or
shared virtual hard disks.
For detailed information on how to install, configure and administer Windows Server 2016 Hyper-V, including
clustering, refer to the Dell SC Series Storage and Microsoft Hyper-V Best Practices (Updated for Server
2016) on TechCenter.
8 Administrative tasks
8.1 Testing failover

One of the primary benefits to using a failover cluster is to keep a role online even if one of the cluster nodes
fails. While pulling the power on one of the nodes can be done to test the failover capabilities, Failover Cluster
Manager can also be used to test failover.
8.1.1 Moving roles

To move a configured Role to another node perform the following steps:
1. From Failover Cluster Manager, expand the cluster and highlight Roles. In the Roles window, right-
click the Role to failover and then choose Move, then Select Node.
2. Select the node to move the Role to and click OK.
3. The Role will move to the other node. Verify the owner node in the Roles window.
8.2 Cluster-Aware updating

Included in Windows Server 2016, Cluster-Aware Updating (CAU) is an automated feature that allows
updates to clustered servers with little or no loss in availability during the update process. During an Updating
Run, CAU transparently performs the following tasks:
Puts each node of the cluster into node maintenance mode

Moves the clustered roles off the node
Installs the updates and any dependent updates
Performs a restart if necessary
Brings the node out of maintenance mode
Restores the clustered roles on the node
Moves to update the next node
CAU is a powerful feature that can be scheduled on daily, weekly or monthly intervals.
For more information on Cluster-Aware updating, refer to TechNet.
8.3 Windows PowerShell
As mentioned previously in this document, Windows Server 2016 and Nano Server both include failover
cluster Cmdlets to allow for the installation, configuration and administration of failover clustering from within
PowerShell.
For a complete listing of available failover clustering Cmdlets, refer to Microsoft TechNet.
8.3.1 Dell Storage PowerShell SDK command set

Dell has provided a PowerShell interface for SC Series arrays for many years. The Dell Storage PowerShell
command set provides cmdlets for many storage tasks, allowing a single script to automate processes that
work with storage as well as other components in the data center. For example, the same script can create
SC Series volumes, present the volumes to a server, format the disks in Windows Server 2016 and then
create Microsoft SQL Server databases on the new disks.
The Dell Storage PowerShell SDK command set is the next-generation PowerShell interface that provides
more functionality than the legacy command set. It is designed to work with Dell Storage Manager. Any task
that can be performed using the Dell Storage Manager Client can be automated in a PowerShell script using
the PowerShell SDK.
For a detailed reference of commands found in the Dell Storage PowerShell SDK command set, refer to the
Dell Storage PowerShell SDK Cookbook found on Dell TechCenter.
Note: At the time of this writing, the Dell Storage PowerShell SDK is not supported on Nano Server.
A Additional resources
A.1 Technical support and resources

Dell.com/support is focused on meeting customer needs with proven services and support.
Dell TechCenter is an online technical community where IT professionals have access to numerous resources
for Dell software, hardware and services.
Storage Solutions Technical Documents on Dell TechCenter provide expertise that helps to ensure customer
success on Dell Storage platforms.
A.2 Related documentation

Referenced or recommended resources
Vendor Resource
Dell Dell Storage SC Series: Microsoft Multipath I/O Best Practices
Dell Dell SC Series Storage: Microsoft Windows Server 2016 & Nano Server Best Practices
Dell Dell SC Series Storage and Microsoft Hyper-V Best Practices (Updated for Server 2016)
Dell Dell Storage PowerShell SDK Cookbook
Microsoft Whats new in Windows Server 2016
Microsoft Getting started with Nano Server
Microsoft Simplified SMB Multichannel and Multi-NIC Cluster Networks
Microsoft Failover Cluster Cmdlets in Windows PowerShell
Microsoft Configure and Manage the Quorum in Windows Server Failover Cluster

ConfiguringWindowsServer2016andNanoServerFailoverClusters (3288 CD WS)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ConfiguringWindowsServer2016andNanoServerFailoverClusters (3288 CD WS)

Uploaded by

Copyright:

Available Formats

Dell Storage SC Series: Configuring Microsoft

Windows Server 2016 and Nano Server

A Dell EMC Configuration Guide

10/27/2016 Initial release

Installation and configuration of Windows Server 2016 and Nano Server

2.2 Active/Active clusters

2.3 Active/Passive clusters

2.4 Cluster quorum

2.4.1 Quorum configurations

1. Node majority (no witness)

2. Node majority with witness (disk or file share)

3. No majority (disk witness only)

2.5 Cluster shared volumes

For more information about CSV cache, refer to TechNet.

2.6 Using MPIO with Microsoft failover clusters

2.7 High availability

2.8 Cluster validate

3.1 Creating a cluster server object in Dell Storage Manager

To create a cluster server object in DSM, perform the following steps:

1. Connect to the SC Series array using the DSM Client.

To add a server to an existing cluster object, perform the following steps:

1. Right-click on the cluster object and select Add Server to Cluster.

To remove a server from a cluster object, perform the following steps:

1. Expand the cluster server object.

3.3 Mapping SC Series Volumes to clustered server object in DSM

1. Connect to the SC Series array using the DSM Client.

4.1 Network configuration

4.1.1 Public interface

4.1.2 Private network

Do not specify a default gateway or DNS servers.

4.1.3 Cluster NIC adapter assignment and NIC teaming

4.2 Installing the failover clustering feature

4.2.1 Installing the failover clustering feature on Windows Server 2016

1. Click Start > Server Manager to open Server Manager.

4.2.2 Installing the failover clustering feature on Nano Server

1. Connect remotely to the Nano Server using a PSSession.

Save-Module -Path $env:ProgramFiles\WindowsPowerShell\Modules `

3. If prompted to install the NuGet provider, click Yes.

5. Obtain a list of available packages by typing:

Install-NanoServerPackage -Name Microsoft-NanoServer-FailoverCluster-

7. Verify the package install by typing:

Get-WindowsOptionalFeature -Online -FeatureName FailoverCluster-NanoServer

4.3 Preparing disks for a new cluster

4.3.1 Preparing disks on Windows Server 2016

2. Right-click on the new disk and choose Online.

v. Leave the box selected to Perform a quick format.

4.3.2 Preparing disks on a Nano Server

1. Connect remotely to the Nano Server using a PSSession.

3. Bring the SC Series disk online by typing:

Set-Disk -Number 1 -IsOffline:$false

4. Initialize the SC Series disk and create a GPT partition by typing:

Initialize-Disk -Number 1 -PartitionStyle GPT

New-Partition -DiskNumber 1 -UseMaximumSize | Format-Volume `

To create a new cluster, perform the following steps:

6. When the Before You Begin dialog appears, click Next.

4. The newly added disk will be shown in Failover Cluster Manager.

5. Optionally, rename the cluster disk.

6.1 Creating a cluster shared volume (CSV) from available cluster

8.1 Testing failover

8.1.1 Moving roles

2. Select the node to move the Role to and click OK.

8.2 Cluster-Aware updating

Puts each node of the cluster into node maintenance mode