In a Windows Server Failover Cluster, the various cluster communication networks are an essential part of the Windows Failover Cluster environment. Without the proper network provisioning, cluster communications are either not possible or are problematic. Generally speaking, the various cluster essential networks are certainly a first place to start when troubleshooting performance and or stability issues. Since Windows Server Hyper-V clusters are built on top of Windows Server Failover Clustering technology, the same is true when looking at the various Hyper-V networks required for appropriate Hyper-V cluster provisioning.

Let’s take a look at Windows Server Hyper-V cluster networks and their various uses and configuration.

Protect Your Data with BDRSuite

Cost-Effective Backup Solution for VMs, Servers, Endpoints, Cloud VMs & SaaS applications. Supports On-Premise, Remote, Hybrid and Cloud Backup, including Disaster Recovery, Ransomware Defense & more!

Table of Contents

  1. Windows Server Hyper-V Cluster Management Network
  2. Windows Server Hyper-V Cluster Network
  3. Windows Server Hyper-V Cluster Live Migration Storage
  4. Windows Server Hyper-V Cluster Storage Network
  5. Concluding Thoughts

Also, we will see how these various networks are an integral part of having a fully functional and performant Windows Server Hyper-V cluster that is able to serve out production workloads with adequate performance, high availability, and stability.

Windows Server Hyper-V Cluster Networks

There are different networks that are required for use in a Windows Server Hyper-V cluster. These various networks serve different purposes as you will see but each play a vital role. There is no single “one size fits all” recommendation for how many networks a Hyper-V host will have. There are a lot of different factors that affect the various networks that will be configured with a Hyper-V cluster, including the storage configuration.

Let’s take a look at the following standard networks that are generally found in many Hyper-V environments and the role they play in a Windows Server Hyper-V cluster environment.

Download Banner
  • Management
  • Cluster
  • Live Migration
  • Storage

Cluster network configuration is essential as you cannot truly have a Windows Server Failover Cluster without the networks that are utilized by the cluster. These ensure all of the required and dedicated network resources that allow cluster resources to communicate.

Windows Server Hyper-V Cluster Management Network

Management traffic is the common type of traffic that you think of when managing a networked Windows operating system, even in a non-clustered form. This includes such activities as establishing RDP connections to the operating system as well as console connections. Additionally, it is the network that is utilized when registering the connection for the Windows operating system in DNS using dynamic DNS registration. The management network connection is the network that all management activities take place on. As far as egress traffic for management purposes, Windows Update as well as Hyper-V backup data protection solutions also will communicate over the management network to each of the Hyper-V hosts in a Windows Failover Hyper-V cluster.

So far, we have only detailed the host specific network communication activities that are carried out over the management network of each host.

Are there any cluster specific network activities that take place on the management network?

A Windows Server Failover Cluster houses a unique computer object called the Cluster Name Object or CNO. This computer object represents the Windows Failover Cluster itself and provides a key role in the successful operation of the Windows Failover Cluster with Active Directory. The Windows Failover Cluster computer object is a standalone computer object that registers its own DNS record with Active Directory. The management network adapter configured on each Hyper-V host will house the cluster name object.

One node in the Windows Server Hyper-V Cluster will own the cluster name object and the IP address associated. During a role drain or a failure of the CNO owner, ownership of these objects are transferred to a healthy host. To determine the node that currently owns the Cluster Name Object, you can use Failover Cluster Manager or PowerShell to quickly see this information.

When you click on the cluster name in Failover Cluster Manager, you will see in the information pane the Current Host Server which indicates the Hyper-V node that owns the CNO.

Windows-Server-Hyper-V-Cluster-Networks

Locating the Cluster Name Object owner in Failover Cluster Manager

Also, it is easy to query the CNO object owner node by using PowerShell.

Get-ClusterResource -Name “Cluster Name” | Format-List *

Windows-Server-Hyper-V-Cluster-Networks

Finding the Cluster Name Object owner using PowerShell

Building the management network for redundancy is a good idea. This means you could make use of adapter teaming to ensure the failure of a single adapter or path wouldn’t take down the management network which would greatly impact the function of the Windows Failover Cluster.

Windows Server Hyper-V Cluster Network

Another extremely important type of network that is configured in a Windows Server Failover Hyper-V cluster is the cluster network. Windows Failover Clusters operate by sending what are called heartbeats to the other nodes in the cluster. Every single second, each node attempts to communicate with the other nodes in the cluster. When Hyper-V nodes in the Failover Cluster are unable to contact the other nodes in the cluster and enough to maintain quorum, it will shut itself down.

The cluster network itself is also extremely important for environments making use of Cluster Shared Volumes or CSVS. The nodes in the Windows Server Hyper-V Failover Cluster make use of the cluster network to exchange metadata concerning files in the cluster shared volume. This metadata exchange is typically a small network operation as the metadata information is extremely small.

However, there is a situation that can potentially develop that Hyper-V administrators want to be aware of. If connectivity for some reason for a node is lost to the cluster shared volume via the normal connection path, the cluster network is used as a “backup” network of sorts in that the connectivity to the CSV is redirected to the cluster network through the coordinator node for the CSV. This can cause an enormous amount of traffic compared to normal operations as the disconnected node from the CSV uses this network to perform I/O with the virtual machines it owns.

Windows Server Hyper-V Cluster Live Migration Network

Management traffic is the common type of traffic that you think of when managing a networked Windows operating system, even in a non-clustered form. This includes such activities as establishing RDP connections to the operating system as well as console connections. Additionally, it is the network that is utilized when registering the connection for the Windows operating system in DNS using dynamic DNS registration. The management network connection is the network that all management activities take place on. As far as egress traffic for management purposes, Windows Update, as well as Hyper-V backup data protection solutions, also will communicate over the management network to each of the Hyper-V hosts in a Windows Failover Hyper-V cluster.

So far, we have only detailed the host-specific network communication activities that are carried out over the management network of each host.

Are there any cluster specific network activities that take place on the management network?

A Windows Server Failover Cluster houses a unique computer object called the Cluster Name Object or CNO. This computer object represents the Windows Failover Cluster itself and provides a key role in the successful operation of the Windows Failover Cluster with Active Directory. The Windows Failover Cluster computer object is a standalone computer object that registers its own DNS record with Active Directory. The management network adapter configured on each Hyper-V host will house the cluster name object.

One node in the Windows Server Hyper-V Cluster will own the cluster name object and the IP address associated. During a role drain or a failure of the CNO owner, ownership of these objects are transferred to a healthy host. To determine the node that currently owns the Cluster Name Object, you can use the Failover Cluster Manager or PowerShell to quickly see this information.

When you click on the cluster name in Failover Cluster Manager, you will see in the information pane the Current Host Server which indicates the Hyper-V node that owns the CNO.

Locating the Cluster Name Object owner in Failover Cluster Manager

Also, it is easy to query the CNO object owner node by using PowerShell.

Get-ClusterResource -Name “Cluster Name” | Format-List *

Finding the Cluster Name Object owner using PowerShell

Building the management network for redundancy is a good idea. This means you could make use of adapter teaming to ensure the failure of a single adapter or path wouldn’t take down the management network which would greatly impact the function of the Windows Failover Cluster.

Configuring Hyper-V Live Migration Priorities

Windows Server Hyper-V Cluster Storage Network

An ultra-critical Windows Server cluster network is the storage network. Windows Server Hyper-V clusters in the traditional shared storage configuration are attached to a shared storage network of some sort. The storage network allows all hosts to communicate with the shared storage so all virtual machine can be accessed by all hosts in the cluster. This allows for many of the Hyper-V cluster features and functionality to take place such as Live Migration and high availability. A few of the storage network design best practices include the following:

  • Use a dedicated storage network
    • Disallow management traffic on the storage network
  • Use redundant adapters and paths to the shared storage from each Hyper-V host
  • Do not route storage traffic – keep it on the same subnet
  • Don’t use NIC teaming for storage traffic, instead use MPIO
  • Use 10gig connections to ensure the most bandwidth and lowest latency for storage traffic serving a large number of VMs. However, 1Gb connections still work quite well with a smaller number of mixed workload VMs

Right-clicking the storage network in Failover Cluster Manager and choosing the properties allows you to configure the Do not allow cluster network communication on this network setting which helps to ensure the storage traffic has full access to dedicated network resources configured.

Windows-Server-Hyper-V-Cluster-Networks

Disallowing management traffic on the storage network in Hyper-V

Concluding Thoughts

Windows Server Failover Clusters that are hosting the Hyper-V role require several network resources to be available to the Windows Failover Cluster. These include the management, cluster, Live Migration, and storage networks. All of these particular networks play an important role in ensuring all cluster resources have the ability to communicate and have the dedicated bandwidth and low latency to communicate well. Additionally, designing each of the networks for redundancy allows for a good deal of high availability to the network. This includes using multiple adapters and redundant paths to critical resources such as storage. By understanding and correctly configuring and utilizing these resources, Hyper-V administrators will ensure proper network communication to all the Hyper-V nodes in the cluster.

Related Posts:

Hyper-V Cluster Network Performance Testing
Hyper-V Cluster iSCSI SAN Storage and Network Considerations
Beginner’s Guide for Microsoft Hyper-V: How to Create Hyper-V Cluster – Part 15

Follow our Twitter and Facebook feeds for new releases, updates, insightful posts and more.

Rate this post