In this video, Robert McMillen talks about how to setup and manage a cluster network. The cluster network is the heartbeat between multiple servers to determine the uptime of each cluster node. Each node in a cluster needs to know that the other nodes are running in a cluster and it does this using cluster network settings. You designate a specific network card on each node to just send a heartbeat request back and forth to the other nodes. This can be on the same or different subnet used by the other network cards.
- [Voiceover] Previously in the course, we set up a failover cluster. Now let's focus on the network connections. The cluster network is the heartbeat between multiple servers to determine the uptime of each cluster node. By creating this heartbeat this allows the cluster to determine if a node has gone down and another node should take over those cluster applications. So let's start by opening up Failover Cluster Manager, and taking a look at our networks.
In previous videos we discussed network load balancing to create the virtual IP, which all your cluster nodes would respond. A cluster network is different. Each node in the cluster needs to know that the other nodes are running in a cluster, and it does this using the cluster network settings. So, you can designate a specific network card on each node to just send a heartbeat request back and forth, or you could have it respond to client traffic and heartbeat on the same card.
While in our cluster we're going to go to the networks option on the left-hand side. And here you can see that our cluster network is up and the cluster use is both cluster and client. By double-clicking on the cluster one properties we can see the subnet it's going to respond to you see the IP version four and the IP version six subnets, and you can change here where it says no not allow cluster network communication on this network, or you could leave it at the default, which is allow cluster network communication, and allow clients to connect as well.
So that's both the heartbeat and the client data. You can't change anything else here so let's go to a different place under cluster manager where we can take a look where you can make some changes. By clicking on the name of the cluster itself, in this case it's called ClusterApp we can see Cluster Core Resources in the center pane. Now we have a couple of options we have both the storage information, which we can expand.
And we have the server name, which we can expand. So let's go ahead and expand the server name. And underneath that we can see the IP address both IP version four and IP version six. Here we can click on the properties and we can take a look at the resource types and cluster permissions. And we can make changes if we want, but what we want to do is make changes to the network itself. Let's go ahead and click OK. Now let's right-click on the IP version four properties and when we bring that up we can see here is where we can make some changes.
The first thing we can do is change the name of the cluster IP address itself to be called something different if we want. And we're just going to call that CSV Name for clustered shared volume name. But again you can make whatever changes you want here. And here's where we can also make changes to the network itself. Now it's going to be on the same network as the card so if we look at the bottom where it says static IP address it's going to have to show the same subnet that's in this static address.
So if we want to change which subnet the network listens on than we would have to change the static IP address as well. At this time let's go ahead and change the static IP address it listens, and change it to 203. And then click apply. Of course we want to make sure that nothing else is responding on 203 before we do this. The properties restored, but not all changes will take effect until IP address is taken offline. Would you like to do this now? And the answer is yes. And now it's back online.
Now we'll click OK and we can see the new address .203 is online. Let's go ahead and ping that address just to make sure that it's working right. And we can see that it is responding. Excellent. Now we want to go into our DNS manager on our domain controller, and update that information if it didn't happen automatically. Here we are in our domain controller called DC1 and we're in our Server Manager, let's go to the Tools menu and from here click on DNS.
Now let's expand our active directory domain, and look for the name ClusterApp, and there it is ClusterApp and it's already updated to our new .203 address. Excellent, so we should no longer have to do anything there. Let's go ahead and test to make sure that we can ping ClusterApp, so let's pull up a command prompt. And we can also do that with PowerShell if we like. And by default when you ping it we're going to get the IP version six address, so if you want to see the IP version four than choose minus four.
And there it is. We can now ping the new address that we just set. Now we're back in our cluster. Under cluster network one we can see that the cluster use is both cluster and client, which once again means the heartbeat and client information can be transmitted through this network connection. When referring to high-availability setting up the proper communication between cluster nodes. The nodes will be able to use the heartbeat communication to determine if all the other nodes are up and running at any given moment.
Note: The topics covered in this course map to the "Configure and manage high availability" domain for the MCSA: Configuring Advanced Windows Server 2012 Services exam (70-412).
- Configuring NLB
- Configuring affinity, port rules, and cluster operation mode
- Configuring failover cluster networks
- Configuring cluster storage
- Upgrading clusters
- Managing clusters
- Interacting with Hyper-V
- Adding virtual machines in Hyper-V
- Managing cluster roles, including ISCSI target, Hyper-V, and generic service roles
- Migrating clusters
- Configuring VM network health protection and drain