High Availability is one of the most sought-after topics these days. When your data centers are struck with a disaster, a reliable High Available setup can actually do wonders. Let’s take a look at the end benefit that it can offer.
Why the Cluster setup?
- Load balancing of hosts
- Reduced risk of data loss
- Prevents business failures
- And hence no fear of losing customers during a downtime!
It’s very important that we manage a Cluster setup that’s robust enough to survive a disaster. Microsoft Hyper-V being a significant player in the virtualization market is the next preferred domain to VMware. Hyper-V is basically a host-based service. It executes all of its functions from the resources(storage, RAM and network resources) that it draws from a single physical physical machine.
Enter the business disaster.
If this physical machine(or the server) gets hit by the disaster, chances of all the data on guest VMs over Hyper-V server getting lost is high. This is where the concept of clustering servers(Hyper-V Cluster setup) can help.
So, let’s deal with building a Cluster setup using Microsoft Hyper-V, i.e., the Hyper-V Failover Cluster.
Setting up the Hyper-V Failover Cluster:
The following is a prototype model of a Hyper-V Cluster setup.
In the end, you can avail a consistent and distributed namespace that the clustered roles can actually access the shared storage from all the nodes of the Cluster Setup with very minimal disruptions.
The hardware, network and the software requirements that a Cluster setup requires is as follows:
- Storage that’s attached to every node of the Cluster Setup
- Physical NIC on each Hyper-V host
- Device controllers or adapters for storage
- Serial Attached SCSI or FC
- Windows servers with a 64-bit processor – a set of matching computers that contain similar components
- Network Adapters and cables(for network communication): For eg, if iSCSI is used each network adapter should be dedicated to network communication or iSCSI, not both
- Make sure you get the right OS for the clustered virtual i.e., guest VMs. Required Installation media – from a physical media or a (.iso) file
- Installation of integration services package- It is important as it improves the connectivity between the physical computer and the VMs
Network Infrastructure Requirements:
- Network: A minimum of 2 networks is required. 1) One- a public network that allows clients to connect to the Cluster. 2) Two – for the internal communication between the nodes in the cluster setup
- DNS: Servers should use the DNS for name resolution. The dynamic update protocol can be used
- Network settings and IP addresses: Compare the configuration settings on the network adapter and the switch it connects to (they should not be conflicting)
- Clients: You can connect as many clients to a node as you want (max of 1024 VMs on a node)
Now, that we are all set with what we need. We should now be focusing on how to build one High Available Hyper-V Failover Cluster Setup.
Steps to building a Highly Available Hyper-V Cluster Setup:
1.Connect the servers to the network and storage.
Ensure that the disks (LUNs) that you want to use in the cluster are exposed to the servers that you will cluster (and only those servers). You can use either of the following interfaces to expose disks or LUNs:
- The interface provided by the manufacturer of the storage
- An appropriate iSCSI interface
2. Install Hyper-V and Failover Clustering feature on all the servers that you group on a cluster setup
3. Create a Virtual Switch
4. Validate the cluster configuration
5. Create the cluster
6. Add a disk as CSV to store the data of VMs:
- To implement certain scenarios for clustered virtual machines, the virtual machine storage, and virtual hard disk file should be configured as Cluster Shared Volumes (CSV)
- CSV also supports live migration of a Hyper-V virtual machine between nodes in a failover cluster
7. Create a highly available VMs.
- Choose the shared storage as the location to store the virtual machine and the virtual hard disk. To make the shared storage available to the virtual machine, you must create the virtual machine on the physical computer that is the node which owns the storage
- specify the CSV volume as the location of both the virtual machine and the virtual hard disk
8. Install guest OS on all the VMs.
9. Test a planned failover.
Try moving the created VM on the cluster to another node. Perform the following migration scenarios.
- Live Migration
- Quick Migration
- Storage Migration
10. Test an unplanned Failover as well.
Move to High Availability with Vembu:
Alright. We’ve now set up a Cluster configuration that would ensure High Availability. But, wait! Does that mean data that resides there are all completely safe?
Vembu upcoming release with Vembu BDR Suite v4.0 has something in store to resolve this problem. Our experts have planned to host a webinar to show you how to manage a High Available Failover Cluster that assures high uptime.
Register Below to join the webinar and learn how to get there.