Microsoft Failover Cluster Manager (MSFCM) is a specific management function within the Windows Server operating system which is used to create, validate, and manage failover server clusters running Windows Server.
A failover cluster is a collection of individual physical servers -- or nodes -- organized into a group capable of sharing the computing workload of an application. When a failure occurs in one node of the cluster, the workload traffic and computing resources are redirected to other nodes -- hence the term failover -- that will continue to operate the application without disruption. The troubled node is then isolated from the cluster until it can be repaired and restored. Once a troubled node is returned to normal operation, the cluster redistributes traffic and computing to accommodate the restored cluster.
Failover across server clusters is a principal data center technology used to improve workload availability and resilience. Failover clusters are most often deployed for stateful applications that require data persistence to avoid application failure. Examples include database operations such as Microsoft SQL Server, financial transactions, or messaging or file servers.
Successful cluster operation requires administrative tools such as Failover Cluster Manager to create the cluster, check the failover cluster configuration, manage the cluster's behavior, and migrate settings to Windows Server cluster nodes.
Typical MSFCM tasks involve taking the cluster's service or application online or offline, pausing or resuming a failover cluster node (such as when the node requires upgrades or other routine service), refreshing the configuration of a virtual machine (VM), migrating VMs between cluster nodes, running common tools on clustered disks (such as chkdsk), validating (troubleshooting) a failover cluster, and accessing event logs and reports produced by a failover cluster.
MSFCM is available as a snap-in through the Microsoft Management Console (MMC). Administrators can open the failover cluster snap-in by clicking Start, Administrative Tools and Failover Cluster Manager. But failover clusters can also be managed through Windows PowerShell cmdlets, and file shares on file server clusters can be managed using File and Storage Services tools.
A failover cluster has certain requirements. For example, the hardware components typically must be certified for the version of Window Server in use. Direct-attached storage (DAS) must be attached to nodes in the cluster to accommodate shared storage and support serial-attached SCSI (SAS), Fibre Channel, Fibre Channel over Ethernet (FcoE) or iSCSI. Once configured, the complete cluster must pass all tests initiated by the Validate a Configuration wizard. And finally, cluster server firmware and the operating system usually require the latest updates.
Failover cluster requirements and capabilities can vary depending on the Windows Server version being used. For example, the move from Windows Server 2003 to 2008 and 2008 R2 worked to simplify cluster setup and management, improve security and stability, and enhance efficiency exchanging storage traffic. The move to Windows Server 2012 increased failover cluster scalability, allowed more flexible cluster architectures and allowed for faster failovers. The current Windows Server 2016 Technical Preview includes site-aware failover cluster that support stretched clusters across locations, allowing for better failover behavior and placement policies. Windows Server 2016 also promises better diagnostics through log files and data dumps.