Microsoft Failover Clustering Service Overview

In Windows 2012 R2, the failover clustering service requires some downtime to migrate clusters from one group to the next. This is no longer a problem for Windows Server 2016. (Source: Microsoft – I forgot the link). Better yet, Windows 2016 clustering service can run at 2012 level until all nodes OS has been upgraded to 2016. Here are some new features of v.2016:

Storage Replica

  • Uses SMB3
  • Storage-agnostic (will work with local DAS and HBA-mounted SAN volumes)
  • Block-level, synchronous replication between servers
  • Asynchronous replication that allows site extension beyond metropolitan ranges

Cloud Witness

  • FC quorum witness using Azure Blob Storage

Virtual Machine Resiliency

  • Resiliency Options: define how failures are handled and how long VMs are allowed to run isolated
  • Quarantine of unhealthy nodes

Another feature such as Cluster site-awareness would enhance key operations during the cluster lifecycle such as failover behavior, placement policies, heartbeat between the nodes, and quorum behavior. To utilize newer features, this command will be necessary:

# Upgrade Cluster functional level to 2016
Update-ClusterFunctionalLevel

While we’re still mostly on 2016 Clustering functional levels, Windows Server 2019 has been used in Production at various sites. Here are the additional features that are added to 2019 cluster functional levels:

  • Cluster Sets
  • Cross-domain cluster migration
  • USB witness
  • CSV caching to increase VM performance
  • Supports Storage Spaces through Cluster Aware Updating (CAU) for restarts orchestration
  • NTLM authentication obsolescence to improve system security posture

Leave a Reply

Your email address will not be published. Required fields are marked *