The number of failures-to-tolerate policy setting is one of the core availability mechanisms with vSAN. This policy controls the number of replicas (mirrors) of a virtual machine component, although the policy can be applied to all of a virtual machine’s disks, or individual VMDKs. This policy is important when planning and sizing storage capacity because it directly relates to the consumption of capacity that the virtual machine has on the storage.

For every n failures-to-tolerate, n+1 copies of the data are needed, and 2n+1 hosts contributing storage are required. For instance, if the Number of Failures-to-Tolerate capability is set to 1, the virtual machine or disk will have two replica mirrored copies of its components created across the cluster. If the number is set to 2, three mirror copies are created, and so on. If a failure occurs, the mirrors take over.

As you can see, this leads to greater storage utilization than is actually configured because of the replicas created, but they also consume additional components. Configure this setting based on the availability requirements of the virtual machine or disk.

These requirements are defined in a storage policy and applied to the appropriate workloads. The default is to have one failure to tolerate unless the policy is changed to be a different value. The maximum number of failures to tolerate is three.

The recommendation is to keep the failures to tolerate setting at the default value of 1, unless there is specific need to configure a separate policy that provides a higher level of fault tolerance. If this is the case, use this policy sparingly with higher priority virtual machines.