Absent failure events are recognized in vSAN whenever I/O failures are detected with any of the following hardware devices:

When an absent failure occurs, all data is not immediately resynchronized to another host in the environment. The resynchronization operation is in fact not initiated for 60 minutes from the detection and acknowledgement of the failure event.

The reason for this is that these types of failures can often be temporary. For instance, the network could be down, or a host might be rebooting for regular maintenance. Therefore, the system waits for these components to come back online before committing to a resynchronization of all the data on the node.

While the default delay is 60 minutes, this is an adjustable system parameter that can be increased or decreased based on business requirements. For example, you might want to adjust the parameter if a host takes longer than the set delay to reboot, or if the host will not be available during a maintenance window that is a longer period of time than the set delay.

See Changing the default repair delay time for a host failure in VMware Virtual SAN (http://kb.vmware.com/kb/2075456) for instructions on modifying the settings without having to restart the hosts.