vSAN Settings

Fault Tolerance Threshold (FTT) support is disabled in PowerChute by default in a vSAN environment. FTT support is only supported in an Advanced UPS configuration where there are vSAN hosts in more than one UPS Setup. If you enable FTT support, PowerChute will take the FTT Level value specified into consideration when starting a maintenance mode task. This value is linked to the number of critical Advanced UPS Setups. The FTT Level configured here should match the Fault Tolerance Threshold in the Storage Policy applied to the vSAN datastore. The default vSAN Storage Policy uses FTT=1.

If the number of critical Advanced UPS Setup(s) is less than or equal to the FTT Level, PowerChute starts a maintenance mode task using the Ensure accessibility flag, and only the critical host(s) will get shut down. This means that VMs can remain running after being migrated to another host in the vSAN Cluster. PowerChute will not disable High Availability and vCLS.

If the number of critical Advanced UPS Setup(s) is greater than the FTT Level, PowerChute will issue a maintenance mode command using the No Action flag. PowerChute will not disable High Availability and vCLS.

In addition to checking for critical events active on other Advanced UPS Setup(s) hosts, PowerChute checks that the hosts are healthy. A host is considered being in an unhealthy state if it is not reachable, not responding, not powered on, or in maintenance mode.

Single UPS Setup(s) (containing one host) in an unhealthy state will be considered when calculating the FTT Level when a critical event occurs. For example:

• Fault Tolerance Threshold is enabled with FTT Level of 1

• PowerChute configured with 4 single UPS Setups

• Critical event occurs on UPS Setup 1

• Host in UPS Setup 2 is in an unhealthy state

• FTT is exceeded as the number of UPS Setups with issues is greater than the FTT Level of 1

The Shut down All Cluster VMs checkbox is selected by default if FTT is enabled. It is not recommended to disable this setting. If the number of critical Advanced UPS Setup(s) is greater than the FTT Level, PowerChute will shut down all VMs in the Cluster and place all critical hosts in maintenance mode using the No Action flag before shutting them down. VMs/vApps running on hosts that are not directly impacted by the critical UPS event or power outage will also be shut down if they reside on the vSAN datastore. These hosts will not be placed into maintenance mode or shut down.

The vSAN Synchronization Duration field specifies a duration, in seconds, that PowerChute will allow any active vSAN data re-synchronization tasks to complete before shutting down the host if Delay Host Maintenance Mode is enabled. For more information, see Host Maintenance Mode.

Re-starting VMs after a shutdown

The All Hosts online prior to startup checkbox in the Virtualization Settings page determines how vSAN Cluster VMs are powered on during the startup sequence.

• If this checkbox is selected, PowerChute will wait for all UPS critical events to be resolved before starting up VMs.

• If this checkbox is unselected, PowerChute will attempt to power on VMs if the FTT Level is sufficient to power them on.

vSAN Cluster Preparation and Recovery Durations

Cluster Preparation – Duration of time for the preparation of the vSAN cluster for shutdown.

Cluster Recovery – Duration of time for the recovery of the vSAN cluster on a startup.

• Default setting of 120 seconds for both fields.
• The values are saved to the pcnsconfig.ini file under the “vsan_prepare” and “vsan_recovery” keys.
• These options are only available for vSAN configurations.