Users Guide

3. Click APPLY.
Hardware component redundancy health—Proactive
HA
Proactive HA is a vCenter feature that works with OMIVV. When you enable Proactive HA, the feature safeguards your
workloads by proactively taking measures based on degradation of redundancy health of supported components in a host.
After assessing the redundancy health status of the supported host components, the OMIVV appliance updates the health
status in the vCenter server. Proactive HA is available only on the platforms that support redundancy on PSU, fan, IDSDM,
and memory. Proactive HA feature is not supported for PSUs for which redundancy cannot be configured (for example, cabled
PSUs).
Memory component is applicable only for iDRAC9-based servers with the Memory Operating Mode BIOS configurations of
Advanced ECC Mode, Mirror Mode, Spare Mode, Spare with Advanced ECC Mode, and Dell Fault Resilient Mode. It is
recommended to have homogeneous cluster and iDRAC9-based hosts in the cluster. OMIVV cannot manage iDRAC8 and earlier
hosts because memory redundancy is not supported.
If memory redundancy is disabled on host, the hosts are moved to memory filter and the memory component is not applicable
for those hosts present in the memory filter. There will not be any action on those hosts present in the memory filter even if the
memory goes bad. However, OMIVV continue to monitor the remaining components.
For the PowerEdge MX chassis hosts, memory component is applicable only if the host is managed through Host Credential
profile in OMIVV.
For memory components, the redundancy health status check happens during initialization, polling, and SNMP events. If the
component redundancy status is disabled on host, the host is moved to memory filter during the next polling cycle or when any
event occurs. The host is moved out of memory filter during the next polling cycle once the component redundancy status is
enabled. After the host is moved out of memory filter or when any event occurs, OMIVV manages the memory components
present in the host. For more information about memory operating modes, see the Mode-specific guidelines topic in the Dell
EMC PowerEdge server Installation and Service Manual.
You can view the events on the Events page in vSphere Client and OMIVV logs page.
The available states of redundancy health status for the supported components (power supply, fans, IDSD, and memory) are:
● Healthy (Information)—component operating normally
● Warning (Moderately degraded)—component has a noncritical error. The moderately degraded states are represented as
Warning in the Type column on the Events page.
● Critical (Severely degraded)—component has a critical failure.
● Unknown—denotes the unavailability of any Proactive HA health update from the Dell Inc provider. An unknown health
status might occur when:
○ All hosts that are added to a Proactive HA cluster may remain in the unknown state for a few minutes until OMIVV
initializes them with their appropriate states.
○ A vCenter server restart may put the hosts in a Proactive HA cluster into an unknown state until OMIVV initializes them
with their appropriate states again.
When OMIVV detects a change in the redundancy health status of supported components (either through Traps or polling),
the health update notification for the component is sent to the vCenter server. Polling runs every hour, and it is available as a
fail-safe mechanism to cover the possibility of a Trap loss.
NOTE:
When configuring events, it is recommended that you select Post all Events option as event posting level. For more
information about configuring events, see Configure events and alarms on page 103.
Proactive HA events
Based on the components supported by VMware for Proactive HA, the following events are registered by the Dell Inc provider
during its registration with vCenter:
Table 5. Dell Proactive HA events
Dell Inc provider event Component type Description
DellFanRedundancy Fan Fan redundancy event
92 Manage OMIVV appliance settings