For a vCenter HA cluster to be healthy, each of the Active, Passive, and Witness nodes must be fully operational and be reachable over the vCenter HA cluster network. If any of the nodes fails, the cluster is considered to be in a degraded state.

Problem

If the cluster is in a degraded state, failover cannot occur. For information about failure scenarios while the cluster is in a degraded state, see Failover Failures.

Cause

The cluster can be in a degraded state for a number of reasons.
One of the nodes fails
  • If the Active node fails, a failover of the Active node to the Passive node occurs automatically. After the failover, the Passive node becomes the Active node.

    At this point, the cluster is in a degraded state because the original Active node is unavailable.

    After the failed node is repaired or comes online, it becomes the new Passive node and the cluster returns to a healthy state after the Active and Passive nodes synchronize.

  • If the Passive node fails, the Active node continues to function, but no failover is possible and the cluster is in a degraded state.

    If the Passive node is repaired or comes online, it automatically rejoins the cluster and the cluster state is healthy after the Active and Passive nodes synchronize.

  • If the Witness node fails, the Active node continues to function and replication between Active and Passive node continues, but no failover can occur.

    If the Witness node is repaired or comes online, it automatically rejoins the cluster and the cluster state is healthy.

Database replication fails
If replication fails between the Active and Passive nodes, the cluster is considered degraded. The Active node continues to synchronize with the Passive node. If it succeeds, the cluster returns to a healthy state. This state can result from network bandwidth problems or other resource shortages.
Configuration file replication issues
If configuration files are not properly replicated between the Active and Passive nodes, the cluster is in a degraded state. The Active node continues to attempt synchronization with the Passive node. This state can result from network bandwidth problems or other resource shortages.

Solution

How you recover depends on the cause of the degraded cluster state. If the cluster is in a degraded state, events, alarms, and SNMP traps show errors.

If one of the nodes is down, check for hardware failure or network isolation. Check whether the failed node is powered on.

In case of replication failures, check if the vCenter HA network has sufficient bandwidth and ensure network latency is 10 ms or less.