This article explains the vSAN/Skyline Health – vSAN cluster configuration consistency check in the vSAN/Skyline Health service and provides details on why it might report an error.
For more information on vSAN cluster configuration consistency, see:
Q: What does the vSAN/Skyline Health – vSAN cluster configuration consistency check do?
This health check validates that all hosts in this cluster and their disks have a configuration consistent with the configuration set at the cluster level, For Example: dedup/compression, Data-at-Rest/Data-in-Transit encryption, performance service.
Q: What does it mean when it is in an error state and how to fix?
If this check fails, it means that there are inconsistent configuration (for example: dedup/compression, encryption) setup on hosts or disks with the cluster. The table included in this health check will report the host and disks with issues, and with some remediation recommendation. Also, user may click "Remediate inconsistent configuration" button to run the cluster configuration remediation action to fix hosts and disks which have inconsistent configurations (e.g. dedup/compression, encryption). Caution should be taken when remediating any inconsistency that has potential to initiate a large amount of data resync (e.g. enabling/disabling dedup/compression, encryption on disks) as this can have a negative impact on the performance of the cluster.
For performance service configuration inconsistency, user can not manually remediate the cluster configuration by click "Remediate inconsistent configuration" button. An auto-remediation will be applied to this cluster when an inconsistent configuration is detected for the vSAN performance between vCenter and ESXi hosts. Note the configuration inconsistency health check for performance service is available since vSphere 6.7 release.
If the vSAN cluster is in compute mode, user may see the health check in a warning state if there's a host with a disk-group. That's because vSAN disk-groups are not supported in compute-only clusters. User can remediate the issue either by removing the host out of cluster or removing the disk-group from host.