CPU|Total Capacity cluster metric is not consistent.

Products

• VCF Operations/Automation (formerly VMware Aria Suite)

Issue/Introduction

'CPU | Total Capacity (MHz)' metric on the Cluster Compute Resource object fluctuates as per chart below:

Environment

Aria Operations 8.18.3

Cause

This issue is related to ESXi host state and can be caused by any of the following:

• Host Shutdown:
• Maintenance Mode:
• Host Restart:
  
  

Note: A host does not need to be explicitly shut down for its capacity to be removed from 'CPU | Total Capacity (MHz)' .

Any instance where vCenter reports a host as 'not responding' – for example, during a restart, a complete shutdown, or due to network connectivity issues between vCenter and the host – will cause a corresponding drop in the metric. In essence, if a host is unreachable by vCenter, its capacity will be impacted.

 

Resolution

This is expected behavior if the ESXi Host is in "not responding" state. 

The Cluster Compute Resource's  'CPU | Total Capacity (MHz)' will decrease until the ESXi host is brought back and is in connected state.