High ESXi Host CPU utilization due to DIMM Faults

Products

• VMware vSAN
• VMware vSphere ESXi

Issue/Introduction

• An ESXi host exhibits abnormally high CPU utilization (e.g., 75-130% consumed CPU) when under load.
• VM CPU utilization is with normal range
• Upon placing the host into maintenance mode and the VMs migrate to other hosts in the cluster CPU utilization doesn't spike up on any other hosts further indicating CPU utilization is not VM related.
• As soon as a load is placed back on the host CPU utilization spiks back up to 75-130%

 

Environment

VMware vSphere ESXi (ALL Versions)

Cause

Out-of-Band Management interface (iLo, iDrac) shows memory modules with errors with the recommendation to either reseat or replace the DIMMs

Resolution

1. Verify the hardware health by reviewing the ESXi host's hardware via Monitor Hardware Health Status in the VMware Host Client or by accessing the physical server's Out-of-Band Management interface (e.g., iLO, iDRAC, XCC) to check for DIMM/memory errors.
2. Place the affected ESXi host into Maintenance Mode with default Ensure Accessibility option
3. Consult the server hardware vendor to perform diagnostics on the physical memory modules.
4. Reseat or replace the affected DIMMs as directed by the hardware vendor.
5. Once the hardware is repaired and the hardware health sensors display a green/healthy state, exit Maintenance Mode.
6. Monitor the host to confirm that CPU usage has stabilized.