ESXi 8.0.3 PSOD with #PF Exception 14 and MicrocodeUpdate Warning

Products

• VMware vSphere ESXi 8.0

Issue/Introduction

An ESXi 8.0.3 host crashes with a Purple Diagnostic Screen (PSOD). The following symptoms are observed:

• Error Message: #PF Exception 14 in world <ID>:<VM_Name>
• Backtrace: Reference to ISched_SysServiceDone, ITimerWheelHandler, or BH_Handler.
• On-Screen Alert: MicrocodeUpdate: 421: This machine’s BIOS should be updated to pick up the latest microcode.
• Environment: The host is typically running on outdated system firmware/BIOS while utilizing vSphere 8.x.

Environment

• Hypervisor: VMware ESXi 8.0.3 [Build 25067014 or later]
• Hardware: Any x86_64 server platform (Intel/AMD) with outdated System BIOS.

Cause

The issue is caused by a critical mismatch between the ESXi 8.0.3 kernel and the CPU microcode provided by the System BIOS.

ESXi 8.x hypervisors require specific CPU microcode levels to maintain stability during high-frequency kernel operations (scheduling and timer interrupts). When the BIOS is significantly outdated (e.g., several versions behind), the CPU may encounter instruction-level faults or timing violations when switching contexts between the VM (World) and the Hypervisor (VMK). This results in a Page Fault (Exception 14) within the scheduler.

Resolution

To resolve this issue, perform a full firmware alignment on the physical host:

1. Identify Current BIOS Version: Log in to the Out-of-Band Management (iDRAC, ILO, IPMI) or check the ESXi summary page to identify the current BIOS version.
2. Verify Compatibility: Consult the VMware Compatibility Guide (VCG) for your server model to find the minimum required BIOS version for ESXi 8.0.3.
3. Apply Updates: Update the System BIOS to the latest vendor-released version.
   Note: It is highly recommended to also update the Management Controller (iDRAC/iLO) and Chipset firmware simultaneously to ensure full subsystem stability.
4. Verify Microcode in ESXi: After rebooting, verify the host no longer displays the "MicrocodeUpdate" alert during the boot sequence or in the vmkernel.log.
5. Configure Core Dumps: Ensure a diagnostic partition is configured to capture data in the event of a subsequent failure: esxcli system coredump file add --datastore="<DATASTORE_NAME>"

Additional Information

If the PSOD persists after BIOS updates, perform a hardware stress test on the System RAM (DIMMs), as Exception 14 can also be triggered by physical memory defects.