Error: Machine Check Exception (MCE) on PCPU resulting in Purple Screen of Death (PSOD)

Products

• VMware vSphere ESXi

Issue/Introduction

• An ESXi host crashes and displays a Purple Screen of Death (PSOD) referencing a Machine Check Exception (MCE)
• The ESXi host enters a Not Responding state in vCenter Server.
• The physical console displays: Machine Check Exception on PCPU## in world ######.
• The error message: System has encountered a Hardware Error - Please contact the hardware vendor.
• Purple screen backtrace may include vSAN-related functions such as BucketlistInsertInNode, Bucketlist_Insert, Rangemap_Insert, or PLOGPrepareWrite.
• The following entries are observed in the /var/run/log/vmkernel.log file on the ESXi host:

YYYY-MM-DDTHH:MM:SS.111Z cpu##:2099051)ALERT: MCA: 191: SRAR Excp G7 B1        ############### ############ M86 ############ Cache Hierarchy: Level 0 Data Cache    DataRead Error.
Machine Check Exception on PCPU## in world 2099051:VSAN_0x4##18
System has encountered a Hardware Error - Please contact the hardware vendor
YYYY-MM-DDTHH:MM:SS.173Z cpu19:2099051)@BlueScreen: Machine Check Exception on PCPU19 in world 2099051:VSAN_0x4##18
System has encountered a Hardware Error - Please contact the hardware vendor

Environment

VMware vSphere ESXi 8.x

Cause

The issue is triggered by the Machine Check Architecture (MCA) within the CPU detecting a hardware anomaly. Specifically, a Data Cache Data Read Error (Level 0 Cache) was detected. Although categorized as a Software Recoverable Action Required (SRAR) event, the crash occurs because the error was encountered by a critical kernel thread (e.g., VSAN_0x45018), leaving the hypervisor with no viable recovery path other than to halt the system to ensure data integrity.

Resolution

To resolve this issue, underlying hardware failure must be addressed. Perform the following steps:

Restart the Host: Reboot the affected ESXi host via the hardware management interface (e.g., iDRAC, ILO, CIMC) to restore temporary service.

Analyze Hardware Logs: Log in to the server's hardware management console and export the System Event Logs (SEL) or IPMI logs. Look for indicators of:

1. Predictive failures on memory modules (DIMMs).
2. CPU internal errors or bank errors.
3. PCI parity or bus errors.

Run Diagnostics: Execute a comprehensive hardware diagnostic test (Vendor-specific) to identify the faulty physical component (CPU or Memory).

Engage Hardware Vendor: Contact your server OEM support team. Provide them with the PSOD screenshot and the exported hardware logs for component replacement.

Update Firmware/BIOS: Ensure the host is running the latest vendor-validated BIOS and firmware versions, as microcode updates often contain fixes for stability issues that can trigger MCEs.

Additional Information

Error: Machine Check Exception (MCE) resulting in a Purple Screen of Death (PSOD)