ESXi Host PSOD: Machine Check Exception (MCE) with Internal Unclassified Error and Cache Hierarchy DataRead Error
search cancel

ESXi Host PSOD: Machine Check Exception (MCE) with Internal Unclassified Error and Cache Hierarchy DataRead Error

book

Article ID: 445180

calendar_today

Updated On:

Products

VMware vSphere ESXi

Issue/Introduction

  • ESXi host experiences a Purple Screen of Death (PSOD). 2. Virtual Machines are restarted by vSphere HA. 3. PSOD screen or logs show: 'Machine Check Exception on PCPU in world'.
  • YYYY-MM-DDTHH:MM:30.097Z cpu91:2651536)ESC[45mESC[33;1mVMware ESXi 8.0.3 [Releasebuild-25205845 x86_64]ESC[0m
    Machine Check Exception on PCPU91 in world 2651536:python
    System has encountered a Hardware Error - Please contact the hardware vendor
  • We may observe below errors in /var/log/vmerknel.log

YYYY-MM-DDTHH:MM:19.053Z cpu71:2546599)ALERT: MCA: 191: SRAR Excp Gf B1 Sbd80000000100134 A61ae208c0 M86 P61ae208c0/40 Cache Hierarchy: Level 0 Data Cache DataRead Error.
} =
MCA error class: SRAR = Uncorrectable, recoverable, action required
Reported by: Excp = Machine Check Exception (#MC) handler.
MCG_STATUS: 0x000000000000000f
   RIPV (restart IP valid)
   EIPV (error IP valid)
   MCIP (machine check exception in progress)
   LMCE_S (local machine check exception signaled)
MCA bank number: 0x1
MCi_STATUS: 0xbd80000000100134
   MCACOD (architectural) 0x0134 = Cache Hierarchy: Level 0 Data Cache DataRead Error.
   MSCOD (model-specific error code) 0x0010

YYYY-MM-DDTHH:SS:43:36.599Z cpu24:2101240)ALERT: MCA: 191: UCNA Intr G0 B1a Sa800000201000405 A0 M1800000000000 P0/0 Internal Unclassified Error.
YYYY-MM-DDTHH:SS:43:38.706Z cpu18:2543938)MCA: 1141: Sending VMCE to monitor worldID 2543938 failed Not initialized
YYYY-MM-DDTHH:SS:43:38.706Z cpu18:2543938)ALERT: MCA: 191: SRAR Excp Gf B1 Sbd80000000100134 A79ba3a340 M86 P79ba3a340/40 Cache Hierarchy: Level 0 Data Cache DataRead Error.
YYYY-MM-DDTHH:SS:44:02.847Z cpu41:2544160)ALERT: MCA: 191: UCNA Poll G0 B1a Sa800000201000405 A0 M1800000000000 P0/0 Internal Unclassified Error.
YYYY-MM-DDTHH:SS:44:06.327Z cpu5:2544157)MCA: 1141: Sending VMCE to monitor worldID 2544157 failed Not initialized
YYYY-MM-DDTHH:SS:44:06.327Z cpu5:2544157)ALERT: MCA: 191: SRAR Excp Gf B1 Sbd80000000100134 A1024628480 M86 P1024628480/40 Cache Hierarchy: Level 0 Data Cache DataRead Error.
YYYY-MM-DDTHH:SS:44:18.408Z cpu26:2107027)ALERT: MCA: 191: UCNA Intr G0 B1a Sa800000201000405 A0 M1800000000000 

  • PSOD backtrace:

#0  0x0000420001391510 in memcpy_unrolled (nbytes=4096, src=0xffff8103c4b7d000, dst=0xffff810251231000) at bora/vmcore/public/x86/vm_libc_memcpy.h:111
#1  Util_CopyMA (destMA=<optimized out>, srcMA=srcMA@entry=16185282560, length=length@entry=4096) at bora/vmkernel/main/util.c:1704
#2  0x00004200017ded7e in UserCow_CopyPage (world=world@entry=0x453a0cc1f000, mpn=mpn@entry=3951485, mpnCopy=mpnCopy@entry=0x453a0cc1b560, vpn=vpn@entry=220431026) at bora/public/address_defs.h:55
#3  0x00004200017fd358 in UserMemBreakSharingAndEnableHdWrite (localFlush=<synthetic pointer>, l1Table=<optimized out>, breakSharing=0 '\000', isWrite=<optimized out>, va=902885482944, pte=<optimized out>, world=0x453a0cc1f000) at bora/vmkernel/user/userMem.c:9226
#4  UserMem_HandleMapFault (world=0x453a0cc1f000, va=va@entry=902885482944, isWrite=<optimized out>, breakSharing=breakSharing@entry=0 '\000', canBlock=canBlock@entry=1 '\001', pageType=pageType@entry=USER_PAGE_UNKNOWN, isFromUser=0 '\000', mpnOut=0x0, newlyPinned=0x0) at bora/vmkernel/user/userMem.c:9879
#5  0x0000420001833b85 in UserArchExceptionHandleMappingFault (excErrorCode=<optimized out>, va=902885482944) at bora/vmkernel/user/x86/user.c:172
#6  User_ArchExceptionHandleFault (vector=vector@entry=14, fullFrame=fullFrame@entry=0x453a0cc1bf40, faultAddr=faultAddr@entry=902885482944) at bora/vmkernel/user/x86/user.c:502
#7  0x00004200017d56ee in User_Exception (currentWorld=currentWorld@entry=0x453a0cc1f000, vector=vector@entry=14, fullFrame=fullFrame@entry=0x453a0cc1bf40) at bora/vmkernel/user/user.c:3074
#8  0x00004200018a8473 in IDTHandleGenericExc (vector=14, fullFrame=0x453a0cc1bf40) at bora/vmkernel/main/x86/idt.c:683
#9  Int14_PageFault (fullFrame=0x453a0cc1bf40) at bora/vmkernel/main/x86/idt.c:731
#10 0x00004200018a20c7 in gate_entry () at bora/vmkernel/main/x86/gates64.S:175
#11 0x000000d2382af293 in ?? ()

Environment

Esx 8.0.3

Cause

The issue is caused by a hardware failure. Specifically, the CPU detected uncorrectable errors (SRAR/UCNA) related to the Level 0 Data Cache and internal system bus, indicating faulty hardware components.

Resolution

This is a hardware specifc crash. Please engage Hardware vendors and perform an intense HW diagnosis. 

  •  Identify specific PCPU reporting 'Internal Unclassified Error' or 'Cache Hierarchy: Level 0 Data Cache DataRead Error'
  •  Replace the failing CPU or motherboard as indicated by vendor diagnostics.
  •  Ensure server BIOS and firmware are updated to the latest supported versions.
Powered by Wolken Software