ESXi 5.5 host becomes unresponsive with a purple diagnostic screen and reports the error #PF Exception 14 in world 33426: vmkeventd IP 0x418002c71507 addr 0x0

Products

• VMware vSphere ESXi

Issue/Introduction

Symptoms:

• Upgrading or installing an ESXi 5.5 host fails.
• After upgrading or installing ESXi 5.5 build 1331820, the host fails unexpectedly with a purpose diagnostic screen.
• You used a VMware standard image or a vendor customized image to install or upgrade the ESXi 5.5 host.
• You see a backtrace similar to:

VMware ESXi 5.5.0 [Relesasebuild-1331820 x86_64]

#PF Exception 14 in world 33426: vmkeventd IP 0x418002c71507 addr 0x0

PTEs:0x12vv96027;0x12bd56027:0x0;

*PCPU1:33426/vmkeeventd

PCPU 0: UUSSHSISISSSUSSUSISS

Code start : 0x418002200000 VMK uptime: 0:00:00:25.561

0x41238a49d9b0: [0x418002c71507]qla2x00_abort_all_cmds@<None>#<None>+0x7f stack: 0x4109cc7fb780

0x41238a49d9f0: [0x418002c75c17]qla2x00_free_device@<None>#<None>+0x23 stack: 0x41238a49da20

0x41238a49db70: [0x418002c763c8]qla2x00_probe_one@<None>#<None>+0x304 stack: 0x41238a49dba0

0x41238a49dc00: [0x41800288fff3][email protected]#9.2+0x93 stack: 0x0

0x41238a49dc30: [0x418002890220][email protected]#9.2+0xc0 stack: 0x0

0x41238a49dc50: [0x418002c8a2c3]qla2x00_module_init@<None>#<None>+0xbf stack: 0x417fc23f2d64

0x41238a49dce0: [0x418002283769]Mod_LoadDone@vmkernel#nover+0x2d5 stack: 0x41238a49dd90

0x41238a49dc50: [0x41800224b571]Elf_LoadDone@vmkernel#nover+0x1c55 stack: 0x410a130f92b0

0x41238a49dec0: [0x4180027b57ea]UWVMKSyscallUnpackLoadVMKModule@<None>#<None>+0x152 stack: 0x41238a4

0x41238a49df00: [0x4180022a9b2d]User_UWVMKSyscallHandler@<None>#<None>+0x246 stack: 0x41238a49df20

0x41238a49df20: [0x4180022f1064]gate_entry@vmkernel#nover+0x64 stack: 0x0

base fs=0x0 gs=0x418040400000 Kgs=0x0

No place on disk to dump data.

No file configured to dump data.

Environment

VMware vSphere ESXi 5.5

Cause

The order of an ESXi host's disk controllers can dictate the driver used during the boot process. This issue occurs if the order of the drivers is changed after installation or upgrade, this can lead to stability issues if the wrong drivers are used.

Resolution

To resolve this issue, change the boot order of the disk controllers in an ESXi host BIOS to boot from the local controller:

1. Enter in to the BIOS of the ESXi host.
   
   Note: Consult your vendors documentation for steps on accessing the BIOS of the host.
   
   
2. Change the boot order for the storage controllers, ensuring that the local controller is first.

Additional Information

To determine the HBA devices loaded correctly, run the lspci command and look for modules related to the vendor of your disk controller:

# lspci

0000:06:00.0 Serial bus controller: QLogic Corp ISP2432-based 4Gb Fibre Channel to PCI Express HBA [vmhba1]
0000:06:00.1 Serial bus controller: QLogic Corp ISP2432-based 4Gb Fibre Channel to PCI Express HBA [vmhba2]



To check all loaded packages, use the esxcli software vib list command.
You see output similar to:

esxcli software vib list | grep qla
ima-qla4xxx 500.2.01.31-1vmw.0.3.040121 QLogic VMwareCertified -
scsi-qla4xxx 634.55.20.0-1OEM.550.0.0.1198610 QLogic VMwareCertified -
scsi-qla2xxx 902.k1.1-9vmw.550.0.0.1331820 VMware VMwareCertified -

To confirm the module is loaded use the vmkload_mod -l command. This example is for Qlogic devices:

vmkload_mod -l | grep -i qla

For more information on checking driver modules, see Determining Network/Storage firmware and driver version in ESXi/ESX 4.x and ESXi 5.x (1027206).

Determining Network/Storage firmware and driver version in ESXi 4.x and later
ESXi 5.5 ホストが応答しなくなり、紫色の診断画面が表示され、エラー #PF Exception 14 in world 33426: vmkeventd IP 0x418002c71507 addr 0x0 が表示される
ESXi 5.5 主机无响应并显示一个紫色诊断屏幕，报告错误 #PF Exception 14 in world 33426: vmkeventd IP 0x418002c71507 addr 0x0