ESXi 8.x encounters PSOD with #PF Exception 14 in world XXXXXX:ql_fcoe_dela

Products

• VMware vSphere ESXi

Issue/Introduction

The ESXi host may go into a PSOD state with the back trace as below :   

YYYY-MM-DDTHH:MM:SS.965Z cpu32:2098142)@BlueScreen: #PF Exception 14 in world 2098142:ql_fcoe_dela IP 0x420035e36d62 addr 0x40
PTEs:0x0;
YYYY-MM-DDTHH:MM:SS.965Z cpu32:2098142)Code start: 0x420034c00000 VMK uptime: 5:01:04:23.924
YYYY-MM-DDTHH:MM:SS.965Z cpu32:2098142)0x45399a09bf60:[0x420035e36d62]_DereferenceSession@(qedf)#<None>+0xa stack: 0x4315c39b8128
YYYY-MM-DDTHH:MM:SS.966Z cpu32:2098142)0x45399a09bf80:[0x420035e329cd]ql_fcoe_do_singlethread_work@(qedf)#<None>+0x76 stack: 0x4315c39b8110
YYYY-MM-DDTHH:MM:SS.966Z cpu32:2098142)0x45399a09bfa0:[0x420034d9fecc]vmkWorldFunc@vmkernel#nover+0x31 stack: 0x420034d9fec8
YYYY-MM-DDTHH:MM:SS.967Z cpu32:2098142)0x45399a09bfe0:[0x4200352dc88e]CpuSched_StartWorld@vmkernel#nover+0xbf stack: 0x0
YYYY-MM-DDTHH:MM:SS.967Z cpu32:2098142)0x45399a09c000:[0x420034d453af]Debug_IsInitialized@vmkernel#nover+0xc stack: 0x0

Environment

ESXi 8.x

Cause

The PSOD happened because qedf driver accessed ql_fcoe_session data structure for the session information, which was NULL. 

Resolution

qedf is an async driver. Marvell is currently working on a fix in the qedf driver. Kindly consult the hardware vendor for additional details and updates regarding the fix.