ESXi host fails with a PSOD "no heartbeat"

Products

• VMware vSphere ESXi

Issue/Introduction

Symptoms:
PSOD Backtrace:

@BlueScreen: PCPU 2: no heartbeat (3/3 IPIs received)
Code start: xxxxxxxxxxxxxx VMK uptime: 56:23:59:56.912
Saved backtrace from: pcpu 2 Heartbeat NMI
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]MemDistributeNUMAPolicy@vmkernel#nover+0x333 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]MemDistribute_Alloc@vmkernel#nover+0x299 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]PagePool_AllocCustom@vmkernel#nover+0x2f0 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]vmk_MemPoolAlloc@vmkernel#nover+0x37c stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]mfiIoctl_SendFWCmd@<None>#<None>+0x1f2 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]mfiIoctl_FWCmd@<None>#<None>+0x115 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx][email protected]#v2_3_0_0+0x1d1 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx][email protected]#v2_3_0_0+0x5ea stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]VMKAPICharDevDevfsWrapIoctl@vmkernel#nover+0x80 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]DevFSIoctl@vmkernel#nover+0x684 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]FSSObjectIoctlCommon@vmkernel#nover+0x9c stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]FSS_IoctlByFH@vmkernel#nover+0xb8 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]UserFile_PassthroughIoctl@<None>#<None>+0x3f stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]UserVmfs_Ioctl@<None>#<None>+0x24 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]LinuxFileDesc_Ioctl@<None>#<None>+0x5c stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]User_LinuxSyscallHandler@<None>#<None>+0xd7 stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]User_LinuxSyscallHandler@vmkernel#nover+0x1d stack: xxxxxxxxxxxxxx
xxxxxxxxxxxxxx:[xxxxxxxxxxxxxx]gate_entry_@vmkernel#nover+0x0 stack: xxxxxxxxxxxxxx

Note: The backtace is only an example.

Environment

VMware vSphere ESXi 6.5
VMware vSphere ESXi 6.7
VMware vSphere ESXi 6.0

Cause

The PSOD is caused by a memory allocation problem.

Resolution

This issue is fixed in VMware ESXi 6.0 P04.