ESXi Host Loses Fiber Channel Datastores Due to Emulex HBA Silent Driver Hang

Products

• VMware vSphere ESXi

Issue/Introduction

An ESXi host abruptly loses connectivity to Fibre Channel datastores, triggering All Paths Down (APD) and Permanent Device Loss (PDL) events. Other hosts in the cluster remain unaffected.

When validating the host hardware, the physical PCIe bus successfully enumerates the Emulex Fibre Channel Adapters. Command: lspci | grep -i "Fibre Channel"

0000:8b:00.0 Fibre Channel: Emulex Corporation Emulex LPe35000/LPe36000 Fibre Channel Adapter
0000:8b:00.1 Fibre Channel: Emulex Corporation Emulex LPe35000/LPe36000 Fibre Channel Adapter

However, the ESXi storage core fails to register these adapters. Command: esxcli storage core adapter list

HBA Name  Driver    Link State  UID                   Capabilities  Description
--------  --------  ----------  --------------------  ------------  -----------------------------------------
vmhba0    bcm_mpi3  link-n/a    fc.################                 PERC H965i Front
vmhba1    vmw_ahci  link-n/a    fc.################                 Sapphire Rapids SATA AHCI Controller vmhba2 vmw_ahci link-n/a fc.################                 Sapphire Rapids SATA AHCI Controller 

(Note the complete absence of the lpfc driver or Emulex HBAs in the logical adapter list).

Environment

VMware ESXi 8.0.3
Emulex LPe35000/LPe36000 Fiber Channel Adapters
Dell Inc. PowerEdge R660

Cause

The Emulex HBA firmware or lpfc driver experiences a silent hang, causing the adapters to abruptly unregister from the ESXi storage core. This severs all logical storage paths and triggers immediate APD timeouts.

Because the hang occurs silently at the firmware/driver layer, the /var/log/vmkernel.log will record the sudden loss of paths (APD) but will not contain hardware exceptions (NMI, MCE) or explicit driver panics.

Example /var/log/vmkernel.log signature showing the APD events without corresponding hardware faults:

2026-03-12T09:49:16.491Z In(182) vmkernel: cpu1:2097646)StorageApdHandlerEv: 120: Device or filesystem with identifier [naa.#########################] has entered the All Paths Down Timeout state after being in the All Paths Down state for 140 seconds.
2026-03-12T09:49:18.403Z In(182) vmkernel: cpu1:2097646)StorageApdHandlerEv: 120: Device or

Resolution

1. 1. 1. Perform a hard reboot of the affected ESXi host to force a hardware PCIe bus reset and reload the lpfc driver, which will temporarily restore datastore connectivity.

      2. Identify the active driver module using the command: esxcli storage core adapter list

      3. Extract the loaded driver version using the command: vmkload_mod -s lpfc | grep Version

      4. Extract the physical HBA firmware version via the hardware management console (e.g., Dell iDRAC) or the host boot logs.

      5. Cross-reference the installed versions against the VMware Compatibility Guide (VCG).

      6. Upgrade the server platform firmware and the Emulex HBA driver/firmware to the latest certified combination for the installed ESXi build.

Additional Information

 

VMware Compatibility Guide

Determining Network/Storage Firmware and Driver Version in ESXi (323110)