The backend LUNs are not visible on the ESXi host despite the configuration from the switch and the storage array being valid.

Products

• VMware vSphere ESXi

Issue/Introduction

Symptoms:

• LUNs from the storage array are not visible on the ESXi host.

• The zoning configuration on the Fibre Channel switch is confirmed as valid.

• The storage array configuration (masking/mapping) is also confirmed as valid.
  
  
• A reboot of the ESXi host did not resolve the issue.
  
  
• A storage rescan did not resolve the issue.

Environment

VMware ESXi 8.x

Cause

This issue happens when the ESXi host fails to establish or maintain a connection with the Fibre Channel fabric or a storage target.

When forcing a Fabric Login reset on Fibre Channel adapters using the command esxcli storage san fc reset -A vmhbaX, the following excerpts can be noticed in the vmkernel logs:

yyyy-mm-ddT##:##:##.###Z Wa(180) vmkernel: cpu###:#######)WARNING: lpfc : vmhba2 lpfc_sp_handle:2390: 0321 Rsp Ring 1 error: Job Data: x2990900 x0 xb0000 x80010000
yyyy-mm-ddT##:##:##.###Z Wa(180) vmkernel: cpu###:#######)WARNING: lpfc : vmhba2 lpfc_cmpl_els_disc_cmd:2932: 4650 ELS cmd x19 error: x9 xB0000

This logging confirms the host failed to establish or maintain a connection with the Fibre Channel fabric or a storage target.

Breakdown of Errors

• Rsp Ring 1 error This is an internal HBA or driver error. The Response Ring (Rsp Ring) is a queue used by the HBA to deliver completion notifications back to the driver for commands sent to the FC fabric. An error here means the HBA encountered an issue while trying to process or post a response. This often suggests a problem with the HBA hardware, firmware, or the driver's communication with the hardware.

• ELS cmd x19 error This identifies the specific Fibre Channel Extended Link Services (ELS) command that failed. ELS command x19 is the code for FLOGI (Fabric Login). FLOGI is the very first step an HBA takes to establish a connection; it is the process of logging into the Fibre Channel switch fabric to receive a permanent address (N_Port ID).

The key takeaway is that the HBA on vmhba failed its initial login attempt (FLOGI) to the Fibre Channel fabric.

If the FLOGI command fails, the HBA cannot get an N_Port ID. Without an N_Port ID, the HBA cannot discover or access any storage LUNs that are zoned to it.

•  

Resolution

The resolution for this issue almost always lies outside the ESXi host itself and involves checking the physical connectivity and the Fibre Channel switch configuration.

• Physical and Logical Connectivity Checks

  • Check Physical Connection: Ensure the cable between the ESXi HBA port (vmhba port) and the Fibre Channel switch port is securely seated and not damaged.

  • Check Switch Port Status: Log into your FC switch and verify the status of the port connected to concerned vmhba. The port should be Online and Enabled. If it's showing Disabled, Faulty, or Isolated, the switch is preventing login.

• Check for Incorrect or missing zoning in the switch configuration.
  
  
• Ensure the HBA firmware and the installed driver version on the ESXi host are compatible with each other and with the installed ESXi version, as per the VMware Hardware Compatibility Guide (HCL). An outdated driver or firmware can sometimes cause low-level communication failures.