Frames dropped and unresponsive Virtual Machine issue with Emulex FC cards and Pure storage array

Products

• VMware vSphere ESXi

Issue/Introduction

This KB article explains an issue seen on certain servers running VMware ESXi 7 and connected to Pure storage array.

Symptoms:

Virtual Machines become unresponsive on ESXi. The following can apply to:

• VMWare ESXi 7(all updates)

• Emulex FC HBA

• Pure Storage //X and //C Arrays

Log snippets with error signature below:

2022-03-31T11:48:18.839Z cpu39:2098402)lpfc: lpfc_handle_status:5615: 0:(0):3271: FCP cmd x28 failed <0/222> sid x010900, did x010d00, oxid x19e6 iotag x1d0c Local Reason x1d Returning Host Busy

2022-03-31T09:36:43.674Z cpu19:2100263)lpfc: lpfc_handle_status:5615: 1:(0):3271: FCP cmd x2a failed <3/42> sid x010900, did x011100, oxid x19f1 iotag x1d17 Local Reason xf Returning Host Busy

2022-03-31T09:35:36.052Z cpu36:2098404)lpfc: lpfc_handle_status:5615: 0:(0):3271: FCP cmd xa3 failed <5/235> sid x010900, did x051100, oxid x19de iotag x1d04 Dropped Frame: Act 3064, Exp 2772 Returning Host Bus Busy

FRAME DROP event has been observed 6 times in the last one minute. This suggests a problem with Fibre Channel link/switch!.

Environment

VMware vSphere ESXi 7.x

Cause

The array is responding incorrectly to traffic from an Emulex HBA.

Resolution

• Below HBA firmware version has the fix for the suppress response.

    LPE3600/Prism: 14.0.639.20
    LPE3200/Lancer G6: 14.0.639.18

Workaround:

Disable the “Suppress Response Feature” in Emulex LPFC driver settings.
    This can be done by issuing the below command and rebooting the ESXi.

# esxcli system module parameters set -p "lpfc_suppress_rsp=0" -m lpfc

 

Additional Information

Impact/Risks:
Virtual Machines may stop running, impacting production.