BMC management network port link down

Products

• VMware vSphere ESXi

Issue/Introduction

A brief network interface link down (interruption) is detected on the BMC (Baseboard Management Controller / iLO, etc.) of an ESXi host.
During this event, there is no record of a link down for the standard ESXi network interfaces (vmnic), and the following IPMI (nipmi) driver timeout errors are recorded in the vmkernel.log:

<YYYY-MM-DD>T<HH:MM:SS> Wa(180) vmkwarning: cpu##:#######)WARNING: nipmi: IpmiKcs_ProcessRequest:307: nipmi: Failure in KCS write start phase. Error: Failure
<YYYY-MM-DD>T<HH:MM:SS> Wa(180) vmkwarning: cpu##:#######)WARNING: nipmi: ProcessRequest:1476: nipmi: A error while sending while sending the IPMI request hardware, sending an IPMI failure response. Error: Failure
<YYYY-MM-DD>T<HH:MM:SS> Wa(180) vmkwarning: cpu##:#######)WARNING: nipmi: ReadPhase:617: nipmi: Timed out waiting for OBF to set in read phase.. Error: Timeout
<YYYY-MM-DD>T<HH:MM:SS> Wa(180) vmkwarning: cpu##:#######)WARNING: nipmi: IpmiKcs_ProcessRequest:338: nipmi: Failure in KCS read phase. Error: Timeout

Environment

VMware vSphere ESXi

Cause

This issue is caused by a failure in the underlying BMC (Baseboard Management Controller / iLO, etc.) hardware, which stops responding or temporarily restarts, resulting in IPMI communication timeouts.

 

Resolution

Verify that the reported network interface link down applies to the management port used for BMC access, and not the ESXi data communication ports (vmnic).

Engage the hardware vendor to investigate the BMC hardware status, perform hardware component diagnostics, and apply BMC firmware updates as necessary.

Additional Information

BMC ネットワークポートのリンクダウン