无法探测到ESXi硬件状态

Products

• VMware vSphere ESXi
• VMware vSphere ESXi 7.0

Issue/Introduction

通过ESXi的client确认到hardware状态是unknow，其他主机都是绿色正常的状态。

sensord日志：
sensord[89916671]: hello_mc: starting communication with bmc over ipmi driver
sensord[89916671]: hello_mc: ipmi returned invalid data block: data_len: 1 ccode 255
sensord[89916671]: ipmi_open: returned descriptor 4 for device /dev/ipmi0
sensord[89916671]: set_address: IPMB address set to: 0x20
sensord[89916671]: ipmi_open: returned descriptor 4 for device /dev/ipmi0
sensord[89916671]: set_address: IPMB address set to: 0x20
sensord[89916671]: ipmi_open: returned descriptor 5 for device /dev/ipmi0
sensord[89916671]: Test 2. verify libipmi ipmi_sdr_iter iterator.
sensord[89916671]: set_address: IPMB address set to: 0x20
sensord[89916671]: ipmi_open: returned descriptor 4 for device /dev/ipmi0
sensord[89916671]: count_sdrs: starting communication with bmc over ipmi driver
sensord[89916671]: count_sdrs: ipmi returned invalid data block: data_len: 1 ccode 255
sensord[89916671]: Test 2 failed initial open failed.

hostd日志有大量以下error：
2024-03-28T03:19:41.421Z error hostd[25215366] [Originator@6876 sub=Default opID=lpdh5k4v-1102081-auto-nmdf-h5:70144863-ce-27-9b50 user=vpxuser:exampledomain\user] IpmiIfcLanInstantiate: Get IP addr failed: cc=0xff
2024-03-28T03:19:43.869Z error hostd[25215366] [Originator@6876 sub=Default opID=lpdh5k4v-1102081-auto-nmdf-h5:70144863-ce-27-9b50 user=vpxuser:exampledomain\user] IpmiIfcLanInstantiate: Get MAC address failed: cc=0xff
2024-03-28T03:19:46.310Z error hostd[25215366] [Originator@6876 sub=Default opID=lpdh5k4v-1102081-auto-nmdf-h5:70144863-ce-27-9b50 user=vpxuser:exampledomain\user] IpmiIfcLanInstantiate: Get netmask failed: cc=0xff
2024-03-28T03:19:51.169Z error hostd[25215366] [Originator@6876 sub=Default opID=lpdh5k4v-1102081-auto-nmdf-h5:70144863-ce-27-9b50 user=vpxuser:exampledomain\user] IpmiIfcRhGetOS_Name: retrieve os_name failed cc=0xff
2024-03-28T03:19:53.619Z error hostd[25215366] [Originator@6876 sub=Default opID=lpdh5k4v-1102081-auto-nmdf-h5:70144863-ce-27-9b50 user=vpxuser:exampledomain\user] IpmiIfcRhGetOS_Version: retrieve os_version failed cc=0xff
2024-03-28T03:19:56.034Z error hostd[25215366] [Originator@6876 sub=Default opID=lpdh5k4v-1102081-auto-nmdf-h5:70144863-ce-27-9b50 user=vpxuser:exampledomain\user] IpmiIfcRhGetOS_Sysname: retrieve sysname failed cc=0xff
2024-03-28T03:20:32.019Z error hostd[25215446] [Originator@6876 sub=Default opID=lpdh5k4v-1103960-auto-nntm-h5:70145368-a-8e-7da9 user=vpxuser:exampledomain\user] count_events: ipmi returned invalid data block: data_len: 1 ccode 255

Environment

VMware vSphere ESXi 7.0

Cause

hostd没有core dump日志，所以排除是hostd crash导致。

Resolution

协调主机硬件厂商进一步排查硬件故障，同时建议Rest BMC，或是在BMC Configuration中将"IPMI over KCS Access" disable 之后，再enable。