CHW-3133：FAS2820 更换主板后出现不可纠正的机器检查错误

问题描述

主板更换后系统崩溃
节点运行数小时并完成 BMC、BIOS 版本升级
PANIC: Uncorrectable Machine Check Error at CPU7. ICL_UBOX Error: STATUS<0xba00000000000e0b>(VALID,UC,EN,MISCV,PCC,CESI(0),CERR_CNT(0),OTHER_INFO(0),MSCOD(0),MCACOD(0xe0b))MISC<0x0000000014020000>(BUS_LOG(0x14),DEVICE_LOG(0),FUNCTION_LOG(0x2),SEGMENT_LOG(0))ErrSrcID(CorrSrc(0x101),UCorrSrc(0)), RPT(0,9,0): Mellanox CX5 Ethernet in slot 1 on Controller, Mellanox CX5 Ethernet in slot 1 on Controller, Mellanox CX5 Ethernet in slot 1 on Controller, Mellanox CX5 Ethernet in slot 1 on Controller, RAS(20,0,2).. in process idle: cpu7 on release 9.15.1P7 (C) on Tue Apr 8 01:49:02 CEST 2025