HPE Cloudline CL3100 Gen10, Руководство пользователя и по обслуживанию

Страница: 173 / 201

Software and configuration utilities 173
Error detected during runtime
Memory error log
The BIOS support runtime memory ECC error detection and log.
With ECC in operation, in the process of every fetch from memory, the data and ECC bits are examined
for each 64-bit data + 8-bit ECC group.
If the ECC computation indicates that a single bit Correctable Error (CE) has occurred, it is corrected and
the corrected data is passed on to the processor. If a multi-bit is detected, it cannot be corrected and is
handled as an Uncorrectable Error (UCE).
For either of these two events is generated, it is logged to BMC as SEL log.
Memory Error Log
Sensor Name
Memory
Generic SEL
Info Value Description
Record Type
0x02 System Event Record
Generator ID
0x0001 Generated by BIOS
Sensor Type
0x0C Memory Sensor Type
Sensor Number
0xAB BMC define
E/R Type
0x6F [7] Assertion [6:0] Event Trigger
Event Data
Usage Available Bit Value Description
Event Data1
Bit[7:4]: 0xA OEM data
in Event Data2 & Event
Data3
Bit[3:0]: Error Type 0xA0 Correctable Memory Error
0xA1 Uncorrectable Memory
Error
0xA5 Correctable ECC error
logging limit reached
Event Data2
[7:1] Memory Controller
Number 0x00 iMC0
0x02 iMC1
Event Data3
DIMM Location:
Bit[7:6]: CPU Socket
Number
Bit[5:4]: Channel
Number
Bit[3:0]: DIMM Number
[7:6]: ‘b00 = Skt0
[5:4]: ‘b00 = Ch10
[3:0] ‘b0000 = Dim0
Notes: Physical Memory
Channel Number =
Memory Controller Number
* 3 + Channel Number