3
2. Features of Machine Check Monitoring Service
This section describes features and characteristics of Machine Check Monitoring Service.
2.1
Features of Machine Check Monitoring Service
For the server that is used in mission critical domain, it is required to identify the failing component,
online degrade it, and online replace it before system down occurs on the server.
If the Machine Check Monitoring Service detects a correctable failure in CPU and memory in Linux
server, it sends log to firmware in the server to identify the failed component. When the correctable
error exceeds threshold value, the Machine Check Monitoring Service degrades CPU or memory page
online (Core Offline, Page Offline). In addition, if the server uses an OS that supports Core Online
feature and spare CPU is equipped in the server, the Machine Check Monitoring Service adds the
spare CPU automatically (Core Online) after Core Offline. Thus the performance deterioration can be
prevented.
Note
Refer to "Capacity Optimization (COPT) User's Guide" for details of Core
Online feature.
Express5800/A1040b does not support Core Offline, Core Online, and Page
Offline.
2.2
System Configuration of Machine Check Monitoring Service
The system configuration of Machine Check Monitoring Service is shown below.
Figure 2-1 System Configuration of Machine Check Monitoring Service
Server
OS
MC Scope
mcemonitor
capmonitor
acpi_call
Firmware
