2
1.3
Terminology
Terms used in Machine Check Monitoring Service are as shown below:
Table 1-2 Terminology
Term
Description
mcemonitor
Software that realizes higher RAS feature.
When mcemonitor receives logs from mce mechanism of Linux kernel,
analyze it, and monitors fault occurrence in cooperation with system.
mcemonitor instructs Core Offline and Page Offline to the kernel.
capmonitor
Software that controls Core Offline for failed core, and Core Online that
COPT feature provides.
Refer to "Capacity Optimization (COPT) User's Guide" for details of COPT
feature.
acpi_call
Driver used to access ACPI
ACPI
Advanced Configuration and Power Interface
Open industry specification related power management and hardware
configuration.
MCE
Machine Check Exceptions
Hardware error detected by CPU
CMC
Corrected Machine Check
Correctable error detected by CPU
CPU socket
Means a single Intel Xeon processor. One CPU socket can have several
cores. With Express5800/A2040, up to 4 CPU sockets can be installed in
the server.
CPU core
Core portion of CPU that performs arithmetic processing and others. One
or more cores can exist in CPU socket.
Physical CPU socket
number
Means physical mounting position of a CPU socket in the server. The
number from No. 1 to No. 4 is assigned for every CPU socket.
Logical processor
Means the processor where OS actually executes task and threads. When
Hyper-Threading feature is enabled, two logical processors exist in one
CPU core. When Hyper-Threading feature is disabled, only one logical
processor exists in one CPU core.
1.4
Access Limitation
Only the privileged user (root account) can use mcemonitor.