
472 IBM BladeCenter JS23 and JS43 Implementation Guide
12.2 System diagnostics
POWER6 processor-based systems contains specialized hardware detection
circuits for detecting erroneous hardware operations, and includes extensive
hardware and firmware recovery logic. IBM hardware error checkers have these
distinct attributes:
Continuous monitoring of system operations to detect potential calculation
errors.
Attempted isolation of physical faults based on runtime detection of each
unique failure.
initiation of a wide variety of recovery mechanisms designed to correct a
problem.
Machine checks are handled by firmware. When a machine check occurs, the
firmware analyses the error to identify the failing device and creates an error log
entry.
In partitioned mode, any error that occurs during partition activity is surfaced to
the operating system running in the partition. If some error occurs during
POWER hypervisor (PHYP) activities, then the system gets rebooted by PHYP.
In case the system degraded to the point where the service processor cannot
reach standby state, then the ability to analyze the error does not exist.
12.2.1 Diagnostic tools
This section brings a list of some tools that can be used to help in diagnostic
hardware problems on IBM BladeCenter JS23 and JS43 Express.
Checkpoints and error codes
During system power-on process, the Power-on self-test (POST) checks out the
hardware, including some system components and interconnections, and
generates 8-digits checkpoint codes to mark the power-on progress.
Important: This section is not intended to be a replacement for the
information provided in the BladeCenter JS23 and BladeCenter JS43 Type
7778 Problem Determination and Service Guide, Part Number: 44R5339. For
detailed steps on how to perform diagnostics tasks, determine the root cause
of an error, and get proper support assistance, refer to this manual.
Comentários a estes Manuais