Technical information

IBM Europe, Middle East, and Africa Hardware
Announcement ZG14-0098
IBM is a registered trademark of International Business Machines Corporation
13
be repaired. An ECC uncorrectable error detected in these caches can also trigger a
purge and delete of cache lines. This results in no loss of operation if the cache lines
contained data unmodified from what was stored in system memory.
Modified data would be handled through Special Uncorrectable Error handling. L1
data and instruction caches also have a retry capability for intermittent errors and a
cache set delete mechanism for handling solid failures.
Special Uncorrectable Error handling
Special Uncorrectable Error (SUE) handling prevents an uncorrectable error in
memory or cache from immediately causing the system to terminate. Rather, the
system tags the data and determines whether it will ever be used again. If the error
is irrelevant, it will not force a check stop. If the data is used, termination may be
limited to the program/kernel or hypervisor owning the data; or the I/O adapters
controlled by an I/O hub controller would freeze if data were transferred to an I/O
device.
PCI extended error handling
PCI extended error handling (EEH)-enabled adapters respond to a special data
packet generated from the affected PCI slot hardware by calling system firmware,
which will examine the affected bus, allow the device driver to reset it, and continue
without a system reboot. For Linux, EEH support extends to the majority of
frequently used devices, although some third-party PCI devices may not provide
native EEH support.
Predictive failure and dynamic component deallocation
Servers with Power processors have long had the capability to perform predictive
failure analysis on certain critical components such as processors and memory.
When these components exhibit certain symptoms that may indicate a failure is
imminent, the system can dynamically deallocate and call home, when enabled,
about the failing part before the error is propagated system-wide. In many cases,
the system will first attempt to reallocate resources in such a way that will avoid
unplanned outages. In the event that insufficient resources exist to maintain full
system availability, these servers will attempt to maintain partition availability by
user-defined priority.
Uncorrectable error recovery
When the auto-restart option is enabled, the system can automatically restart
following an unrecoverable software error, hardware failure, or environmentally
induced (ac power) failure.
Serviceability
The purpose of serviceability is to efficiently repair the system while attempting to
minimize or eliminate impact to system operation. Serviceability includes system
installation, MES (system upgrades/downgrades), and system maintenance/repair.
Depending upon the system and warranty contract, service may be performed by
the customer, an IBM representative, or an authorized warranty service provider.
The serviceability features delivered in this system provide a highly efficient service
environment by incorporating the following attributes:
• Design for Customer Set Up (CSU), Customer Installed Features (CIF), and
Customer Replaceable Units (CRU)
• Detection and Fault Isolation (ED/FI)
• First Failure Data Capture (FFDC)
• Lightpath service indicators:
– Service labels and service diagrams available on the system and delivered
through IBM Knowledge Center
– Step-by-step service procedures documented in IBM Knowledge Center or
available through the Hardware Management Console