Sun Microsystems Netra T5220, Руководство по обслуживанию

Страница: 17 / 178

Chapter 1 Server Diagnostics 1-7
1.2.1.1 Memory Configuration
In the server memory there are 16 slots that hold DDR-2 memory FB-DIMMs in the
following FB-DIMM sizes:
■
1 Gbyte (maximum of 16 Gbyte)
■
2 Gbyte (maximum of 32 Gbyte)
■
4 Gbyte (maximum of 64 Gbyte)
FB-DIMMs are installed in groups of 8, called ranks (ranks 0 and 1). At minimum,
rank 0 must be fully populated with eight FB-DIMMs of the same capacity. A second
rank of FB-DIMMs of the same capacity can be added to fill rank 1.
See Section 4.6, “Replacing FB-DIMMs” on page 4-23 for instructions about adding
memory to a server.
1.2.1.2 Memory Fault Handling
The server uses an advanced ECC technology, called chipkill , that corrects up to 4 bits
in error on nibble boundaries, as long as all of the bits are in the same DRAM. If a
DRAM fails, the FB-DIMM continues to function.
The following server features independently manage memory faults:
■
POST – Based on ILOM configuration variables, POST runs when the server is
powered on.
For correctable memory errors (CEs), POST forwards the error to the Solaris
Predictive Self-Healing (PSH) daemon for error handling. If an uncorrectable
memory fault is detected or if a “storm” of CEs is detected, POST displays the
fault with the device name of the faulty FB-DIMMs, logs the fault, and disables the
faulty FB-DIMMs by placing them in the ASR blacklist. Depending on the memory
configuration and the location of the faulty FB-DIMM, POST disables half of
physical memory in the system, or half the physical memory and half the
processor threads. When this offlining process occurs in normal operation, you
must replace the faulty FB-DIMMs based on the fault message. You then must
enable the disabled FB-DIMMs with the ALOM CMT CLI
enablecomponent
command.
■
Solaris Predictive Self-Healing (PSH) technology – A feature of the Solaris OS,
uses the fault manager daemon (
fmd
) to watch for various kinds of faults. When a
fault occurs, the fault is assigned a unique fault ID (UUID), and logged. PSH
reports the fault and provides a recommended proactive replacement for the
FB-DIMMs associated with the fault.