Fujitsu M10 Product Notes page 46

Table 3-3
Problems resolved in versions prior to XCP 2043 (continue d)
SPARC
RTI No. M10-1
RTIF2-121219-
018
RTIF2-121219- x
019
RTIF2-130109-
002
RTIF2-130109- x
005
Fujitsu M10/SPARC M10 Systems Product Notes for XCP Version 2043 ･ October 2013
34
SPARC SPARC
M10-4 M10-4S Description
x In a system in which two or
more SPARC M10-4S cabinets
comprise multiple physical
partitions (PPARs), if you
power on the PPAR after
switching of the master XSCF
due to any of the following
events, error logs of
"STICK does not start (CPU)",
"STICK does not start (MBC)",
or "STICK count up error" may
be registered, causing
degradation of a CPU memory
unit (CMUU/CMUL).
Performing firmware update
■
Detecting an XSCF failure
■
Executing the switchscf(8)
■
command
x x When a PCI card is inserted to
or removed from a PCI
expansion unit, a number that is
different from the relevant PCI
card number (PCI#) is displayed
in the log. In addition, when an
error log "PCICARD failed" is
registered due to an error in the
PCIe slot on the PCI expansion
unit or the PCI card, a number
that is different from the
relevant PCI card number
(PCI#) is displayed in the error
log.
x If "Console path is switched" is
registered in the event log in a
system where a physical
partition (PPAR) consists of
multiple system boards (PSBs),
the PPAR-ID may show an
invalid value.
x x If you use the replacefru(8)
command to replace a power
supply unit (PSU), "Indispensable
parts are not installed (PSU)"
may be registered in the event
log.
Workaround
There is no effective workaround.
After switching of the master
XSCF is performed, power off
the input power supply of all
SPARC M10-4Ss and power
them on again, without
powering on the PPAR.
When any of error logs "STICK
does not start (CPU)", "STICK
does not start(MBC)", or "STICK
count up error" is registered
and a failure mark (*) appears
on a part when you power on
the PPAR after switching of the
master XSCF, contact field
engineers.
Read it as the number obtained
by subtracting one from the PCI
number (PCI#) displayed in the
log.
There is no effective workaround.
This is an error log that is
registered because a PSU is
removed. Ignore it.