Fujitsu M10 Product Notes page 162

Table 3-21 Problems resolved in versions prior to XCP 2050 (continued)
SPARC M10-
RTI No. 1 4 4S
RTIF2- x x x
121129-002
RTIF2- x x x
121129-004
RTIF2- x x x
121130-001
RTIF2- x x x
121204-001
RTIF2- x x x
121204-002
RTIF2- x x x
121204-003
RTIF2- x x x
121204-004
Fujitsu M10/SPARC M10 Systems Product Notes for XCP Version 2240 ･ March 2015
148
Description
When you replace the CPU memory unit
lower (CMUL) or motherboard unit
(MBU), the XSCF user account
information may be deleted.
To recover the XSCF user account
information, you need to use the
restoreconfig(8) command to restore the
information saved by the dumpconfig(8)
command.
The restoredefaults(8) cannot initialize
the following information. The setting
information remains as it is.
- Timezone for XSCF
- Server certificate for HTTPS
When you enable and disable the power
supply interlocking of RCIL, and then
enable it again, the power supply
interlocking of ETERNUS DX80/DX90/
DX410/DX440/DX8100/DX8700 S2 may
not work.
The dynamic resource management
software (ServerView Resource
Orchestrator) sometimes fails to register
a server to be monitored.
If a node of PRIMECLUSTER stops, node
switching does not occur automatically.
When you set or change environment
variables of OpenBoot PROM by setenv
or nvramrc of OpenBoot PROM, or by
the eeprom(1M) or ldm set-var
commands of Oracle Solaris, the setting
or changes may not be retained after
recycling the power supply to the system.
When you enable the auto boot function
of the guest domain by the setpparmode
(8) command of XSCF and then start the
control domain and the guest domain
simultaneously, you may encounter the
situation where the following error
message is recorded and Oracle Solaris
cannot be started in the guest domain.
Code: 20000000-00ffff0000ff0000ff-
030000020000000000000000
Status: Notice Occurred: Nov 16
16:55:25.518 JST 2012
FRU: /UNSPECIFIED
Msg: Boot process failed
Workaround
After you have conducted the replacement,
use the restoreconfig(8) command to
restore the information saved by the
dumpconfig(8) command, or set the
XSCF user account information again.
There is no effective workaround.
Reset by the settimezone(8), sethttps(8),
and setssh(8) commands.
When you enable and disable the power
supply interlocking of RCIL, restart XSCF.
There is no effective workaround.
Register the server to be monitored again
by the dynamic resource management
software.
There is no effective workaround.
Switch nodes manually in PRIMECLUSTER.
After you update an environment
variable of OpenBoot PROM, execute the
ldm add-config command to save the
domain configuration information to
XSCF.
Use the XSCF setpparmode(8) to disable
the auto boot function of the guest
domain, and then start Oracle Solaris of
the control domain. Then use the ldm
start command of Oracle VM Server for
SPARC to start the guest domain.
[How to restore]
Use the ldm stop command of Oracle VM
Server for SPARC to stop the guest
domain, and then use the ldm start
command to start the guest domain.