Skip to main content

Problems Resolved in XCP 2320


Problems Resolved in XCP 2320
The following table lists the problems resolved in XCP 2320.
Table 3-24  Problems Resolved in XCP 2320
RTI No. RTIF2-201116-009
Model SPARC M10-4
Description If the ioxadm command is executed to update the firmware for the PCI expansion unit connected to PCI slot #10 in the SPARC M10-4, the LINKCARD number of the event log will not be correctly displayed.
Workaround There is no effective workaround.
RTI No. RTIF2-201116-012
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description If the fan unit or power supply unit is replaced without using the maintenance menu, an error will be detected and the event log will be registered, but the CHECK LED on the operation panel will not go on.
Workaround There is no effective workaround.
RTI No. RTIF2-201109-012
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description Hypervisor might abort or logical domains might panic when a diag report is issued in TOD degradation.
Workaround There is no effective workaround.
RTI No. RTIF2-201109-013
Model SPARC M10-4S
Description In a configuration with multiple PPARs, logical domains may panic or Hypervisor may abort when a PPAR is powered on while another PPAR is being powered off.
Workaround There is no effective workaround.
RTI No. RTIF2-161108-002
Model SPARC M10-4
Description If the PCI expansion unit is connected to PCI slot #9 of the SPARC M10-4, logical domains using the resources of the PCI expansion unit for PCI slot #9 may take a few more minutes to restart than other logical domains.
Workaround There is no effective workaround.
RTI No. RTIF2-160613-002
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description Security fixes.
(CVE-2016-0800)
For details, see the Oracle website relating to the following: Critical Patch Updates
Workaround Update the XCP firmware to XCP 2320 or later.
RTI No. RTIF2-160607-001
Model SPARC M10-4
Description Suppose that the PCI expansion unit is connected to PCI slot #9 of the SPARC M10-4. It is not possible to execute the ioxadm command to cause a chassis LED (locater) to blink to point to the PCI expansion unit connected to PCI slot #9. Nor is it possible to perform active replacement of the fan unit and power supply unit (PSU) of the PCI expansion unit.
Workaround There is no effective workaround. If the ioxadm command was executed for a PCI expansion unit connected to a slot other than PCI slot #9, execute the ioxadm command again. To replace a fan unit or PSU of a PCI expansion unit connected to PCI slot #9, power off the physical partition connecting the PCI expansion unit and then replace the fan unit or PSU.
RTI No. RTIF2-160520-001
Model SPARC M10-4
Description If the firmware of the PCI expansion unit is updated when the unit is connected to PCI slot #9 in the SPARC M10-4, the update may fail with the following message displayed.
Note - The update will fail even on a PCI expansion unit connected to a slot other than PCI slot #9.
 
[Example of output message]
XSCF> ioxadm -c update PCIBOX#8002 -s 1200
 Firmware update is started. (version=1200) Operation was not successful.
 
In a PCI expansion unit firmware update for the PCI expansion unit connected to PCI slot #9, it may take two hours or more until the ioxadm command ends with an error. Even though the firmware update of the PCI expansion unit failed, it does not affect the operation of the physical partition.
Workaround There is no effective workaround. If the ioxadm command was executed for a PCI expansion unit connected to a slot other than PCI slot #9, execute the ioxadm command again. To replace a fan unit or PSU of a PCI expansion unit connected to PCI slot #9, power off the physical partition connecting the PCI expansion unit and then replace the fan unit or PSU.
RTI No. RTIF2-160520-002
Model SPARC M10-4, SPARC M10-4S 
Description The connection of a PCI expansion unit may not be identifiable from the XSCF by executing the ioxadm -v list command after the physical partition (PPAR) is powered on, although it is identifiable from Oracle Solaris or OpenBoot PROM on the physical partition.
In this case, it is not possible to perform active replacement of the LINK card used with the XSCF functions, or a fan unit or power supply unit of the PCI expansion unit.
Workaround There is no effective workaround.
Power off the physical partition and power it on again.
RTI No. RTIF2-160520-003
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description If the graphics card (SP0X7GR1F) is used to start GNOME Display Manager (GDM), nothing may appear on the display unit.
This problem occurs when Oracle Solaris is in the factory-default configuration.
Workaround Perform either [Workaround 1] or [Workaround 2] below.

[Workaround 1]
Comment out the BusID in the desktop environment information used by GNOME.
This operation does not affect system performance, etc.
Perform the following procedure.
1. Edit BusID in /etc/X11/xorg.conf to make it a comment line (add "#" to the beginning of the line).

[Example] Comment out BusID written between the # First card start line and the # First card end line.
# First card start
(Omitted)
# BusID "PCI:8:0:0"
(Omitted)
# First card end
2. Apply the changed desktop environment information to GNOME.

- If you are logged in to GNOME, log off from GNOME, and log in again.

- If you are not logged in to GNOME, restart GDM, and log in to GNOME.

[Example] If you are not logged in to GNOME, restart GDM.
# /usr/sbin/svcadm restart gdm
 
[Workaround 2]
After saving logical domain configuration information, operate the system.
GFX 550e Driver Software must be reinstalled because this problem be the cause of incorrectly configured PCI information used by the graphics card. If the hardware configuration has changed (including a change to the CPU Activation settings), this operation needs to be performed again.
Perform the following procedure.
1. Start Oracle Solaris with the current configuration (factory-default) left as is.

2. Execute the ldm add-spconfig command to save the current logical domain configuration information.

[Example]
# /usr/sbin/ldm add-spconfig <config>
3. Stop Oracle Solaris.

[Example]
# /usr/sbin/shutdown -y -g0 -i5
4. Start Oracle Solaris, and reinstall the GFX 550e driver software.

After performing the procedure, operate the system with the logical domain configuration information saved in step 2.
RTI No. RTIF2-160520-004
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description If you execute the reset command for a logical domain while executing the reset command for another logical domain, the later reset command may be blocked. For this reason, if cluster software tries to switch clusters by executing the reset command, the cluster switching may fail.
The clusters are switched successfully in the end because the cluster software changes paths in sequential order and executes the reset command to switch the clusters.
Workaround There is no effective workaround.
RTI No. RTIF2-160512-001
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description The ldmd service may enter maintenance mode 734 days after starting the physical partition (PPAR). Consequently, none of the following will be available: Oracle VM Server for SPARC ldm command, logical domain operations (start/stop/change configuration/migration), automatic replacement of CPU cores, dynamic CPU/memory degradation, and monitoring (Host Watchdog) between a hypervisor and logical domain. In addition, when 1,101 days have elapsed after starting the PPAR, the logical domain may panic or a hypervisor may abnormally end (Hypervisor Abort).
Workaround There is no effective workaround. After starting a physical partition (PPAR), restart (stop/start) the PPAR before 734 days have elapsed.
[How to restore]
Restart (stop/start) the physical partition (PPAR).
RTI No. RTIF2-150521-002
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description When the ioxadm(8) command is executed, the message "Operation was not successful." is displayed and the command may fail.
[Example]
XSCF> ioxadm -c update PCIBOX#0000 -s 1234
Firmware update is started. (version=1234)
Operation was not successful.

When the prtfru(8) command is executed, the message "An internal error has occurred. Please contact your system administrator." is displayed and the command may fail.
[Example]
XSCF> prtfru
An internal error has occurred. Please contact your system administrator.
Workaround There is no effective workaround.
[How to restore]
Execute the command again.