Skip to main content

Problems Resolved in XCP 2230


Problems Resolved in XCP 2230
The following table lists the problems resolved in XCP 2230.
Table 3-35  Problems Resolved in XCP 2230
RTI No. RTIF2-201118-007
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description If a POST diagnosis detects a partial failure of a DIMM, the DIMM may be excessively degraded.
Workaround There is no effective workaround.
RTI No. RTIF2-201118-008
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description When the NTP client function is used with the XSCF, the memory area of ​​the XSCF may become full with NTP-related logs, which may cause firmware update failures due to insufficient free space.
[Example of message]
XCP update is started (XCP version=yyyy:last version=xxxx)
XCP image is invalid or corrupt (XCP version=yyyy)
XCP update is failed (XCP version=yyyy:last version=xxxx)
Workaround There is no workaround.
[How to restore]
Perform the following procedure.
1. Execute the getflashimage -d command to delete all XCP files.

2. Execute the getflashimage command again to import XCP files.

3. Execute the flashupdate command to update the firmware.

If the failure recurs, perform the following steps to initialize the memory area.
4. Use the dumpcodactivation command to save the CPU Activation key.

5. Execute the restoredefaults -c xscf command to clear XSCF log information and the CPU Activation key.

6. Use the restorecodactivation command to restore the saved CPU Activation key.
RTI No. RTIF2-201118-010
Model SPARC M10-4S
Description If a Fatal error occurs while the physical partition dynamic reconfiguration (PPAR DR) function is being used to delete a PSB, the PPAR may hang without a reset for automatic system restore.
Workaround There is no effective workaround.
RTI No. RTIF2-201109-015
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description Hypervisor might abort when multiple hardware errors occur simultaneously.
Workaround There is no effective workaround.
RTI No. RTIF2-140910-001
Model SPARC M10-4S
Description If Oracle Solaris 11.2 and later is running and the system board is deleted by dynamic reconfiguration of physical partitions, Hypervisor may abort.
Workaround There is no effective workaround.
RTI No. RTIF2-140819-001
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description When the XSCF is rebooted or the input power is turned on again, the following problem may, in rare cases, occur: ntpd is left inactive and, as a result, synchronization of the XSCF time with the NTP server time fails. To check whether ntpd is inactive, execute the showntp -l command. If the message "NTP is unavailable." appears, ntpd is inactive.
This problem does not occur if the setting specifying the XSCF as an NTP client or NTP server is disabled.
Workaround There is no effective workaround.
[How to restore]
If the setting specifying the XSCF as an NTP client or NTP server is enabled and ntpd is inactive, execute the rebootxscf(8) command to reboot the XSCF.
RTI No. RTIF2-140815-001
Model SPARC M10-4S
Description If the system board is deleted by the dynamic reconfiguration of physical partitions (DR), the monitoring of Host Watchdog for logical domains may stop.
Workaround After the system board is deleted, to restart the monitoring of Host Watchdog, reboot one of the logical domains in the physical partition (PPAR).
RTI No. RTIF2-140813-001
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description When an attempt is made to install Oracle Solaris via vnet, installation menu activation may take one hour or more in the following case: the network speed between the SPARC M10 system chassis and the installation server exceeds 1 Gbps or the system consists of multiple SPARC M10-4S chassis.
[Example] Network installation on a SPARC M10-4S via vnet
{0} ok boot net:dhcp - install
Boot device: /virtual-devices@100/channel-devices@200/network@0:dhcp File and args: - install
<‌time unavailable>‌ wanboot info: WAN boot messages->console
<‌time unavailable>‌ wanboot info: configuring /virtual-devices@100/channel-devices@200/network@0:dhcp
<‌time unavailable>‌ wanboot info: Starting DHCP configuration
<‌time unavailable>‌ wanboot info: DHCP configuration succeeded
<‌time unavailable>‌ wanboot progress: wanbootfs: Read 368 of 368 kB (100%)
<‌time unavailable>‌ wanboot info: wanbootfs: Download complete
Wed Aug 13 06:24:51 wanboot progress: miniroot: Read 4864 of 243471 kB (1%)
(*1)
  *1 This step takes time.
Workaround Install Oracle Solaris from a vdisk or DVD-ROM created using an ISO image.
RTI No. RTIF2-140808-001
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description On a SPARC M10 system connected to an uninterruptible power supply (UPS), the following problem may occur: after recovery from a power failure occurring while Oracle Solaris is inactive or the physical partition (PPAR) is powered off, the PPAR may erroneously reboot even though the PPAR should remain powered off.
Note - When a power failure occurs while Oracle Solaris is active and a UPS is connected, Oracle Solaris will automatically reboot upon power recovery.
This phenomenon may occur when the operation below is followed by a power failure recurrence and then a recovery therefrom.
- A power failure occurs while Oracle Solaris is active. After Oracle Solaris boots automatically upon power recovery, the shutdown command is executed on Oracle Solaris to stop Oracle Solaris (display the ok prompt).

[Example]
# shutdown -y -g0 -i0
- A power failure occurs while Oracle Solaris is active. After Oracle Solaris boots automatically upon power recovery, the shutdown command is executed on Oracle Solaris to power off the PPAR.

[Example]
# shutdown -y -g0 -i5
Workaround If you want to stop Oracle Solaris, there is no effective workaround. To power off the PPAR, execute the poweroff(8) command on the XSCF.
[How to restore]
Execute the poweroff(8) command on the XSCF to power off the PPAR.
RTI No. RTIF2-140808-004
Model SPARC M10-1
Description A PCI Express correctable error may occur if you connect a PCI Express expansion unit (connected by using PCI Express Gen1), which is an IHV product, to a SPARC M10-1.
Workaround There is no effective workaround.
RTI No. RTIF2-140805-001
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description
1. If the building block (PSB) is deleted by a dynamic reconfiguration of physical partitions (PPAR DR), a process on the logical domains may produce a core dump due to SIGILL (Illegal Instruction). (SPARC M10-4S only)

2. After a live migration of a guest domain, some commands using the libcrypt library (e.g., ssh, pkg, zonename) on the guest domain may produce a core dump due to SIGILL (Illegal Instruction) if the guest domain is Oracle Solaris 11.1 or Oracle Solaris 11.2 with PRIMECLUSTER GD or PRIMECLUSTER GDS applied to the guest domain.
Workaround There is no effective workaround.
RTI No. RTIF2-140731-001
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description An Event Log ("Host stopped") is not displayed by a showlogs(8) event and showmonitorlog(8) when the primary domain is shutdown.
Workaround There is no effective workaround.
RTI No. RTIF2-140623-002
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description If you specify "pop" as the authentication method by executing the setsmtp(8) command, e-mail sending may fail even after you execute the setemailreport(8) command to enable e-mail notification.
Workaround When executing the setsmtp(8) command to specify an authentication method, specify either "smtp-auth" or "none" instead of "pop".
RTI No. RTIF2-140623-003
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description Suppose that the XSCF STANDBY LED on the operation panel blinks immediately after turning on the input power (AC ON). If the power switch on the operation panel is pressed, powering on/off the physical partition (PPAR) becomes disabled. The power to the PPAR is turned on/off when:
- The power switch on the operation panel is pressed.

- The poweron(8) or poweroff(8) command is executed.

- PPAR power on/off operation is performed on XSCF Web.

- Remote power management using RCIL takes place.

- The auto power control system (APCS) issues a power on/off instruction.
Workaround To operate the power switch on the operation panel immediately after turning on the input power, wait at least 30 seconds after the STANDBY LED switches from blinking to being constantly on.
RTI No. RTIF2-140606-002
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description Even though the "AEV_AUTHENTICATE" audit event is enabled, XSCF Web login failure events are not registered in the audit log.
Workaround There is no effective workaround.
RTI No. RTIF2-140606-003
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description When an attempt is made to look at an audit log concerning XSCF Web operation using the viewaudit(8) command, it displays incorrect values. The displayed values of the port number and IP address at the time of XSCF Web access are "0" and "0000".
Workaround There is no effective workaround.
RTI No. RTIF2-140606-006
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description When specifying a system board (PSB) for a logical system board (LSB) on the [PPAR Operation] - [PPAR Configuration] screen of XSCF Web, you need to note the following: A PSB cannot be specified if it has already been assigned to another physical partition (PPAR).
Workaround For the setting, use the setpcl(8) command in the XSCF shell.
RTI No. RTIF2-140606-007
Model SPARC M10-4S
Description Suppose that the system is made up of multiple physical partitions (PPARs) consisting of multiple SPARC M10-4S chassis. If you turn on the power to multiple PPARs at the same time, for the first time after turning on the input power or rebooting the XSCF, the following event may occur: A PPAR ID error is registered, the self-diagnosis test (POST) on some PPARs is not completed, and these PPARs fail to start.
You can confirm this event by executing the showpparstatus command.
[Example] Only PPAR #4 is indicated with "Initialization Phase" and fails to complete POST processing.
XSCF> poweron -a
XSCF> showpparstatus -a
PPAR-ID       PPAR Status
00            Initialization Complete
01            -
02            -
03            -
04            Initialization Phase
05            -
06            Initialization Complete
07            -
08            -
09            -
10            -
11            Initialization Complete
12            -
13            -
14            -
15            -
XSCF>
Workaround Do not use the poweron -a command to turn on the power to multiple PPARs at the same time. Use the poweron -p command to turn on the power to PPARs one at a time.
[How to restore]
Turn off the power to PPARs by using the following procedure and then turn it on again:
1. Execute the poweroff -f command to forcibly turn off the power to the PPARs that have failed to start.

XSCF> poweroff -y -f -p ppar_id
2. Confirm that the status of each PPAR specified in step 1 is "Powered OFF".

XSCF> showpparstatus -p ppar_id
--- Omitted ---

04 Powered Off
--- Omitted ---

3. Turn on the power to the PPAR again.

XSCF> poweron -y -p ppar_id
RTI No. RTIF2-140527-001
Model SPARC M10-4S
Description On a system that is configured with several SPARC M10-4S, if an XSCF reboot is not performed with the rebootxscf(8) command or if the input power is not turned off for about 10 days in respect to the XSCF in the following status, the "XSCF self-diagnosis warning detection" is registered in the error log:
- A standby XSCF which did not perform master/standby switchover for about 10 days.

- Slave XSCF.
Workaround There is no effective workaround.
Ignore this error log entry.
RTI No. RTIF2-140212-009
Model SPARC M10-1, SPARC M10-4, SPARC M10-4S
Description When the XSCF is under high load due to a high volume of packets sent to the XSCF network by the L2 loop etc., the false error "RTC read data error" or "Hardware access error" regarding the power unit (PSU) is detected. Moreover, the PSU is marked as failed due to this error.
Workaround If the reason for the high load is the L2 loop, eliminate the root cause by reviewing the network connection etc.
[How to restore]
To remove the "Failure" mark from the PSU, set the mode switch on the operation panel to Service mode, turn off the input power, and then turn it on again. After that, restore the mode switch to its original state.