Software Issues

C H A P T E R 5

Software Issues

This chapter describes software issues related to Oracle’s Sun Blade X6250 server module.

Delay Before Server Can Be Powered On After Software Upgrade (6893973)

After performing a software upgrade, there is a delay of about three minutes before the server can be powered on.

Workaround

Wait three minutes or more before powering on a server after a software upgrade.

PCCheck Cannot Run Multiple Times (6875913)

If you stop and start the host more than once, PCcheck does not run after the first time, and an error message appears in dmesg:

BotReqHandler() : BOT Reset Request Called
ERROR: UsbWriteData(): Transmission timeout for Device 0 EP 2 for DataSize 0xD
Error in Sending CSW for EP2
WARNING:Wakeup for Sleep Struct(0xc71be6dc) Called after timeout
ERROR: UsbWriteData(): Transmission timeout for Device 0 EP 2 for DataSize 0x24
Error in Sending SCSI_INQUIRY

Workaround

Retry several times until it runs.

Upgrading to ILOM 3.0.3.33 Changes Onboard NIC Numbering and Breaks VMware Network Configuration (6853562)

Upgrading from ILOM 2.0.5 to 3.0 renumbers the onboard NIC devices, which breaks VMware network configuration.

Workaround

Reinstall VMware.

Multiple RKVM Sessions Hang and Reconnect Automatically (6846082)

This issue includes two related problems:

1. If you connect multiple RKVM sessions to a server module, then reset the server module’s SP, the sessions hang sporadically instead of continuing without interruption.

2. If you connect multiple RKVM sessions to a server module and disconnect one or more of the sessions, they reconnect automatically.

These issues are fixed in SW3.1.

SUSE Linux With CF Card Reports HD Errors (CR 6850576)

On systems with a CF card, SUSE 10/9 reports errors.

For example:

dmesg |grep -i fail
hda: drive_cmd: status=0x51 { DriveReady SeekComplete Error }
hda: drive_cmd: error=0x04 { DriveStatusError }

or you might see...

Error attaching device data
aer: probe of 0000:00:02.0:pcie01 failed with error 1
aer: probe of 0000:00:03.0:pcie01 failed with error 2

These messages can be ignored. They do not affect system performance.

Creating SSH Key During Upgrade to SW3.0 Takes Awhile (CR 6805802)

While upgrading to SW3.0, it can take about five minutes to create the SSH key. The screen displays:

SSH RSA Host Key files don't exist. Creating... 
Starting sshd

Workaround

While this might take awhile, it should still work. No action is necessary.

Intel Errata AX30 and AX52 Are Resolved with SW2.0.3 (CR 6826490)

Intel Errata AX30

Intel Errata AX30 states that programming the digital thermal sensor (DTS) threshold in the Intel 5400-series processor can cause an unexpected thermal interrupt.

This issue has been resolved by a software change in SW2.0.3.

This interrupt might not happen in all previously-released software versions. However it is recommended as a standard practice that all systems be upgraded to the latest software version to ensure that the latest fixes are installed.

Intel Errata AX52

Intel Errata AX52 states that a rare set of timing conditions and address alignments of instructions in a short nested loop sequence might cause a machine check exception or a system hang in software that contains multiple conditional jump instructions and spans multiple 16-byte boundaries.

This issue has been resolved by a software change in SW2.0.3.

OS Fails to Detect 146 or 300GB Hard Drives During Warm Boot

On systems with a Sun Blade RAID 5 expansion module, the operating system might fail to detect one or more 146 or 300 GB hard drives during a warm boot.

Workaround

If your server module has a Sun Blade RAID expansion module, and you upgrade to either 146GB (390-0448-02) or 300GB (390-0449-02) hard drive(s), you must update the Sun Blade RAID expansion module firmware to 15872 or newer.

The firmware is included in the Tools and Drivers DVD.

62GB of DIMM Available to OS on Systems with 64GB of DIMMs Installed

On systems with 64GB of DIMMs installed, only 62GB is available to the operating system.

In SW2.0.3, some enhancements were made to support PCIe EM hotplugging. These included increasing the option ROM space to 256MB per device. Because of this change, systems with 64GB of memory only show 62GB of memory available to the OS.

Note - On systems with less than 64GB of DIMMs, the full memory is available to the OS.

Oracle Solaris Reboots When PCIe EM is Hotplugged in Slot 0 (CR 6794631)

If you hotplug a PCIe EM in slot 0 on a system running Oracle Solaris, the system reboots. This only occurs if the system did not previously have a PCIe EM card in slot 0.

This problem only appears on systems equipped with Oracle Solaris 10u6 or later.

Workaround 1

Hotplug a PCIe EM card in slot 0, and allow the system to reboot.

Workaround 2

Download and install SW2.0.4.

System Hangs when Booting

Single-processor systems equipped with X5260 or X5270 processors and Oracle Solaris 10u5 or 10u6 hang when booting.

Workaround

Disable the core multiprocessing in the BIOS.

1. Start the BIOS.

2. Select Advanced -> CPU Configuration

3. Select Disable Core Multiprocessing.

4. Save your changes and exit.

Server Modules Equipped with Additional Memory Take Longer to Run POST

A server module that is fully populated with DIMMs can take up to three additional minutes to boot, due to the time required to run POST tests on the DIMMs.

VMware ESX/ESXi 3.5 u2 Health Status is Sometimes Reported Inconsistently

VMware ESX/ESXi releases provide a health status and monitoring feature. Incompatibility between the ILOM and VMware IPMI interfaces causes ILOM IPMI sensors to be partially displayed in the VMware health status window. Sensors might also appear and disappear, and updates occur slowly.

Workaround

Use the ILOM GUI, CLI, SNMP, or IPMI interfaces to monitor server health. Also, the ILOM provides remote syslog and remote alert interfaces to enable remote monitoring.

Arrow Keys Do Not Work When Using RKVM on SunRay

The arrow keys do not work when using an RKVM on a SunRay.

Workaround

Use alternative keystrokes, or remap the arrow keystrokes in the SunRay X session.

Keyboard Alt Key Appears Stuck in RKVM on RHEL4 and ESX 3.5 u2

During RKVM sessions with RHEL4 and ESX 3.5 u2, the Alt key appears stuck.

Workaround

From the javaRconsole menu:

Select Keyboard ->Left Alt Key.

Deselect Keyboard ->Left Alt Key.

Platform-Specific Drivers Installation Fails During Windows Installation

On older server modules, during Windows installations when you are installing platform-specific drivers, you might see the following error message:

Setup failed to detect the correct model and cannot continue

This occurs when the BIOS manufacturer string is not Sun Microsystems.

Workaround

Use DriverPack.zip and Device Manager, or OptPack.zip to install all drivers and applications manually, as described in the Sun Blade X6250 Windows Operating System Installation Guide.

Networking Performance Regression With Oracle Solaris 10 5/08 (CR 6666472)

Servers running Oracle Solaris 10 5/08 you might experience reduced network performance.

Workaround:

Install patch 137112-06.

The Oracle Solaris OS Will Not Install On Systems Configured with RAID Arrays Larger Than One Terabyte

The Oracle Solaris OS cannot be installed on systems configured with RAID arrays larger than one terabyte because Oracle Solaris installer doesn't support it.

Error Message Appears When Finished Updating Platform-Specific Drivers and Optional Components (CR 6701068)

Upon completing the platform specific drivers and optional components update, an error message states that some drivers could not be installed. FIGURE 5-1 shows an example.

FIGURE 5-1 Sample of Error Message

Screen shot showing example of error message

Workaround

This message can be ignored.

Systems with Red Hat Linux Might Hang

Some systems running Red Hat Enterprise Linux 4.5 or older might hang. This is because of a change in the Intel addressing scheme, allowing it to address greater than 36 bits.

Workaround

Update Red Hat Enterprise Linux to one of the following:

4.5 to kernel-2.6.9-55-0.12.EL (released November 1, 2007) or newer

Red Hat Enterprise Linux 4.6 (released November 14, 2007)

Red Hat Enterprise Linux 5

For more details, see:

http://kbase.redhat.com/faq/FAQ_85_11696.shtm

Question Mark in Device Manager When Sun Blade RAID 5 Expansion Module is Present

If you install the Windows OS on a server that includes a Sun Blade RAID 5 Expansion Module, a question mark might appear in the device manager next to the Adaptec Virtual SPGIO 0 Device.

Workaround

This is harmless. However you can get rid of the question mark by reinstalling the driver.

1. Right click the Adaptec Virtual SPGIO 0 Device entry in the device manager.

2. Select Update Driver Software.

The operating system re-installs the driver.

Driver Download Required Before Using StorageTek RAID Manager (CR 6606963)

Before you can use the Adaptec Storage Manager, you must download and install the latest aac driver.

The driver is included in the Tools and Drivers DVD.

To obtain the latest Tools and Drivers DVD, go to:

http://www.sun.com/servers/blades/downloads.jsp

PC-Check Hangs on Startup

PC-Check cannot start, and hangs at “Detecting Hardware.”

Workaround

1. Login to the web GUI and reset the system:

For systems with ELOM:

a. Select Remote Control => Remote Power Control.

b. Select Do not boot from PCCheck and click Submit.

c. Again select Remote Power Control => Reset, then click Submit.

For systems with ILOM:

a. Select Remote Control.

The Remote Control tabs appear.

b. Select Remote Power Control.

The Host Control page appears.

c. Select BIOS from the drop down list, then select Save.

d. Select the Remote Power Control tab, then select Reset from the drop-down menu.

Select Save.

The host resets and messages appear on the screen.

2. Press F2 to enter BIOS setup.

3. Navigate to Server => Remote Access Configuration and press Enter.

4. Select Remote Access and change it to Disabled.

5. Press F10 to save changes and exit BIOS.

You can now successfully start PCCheck tool.

VMware ESX 3.5 Process Uses 100% of Core (CR 6692155)

The VMware ESX 3.5 process occasionally uses 100% of a core. This is fixed in SW2.0.3.

For example:

# cat /proc/vmware/interrupts
Vector PCPU 0   PCPU 1   PCPU 2   PCPU 3   PCPU 4   PCPU 5   PCPU 6   PCPU 7
0x21:  7442833       0        0        0        0        0        0        0 VMK ACPI Interrupt

Workaround

1. Edit the file /etc/vmware/esx.conf and add the line /vmkernel/mpsIntRouting = "TRUE".

2. Reboot the system.

This issue is fixed in BIOS version 1ADPI40 of SW1.3.1.