C H A P T E R 1 - Introduction to the Sun Fire V20z and Sun Fire V40z Servers

C H A P T E R 1

Introduction to the Sun Fire V20z and Sun Fire V40z Servers

Systems Overview

Strong server-management capabilities are crucial to maintaining mission-critical servers. Advance notification of problems and rapid diagnosis and correction are critical functions to an environment in which a few servers bear the bulk of the workload. The Sun Fire V20z and Sun Fire V40z servers and their extensive server-management capabilities lower costs by reducing failure and by potentially eliminating hands-on management.

This document describes how to perform remote management on the Sun Fire V20z and Sun Fire V40z servers.

The Sun Fire V20z server is an AMD Opteron processor-based, enterprise-class one rack unit (1U), two-processor (2P) server. The Sun Fire V40z server is also an AMD Opteron processor-based server, but it is a three rack unit (3U), four-processor (4P) server.

These servers include an embedded service processor (SP), flash memory, RAM, a separate Ethernet interface, and server-management software. They come equipped with superior server-management tools for greater control and minimum total cost of ownership. You can use the command-line interface (CLI), SNMP integration with third-party frameworks, or IPMI to configure and manage the platform with the SP. The dedicated SP provides complete operating-system independence and maximum availability of server management.

User Documentation

For the most up-to-date user documentation, for both the Sun Fire V20z and Sun Fire V40z servers, visit the following web site:

http://www.sun.com/products-n-solutions/hardware/docs/ Servers/Workgroup_Servers/Sun_Fire_V20z/index.html

To verify whether a document on the site is more recent than the document that you have, refer to the final two digits (the dash-roll) of the part number for that document or review the release date of the document.

Note - A document explaining the differences among the released versions of the Sun Fire V20z and Sun Fire V40z servers is also available at this web site. Refer to part number (PN) 817-7185.

Acronyms

TABLE 1-1 defines the acronyms found in this document.

TABLE 1-1 Acronyms Used in This Manual
Acronym	Explanation
ACPI	Advanced Configuration and Power Interface
ADS	Active Directory Service
ARP	Address Resolution Protocol
BMC	baseboard management controller
CRU	customer-replaceable unit
DPC	Direct Platform Control
FRU	field-replacement unit
grub	Grand Unified Bootloader
HDT	hardware debugging tool
ICMB	Intelligent Chassis Management Bus
IPMB	Intelligent Platform Management Bus
IPMI	Intelligent Platform Management Interface
KCS	Keyboard Controller Style
KVM	keyboard, video, and mouse
LAN	local area network
LILO	Linux Loader
LOM	lights out management
MIB	Management Information Base
NNM	Network Node Manager
NPS	Newisys Platform Software
PEF	platform event filter
PET	platform event trap
PRS	Power and Reset Sequencer
PXE	Preboot Execution Environment
RMCP	Remote Management Control Protocol
RPM	Red Hat Package Manager
RTC	real-time clock
SDR	sensor data record
SDRR	sensor data record repository
SEL	system event log
SMTP	Simple Mail Transfer Protocol
SNMP	Simple Network Management Protocol
SOL	serial-over-LAN
SP	service processor
SSL	Secure Sockets Layer
SSU	System Setup Utility
SunMC	Sun Management Center
UART	universal asynchronous receiver/transmitter
UDP	User Datagram Protocol
UMB	Upper Memory Block
UTC	Coordinated Universal Time
VPD	vital product data
WAN	wide area network

Server Management

There are two options for remotely managing a Sun Fire V20z or Sun Fire V40z server:

lights out management (LOM) through IPMItool

Simple Network Management Protocol (SNMP)

Service Processor

The Sun Fire V20z and Sun Fire V40z servers include a dedicated chipset for complete operating-system independence and maximum availability of server-management functions. This chipset, called a service processor (SP), is an embedded PowerPC chip providing the following:

Environmental monitoring of the platform (such as temperatures, voltages, fan speeds, and panel switches)

Alert messages when problems occur

Remote control of server operations (boot, shutdown and reboot of the server's operating system, turning the server's power on and off, stopping the server's boot process in BIOS, and upgrading the BIOS)

Note - In this document, you might see references to a baseboard management controller (BMC). A BMC is a dedicated IPMI controller. The SP found in these servers is a general-purpose, embedded CPU that contains software to emulate a BMC.

The SP runs an embedded version of Linux software, and all the server-management functions are developed as standard Linux applications. Its sole purpose is to support server management; therefore, the full functionality of the operating system is not available in the SP. Many familiar applications, such as FTP and Telnet, are not provided as they are not required to support the server-management feature set.

FIGURE 1-1 shows the back panel of the Sun Fire V20z server.

FIGURE 1-2 shows the back panel of the Sun Fire V40z server.

FIGURE 1-1 Sun Fire V20z Back Panel

Graphic showing the back panel of the Sun Fire V20z server.

FIGURE 1-2 Sun Fire V40z Back Panel

Graphic showing the back panel of the Sun Fire V40z server.

Server-Management Interfaces

These servers include local and remote server-management capabilities through the SP; the SP supports four server-management interfaces:

IPMI using a Keyboard Controller Style (KCS) interface and an IPMI kernel driver (in-band)

IPMI over local area network (LAN) (out-of-band)

SNMP integration with third-party SNMP management consoles

Command-line interface (CLI) lights out management (LOM)

Command-Line Interface

Server-management capabilities are available from the command line.

See Appendix A for a list of server-management commands that you can use with these servers, as well as a description, the command format, a list of arguments, and a list of return codes for each command.

SSH and Scripting Capabilities

A system administrator can log in to the SP using Secure Shell (SSH) and issue commands, or more commonly, write a shell script that remotely invokes these operations.

The server-management commands enable you to efficiently manage each area of the server. From the command line, you can write data-driven scripts that automate the configuration of multiple machines. For example, a central management system can cause many servers to power on and boot at a specified time, or when a specific condition occurs.

For more information about scripting, see Further Management Information.

SNMP Integration

Simple Network Management Protocol (SNMP) management provides remote access by SNMP-compliant entities to monitor the health and status of the server. The SP sends SNMP alerts to external management functions when warranted.

For more information about SNMP, refer to SNMP Server Management.

The diagram in FIGURE 1-3 illustrates the communication paths for the different server-management options.

FIGURE 1-3 Diagram of the Server-Management Options

This diagram shows how the various server-management options work together. [ D ]

Operator Panel

You can use the operator panel to configure network settings for the SP. See FIGURE 1-4 or FIGURE 1-5 for the operator panel location on your server.

Note - The SP defaults to Dynamic Host Configuration Protocol (DHCP) networking if the operator panel is not interactively engaged on the first power-up.

FIGURE 1-4 Sun Fire V20z Server Operator Panel and Buttons

Operator panel located on the front of the Sun Fire V20z server, with the LCD operator panel in the lower-right. The buttons are Back, Select, Forward, Cancel, and Enter.

FIGURE 1-5 Sun Fire V40z Server Operator Panel and Buttons

Operator panel located on the front of the Sun Fire V40z server, with the LCD operator panel in the upper-right. The buttons are Back, Select, Forward, Cancel, and Enter.

The operator panel displays information on the operator panel display in two lines; you respond to prompts or initiate actions using the following buttons.

TABLE 1-2 Operator-Panel Buttons
Buttons	Function
	Back-Move or step backward through data options for a field. Moves through the bottom line of text, only.
	Select-Move or step forward through menus and fields that display in the top line of text and through field values for octets that display in the bottom line of text. Confirm and save a selected data option that displays in the bottom line of text. (To confirm submenu fields that require octets, use the Enter button combination.)
	Forward-Move or step forward through data options for a field. Moves through the bottom line of text, only.
	Enter-(Select plus Forward, the checkmark combination.) Confirms and saves a selected data option in submenu fields that require octets, such as IP address, netmask, or gateway.
	Cancel-(Back plus Select, the X combination.) Cancels the previous confirmation and steps backward to the previous display.

Characteristics of Operator Panel Displays

The Enter combination (Select plus Forward) is indicated by a checkmark. This combination confirms a data choice in submenu fields that require octets, such as an IP address. You must press both buttons simultaneously, and you must release both buttons simultaneously. (In most fields, you can press Select to confirm a data choice.)

The Cancel combination (Back plus Select) is indicated by an X. This combination cancels an action, backs up in a menu, and undoes other actions, depending on the menu. You must press both buttons simultaneously, and you must release both buttons simultaneously.

For numerical value in octets, such as an IP address, you can press and hold the Back or Forward button to activate the auto-scrolling feature. This enables you to move through the range of numbers more quickly.

A menu or data entry screen that displays for more than 30 seconds with no action taken is canceled, and the display returns to the idle/background state.

For every action that you confirm, feedback is displayed to indicate success, failure, or that the action has been initiated.

For a complete list of the menu options on the operator panel, refer to "Operator Panel" in Chapter 1 in the Sun Fire V20z and Sun Fire V40z Servers--User Guide.

User Groups

Administrators can define several different user groups, or types, on the server. Capabilities of the different user types are defined in TABLE 1-3.

For example, when you log in to the system the first time using the setup account, you must set up the initial manager account so that other user accounts can be managed. (See Creating the Initial Manager Account for details.)

[ D ]

TABLE 1-3 User Types
User Type	Capability
monitor	Read-only access for sensor data and log displays
admin	All capabilities except user-account management and SP field upgrades
manager	All capabilities except SP field upgrade
service	SP field upgrades

Users

There are two classes of SP users: one class of users can log in to the SP through SSH; the other class of users can establish IPMI sessions to the SP.

These two classes of users are managed independently:

Users who are created using the IPMI cannot access the SP through SSH.

Users who log in through SSH cannot access the SP through the IPMI.

It is possible to configure the SP so that directory-services (ADS/NFS) users can log in to the SP through SSH. However, these directory-services users cannot log in to the SP through the IPMI.

Passwords Files

Passwords for local, non-IPMI users are stored in a standard Linux shadow-password file, which enhances the security of the system. The hashed passwords are in a file that is not readable by users.

Passwords for IPMI users are stored separately. The IPMI password file is not readable by users, but passwords are stored unencrypted because of limitations imposed by the IPMI authentication algorithms.

Systems Management Tasks

To accomplish most systems management tasks, you can use any of the systems management tools that are included with your server. TABLE 1-4 lists some common systems management tasks and the tools that you can use to accomplish each task. For information about the systems management commands, see the appendixes in this guide. Commands documented in the online help are noted in the table. For information about running diagnostic tests, see the Sun Fire V20z and Sun Fire V40z Servers--Troubleshooting Techniques and Diagnostics Guide.

TABLE 1-4 Systems Management Tasks
Task	Systems Management Console	Operator Panel	Systems Management Command	SNMP	IPMI
Analyze events	Yes: online help	Yes, minimal	Yes	Yes	Yes
Autoconfigure SP	Yes	Yes	Yes	No	No
Configure directory services	Yes	No	Yes	No	No
Configure external file system	Yes	No	Yes	No	No
Configure network settings	Yes	Yes	Yes online help	No	Yes
Configure scripting capabilities	Yes	No	Yes online help	No	No
Configure SMTP event notification	Yes	No	Yes	No	N/A
Configure SP date and time settings	Yes	No	Yes	No	N/A
Configure SSL	Yes	No	Yes	No	N/A
Create initial manager account	Yes	No	Yes	No	N/A
Define default system name in Operator Panel	No	Yes	Yes	No	No
Monitoring system status	Yes: online help	Yes	Yes	Yes	Yes
Power on and off	Yes	Yes	Yes	N/A	Yes
Remove software	N/A	N/A	Yes	No	N/A
Run diagnostics tests	Yes	No	Yes	No	No
Run Troubleshooting Dump Utility	No	Yes	Yes	No	No
Set SP hostname	Yes	Yes	Yes	No	No
Set up network share volume	Yes	No	Yes: online help	No	N/A
Start and stop platform OS	Yes	Yes	Yes: online help	No	Yes
Update software	Yes	Yes	Yes	No	Yes: only the SP
Update SP software	Yes	Yes	Yes	No	Yes

Initial Setup of the SP

This procedure describes the steps for the initial setup of the SP.

Task I: Assigning Network Settings to the SP

This section contains two alternative methods you can use to define SP network settings:

Assigning SP Network Settings Using DHCP

Assigning Static SP Network Settings

Note - As an alternative, if no DHCP server or physical access is available, you can configure the SP using IPMItool in conjunction with an IPMI kernel driver. To configure your server for IPMI, perform the correct procedures for your operating system in Task III: Enabling IPMI Access on the Server, then Task IV: Enabling IPMI LAN Access.

Assigning SP Network Settings Using DHCP

The following procedure describes how to configure the SP network settings using DHCP from the operator panel. If your network does not use DHCP, or you want to assign a static IP address to the SP, follow the instructions in Assigning Static SP Network Settings.

Note - This procedure assumes that you have cabled the server and powered it on as described in the Sun Fire V20z and Sun Fire V40z Servers--Installation Guide. At least one of the server's SP ports must be connected to a LAN.

1. Press any operator-panel button on the server front panel (see FIGURE 1-6).

The operator panel displays the first menu option:

Menu: Server Menu

FIGURE 1-6 Operator-Panel Buttons

The three operator-panel buttons: from left to right, Back, Select, and Cancel. Press Back and Select for Cancel; press Forward and Select for Enter.

2. Press the Forward button until you reach the SP menu.

Menu: SP menu

3. Press the Select button to display the SP menu options.

SP Menu: Set SP IP info?

4. Press the Select button.

The following prompt appears with the default response:

SP use DHCP? No

5. Press the Forward button to change to Yes, then press the Select button.

6. Press the Select button at the confirmation prompt.

SP use DHCP: Yes?

The server attempts to contact a DHCP server for an IP address. When the server receives a DHCP response, the operator panel displays the DHCP-assigned SP IP address. The SP IP address is configured and the server is ready for use.

Note - Depending on your network conditions, it may take five to ten seconds for the new IP address allocated by the DHCP server to appear in the operator panel.

Note - A prompt appears that asks whether you want to perform autoconfiguration. As an alternative to configuring an SP manually, you can run autoconfiguration, which replicates the configuration of one SP to another. Refer to Autoconfiguring the SP for instructions on autoconfiguration.

7. Continue with Task II: Securing the SP for instructions on creating the initial manager account.

Assigning Static SP Network Settings

From the operator panel, follow these steps to configure the SP network settings using a static IP address. You must specify a subnet mask and default gateway. This example uses the following sample settings:

IP Address: 10.10.30.5
Subnet Mask: 255.255.255.0
Default Gateway: 10.10.30.254

1. Press any operator-panel button on the server front panel (see FIGURE 1-6).

The operator panel displays the first menu option:

Menu: Server Menu

2. Press the Forward operator-panel button until you reach the SP menu.

Menu: SP menu

3. Press the Select operator-panel button to display the SP menu options.

SP Menu: Set SP IP info?

4. Press the Select operator-panel button.

The following prompt is displayed with the default response:

SP use DHCP? No

5. Press the Select operator-panel button.

The operator panel displays the following:

SP IP Address: 0.0.0.0

6. With the cursor in the first field, increase or decrease the value using the Back or Forward operator-panel button.

This field can hold a value between 0 and 255.

SP IP Address: 10.0.0.0

7. After reaching your desired value, press the Select operator-panel button to advance the cursor to the next field.

SP IP Address: 10.0.0.0

8. Repeat Step 6 and Step 7 for each field until the desired IP address is displayed, then use the Enter button combination to save the IP Address.

The process continues to the next network setting, the subnet mask. The operator panel displays the following:

SP netmask: 255.255.255.0

9. Edit the subnet mask setting in the same manner as you did for the IP address. When finished, use the Enter button combination to save the subnet mask.

The process continues to the next network setting, the default gateway. The operator panel displays the following:

SP IP Gateway 10.10.30.1

10. Edit the default gateway setting in the same manner as you did for the IP address and the subnet mask. When finished, use the Enter button combination to save the default gateway.

The operator panel displays the following confirmation prompt:

Use new IP data: Yes?

11. Press the Select operator-panel button to use the new data, or use the Cancel button combination to disregard the new information.

The SP address is now configured and the server is ready for use.

12. Continue with Task II: Securing the SP.

Task II: Securing the SP

After you install the server and configure the SP's network settings, you must create the initial manager account. You can then perform initial configuration of the server and create additional user accounts. Only the administrator who does the initial system configuration can create the initial manager account.

Caution - The SP must be secured with a user name and password when the server is first deployed. Failure to secure the SP can expose the server to a potential denial-of-service attack through the SP network interface.

Creating the Initial Manager Account

A setup account is included with each server. This setup account has no password. When you log in to the SP the first time using the setup account, you are prompted to define the initial manager account with a password and an optional public key.

User names and passwords are strings that consist of any alphanumeric character, underscore, hyphen, or period.

User names must be unique and must begin with an alphabetic character.

Passwords can contain any printable character and are case sensitive.

A user name or a password is limited to 32 characters and cannot be a null or an empty string.

There are two methods you can use to create the initial manager account:

From a command line: see Creating the Initial Account From a Command Line.

From the Systems Management (SM) console: see Creating the Initial Account From the SM Console.

Creating the Initial Account From a Command Line

1. Using an SSHv1 or SSHv2 client, connect to the IP address of the SP.

2. Authenticate as the user setup with no password required.

# ssh spipaddress -l setup

3. Follow the on-screen prompts to create the initial manager account.

After you create the initial manager account, the setup account is deleted and you are logged out of the server. You can then log in using the new initial manager account, from which you can create other user accounts.

Note - If you are prompted for a password, this indicates that the SP has already been secured with an account. If you do not know the management user name and password, you can reset the SP from the operator panel by navigating to the SP menu and selecting the Use defaults option. Note that all current settings for users and networks will be lost and the SP will reboot.

Creating the Initial Account From the SM Console

For information about the SM Console features, see Systems Management Console Features.

To create the first manager account from the SM Console:

1. Type the SP name or IP address as the URL or address in a browser, to enter the SM Console.

The main SM Console screen appears.

Note - When you create the initial manager account, you are prompted to accept a license agreement. After you create the initial manager account, this prompt no longer appears.

2. At the Create Initial Manager-Level User ID screen, type a user name for this account.

3. Type a password for the account.

4. Re-enter the password to confirm the password.

5. Click the checkmark button.

6. Use the SM Console to select initial configuration options.

After you create the initial manager-level user, the Initial Configuration Checklist screen is displayed in the SM Console. This screen enables you to determine the options you want for the initial setup of the SP.

The Initial Configuration Checklist is a table that lists the SM Console menu options and the commands you use to configure each option. It also includes links to the online help that provides instructions for each option.

Note - This table is displayed only after you create the initial manager-level user. Therefore, only the administrator who initially configures the account or who resets it via the operator panel can access it.

Note - The IP address, user name, and password that you configure are referred to in subsequent examples as the spipaddr, spuser, and sppasswd.

Task III: Enabling IPMI Access on the Server

This section contains two alternative procedures: one for a Linux software-based server and one for a Solaris x86-based server. Use the procedure that corresponds to your OS:

Enabling IPMI Access on a Linux-Based Server (In-Band)

Enabling IPMI Access on a Solaris x86-Based Server (In-Band)

Enabling IPMI Access on a Linux-Based Server (In-Band)

1. Log in to the server and authenticate as the root user.

2. Install the custom OpenIPMI Linux kernel driver from the Sun Fire V20z and Sun Fire V40z Servers Documentation and Support Files CD.

The drivers are located in the CD directory /support/sysmgmt/.

Browse to the OS variant installed on your server. The options are:

redhat/rhel3 for Red Hat Enterprise Linux 3 (32-bit mode uses the architecture type "i386"; 64-bit mode uses the architecture type "x86_64")

suse/sles8 for SUSE Linux Enterprise Server 8 (32-bit mode uses the architecture type "i386"; 64-bit mode uses the architecture type "x86_64")

suse/sles9 for SUSE Linux Enterprise Server 9 (64-bit mode uses the architecture type "x86_64")

suse/suse9 for SUSE Linux Professional Server 9

3. Ensure that the kernel-source RPM is already installed on your distribution by running the command:

# rpm -qvi kernel-source

If this utility reports that the kernel-source software package is not installed, install the kernel-source RPM that is current for your installed Linux distribution.

On SUSE distributions, install the kernel-source RPM by running the command:

# yast2

On Red Hat distributions, download the current kernel-source RPM to a temporary directory (such as /tmp). Install the package by running the command:

# rpm -ivh /tmp/kernel-source*.rpm

4. Install the OpenIPMI Linux kernel driver RPM.

a. Browse to the OS variant installed on your server. The options are:

redhat/rhel3 for Red Hat Enterprise Linux 3 (32-bit mode uses the architecture type "i386"; 64-bit mode uses the architecture type "x86_64")

suse/sles8 for SUSE Linux Enterprise Server 8 (32-bit mode uses the architecture type "i386"; 64-bit mode uses the architecture type "x86_64")

suse/sles9 for SUSE Linux Enterprise Server 9 (64-bit mode uses the architecture type "x86_64")

suse/suse9 for SUSE Linux Professional Server 9

b. Install the OpenIPMI RPM file by running the command:

# rpm -ivh openipmi*.rpm

Note - The kernel driver will be compiled using the kernel-source code during installation.

5. Install IPMItool.

IPMItool is the command-line interface (CLI) server-management client.

If the installed Linux distribution uses the 32-bit "i386" architecture, run the following command:

# rpm -ivh ipmitool*.i386.rpm

If the installed Linux distribution uses the 64-bit "x86_64" architecture, run the following command:

# rpm -ivh ipmitool*.x86_64.rpm

6. Test the IPMI kernel device driver and client application by running the following command:

# ipmitool -I open chassis status

Successful output looks similar to the following:

" System Power: on Power Overload: false Power Interlock: inactive Main Power Fault: false Power Control Fault: false Power Restore Policy: unknown Last Power Event: Chassis Intrusion: inactive Front-Panel Lockout: inactive Drive Fault: false Cooling/Fan Fault: false "

Note - On a subsequent reboot, the IPMI kernel driver might have to be loaded with the following command:
# modprobe ipmi_kcs_drv

Note - If you upgrade your Linux kernel, refer to Upgrading the Linux Kernel.

Enabling IPMI Access on a Solaris x86-Based Server (In-Band)

1. Log in to the server and authenticate as the root user.

2. Run the following command to install the LIPMI Solaris x86 kernel driver and the IPMItool management control application.

These files are located on the Documentation and Support Files CD in the /support/sysmgmt/solaris9 directory.

# pkgadd -d ./

Confirm installation of all packages when prompted.

3. Reboot the server.

Task IV: Enabling IPMI LAN Access

This section contains three alternative procedures: two in-band procedures and one out-of-band procedure. Use the procedure that corresponds to your OS:

Enabling IPMI LAN Access on a Linux-Based Server (In-Band)

Enabling IPMI LAN Access on a Solaris x86-Based Server (In-Band)

Enabling IPMI LAN Access Using an Alternative Method (Out-of-Band)

Enabling IPMI LAN Access on a Linux-Based Server (In-Band)

1. If the server is powered off, boot the local OS.

2. Log in to the server and authenticate as the root user.

3. Load the OpenIPMI kernel device driver (as installed in Step 3 of Enabling IPMI Access on a Linux-Based Server (In-Band)).

# modprobe ipmi_kcs_drv

4. Using the following commands in IPMItool, configure the network settings for the SP.

Note - For more information about the syntax for IPMItool commands, refer to Syntax.

# ipmitool -I open lan set 6 ipaddr ipaddr # ipmitool -I open lan set 6 netmask netmask # ipmitool -I open lan set 6 defgw ipaddr gwipaddr # ipmitool -I open lan set 6 password ipmipasswd

Enabling IPMI LAN Access on a Solaris x86-Based Server (In-Band)

1. If the server is powered off, boot the local OS.

2. Log in to the server and authenticate as the root user.

3. Using IPMItool, configure the network setting for the SP by using the following commands.

Note - For more information about the syntax for IPMItool commands, refer to Syntax.

# ipmitool -I lipmi lan set 6 ipaddr ipaddr # ipmitool -I lipmi lan set 6 netmask netmask # ipmitool -I lipmi lan set 6 defgw ipaddr gwipaddr # ipmitool -I lipmi lan set 6 password ipmipasswd

Enabling IPMI LAN Access Using an Alternative Method (Out-of-Band)

1. Using an SSHv1 client or SSHv2 client, log in to the IP address of the SP.

2. Authenticate as the newly created manager user (see Task II: Securing the SP").

# ssh spipaddr -l spuser

3. Enable IPMI LAN access and assign a password when prompted.

# ipmi enable channel lan # exit

Note - This password will be referred to as ipmipasswd in subsequent examples.

4. Using IPMItool, test the IPMI LAN access.

# ipmitool -I lan -H spipaddr -P ipmipasswd chassis status

Upgrading the Linux Kernel

Upgrading the installed Linux kernel to a newer version requires you to recompile the upgraded IPMI kernel device driver.

1. Install the kernel-source RPM that matches the version of the upgraded kernel binary RPM package.

2. Log in to the server and authenticate as the root user.

3. Change to the following directory.

# cd /usr/src/kernel-modules/openipmi

4. Recompile the module by running the following commands.

# make clean # make # make install

5. Re-test the IPMI kernel device driver and client application by running the following command.

# ipmitool -I open chassis status

Successful output should look similar to the following:

Note - On a subsequent reboot, the IPMI kernel driver might have to be loaded with the following command:
# modprobe ipmi_kcs_drv

Site Integration

When deploying your server, ensure that you determine the best integration strategy for your environment.

The Sun Fire V20z and Sun Fire V40z servers include network connections for the SP that are separate from network connections for the platform. This enables you to configure the server so that the SP is connected to an isolated, management network and is not accessible from the production network.

Daisy-Chaining the Servers

You can interconnect multiple servers in different daisy-chain configurations by using the SP connectors to form a management LAN, as shown in FIGURE 1-7, FIGURE 1-8, and FIGURE 1-9. The figures also show how the servers are connected to external LANs using the platform Gigabit Ethernet connectors.

Note - It is a best practice to use cross-over cables of at least one meter in length for daisy-chaining the servers.

FIGURE 1-7 Daisy-Chain Architecture With Redundancy at Switch Level on the Management LAN

This graphic shows a diagram of servers in a daisy-chain configuration with redundancy in the Management LAN at the switch level.

To interconnect the servers, you must use an RJ-45 cross-over cable. Cables can be connected to either the top or bottom SP port. To configure servers in a daisy chain, connect the first and last server in the chain to different switches.

In the configuration shown in FIGURE 1-7, two managed switches that are capable of spanning-tree discovery are required to redundantly connect both the top and bottom of the chain. If the switches are not capable of spanning-tree discovery, then only connect either to the top or the bottom of the chain, but not both.

FIGURE 1-8 Daisy-Chain Architecture With Redundancy at Port Level on the Management LAN

This graphic shows a diagram of servers in a daisy-chain configuration with redundancy in the Management LAN at the port level.

In the configuration shown in FIGURE 1-8, a managed switch that is capable of spanning-tree discovery is required to redundantly connect both the top and bottom of the chain. If the switch is not capable of spanning-tree discovery, then only connect either to the top or the bottom of the chain, but not both.

FIGURE 1-9 Daisy-Chain Architecture With No Redundancy on the Management LAN

This graphic shows a diagram of servers in a daisy-chain configuration with no redundancy in the Management LAN.

To interconnect the servers, you must use an RJ-45 cross-over cable. Cables can be connected to either the top or bottom SP port.

In the configuration shown in FIGURE 1-9, no redundancy is provided on the Management LAN.

Platform Drivers and Applications

Installation of the platform drivers and applications provides the following capabilities:

Enables communication between the SP and the platform. This allows for better control of the platform. For example, the platform can be shut down or rebooted properly, rather than forcefully, via power-downs and resets.

Allows the platform SNMP traps to be forwarded through the SP's SNMP daemon.

Allows the SP to monitor the health of the platform operating system when platform difficulties occur, and to attempt to handle machine check errors.

Allows the SP to gather additional vital product data about system components.

Allows the SP to gather inventory information about operating system software.

Allows updates to the platform BIOS from the SP.

If you do not install the platform software, the following features will not be available from the SP:

Ability to gracefully shut down and reboot the platform.

Ability to receive notices of recoverable machine check events and ECC errors.

Ability to obtain platform hostname.

Ability to determine the current OS status.

Ability to determine the current version and inventory of platform software.

Ability to obtain CPU vital product data and inventory information.

Ability to determine whether the platform is running, via the platform heartbeat.

Ability to obtain platform-side SNMP information, if attached to the SP's SNMP server.

Ability to set the platform JNET address with the sp set jnet command.

The features or characteristics below are available without installation of the platform drivers. However, they require that the SP was fully booted during the last BIOS boot:

BIOS inventory information is available from the SP.

SP time is synchronized to the platform.

Optimized thermal management is available via the SP.

Also note these other important points about platform software:

When you install a platform operating system, you can configure the language support. If you choose a language other than English, ensure that you also install the appropriate version of third-party drivers.

When you install a platform operating system, you can configure the power state. When you choose a power state, turn off Suspend and Hibernate.

There is a private network between the SP and the platform that supports internal communication.

The link-local address 169.254.101.2 is assigned to the SP.

The link-local address 169.254.101.3 is assigned to the platform for communication over this private network. These addresses are physically assigned, not randomly generated, and probed for conflict. You can use the sp set jnet command to change these IP addresses. The platform drivers must be installed in order for JNET communication to work.

Updating Software

Note - For complete information about the menu options available through the operator panel, refer to the Sun Fire V20z and Sun Fire V40z Servers--User Guide.

If you attempt to update the SP software using the operator panel when the IP address for the SP has not been set, the update fails. Ensure that the IP address has been set prior to attempting an update. For more information, refer to the Sun Fire V20z and Sun Fire V40z Servers--Installation Guide.

A new network share volume (NSV) that is installed on your network contains firmware packages. You can make these firmware packages available to a SP in either of these ways:

The recommended method is through the Update Server, a Java application that transfers the packages from the NSV to the SP.

You can update multiple SPs, simultaneously, if you use the Update Server application.

You must use the Update Server application to update the SP base package.

Another method is to use the SP to create a network file system (NFS) mount to the NSV. Once you accomplish this, the NSV and the packages it contains appear to be local to the SP and are available for update.

Note - The latest BIOS version number is never the same as the latest NSV version number, as represented in the configuration file example data lines in Configuring and Starting the Update Server Application.

FIGURE 1-10 NSV on Network-Accessible Server

Graphic showing importance of using Update Server application to obtain new NSV. [ D ]

In the illustration above, the NSV has been unzipped and saved to a server that is network-accessible by the SPs that require updated packages. SP1 has mounted the NSV directly. SP2, SP3, and SP4 access the NSV through the Update Server.

Note - In this example, SP1 cannot update the SP base package without using the Update Server application.

Selecting and Setting Up the File Server

Select a server that is network-accessible to the SP(s).

Note - The Update Server application requires Java Version 1.4 software or a subsequent release that is compatible with the software. If you plan to use the Update Server, open a shell or command prompt window and type java -version to verify the version.

To install the NSV, follow the steps below.

1. Download the latest version of the NSV from http://www.sun.com/servers/entry/v20z/downloads.html.

2. Extract files from the compressed NSV to a location on your selected file server.

Note - When you unzip a compressed file on Linux, use the -a switch (for example, unzip -a filename.zip) to force text files to convert to the target operating system's appropriate end-of-line termination.

A new release manifest file (releaseVersion.xml) is added to the root directory of the NSV. See Network Share Volume for details about the NSV structure.

3. For Linux-based systems, ensure that the NSV directory has been exported.

4. Decide whether you will update by using the Update Server application or by using an NFS mount:

If you plan to use the Update Server application, go to Configuring and Starting the Update Server Application.

If you plan to update by using the NFS mount method, log in to the SP and mount the NSV.

For example, if the IP address of the machine with the new NSV is 10.10.20.100 and you extracted the NSV files to a directory named newNSV, you would run the command:

sp add mount -r 10.10.20.100:/newNSV -l /mnt

The NSV will then be available to the SP at /mnt/sw_images/.

Continue with Identifying Packages for Update.

Configuring and Starting the Update Server Application

The Update Server configuration file enables you to export multiple packages with multiple versions to one or more SPs. To select the appropriate updates, follow the instructions below.

1. Navigate to NVS/update_server/Vx.xx (where Vx.xx is the version you want) to find the configuration file.

The configuration file includes example data lines, shown below.

SP_BASE	V2.0.0.38	/nsv/sw_images/sp/spbase/V2.0.0.38/install.image
SP_BASE	V2.0.0.40	/nsv/sw_images/sp/spbase/V2.0.0.40/install.image
SP_VALUE_ADD	V2.0.0.38	/nsv/sw_images/sp/spvalueadd/V2.0.0.38/install.image
SP_VALUE_ADD	V2.0.0.40	/nsv/sw_images/sp/spvalueadd/V2.0.0.40/install.image
BIOS-X250Alpha	V1.27.9	/nsv/sw_images/platform/firmware/bios/V1.27.9/bios.sp

Note - The latest BIOS version number is never the same as the latest NSV version number, as seen in the example above.

Each data line contains three space-delimited values:

Package type: SP-BASE, SP_VALUE_ADD, BIOS

In order to support BIOS updates for several products that each require unique BIOS firmware, the BIOS package must include the product ID. The product ID is the value that is returned by the platform get product-id command. It is also found in the BIOS software manifest (swinventory.xml) that is included in an NSV. The actual product ID used in the example above is x250 Alpha. When you include this in the BIOS package type in the configuration file, you must add the hyphen between BIOS and the product ID, and you must remove all spaces from the product ID string.

Version, in standard version format: V[major].[minor].[patch].[build].

File path: actual path and file name of an update file.

2. In the configuration file, each data line is preceded by a # sign. To indicate a file that should be updated, add the correct version number and remove the # sign at the beginning of the data lines.

3. Navigate to the NSV folder that contains the Update Server application and start the server via the command line:

java -jar updateServer.jar -c updateServer.config -p <port> -l logfile.log

The updateServer.jar file is located in the update_server folder of the NSV.

It is recommended that you use the -l flag to create a log file.

Only the start and the end of an update transaction will be sent to the console.

Detailed information about the update process is sent to the log file, which can be useful if you need to troubleshoot a failed update.

By default, the server uses port number 52708.

If this port number is in use already, use the optional -p flag to specify a different port.

The Update Server application does not start if the file is not found in the specified path. Otherwise, the server is ready to receive update requests from any SP.

The Update Server application can simultaneously accept multiple update requests from different SPs.

Identifying Packages for Update

1. To determine which packages currently are installed on an SP, run this command from the SP:

inventory get software

2. To determine which packages are available from a running Update Server application, run this command from the SP:

inventory get remote-software -i <server_ipaddress> -p <server_port>

Note - Some older versions of the SP do not accept the -i or -p options. These older versions accept only these arguments: [{-a|--all}], [{-D|--Delim}], and [{-H|--noheader}].

3. To compare currently installed packages with packages that are available on a mounted NSV, run this command from the SP:

inventory compare versions -f <manifest_filename>

4. To compare currently installed packages with packages that are available on a running Update Server, run this command from the SP:

inventory compare versions -i <server_ipaddress> -p <server_port>

Note - Some older versions of the SP do not accept the -i or -p options. These older versions accept only these arguments: [{-a|--all}], [{-D|--Delim}], and [{-H|--noheader}].

Updating the SP Base Package

Note - You can use the Update Server application to install this package, or you can use the SP Update Flash option in the operator panel's SP menu.

The SP base component includes the SP Value-Add component, so it also is updated as part of this process.

Note - Because the Value-Add package can contain all feature updates in a new release, check the Sun Fire V20z and Sun Fire V40z Servers--Release Notes to determine which package you should update.

1. Log in to the SP.

2. Execute the SP command to start the update process on the SP:

sp update flash all -i <server_ipaddress> -p <server_port> -r <version>

The optional -p flag indicates that the server is running on a port other than the default port. This command pings the Update Server application to determine whether it is running. If it is successful, your connection is closed when the SP reboots and the update process begins.

The -r flag indicates the version of the remote package that is requested. If LATEST is specified, the latest available version of the package is requested.

Note - Older versions of the SP do not support the -r option. If you run the sp update flash all command with an old version of the SP, the Update Server application will automatically update your software to the most recent version.

3. Monitor the update process on the server.

Messages are displayed as the installation process begins and ends. (More details of the update processes are in the Update Server log file.) When the update is complete, the SP reboots with the new version.

Note - If you update to a new version of the SP Base or Value-Add package, but do not install the associated documentation in the NSV, the online help will not work. After you mount the file system, check the software inventory for the version of the SP Value-Add package. Ensure that the latest version of the documentation is installed in the /docs directory.

Updating the SP Value-Add Package

The SP Value-Add component can contain all the new features in a new release. Check the Sun Fire V20z and Sun Fire V40z Servers--Release Notes to determine whether to update the Value-Add package or the SP Base package.

Note - You do not have to perform this upgrade if you already updated the SP Base package.

1. Log in to the SP.

2. Run this command:

sp update flash applications -i <server_ipaddress> -p <server_port>-r<package_version>

Note - If you use an NFS mount, execute this command:
sp update flash applications -f <filename>

Updating the BIOS

There are three methods available for updating the BIOS, as shown in the procedures in this section:

Use the Update Server application.

Mount the NSV.

Copy the BIOS image directly.

Using the Update Server Application to Update the BIOS

1. Follow the steps in Configuring and Starting the Update Server Application to use the Update Server application.

2. At the SP prompt, type this command:

platform set os state update-bios -i <server_ipaddress> -p <server_port> -r <package_version>

Mounting the NSV to Update the BIOS

1. Log in to the SP and mount the NSV.

For example, if the IP address of the machine with the new NSV is 10.10.20.100 and you extracted the NSV files to a directory named newNSV, you would run the command:

sp add mount -r 10.10.20.100:/newNSV

The NSV will then be available to the SP at /mnt/sw_images/.

2. At the SP prompt, type this command:

platform set os state update-bios /mnt/sw_images/platform/firmware/bios/Vx.x.x.x/bios.sp

Where Vx.x.x.x is the version number you want.

Copying the BIOS Image to Update the BIOS

1. Copy the BIOS image directly from the NSV to the /tmp folder on the SP file system.

2. At the SP prompt, type this command:

platform set os state update-bios /tmp/bios.sp

Updating the Diagnostics

The SP-based diagnostics tests are stored in the NSV and are referenced by the /diags symbolic link in the SP. The SP software references a default version of the diagnostics. However, if a new version is released and stored on the NSV, you must point to that new version in order to use it.

1. Log in to the SP.

2. Mount the NSV, using the sp add mount command. For example:

sp add mount -r <network_path>

This mounts the directory specified by network_path on /mnt.

3. To verify that the mount was successful, type ls /mnt/diags. For example:

ls /mnt/diags V2.4.1.0

4. Use the sp update diags command to establish a soft link from /diags to the desired diagnostics directory. For example:

sp update diags -p /mnt/diags/V2.4.1.0

5. To verify the new soft link, type ls -l /diags. For example.

ls -l /diags /diags -> /mnt/diags/V2.4.1.0

6. To verify that the diagnostics subsystem is available, type this command:

diags

The output lists all the subcommands of the diags command.

Note - See the Sun Fire V20z and Sun Fire V40z Servers--User Guide for a complete list of diagnostics modules and sample output.

Autoconfiguring the SP

You can also use the autoconfiguration option to update the SP. Autoconfiguration replicates the majority of configuration files from an SP that has already been configured to another SP, so that the two servers have identical configurations, except for the host name and IP address.

For example, after you configure a single SP (set up users, hosts, certificates, mounts and so on), you then run autoconfiguration on each additional SP so that the settings are identical. In addition, if you modify the configuration of one SP, you can update all of them by re-running autoconfiguration on each one. (For this reason, set the IP address of the autoconfigure server to x.x.x.1.)

For a list of files that are copied or not copied during the autoconfiguration process, see Files Copied During the Autoconfiguration Process.

Autoconfiguration has these limitations:

Autoconfiguration does not merge configurations, it overwrites the existing configuration.

Autoconfiguration does not work across different server platforms. That is, you cannot configure a Sun Fire V40z SP using settings on a Sun Fire V20z SP.

Autoconfiguration also does not work across different SP software versions. The servers must be running the same version of the SP software.

You can start autoconfiguration either when you are prompted at the completion of setting the IP address of the SP, or by selecting Autoconfigure from the SP menu option on the operator panel at any time.

To perform autoconfiguration of an SP, follow these steps.

1. On the operator panel, press the Forward or Back button until the following prompt appears.

SP Autoconfigure?

2. Press the Select button.

The following prompt appears:

SP Auto Setup? No

3. Press the Forward or Back button to change the prompt to Yes.

For instructions on setting an IP address, refer to the Sun Fire V20z and Sun Fire V40z--Server Installation Guide.

4. Press the Select button.

The SP attempts to locate an IP address.

If the SP successfully locates an IP address, the following prompt appears, displaying an IP address for this SP:

Setup Server IP:x.x.x.1

Where x.x.x is the first three octets of the SP IP address. For example, if the address is 10.10.30.19, the address that displays in the prompt appears as 10.10.30.1.

In this case, press the Select button on the operator panel to start the autoconfiguration.

If the SP does not locate an IP address, the following message appears:

Unable to get SP IP address

In this case, you must manually enter an IP address before you press the Select button on the operator panel to start the autoconfiguration.

5. Wait until the autoconfiguration is complete, at which point the SP automatically reboots.

The following message is displayed when the autoconfiguration is running:

SP AutoConfigure in progress

Note - If the autoconfiguration is unsuccessful, a failure message displays. Press any button on the operator panel to clear it.

Files Copied During the Autoconfiguration Process

TABLE 1-5 provides the list of files copied during the autoconfiguration process. TABLE 1-6 provides the list of files that are not copied during the process.

The autoconfiguration process performs some safety checks on some files.

The passwd and shadow files are processed to pass only user accounts that can be created through the access command.

The root account, enabled or not, is not cloned.

The fstab file passes only the mount point information for /mnt.

Only the SSH key files for those users who are authorized on the system are replicated on the target. Key files for users who have been removed from the target are removed.

Under SP software version 2.1.*, the files IPMIConfig.xml and SystemStruct.xml are copied only if the product ID and board revision are identical. These files cannot be transferred from a server running SP software version 2.1.* to a server running SP software version 2.2.*, and vice-versa.

All files are transferred between the two hosts through an SSL socket connection. This is true even if the option ssl_not_enforced is enabled.

TABLE 1-5 Files Copied During the Autoconfiguration Process
File	Purpose
`/pstore/passwd`	User account list
`/pstore/group`	User group list
`/pstore/shadow`	User account passwords (local users only)
`/pstore/fstab`	`/mnt` file system information
`/pstore/smb.creds`	User/password information for SMB mount
`/pstore/evcfg.xml`	Event manager configuration file
`/pstore/seccfg.xml`	Security manager configuration file
`/pstore/oppanelcfg.xml`	Operator panel configuration file
`/pstore/snmpd.conf.template`	SNMP configuration file
`/pstore/snmp_proxy_community.txt`	SNMP configuration file
`/pstore/resolv.conf`	Directory Name Service configuration
`/pstore/jnet_config`	JNET network configuration
`/pstore/krb5.keytab`	Kerberos authentication configuration (for Windows authentication)
`/pstore/ssl_not_enforced`	Disables SSL requirement for SM GUI console
`/pstore/user_ssl_server.key,.crt`	SSL key and certification for SM GUI console
`/pstore/ssh_known_hosts`	SSH host keys (trusted hosts)
`/pstore/ssh_authorized_keys/*`	SSH user keys (trusted users)
`/pstore/IPMI/IPMIConfig.xml`	IPMI configuration
`/pstore/IPMI/ipmiusers`	IPMI user list (Note: Not copied by servers or clients running SP software version 2.1.)
`/pstore/SystemsStruct.xml`	User-modified sensor thresholds
`/dev/mtd/custom`	Custom configuration area

TABLE 1-6 Files Not Copied During the Autoconfiguration Process
File	Purpose
`/pstore/mc.conf`	Machine check configuration
`/pstore/hostname`	Local SP hostname
`/pstore/ifcfg2-eth0`	Local SP IP configuration
`/pstore/dimm.map`	Platform DIMM configuration
`/pstore/edstatefile`	Local SP event log
`/pstore/emstatefile`	Local SP event log
`/pstore/hwinventory`	Hardware inventory list
`/pstore/inv_manifests/*`	Software inventory list
`/pstore/snmpd.conf`	SNMP engine unique ID
`/pstore/sp_uuid`	SP unique ID
`/pstore/ssh/ssh_host*`	SSH host keys
`/pstore/IPMI/sdrr`	IPMI sensor data repository
`/pstore/IPMI/SEL`	IPMI sensor event log

Determining SP and Platform Network MAC Addresses

Use the following commands if you need to determine the MAC address of your server's SP or platform:

# ssh spipaddress -l spusername sp get mac

# ssh spipaddress -l spusername platform get mac

Systems Management Console Features

You can manage the servers by using line commands or by using the web-based Systems Management (SM) Console graphical interface. This section gives an overview of the actions you can do with the SM Console interface.

Note - For full information on the line commands, see the appendixes in this guide.

Configuring Network Settings

Admin-level and manager-level users can use the SM Console to configure the SP network settings to define the IP address method (static or DHCP), and other network settings such as the host name, DNS server address, and domains.

Note - As discussed in the previous section, you also can configure network settings from the operator panel, or you can use the sp ip commands, which are explained in the online help.

FIGURE 1-11 Network Configuration Screen

Figure showing SP Network Configuration screen.

To configure the SP network settings from the SM Console:

1. Click Configuration>SP Network from the menu bar.

2. In the Settings table, select which IP Address Method (DHCP or Static IP Address) that you want to use.

3. If you choose Static IP Address, type the IP address, gateway address, and subnet mask.

4. Identify other network settings.

The host name of the SP

A single IP address of a DNS server (if available)

A space-separated list of search domains (if applicable)

5. Click the checkmark button to save the settings.

Note - If you choose DHCP, the SP broadcasts for a DHCP server to obtain a dynamic IP. The IP address information is displayed, but you cannot edit it.

Starting and Stopping the Platform OS

Admin-level and manager-level users can start and stop the platform operating system from the SM Console. Choose Management>Platform Operations from the menu bar, then choose one of the options listed in the table below.

TABLE 1-7 Stop and Start Options for the Platform OS
Option	Description
Power On / Restart	The Power On/Restart option starts the platform operating system.
	The Boot into BIOS Setup option boots the platform and causes BIOS to enter setup mode. This enables you to modify the BIOS settings from the platform console. After you select this option, you must access the BIOS configuration screen to change BIOS settings. See the Sun Fire V20z and Sun Fire V40z Servers--User Guide for details about configuration of BIOS settings. For remote access, log in via the SM Console. Choose Troubleshooting>SP SSH Console. Then execute the `platform console` command.
	The Forced Restart option bypasses the operating system shutdown stage during a system restart. It can cause loss of data.
Shutdown / Power Off	The Shutdown/Power Off option shuts down the platform operating system and powers off the machine.
Shutdown / Power Off	The Forced Power Off option bypasses the operating system shutdown stage. It might cause loss of data. Use the Forced Power Off option if you must force a shutdown.

After you choose an option and click the checkmark button, the operation is initiated on the server. The help text displays any processing and results messages. The current state (normal, warning, or critical) is reflected in the System Status button, so you can monitor the progress.

When you move the cursor over the Platform Operating System button, one of these states is displayed in the help panel:

Communicating

Diagnostics

Sleeping

BIOS booting

BIOS setup

OS booting

OS shutting down

Note - You can perform platform state management from the command line with the platform subcommands. See Appendix F for more information.

If the power currently is off, the Shutdown/Power Off option returns a message that the option was not executed, due to the current state.

Configuring SMTP Event Notification

Admin-level and manager-level users can configure the system to:

Send email for generated events, via a Simple Mail Transfer Protocol (SMTP) server.

Route email based on the severity of events.

Send email that contains a subject and content (long format), or only a subject (short format) to support target devices such as phones, pagers, and so on.

SMTP notification ensures rapid notification about events and rapid response to critical situations. You can use the SM Console or the sp smtp commands to configure SMTP event notification.

Follow these steps to configure automatic SMTP email alerts from the SM Console.

1. Click Configuration>SMTP Event Notification from the menu bar.

2. Type the SMTP server name (either the host name or IP address of the SMTP server where you want email to be routed). Use an IP address unless DNS is configured on the SP.

3. For each level of severity, type a comma-separated list of email addresses.

These are the addresses that will receive email for each level of severity. Severity levels are:

Informational

Warning

Critical

FIGURE 1-12 SMTP Event Notification Screen

Screenshot showing SMTP Event Notification Configuration screen.

Note - Type separate address lists for short and for long email formats. Type a comma between each email address. Type a separate list for pager email addresses that require shorter text.

4. Click the checkmark button to save the settings.

Configuring Directory Services

Configure the directory services options to define how SP user name and password information is stored or accessed.

TABLE 1-8 Directory Services Options for the Service Processor
Option	Description
NIS	Network Information Service: A UNIX-originated solution to directory service. Both local files and a remote NIS server authenticate users.
ADS	Active Directory Service: Microsoft's directory service. Both local files and a remote ADS server authenticate users.

You can use the SM Console or the access subcommands to configure the directory services options. See Appendix B for more information.

To configure directory services from the SM Console:

1. Click Access Control>Directory Services from the menu bar.

Select the type of directory service you want to use:

To use local authentication only, select Use local etc/pw files in the Settings table.

To use directory services, select Use NIS or Use ADS for the network directory services database you want to use.

FIGURE 1-13 Directory Services Configuration Screen

Screenshot showing SP Directory Services Configuration screen.

2. Type the domain name for the option you selected in Step n.

3. Type the server name for the option you selected in Step n. (For multiple servers, type a comma between each server name.)

4. If you chose ADS, also type the organizational unit, ADS Logon ID, and the location of the keytab file.

See Creating Keytab Files for ADS.

5. Click the checkmark button to save the settings.

Note - If you use ADS, the clock on the service processor must be synchronized with the clock on the ADS server. Also, the service processor and the ADS server must be able to resolve each other's host names using DNS.

Remote users who are authenticated via directory services have access to the SP only via a group mapping that maps the user's remote group to an SP administrative group.

To simplify configuration on the SP, manager-level users can map directory service groups to predefined groups. When you map those users (members of directory services groups) to an SP administrative group, they automatically have appropriate access rights.

Mapping Directory Service Groups

The Directory Service Group Mappings table identifies existing group mappings. This table also provides options for mapping other directory service groups to an SP group. For example, if Directory Services group 5 is mapped to manager, all members of group 5 are granted manager-level privileges on the SP.

To map directory service groups from the SM Console:

1. Click Access Control>DS Group Mappings from the menu bar.

The current group mappings table is displayed. Directory service groups are listed alphabetically.

FIGURE 1-14 Directory Service Group Mappings Screen

Screenshot showing Directory Service Group Mappings screen.

2. Choose one of the following:

Select which mapping you want to create (monitor, admin, or manager).

Select No Mapping to disable access to the SP.

Enter a new group in the text field and select one of the corresponding options to indicate the SP group for the new entry.

Select the Verify checkbox so that an error will display if a new group name is not located in the directory service.

3. Click the check mark button.

Creating Keytab Files for ADS

To use Active Directory Service (ADS) as a directory service on the SP, you must create an active directory account. The name service library on the SP uses this account to authenticate itself to the LDAP interface of the active directory server.

ADS Server Requirements

The ADS server must have Certificate Services and the High Encryption Pack installed.

The Microsoft Windows administrator must create an Active Directory account and a keytab (for that account) that the SP(s) will use to conduct LDAP queries. You can create keytab files with the ktpass command that is located in the Microsoft Windows 2000 resource kit:

ktpass -princ <logon>@<domain> -pass <password> -mapuser <logon> -out <output filename>

Note - The keytab you create with this command can be uploaded to the SP with the scp command, or can be accessed from an exported file system that is mounted by the SP. See your Microsoft documentation for details about this command.

ADS SP Requirements

You must configure DNS.

The canonical name of each host must be the fully-qualified host name (including the domain).

The IP address of each host must reverse-resolve to the canonical name.

The time on the SP must be accurate to within five minutes of the time on the ADS server (domain controller). When the platform is started, the SP clock is synchronized with the platform clock.

You must configure ADS properly. From the SM Console, type:

The ADS domain

The ADS server name

The organization unit (OU) under which the SP searches for group information

The ADS log in ID (the name of the account that was created for the SP to use)

The keytab file that was uploaded and installed on the SP

Configuring Date and Time

Admin-level and manager-level users can configure the date and time setting for the SP clock. Use the sp date command from the command line or configure date and time from the SM Console.

The clock is synchronized automatically when platform drivers are installed. If the platform is running (and the drivers are installed) the platform time takes precedence over the SP time.

The platform time must be set correctly for ADS to function.

If you configure the SP before you load the platform operating system and you want to set the time to synchronize with ADS and other network services, follow the procedure below.

From the SM Console:

1. Click Configuration>SP Date/Time from the menu bar.

2. Identify the date and time on the SP clock.

The current SP time displays in yyyy:mm:dd hh:mm:ss format. FIGURE 1-15 illustrates an example of this format.

FIGURE 1-15 Date/Time Configuration Screen

Screenshot showing SP Date/Time Configuration screen.

3. Click the checkmark button to save the settings.

Configuring SSL

Set up web access to the SP with either an encrypted or a non-encrypted communication method.

By default, all messages between your browser and the SP are encrypted according to Hypertext Transfer Protocol over Secure Sockets Layer (HTTPS). Version 0.9.6j is supported.

1. Allow browsers to communicate with the SP via non-encrypted messages by one of the following two methods:

From the command line, use this command:

sp disable ssl-required

From the SM Console, select Optional (disable) or Required (enable) from the SSL Configuration screen as described in Configuring the SSL Certificate From the SM Console.

With SSL disabled, HTTP requests are serviced directly without any redirection to HTTPS. HTTPS requests continue to be secure.

Note - If you enable the HTTPS protocol, your browser will display a warning message that states that it cannot verify the validity of the Server Certificate. This warning is informational and you can ignore it safely. To upload a certificate that is signed by your own organization or by an independent certificate authority, select User Supplied, as described in Configuring the SSL Certificate From the SM Console.

2. To revert to the default behavior, type this command:

sp enable ssl-required

With SSL enabled, HTTP requests are redirected automatically to equivalent HTTPS requests to maintain site security.

Configuring the SSL Certificate From the SM Console

Admin-level and manager-level users can enable or disable SSL encryption and can define the SSL certificate that is used to manage transmission security.

Note - You also can use sp ssl commands to configure the SSL certificate. See Appendix H or the SM Console online help for details about commands.

Follow these steps to configure the SSL certificate from the SM Console:

1. Click Configuration>SSL Certificate from the menu bar.

2. Select Required or Optional to designate SSL access.

FIGURE 1-16 SSL Configuration Screen

Screenshot showing the SSL Configuration screen.

3. Do one of the following, depending on whether you selected Required or Optional:

If you selected Required, select the option for the type of SSL certificate configuration you want to use, factory-installed or your own in-house certificate management.

If you selected User-supplied Certificate:

a. Type the name of your generated certificate file to be installed with Apache on the SP, or click the Browse button to search for a file.

b. Type the name of your generated key file to be installed with Apache on the SP, or click the Browse button to search for a file.

4. Click the checkmark button.

Monitoring System Status

The System Status window displays an image that represents the physical layout and status of all hardware components and sensors. You can use this window to identify components that have problems, or failed components that must be replaced. To access this window, click the System Status button from the toolbar in the SM Console.

FIGURE 1-17 System Status Window, Bottom Tab, Sun Fire V40z Server

Screenshot showing System Status window.

The component images represent the actual physical hardware components including their approximate location, size, and status. The hardware layout for the Sun Fire V40z server is represented in two layers. (FIGURE 1-17 illustrates the default, bottom tab view for the dual-core Sun Fire V40z server.) Click the Bottom and Top tabs at the top of the image to change views.

FIGURE 1-18 System Status Window, Top Tab, Sun Fire V40z Server

Screenshot showing System Status window.

You also can display component details for troubleshooting purposes. To view details about a component, click on the image of that component. You also can use the pulldown menus in the top right corner of the window to locate specific component types (CD-ROM drives, CPUs, disk drives, fans, and so on) and specific sensor types (fan, power, and temperature sensors).

Sensor images represent the approximate location, current value, and warning or critical thresholds of system sensors. Current information about the sensor (name, type, current value, low and high warning, critical thresholds, and status) is displayed in a gauge component in the top right corner of the image.

FIGURE 1-19 Temperature Sensors Displayed, Bottom Layer

Screenshot showing System Status window.

To view values for a sensor, click on the image of the sensor. In FIGURE 1-19, which illustrates the bottom layer, the temperature sensor gauge for CPU 0 is displayed.

FIGURE 1-20 Ambient Air Temperature Gauge Displayed, Top Layer

Screenshot showing System Status window.

In FIGURE 1-20, which illustrates the top layer, the sensor gauge for the ambient air temperature is displayed.

Note - See the SM Console online help for additional instructions and details about management tasks.

System Events

Administrators with appropriate privileges can use the SM Console's System Events table to view detailed information about all active events. They also can perform various actions that are related to each event.

Each active event is displayed on its own row in the table, as illustrated in FIGURE 1-21.

FIGURE 1-21 System Events Table

Screenshot showing System Events window.

TABLE 1-9 describes the columns in the System Events table.

TABLE 1-9 System Events Table
Column	Description
Component	The unique name of the component that has caused this event. Components can be hardware or software.
Detail	Displays details for the component.
ID	Contains a unique event ID for each event enabling you to track the event in your external trouble-ticket system and query the log files for all actions related to the event.
All	Displays the highest severity that this event has ever achieved: informational (green), warning (yellow), and critical (red) icons. Click the icon to view details for the event.
Current	Displays the current severity: informational (green), warning (yellow), and critical (red) icons. Also displays a descriptive message. These descriptions might be lengthy, (for example, describing the nominal, warning, critical, and current temperature for a fan) and therefore are displayed in the help panel.
Type	Identifies the event type. Refer to TABLE 1-10 for a description of the icons representing each type.
First	Lists the date and time at which the event was initially generated. Only the time is displayed in the table; help text at the bottom of the window displays the entire date and time.
Last	The date and time at which the event was most recently generated. Only the time is displayed in the table; help text at the bottom of the window displays the entire date and time.
Count	Lists the number of times the event has occurred. If a new event has the same component and event type as a current, uncleared event, a new event is not created but the count for the current event is incremented and the current severity is updated in the Current column.
Clear	Clicking this button clears the event. You must manually clear all events. A cleared event is deleted from the server and automatically cleared from any other connected SM Console System Events screen. Only admin-level and manager-level users can clear events.
Help	Clicking this button displays online help for this table.

Note - You also can use the sp get events command to obtain information about events. See Appendix H or the online help for more information.

Icons Used to Identify System Events

The SM Console displays the icons that are illustrated in TABLE 1-10 to represent specific types of events.

TABLE 1-10 System Event Type Icons
Event Type	Icon
BIOS Events
Fan Speed Events
Machine Check Events
Miscellaneous Operating System Events
Platform State Events
Switch Events
Temperature Events
Unknown Event
Voltage Events

Note - See the Sun Fire V20z and Sun Fire V40z Servers--Troubleshooting Techniques and Diagnostics Guide for a table of system events and troubleshooting suggestions.