Skip Headers
Oracle® Enterprise Manager Ops Center Feature Reference Guide
12c Release 1 (12.1.4.0.0)

E27511-09
Go to Documentation Home
Home
Go to Book List
Book List
Go to Table of Contents
Contents
Go to Index
Index
Go to Feedback page
Contact Us

Go to previous page
Previous
Go to next page
Next
PDF · Mobi · ePub

11 Hardware

This chapter describes hardware management features that are available in Oracle Enterprise Manager Ops Center:

11.1 Introduction to Managing Hardware Assets

Oracle Enterprise Manager Ops Center provides comprehensive lifecycle management for the hardware assets in your data center. The hardware assets can be handled individually or as a group. After the discovery and management of the hardware assets, as described in Asset Management, you can configure and then monitor them to gather the information to maintain them.

11.1.1 Configuring Hardware Assets

Use these deployment plans:

  • Install Server

  • Update BIOS Configuration

  • Configure Service Processor

  • Configure Server Hardware and Install OS

  • Configure RAID

  • Configure M-Series Hardware, Create and Install Domain

  • Configure and Install Dynamic System Domain

All the plans are based on hardware resource profiles. See Hardware Resource Profiles

11.1.2 Monitoring Hardware Assets

As soon as a hardware asset is managed, Oracle Enterprise Manager Ops Center starts to monitor it, according to the asset type's monitoring profile. The center pane displays information for a selected asset in a series of tabbed windows. The tabs and the type of information is specific for the asset type but, in general, Oracle Enterprise Manager Ops Center reports the following:

  • Health status

  • Power state

  • Power usage

  • Hardware variables and connectivity

You can change the monitoring thresholds in the standard profile to specify the conditions that generate an alert. You can also create custom profiles with different rule sets and alert parameters and apply the profile to a specific system, a group of homogeneous systems, or a group that you define.

For more information about monitoring policies and rules, see Chapter 4.

For information about monitoring power consumption, see Energy Tab.

11.1.3 Maintaining Hardware Assets

Based on your observations, you can control your hardware assets and do the following actions:

Use these deployment plans or customize them:

  • Update Firmware

  • Update Storage Appliances

  • Update BIOS Configuration

11.2 Roles for Hardware Management

Table 11-1 lists the tasks and the role required to complete the task. Contact your administrator if you do not have the necessary role or privilege to complete a task. See the Oracle Enterprise Manager Ops Center Administration Guide for information about the different roles and the permissions they grant.

Table 11-1 Hardware Tasks and Roles

Task Role

Configure and Deploy Server

Server Deploy Admin

Install Server

Server Deploy Admin

Configure RAID

Server Deploy Admin

Update Management Credentials

Security Admin

Importing and uploading firmware images

Storage Admin

Edit Attributes

Asset Admin

Power On, Power Off, Power on with Net Boot

Asset Admin

Set Power Policy

Asset Admin

Reset Servers, Reset Service Processors, Refresh

Asset Admin

Locator Light On/Off,

Asset Admin

Snapshot BIOS Configuration, Update BIOS Configuration

Asset Admin

Update Firmware

Update Admin

Simulate a firmware update

Update Admin

Update Sim Admin

Launch LOM Controller

Asset Admin

Edit Tags

Asset Admin


11.3 Actions for Hardware Management

After you manage your assets, you can perform the following actions:

  • Use a resource profile to configure a hardware asset

  • View utilization of systems

  • Modify energy consumption

  • Reset a server

  • Power systems on and off, including a forced power off

  • Place in maintenance mode

  • Get access to the serial console

  • Use locator lights to identify a specific asset

  • Check firmware compliance

  • Update firmware

  • Use a provisioning profile to update firmware

11.4 Location of Hardware Information in the User Interface

Hardware assets are visible in the All Assets section of the user interface. For assets with a service processor, both the service processor and the system are included, as shown in Figure 11-1.

Figure 11-1 Managed System

Description of Figure 11-1 follows
Description of "Figure 11-1 Managed System"

Each type of hardware is also added to the appropriate group in the Resource Management view, as shown in Figure 11-2.

Figure 11-2 Example of Resource Management View

Description of Figure 11-2 follows
Description of "Figure 11-2 Example of Resource Management View"

Table 11-2 shows where to find information.

Table 11-2 Location of Hardware Information in the BUI

To See Location

All servers

Expand Servers in the Assets pane. The center pane lists up to 50 servers.

Details for a specific server

Expand Servers in the Assets pane. Then select one of the servers. The center pane includes a series of tabbed displays.

Monitoring Rules for a specific server

Expand Servers in the Assets pane. Then select one of the servers. Then select the Monitoring tab.

Monitoring rules for a hardware type

Expand Plans and then Operational Plans and then Monitoring Policies. Select a type. The center pane shows all of the rules and thresholds.

Current firmware version

Expand Servers in the Assets pane. Then select one of the servers. Then select either the Summary tab or the Hardware tab. Each one includes the Firmware table.

A report of all attributes

Expand Reports and then select System Information Report.


11.5 Profiles for Hardware Management

The work flow of an asset deployment can be captured and enacted in a repeatable fashion using plans and the profiles included in the plans. Oracle Enterprise Manager Ops Center provides default profiles for configuring hardware asset types consistently. You can use the default profiles, make copies of the profile to edit, or create new ones.

Deployment proceeds from configuring the hardware, installing the correct firmware, provisioning the OS, and applying the required updates. The deployment is not only for servers with an OS installed but also for chassis, racks, power distribution units, and M-Series servers.

11.5.1 Hardware Resource Profiles

Use hardware resource profiles in a deployment plan to configure, install, or update systems. The Oracle Enterprise Manager Ops Center software provides the following hardware resource profiles:

You can use the default profiles, make copies of the profile to edit, or create new ones. Use the profiles in a deployment plan to configure your hardware assets.

11.5.2 Firmware Provisioning Profiles

Hardware depends on firmware to perform operations. Part of monitoring and managing a hardware asset is to make sure that it has the appropriate version of firmware. Oracle Enterprise Manager Ops Center uses firmware profiles to provision, or update, firmware on each type of asset.

A firmware profile is a set of actions and values that define how to provision one or more assets and specifies one or more firmware images. A firmware profile updates existing firmware assets completely and consistently. When you apply a deployment plan that contains a firmware profile, Oracle Enterprise Manager Ops Center compares the versions of each firmware image specified in the profile with the versions of the existing firmware on the asset and then takes the action you specify in the profile.

See Firmware Compliance Reports and Firmware Provisioning.

11.6 Configuring the Service Processor

The Declare Unconfigured Asset action includes the service processor in the Oracle Enterprise Manager Ops Center environment. You can configure only unconfigured service processors, that is, a processor in its factory default state. Use a deployment plan to configure the service processor:

  • Configure Service Processor

  • Update BIOS Configuration

  • Configure and Deploy Server

  • Install Server

An alternative to specifying the configuration in the profile, is to duplicate an existing configuration. You can create a snapshot of the BIOS of a working service processor, which creates also creates profile. You then apply the profile to an unconfigured service processor.

11.6.1 Creating a Service Processor Configuration Profile and Plan

  1. Expand Plans in the Navigation pane and the select Profiles and Policies.

  2. Expand Service Processors.

  3. Click Create Profile. The first step of the wizard is displayed, as shown in Figure 11-3. Depending on the subtype and target you select, more steps are added.

    Figure 11-3 Create Profile - Service Processor

    Description of Figure 11-3 follows
    Description of "Figure 11-3 Create Profile - Service Processor"

  4. Complete the specification of the service processor and click Finish.

The new profile and plan are available from the Assets pane.

11.6.2 Creating a BIOS Configuration Profile and Plan

  1. Expand Plans in the Navigation pane and the select Plans.

  2. Expand Update BIOS Configurations.

  3. Click Create Plan from Template.Figure 11-4 shows the first step of the wizard, which includes the Update BIOS profile.

    Figure 11-4 Create Plan - BIOS

    Description of Figure 11-4 follows
    Description of "Figure 11-4 Create Plan - BIOS"

  4. Click the Update BIOS profile.

  5. Click the Create Profile icon. Figure 11-5 shows the window for creating the profile.

    Figure 11-5 Create Profile-BIOS Configuration

    Description of Figure 11-5 follows
    Description of "Figure 11-5 Create Profile-BIOS Configuration"

  6. Enter a name and description and then select the type of BIOS and the type of target.

  7. Click Next to review and then click Finish to submit the job.

  8. When the job is completed, return to the Create a Deployment Plan window.

  9. Specify a name for the plan and select the new profile.

  10. Click Save to submit the job.

The new profile and plan are available from the Assets pane.

11.6.3 Creating a Snapshot of a Service Processor Configuration

  1. Expand Assets in the Navigation pane and select Servers.

  2. Select the configured server.

  3. Click the Hardware tab in the center pane.

  4. In the Component Navigation section, select Service Processor. The Service Processor Configuration section displays the configuration attributes and values.

  5. In the Service Processor Snapshots section, click the Create Snapshot icon to display the window shown in Figure 11-6.

    Figure 11-6 Create Service Processor Configuration Snapshot

    Description of Figure 11-6 follows
    Description of "Figure 11-6 Create Service Processor Configuration Snapshot"

    Note:

    An alternative method of displaying this window if you are familiar with the current configuration is to select the server and then choose Snapshot BIOS Configuration in the Actions pane.
  6. Specify a name for this snapshot and accept the default of creating a profile from the snapshot.

  7. Click Create Snapshot to submit the job.

  8. When the job completes, the snapshot is located in the EC local library.

11.6.4 Applying a Snapshot of a Server Processor Configuration

This procedure applies a snapshot of one server's service processor to another, unconfigured service processor. Use the Update BIOS Configuration action or use the snapshot profile as a step in the Install Server deployment plan.

View a Service Processor Snapshot

To verify the snapshot configuration before you attempt to configure a new server, view the snapshot.

  1. Expand Libraries in the Navigation pane and then select the current EC local library.

  2. In the center pane, scroll down to the BIOS Configuration section of the library.

  3. Select the snapshot and then select the View Snapshot icon.

  4. You can apply the snapshot directly to a server or you can use a deployment plan.

Apply a Service Processor Snapshot to a Server

  1. Expand Assets in the Navigation pane and then select Servers.

  2. Select the unconfigured server.

  3. Click Update Service Processor in the Action pane. A one-step wizard is displayed, including the list of snapshots in the EC local library that are appropriate for the selected server.

  4. Select the snapshot from the list.

  5. Click Next to review the Summary and then click Finish to submit the job.

When the service processor is configured, the server reboots.

Apply a Service Processor Snapshot Profile to a Server

  1. Expand Plans in the Navigation pane and then select Profiles.

  2. Expand Update BIOS Configuration and select the profile for the snapshot.

  3. Review the profile in the center pane. To modify the profile, click the Edit Profile icon.

  4. Create a plan that includes this profile as its only step or edit an existing deployment plan to include this profile.

11.7 Configuring a RAID Controller

Oracle Enterprise Manager Ops Center provides a default profile for configuring RAID controllers. You can configure and update the hardware devices that can be done only through the host OS. This profile provisions a reduced OS image along with the management pack on the target to configure RAID.

Caution:

When you reconfigure an existing RAID controller, all the data on the disk is lost.

11.8 Configuring a Dynamic System Domain

You can create domains on a M-series server.

  • Verify that a user account with platadm privilege exists on the XSCF processor and is included in the profile. Disable the audit policy for this user account, using the following command:

    setaudit -a opsadm=disable
    
  • Verify that you can use ssh to log in to the XSCF processor.

To configure a Dynamic System Domain, you apply a deployment plan to an M-Series server, as described in Chapter 8. The plan includes the profile for the Dynamic System Domain.

11.9 Configuring a Rack and Placing Components

The rack asset is a group that includes the other managed assets installed in the physical rack, such as servers or compute nodes, storage arrays or appliances, switches, and power distribution units (PDU). After you create the rack, you put assets in the rack so that Oracle Enterprise Manager Ops Center can present them and manage them as if it were a physical asset.

Creating a Rack

  1. Expand Assets in the Navigation pane and select Racks.

  2. Click Create Rack in the Action pane.

  3. Enter the name for the rack.

  4. For a rack containing an Oracle Engineered System, specify the serial number of the rack.

  5. Enter the total number of slots in the rack. A full rack has 42 slots. Default to 42.

  6. You have the option to place assets into the rack immediately after this procedure ends or to perform this task at a later time.

  7. Enter a description for this rack.

  8. Add semantic tags that are appropriate for this rack.

  9. Click Create Rack to submit the job.

Use the Place/Remove Assets in Rack action and the Place/Remove PDU in Rack action to specify each asset in the physical rack and its location in the rack so that Oracle Enterprise Manager Ops Center can represent the physical rack accurately to remote users.

Placing Assets in a Rack

The rack asset is a group that includes the managed assets installed in the physical rack, such as servers or compute nodes, storage arrays or appliances, switches, and power distribution units (PDU).

Use the Place/Remove Assets in Rack action and the Place/Remove PDU in Rack action to specify each asset in the physical rack and its location in the rack so that Oracle Enterprise Manager Ops Center can represent the physical rack accurately to users who cannot examine the rack. Update the type and location of the assets in the rack asset when the configuration of the physical rack changes.

Placing a Power Distribution Unit in a Rack

  1. Expand Assets in the Navigation pane and select Racks from the Resource Management Views.

  2. Select the rack.

  3. Click Place/Remove Assets in Rack in the Actions pane or click the Details tab to navigate to the icon. The Place Assets in name Rack window opens, which displays a list of all managed assets in the physical rack. You can filter the list by type or attribute.

  4. Click an asset and then enter its slot number.

  5. Click Place Asset in Rack. The Assets in the Rack pane is updated to show the location.

  6. Continue to select assets and place them in the rack.

  7. To change an asset's slot, select it and click Edit Placement. Enter a new slot number.

  8. To remove an asset, select it and click Remove Asset. The asset is deleted from the Assets in the Rack pane and is available to be placed.

  9. When you are satisfied with the configuration displayed in the Assets in the Rack pane, click Submit.

To add PDUs to the rack asset, use the Place/Remove PDUs action in the same way. Select the rack and the action. The assets you have placed in the rack are shown in the Assets in Rack pane. Add each PDU.

11.10 Hardware Monitoring

The Oracle Hardware Management Agents use Simple Network Management Protocol (SNMP) to monitor your Oracle hardware and storage devices. The software uses the Intelligent Platform Management Interface (IPMI) protocol to accesses the Oracle ILOM service processors.

A monitoring policy is a set of rules applied to an asset. If a status changes or a threshold is crossed, an alert is created. Oracle Enterprise Manager Ops Center provides default policies for each asset type. You can create new policies or modify existing policies.

11.10.1 Hardware Status

If a hardware asset can report a value for a hardware variable, Oracle Enterprise Manager Ops Center reports its current state and compares it to the threshold value.

  • Good – The hardware asset is working properly.

  • Unknown – Oracle Enterprise Manager Ops Center is unable to retrieve information from the sensor. The hardware asset is connected but is not reporting information.

  • Unreachable – The hardware asset cannot be contacted. This state indicates a network problem.

  • Warning Failure – Oracle Enterprise Manager Ops Center has detected a potential or impending fault condition. Take action to prevent the problem.

  • Critical Failure – A fault condition has occurred. Take corrective action.

  • Nonrecoverable Failure – The hardware asset has failed. Recovery is not possible.

  • Faulted – The hardware asset reports a fault. Contact service personnel to repair.

11.10.2 Groups of Hardware Assets

Oracle Enterprise Manager Ops Center monitors hardware assets according to the monitoring profile for that type of asset. To see the default profile for monitoring a hardware type, see Hardware Monitoring.

To change a threshold value, see Editing Monitoring Rules in Chapter 4.

11.10.3 Connectivity Status

Connectivity is the network interface of the system. You can view information about a hardware asset's Network Interface Card (NIC).

11.10.4 Service Processor Details

Use the Hardware tab to view information about each component of the system:

  • Name and SNMP Community

  • Whether Auto DNS through DHCP is in use, and any DNS Servers

  • Search Path, if any

  • Time Zone

  • Whether an NTP server is in use an its identifier

11.10.5 RAID Controller Details

  • RAID volume name

  • RAID level

  • Number of disks

  • Stripe zone

  • RAID Controller ID

11.10.6 Sun ZFS Storage Appliance Details

The Sun ZFS Storage Appliance support both file storage and application use.

The Dashboard tab reports the following hardware information:

  • Name

  • Description

  • Current Alert Status

  • Model

  • Serial Number

  • Management IP

  • Memory

  • Power

  • Locator Light

  • Appliance Kit Version

  • Running Time

  • Processor

The Hardware tab displays the appliance's firmware version and the following information for each component:

  • CPU: Name, Model, Architecture, Speed, Manufacturer

  • Memory: Name, Type, Size in bytes, Manufacturer, Part number, Serial number

  • Network Adapters: Name or each, MAC Address, Description, Manufacturer, Part number, Serial number

  • Disks: Name, Size in bytes, Manufacturer, Part number, Serial number

  • Power Supply: Name, Manufacturer, Part number, Serial Number

  • Fan Tray: Name, Manufacturer, Part number, Serial number

Note:

The Create Filesytem and Delete Filesystem actions are disabled for the appliance installed in Exalogic systems.

See Sun ZFS Storage Appliance in Chapter 2 for discovery requirements.

11.10.7 Server Details

For server hardware, the Summary page displays:

  • Name

  • Description and Tags

  • Current Alert Status

  • Model

  • Serial Number

  • Management Interface IP

  • MAC Address

  • Processor

  • Memory

  • Power state

    • On – The server is powered on and running.

    • Standby – The server is powered off but responds to commands.

    • Unknown – An error occurred while attempting to retrieve the power status of the hardware. The server is connected but is not returning any information on power status.

    • Unreachable –The server cannot be contacted for information about its power state. This indicates a network problem or that the server is in standby mode.

  • Locator Lights state

  • Notification

Use the Hardware tab to view information about each component of the system:

  • System: Description, type, and version of all firmware installed except for disk firmware. See the Disk tab for firmware version.

  • CPU: Name, Model, Architecture, Speed, Manufacturer

  • Memory: Name, Type, Size in bytes, Manufacturer, Part number, Serial number

  • Network Adaptors: Name or each, MAC Address, Manufacturer, Part number, Serial Number

  • Disks: Name of each, Model, Size in bytes, Slot ID, Node ID, Firmware Version, Manufacturer, Root Disk, RAID Disk

  • Power Supply: Name, Manufacturer, Part number, Serial number

  • Disk Controller: Name, Model Number, Firmware Version, BIOS Version, PCI Address, PCI Version ID

  • Disk Expander: Name, Manufacturer, Version, Model Number, Firmware Version, Chassis ID

  • Fan Tray: Name, Manufacturer, Part number, Serial number

  • Fans: Name, Speed

11.10.8 Chassis Details

For chassis hardware as a group, the Summary page shows:

  • The five largest consumers of CPU

  • The five largest consumers of memory

  • The five largest consumers of the network

For chassis hardware, the Summary page displays:

  • Group Name

  • Description

  • Location

  • Type

11.10.9 M-Series Details

The hardware resources in a SPARC Enterprise M-Series Server are divided into one or more logical units, called dynamic system domains. Oracle Enterprise Manager Ops Center can monitor each domain, in addition to the server hardware.

For an M-Series server, the Dashboard tab displays:

  • Number of Dynamic System Domains it is supporting

  • Model

  • Product Serial Number

  • Description

  • Support contract

  • XCP Firmware version

  • OBP Firmware version

  • XSCF Firmware version

  • Operator Panel Switch Status: Locked

  • Current Alert Status

The Summary tab adds details to the information in the Dashboard tab. For the Power status, the reported status is for the server's domains. When any domain is powered on, the status is reported as powered on. When all domains are powered off, the Summary tab shows a status of Powered Off; the M-Series server itself remains powered on.

  • Name

  • Model

  • Product Serial Number

  • Management IP

  • MAC Address

  • Current Alert Status

  • Power

  • Locator Light

  • Notification

  • All firmware versions

  • For each domain: Name, Model, Health, Power, Locator Light, Notification

The Hardware tab shows the state of the server or, if a Dynamic System Domain is selected, the state of that domain. The Hardware tab reports the following:

  • Model

  • Serial Number

  • State

  • Power

  • Locator Light

  • Notification

  • Operator Panel Switch State

At the System level, the Hardware tab include the following information.

  • The Unallocated Resources table lists all the physical system boards and their status: PSD ID, Assignment Status, Power Status, Connection Status, Diagnostics Status, and Operational Status

  • The Allocated Resources table lists all domains that are using the physical system boards and their status: Domain ID, PSB ID, XSB ID, LSB ID, Assignment Status, Power Status, Connection Status, Diagnostics Status, and Operational Status

  • The Dynamic System Domain table lists all the domains and their details: Domain ID, MAC Address, Autoboot Policy, Secure Mode Policy, CPU Mode, Diagnostics Level, Domain Degradation Policy, and Operational Status

You change the display to show information about each component of the system:

  • CPU: Name, Architecture, Type, Manufacturer, Speed, Core Count, Thread Count, Serial Number, Part Number, Version, Status For Sensors: Name, Description, Type, Value

  • Memory: Name, Type, Size in bytes, Serial number, Part number, Status For Sensors: Name, Description, Type, Value

  • Board: Name, Serial number, Part number, Memory mirrored, Version, Status

  • Power Supply: Name, Serial number, Part Number, Status For Sensors: Name, Description, Type, Value

  • Board: Name, XSB Mode, Memory Mirrored, Serial Number, Part Number, Version, Status For Sensors: Name, Description, Type, Value

  • IO Unit: Name, Serial Number, Part Number Version, Status For Sensors: Name, Description, Type, Value

  • XSCF: Name, Host Name, Serial Number, Part Number, Version, Status

  • Fan Tray: Name, Manufacturer, Part number, Serial number

  • Fans: Name, Speed For Sensors: Name, Description, Type, Value

Oracle Enterprise Manager Ops Center monitors the voltage for the Board and IO Unit components and the speed for the Fan components. The Monitoring tab shows the actual value and the threshold values.

See Discovering a SPARC Enterprise M-Series Server in Chapter 2 for discovery requirements.

11.10.10 Switch Details

Oracle Enterprise Manager Ops Center can manage Sun Ethernet 10GbE Fabric switches, Sun Datacenter InfiniBand switches, and Cisco Catalyst 4948 Ethernet switches. These switches reside in the system or blade system and provide the switch fabric.

Oracle Enterprise Manager Ops Center reports hardware information on the Summary tab:

  • Name

  • Description

  • Model

  • Port count

  • Serial number

  • Management Interface IP

  • MAC Address

  • Fabric Manager: true or false

  • Fabric Manager Address

  • Power state

  • Locator lights state

  • Notification state

  • Current Alert Status

  • Firmware types and versions

At the System level, the Hardware tab includes:

  • Model

  • Server Name

  • Serial Number

  • State

  • Power

  • Firmware versions

  • Sensors: temperature and voltage

You change the display to show information about each component of the switch:

  • Network Adaptors: Name or each, MAC Address, IP Address, Description

  • Power Supply: Name, Manufacturer, Part number, Serial Number For Sensors: Description, Type, Status

  • Fan Sensors: Description, Type, Value, Status, Warning Threshold (Lower), Warning Threshold (Upper), Critical Threshold (Lower), Critical Threshold (Upper), Non-Recoverable Threshold (Lower), Non-Recoverable Threshold (Upper)

11.10.11 Rack Details

The rack's Dashboard displays the following information:

  • Name

  • Rack ID

  • Description

  • Number of Slots

  • Tags

  • Support

The Details tab shows the configuration of each slot in the rack. For each component in the rack, this tab displays the position, name, description, type, model and health status. The Power Distribution Units are included and their current status is displayed.

The Firmware tab displays the name, description, and version of the firmware for each component and the slot for each component:

  • Compute Nodes

  • Switches

  • Storage Appliances

  • Power Distribution Units

The rack's Charts tab displays the following plots, as described in Charts Tab:

  • Aggregate Power Usage

  • Power Usage

  • Average Fan Speed

The rack's Energy tab displays the information described in Energy Tab

Note:

The Power Off and Reset actions are disabled for Exalogic systems and for single-rack configurations.

11.10.12 PDU Details

The Dashboard tab shows:

  • Name and description

  • Model and serial number

  • Management IP address and MAC address, if any

The Details tab reports the same information and adds whether SNMP and HTTP is enabled and the version of the firmware.

11.10.13 Oracle Solaris Cluster Details

The Dashboard tab shows:

  • Name, description, and ID

  • Number of possible quorum votes and the current quorum votes

The Network tab shows the public and private interconnects used by the cluster.

The Quorum tab shows the status of each member of the quorum and the number of quorum votes for each member.

11.10.14 Oracle Exalogic Details

Oracle Enterprise Manager Ops Center supports bare metal provisioning and discovery of hardware assets of an Exalogic machine. The Oracle Exalogic architecture includes InfiniBand and Ethernet switches and the Sun ZFS Storage 7320 Appliance. Using Oracle Enterprise Manager Ops Center, you can manage these components individually and as a group, as well as get access to the browser user interface to the storage appliance and serial console access to the compute nodes and switches.

For detailed information about how to use Oracle Enterprise Manager Ops Center with Oracle Exalogic, see the Oracle Exalogic documentation: http://docs.oracle.com/cd/E18476_01/doc.220/e18478/toc.htm

11.11 Monitoring Power Utilization

Input power is the power pulled into a power supply from an external resource. The power consumption of a hardware asset is the sum of the input power consumed by each power supply of the asset. Output power is the amount of power provided from the power supply to the system components, measured at the power supply output. Input power is calculated from output power by applying an efficiency function to the output power from each power supply.

Calculating power compensation for the blades is difficult because the power supplies are shared. Each blade gives a report based on the power consumption of the local components, but this is not an accurate power consumption value for an individual blade.

To measure the input power, the interfaces must be exposed and the service processors must be able to retrieve and report data with one-minute accuracy. Servers that can report power usage have a Charts tab.

You can see current power usage and change the display of power graphs using the controls on the Energy tab and the Charts tab.

11.11.1 Energy Tab

The asset's Energy tab reports power consumption as the current value and for a period of time, as well as attributes of the fan and power supplies. The current values are reported for the following attributes:

  • Wattage

  • System Load for an OS

  • Utilization Per Cent for an Oracle VM Server for SPARC

  • Incoming air temperature and outgoing air temperature

  • Power Policy

  • Cost Per KiloWatthour

  • Currency units used to compute cost. The price per currency unit is set by the Edit Energy Cost action in the Administration section of the Navigation pane. See the Oracle Enterprise Manager Ops Center Administration Guide for information.

  • Total Power Cost for one day

The data over time is represented in the following graphs:

  • Power Consumption and Utilization: By default, the graph shows the power consumed in the last day in watts. If the server is shut down, the graph shows any existing historical data.

  • Temperature and Fan Speed: By default, the graph shows the incoming air temperature and the outgoing air temperature in Celsius, and the average fan speed in RPM. Click any point on the graph to see that data for that point in time.

By default, the graphs are in Live mode, which reports new information every five seconds. Click the Live button in the graph's toolbar to make the information static so that you can set the period of the graphs to one of the following:

  • One hour (1H) – One point for every five minutes

  • One day (1D) – One point for every five minutes

  • Five days (5D) – One point for every five minutes

  • Three weeks (3W) – One point every hour

  • Six weeks (6W) – One point every 12 hours

  • Six months (6M) – One point for every day

To make a graph with the minimum of two points, a hardware asset must have been managed for at least 10 minutes to view a one-hour graph and for at least two days to view the six-months graph.

The data for these time periods is stored separately. For example, if a server has been managed for two hours and you select the 6W view, the graph cannot be displayed because only one point of data of that type has been stored; the second point has not yet occurred. If you then select the 1D view, the graph can display 24 points of data (120 minutes at 5-minute intervals). However, the graph displays these points over a 24-hour period and not over the actual two-hour period. For the most accurate representation of the data, choose a time period that is less than or equal to the time that the hardware asset has been managed.

You can export the data for either the current view or all available data to a file in either CSV or XML format. Use the Export Chart Data toolbar icon to choose options for exporting the data.

If the graph is blank, one of the following conditions has occurred:

  • The server does not have the appropriate ILOM version.

  • The server has not been discovered through the ILOM driver.

  • The server is unreachable.

11.11.2 Charts Tab

The Chart tab provides more ways to display the power utilization data. You can change the graphed data to a bar chart or an area chart. You can also export the data for either the current view or all available data to a file in either CSV or XML format. Use the Export Chart Data button to choose options for exporting the data.

For groups and virtual pools, the following options are available:

  • Select Order: The five highest or five lowest historical power utilization.

  • Select Resource: Select the Power or Aggregate Power option for a homogeneous or heterogeneous group of servers.

    • The Power option displays power utilization for the five highest or lowest power consumers in the group or virtual pool.

    • The Aggregate Power option displays the power utilization, using the sum of all members that report power consumption. The number of systems in the aggregate is included. For heterogeneous group, the Chart tab includes a table of all systems in the group and their various power attributes for the selected time period. From this table, you can power off and power on selected servers to conserve power.

11.12 Maintaining Hardware Assets

  • Update management credentials. See Updating Management Credentials in Chapter 2

  • Set power policy

  • Power systems on and off

  • Place in maintenance mode

  • Reset a server

  • Get access to the serial console

  • Enable and disable ports

  • Use locator lights to identify a specific asset

  • Check firmware compliance and update firmware. See Firmware Provisioning

11.12.1 Setting and Changing the Power Policy

An asset's use of power can be optimized for performance or for conservation:

  • Performance – Disable CPU power management to use of more power.

  • Elastic – Enable CPU power management to conserve power.

View an Asset's Power Policy

  1. Expand the Assets in the Navigation pane.

  2. Select a hardware asset.

  3. Click the Energy tab in the center pane.

Changing an Asset's Power Policy

  1. Expand the Assets in the Navigation pane.

  2. Select a hardware asset.

  3. Click Set Power Policy in the center pane to display the window shown in Figure 11-7.

    Figure 11-7 Set Power Policy

    Description of Figure 11-7 follows
    Description of "Figure 11-7 Set Power Policy"

  4. The current power policy is displayed. Choose the alternative policy.

  5. Click Submit to create a job that sets the power policy.

11.12.2 Replacing a Failed Power Distribution Unit in a Rack

  1. Expand Assets in the Navigation pane and then expand Racks.

  2. Navigate to the rack type and expand to show its components. Expand Power Distribution Units.

  3. Select the failed power distribution unit.

  4. Click Place in Maintenance Mode in the Action pane.

  5. Go to the rack's location and remove the failed PDU.

    Install the new PDU and connect it to the network, according to the procedures in the Power Distribution Units User's Guide.

  6. Verify that the new PDU has the same IP address as the failed PDU.

  7. Use the PDU's web interface to configure the administrator user account, according to the procedure in the Power Distribution Units User's Guide.

  8. Return to the Oracle Enterprise Manager Ops Center user interface and navigate to the same rack and the new PDU. Select the PDU.

  9. Click Remove From Maintenance Mode in the Action pane.

  10. To update the credentials for the administrator user, click Update Management Credentials in the Action pane. The wizard opens.

  11. At the Management Type step, select HTTP credentials and then select the Create a new set of credentials option. Click Next.

  12. Enter new credentials for the administrator user.

  13. Click Apply to update credentials.

  14. To create new SNMP credentials, click Update Management Credentials in the Actions pane again.

  15. Select SNMP credentials and Create a new set of credentials.

    Specify a community string that is different than the previous community string.

  16. Press Apply to update credentials.

11.12.3 Installing and Upgrading Oracle Solaris Cluster

You use cluster profiles in a deployment plan to perform the following operations:

  • Install Oracle Solaris Cluster software

  • Upgrade Oracle Solaris Cluster software

See Appendix B for instructions on obtaining the current profiles and supporting scripts.

11.13 Firmware Provisioning

The Oracle Enterprise Manager Ops Center provisioning feature installs firmware on the managed hardware assets, using the protocols and ports listed in Table 7-6. You initiate the installations from the UI, rather than from the asset itself.

Oracle Enterprise Manager Ops Center provides default profiles for configuring firmware for servers and for disk storage. An alternate procedure is to create a Firmware Report. You then use the report results to update the firmware.

Note:

The Update Firmware and Firmware Reports actions are disabled for Exalogic systems, except for its storage resource, the Sun ZFS Storage Appliance.

The benefit of using a profile to install firmware is that the firmware is installed consistently, no matter how many assets you provision. The benefit of using the Firmware Compliance Report is to identify the firmware on a specific asset or set of assets.

11.13.1 Firmware Profiles

The general procedure for provisioning firmware has the following steps:

  1. Import a file with the firmware and the associated metadata into a software library, according to the procedure in Keeping Your Firmware Up-to-Date in the How To library at http://docs.oracle.com/cd/E27363_01/nav/howto.htm.

  2. Create a firmware profile, based on one or more firmware images, according to the procedures in Keeping Your Firmware Up-to-Date in the How To Library at http://docs.oracle.com/cd/E27363_01/nav/howto.htm.

  3. Shut down the server gracefully. Most firmware requires that the server is not running when the firmware is updated. Most firmware images include a power-off command for a running server, which causes a hard shutdown of the server.

  4. Apply the firmware profile.

11.13.2 Firmware Compliance Reports

The Firmware Report feature compares the firmware images specified in a firmware profile to the firmware images installed on one or more hardware assets. Use the Make Targets Compliant button in the Interactive type of report to update the firmware on any non-compliant assets.

The report shows whether a target asset is compliant, not compliant, or not applicable:

  • A compliant asset has the firmware images specified in the profile.

  • A non-compliant asset does not have the same firmware images as specified in the profile. Update the firmware by either clicking the Make Targets Compliant button in the Interactive report or using the procedure in Updating Firmware.

  • A non-applicable asset indicates that a firmware image in the profile does not match the model of service processor in the asset. This condition can occur when either the profile does not recognize the model of the service processor or the profile includes firmware images that are not designed for the service processor.

    1. Compare the model of the service processor displayed in the asset's Summary tab with the model of the service processor included in the profile. If they are different, add the name in the profile to the asset's data.

    2. When the firmware profile was created, only images that matched the service processor could be included. However, if the service processor did not report all the firmware types it supported, an image that did not match the service processor could have been included in the profile. To update the Oracle Enterprise Manager Ops Center software with all the service processor's supported firmware types, use the Refresh action. When the job is completed, view the service processor's Summary tab to see all firmware types.

    3. Create a new firmware compliance report.

See Creating a Firmware Report for the procedure to create the report.

11.13.3 Updating Firmware

To update the firmware on one or many assets, you use a deployment plan to apply a firmware profile for the type of asset. For a server, the profile updates the firmware on a service processor, and restarts the service processor and operating system. For storage components, profiles update firmware on a RAID controller, an expander, or disk.

To see the deployment plans that update firmware, expand the Deployment Plans section of the Navigation pane and then click Firmware. A list of existing plans and profiles is displayed.

To update the firmware of one asset, an alternative to a deployment plan is to use the Update Firmware action. Select the asset from the Asset section of the Navigation pane and then click Update Firmware in the Actions pane.

Before You Begin

The software library must contain the images that provision the firmware. Perform the Uploading a Firmware Image procedure in the Keeping Your Firmware Up-to-Date document in the How To library at http://docs.oracle.com/cd/E27363_01/nav/howto.htm.

If you are updating the firmware on a server, shut down the server before you update the firmware. A firmware update to a server's service processor usually requires that the server is not running. If you start to update the firmware on a running server's service processor, the procedure performs a hard shutdown of the server.

Requirements for ALOM Service Processors and M-Series Servers

The firmware provisioning process for M-Series servers and servers that have ALOM service processors relies on a temporary account that performs an FTP operation.

Note:

Advanced Lights Out Management (ALOM) is a Sun Microsystems standard for these systems: SunFire V125/V210/V215/V240/V245/V250/V440/T1000 and Sun Netra 210/240/440.

If your site does not allow a temporary account, use the following procedure to prepare for the provisioning operation:

  1. On the Enterprise Controller, open the /var/opt/sun/xvm/hal.properties in an editor.

  2. Add the following properties to the file:

    ftp.user.name=username
    ftp.user.password=password
    
  3. Restrict access to the file to root user:

    chmod 600 /var/opt/sun/xvm/hal.properties
    
  4. On the Proxy Controller that provisions the firmware, enable the ftp service on Oracle Solaris or the vsftpd service for Oracle Linux systems.

You can now apply the firmware provisioning deployment plan. The FTP operation retrieves the credentials from the file.

If a network failure occurs while updating the firmware, repeat the firmware update procedure. If you do not repeat the procedure, the firmware inventory list might be incomplete.

Option for Deferring the Stop and Restart of the Operating System and Server

The procedure is available for Oracle Solaris 10 Update 10 operating systems running on servers with the ILOM x86 (3.0 and higher) service processor.

When you update the firmware of a service processor, the procedure stops the operating system and the server before the update and restarts them after the update so that the new BIOS takes effect. If you prefer to stop and restart at a convenient time, keeping the current BIOS in effect, use the following procedure to change the action of both Oracle Enterprise Manager Ops Center and the firmware's metadata:

  1. On the Enterprise Controller, open the /var/opt/sun/xvm/hal.properties in an editor.

  2. Add the following property to the file:

    ilom.fwp.skipAutoReboot=true 
    
  3. On the Proxy Controller that provisions the firmware, enable the ftp service on Oracle Solaris or the vsftpd service for Oracle Linux systems.

  4. At a later time, reboot the servers.

11.14 Related Resources for Hardware Management

For instructions in performing actions or to learn more about the role of this feature, go to one of the following resources.