Managing a node remotely by using the netapp SP

Managing a node remotely by using the Service Processor

The Service Processor (SP) is a remote management device that enables you to access, monitor, and troubleshoot a node remotely.

The SP provides the following capabilities:

• The SP enables you to access a node remotely to diagnose, shut down, power-cycle, or reboot the node, regardless of the state of the node controller.

  The SP is powered by a standby voltage, which is available as long as the node has input power to at least one of its power supplies.

  You can log in to the SP by using a Secure Shell client application from an administration host. You can then use the SP CLI to monitor and troubleshoot the node remotely. In addition, you can use the SP to access the serial console and run Data ONTAP commands remotely.

  You can access the SP from the serial console or access the serial console from the SP. The SP allows you to open both an SP CLI session and a separate console session simultaneously.

  For instance, when a temperature sensor becomes critically high or low, Data ONTAP triggers the SP to shut down the motherboard gracefully. The serial console becomes unresponsive, but you can still press Ctrl-G on the console to access the SP CLI. You can then use the system power on orsystem power cycle command from the SP to power on or power-cycle the node.

• The SP monitors environmental sensors and logs events to help you take timely and effective service actions.

  The SP monitors the node temperatures, voltages, currents, and fan speeds. When an environmental sensor has reached an abnormal condition, the SP logs the abnormal readings, notifies Data ONTAP of the issue, and sends alerts and "down system" notifications as necessary through an AutoSupport message, regardless of whether the node can send AutoSupport messages.

  Other than generating these messages on behalf of a node that is down and attaching additional diagnostic information to AutoSupport messages, the SP has no effect on the AutoSupport functionality. The AutoSupport configuration settings and message content behavior are inherited from Data ONTAP.

  Note: The SP does not rely on the system node autosupport modify command's -transport parameter setting to send notifications. The SP uses the Simple Mail Transport Protocol (SMTP).

  If SNMP is enabled for the SP, the SP generates SNMP traps to configured trap hosts for all "down system" events.

  The SP also logs events such as boot progress, Field Replaceable Unit (FRU) changes, Data ONTAP-generated events, and SP command history.

• The SP has a nonvolatile memory buffer that stores up to 4,000 events in a system event log (SEL) to help you diagnose issues.

  The SEL stores each audit log entry as an audit event. It is stored in onboard flash memory on the SP. The event list from the SEL is automatically sent by the SP to specified recipients through an AutoSupport message.

  The SEL contains the following data:

  • Hardware events detected by the SP—for example, sensor status about power supplies, voltage, or other components
  • Errors detected by the SP—for example, a communication error, a fan failure, or a memory or CPU error
  • Critical software events sent to the SP by the node—for example, a panic, a communication failure, a boot failure, or a user-triggered "down system" as a result of issuing the SP system reset or system power cycle command
• The SP monitors the serial console regardless of whether administrators are logged in or connected to the console.

  When messages are sent to the console, the SP stores them in the console log. The console log persists as long as the SP has power from either of the node power supplies. Because the SP operates with standby power, it remains available even when the node is power-cycled or turned off.

• Hardware-assisted takeover is available if the SP is configured.

  For more information about hardware-assisted takeover, see the Clustered Data ONTAP High-Availability Configuration Guide.

More information

• Configuring the SP network
  Before you can access the SP of a node, the SP network must be configured and enabled. You can configure the SP to use IPv4, IPv6, or both. The SP IPv4 configuration supports static and DHCP addressing, and the SP IPv6 configuration supports static addressing only.
• Accounts that can access the SP
  Cluster user accounts that are created with the service-processor application type have access to the SP CLI on any node of the cluster that supports the SP. SP user accounts are managed from Data ONTAP and authenticated by password.
• Accessing the SP from an administration host
  You can log in to the SP of a node from an administration host to perform node management tasks remotely.
• Accessing the SP from the serial console
  You can access the SP from the serial console to perform monitoring or troubleshooting tasks.
• Accessing the serial console from the SP
  The SP's system console command enables you to log in to the serial console from the SP.
• Relations among the SP CLI, SP console, and serial console sessions
  You can open an SP CLI session to manage a node remotely and a separate SP console session to run Data ONTAP commands remotely. The SP console session mirrors output displayed in a concurrent serial console session. The SP and the serial console have independent shell environments with independent login authentication.
• Using online help at the SP CLI
  The SP online help displays the SP CLI commands and options when you enter the question mark (?) or help at the SP prompt.
• Commands for managing a node at the SP admin privilege level
  The SP commands at the admin privilege level enable you to display events, logs, and status information for node power, batteries, sensors, field-replaceable units (FRUs), or the SP itself. The commands also enable you to reboot the node or the SP and create a core dump.
• Commands for managing a node at the SP advanced privilege level
  You can use the SP advanced privilege level to display the SP command history, SP debug file, SP messages file, and data history for field-replaceable units (FRUs). You can also manage the battery firmware and automatic update.
• How to determine the status of a threshold-based SP sensor
  Threshold-based sensors take periodic readings of a verity of system components. The SP compares the reading of a threshold-based sensor against its preset threshold limits that define a component’s acceptable operating conditions. Based on the sensor reading, the SP displays the sensor state to help you monitor the condition of the component.
• Understanding the status of a discrete SP sensor
  The Status column of the system sensors command output in the SL CLI shows the discrete sensors' conditions in hexadecimal values. To interpret the status values of most discrete sensors, you can use the system sensors get sensor_name command in the SL CLI.
• Troubleshooting a node by using the SP
  When you encounter a problem with a node, you can use the SP to display information about the problem, create a core dump, and reboot the node, even if the node's firmware is corrupted.
• Managing the SP with Data ONTAP
  You can use Data ONTAP to set up and display the SP configuration, display the SP status, reboot the SP, manage the SP firmware image, and manage access to the SP.

Parent topic: Managing a node remotely

Copyright © 1994-2013, NetApp, Inc. All rights reserved.
Part No. 215-07956_A0
May 2013