F5 BIG-IP平台上的LCD消息说明

This article applies to BIG-IP iSeries hardware platforms.

Description

The LCD panel provides the ability to manage and monitor the unit without attaching a console or network cable. The different sensors that exist in the hardware platform provides information regarding the hardware health of the system by way of messages on the LCD screen. The following tables list messages that can appear on the LCD screen of iSeries platforms.

PSUx Status sensor messages

The PSUx sensor is a discrete sensor based on the Intelligent Platform Management Interface (IPMI) standard sensor type code of 0x8. The sensor indicates various power supply unit (PSU) states.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
PSUx Status present	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	PSU# Status present	.1.3.6.1.4.1.3375.2.4.0.190
PSUx Status absent	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	PSU# Status absent	.1.3.6.1.4.1.3375.2.4.0.190
PSUx Status input lost	Warning	BIGIP_LIBHAL_AOM_SENSOR_PWR_WARN	012a0040	PSU# Status input lost	.1.3.6.1.4.1.3375.2.4.0.185
PSUx Status input active	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	PSU# Status input active	.1.3.6.1.4.1.3375.2.4.0.190
N/A	Notice	BIGIP_LIBHAL_CHASSIS_PS1_POWERED_ON BIGIP_LIBHAL_CHASSIS_PS2_POWERED_ON BIGIP_LIBHAL_CHASSIS_PS3_POWERED_ON BIGIP_LIBHAL_CHASSIS_PS4_POWERED_ON BIGIP_LIBHAL_CHASSIS_PS_POWERED_ON BIGIP_LIBHAL_CHASSIS_PS_IS_POWERED_ON	012a0046 012a0047 012a0048 012a0049 012a0017 012a0019	Chassis power module 1 turned on. Chassis power module 2 turned on. Chassis power module 3 turned on. Chassis power module 4 turned on. Chassis power module # turned on. Chassis power module # is on.	.1.3.6.1.4.1.3375.2.4.0.147
N/A	Warning	BIGIP_LIBHAL_CHASSIS_PS1_POWERED_OFF BIGIP_LIBHAL_CHASSIS_PS2_POWERED_OFF BIGIP_LIBHAL_CHASSIS_PS3_POWERED_OFF BIGIP_LIBHAL_CHASSIS_PS4_POWERED_OFF BIGIP_LIBHAL_CHASSIS_PS_POWERED_OFF BIGIP_LIBHAL_CHASSIS_PS_IS_POWERED_OFF	012a0050 012a0051 012a0052 012a0053 012a0018 012a0020	Chassis power module 1 turned off. Chassis power module 2 turned off. Chassis power module 3 turned off. Chassis power module 4 turned off. Chassis power module # turned off. Chassis power module # is off.	.1.3.6.1.4.1.3375.2.4.0.148
N/A	Notice	BIGIP_LIBHAL_CHASSIS_PS1_ABSENT BIGIP_LIBHAL_CHASSIS_PS2_ABSENT BIGIP_LIBHAL_CHASSIS_PS3_ABSENT BIGIP_LIBHAL_CHASSIS_PS4_ABSENT BIGIP_LIBHAL_CHASSIS_PS_ABSENT	012a0054 012a0055 012a0056 012a0057 012a0021	Chassis power module 1 absent. Chassis power module 2 absent. Chassis power module 3 absent. Chassis power module 4 absent. Chassis power module # absent.	.1.3.6.1.4.1.3375.2.4.0.149
Power supplies do not match	Critical	BIGIP_LIBHAL_CHASSIS_PSU_MISMATCH	012a0060	Power supplies do not match. PSU #: modelnumber PSU #: modelnumber	N/A
PSUx Status config mismatch	Critical	BIGIP_LIBHAL_AOM_SENSOR_PWR_CRIT	012a0043	PSU# Status config mismatch	.1.3.6.1.4.1.3375.2.4.0.188
PSUx Status config match	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	PSU# Status config match	.1.3.6.1.4.1.3375.2.4.0.190
PSUx Status output lost	Critical	BIGIP_LIBHAL_AOM_SENSOR_PWR_CRIT	012a0043	PSU# Status output lost	.1.3.6.1.4.1.3375.2.4.0.188
PSUx Status output active	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	PSU# Status output active	.1.3.6.1.4.1.3375.2.4.0.190
PSUx Status PSU unidentified	Critical	BIGIP_LIBHAL_AOM_SENSOR_PWR_CRIT	012a0043	PSU# Status PSU unidentified	.1.3.6.1.4.1.3375.2.4.0.188
PSUx Status identified	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	PSU# Status PSU identified	.1.3.6.1.4.1.3375.2.4.0.190

Power Fault sensor messages

The Power Fault sensor is a discrete sensor indicating that a power fault has occurred.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
Power Fault host asserted	Critical	BIGIP_LIBHAL_AOM_SENSOR_PWR_CRIT	012a0043	Power Fault host asserted	.1.3.6.1.4.1.3375.2.4.0.188
Power Fault host deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	Power Fault host deasserted	.1.3.6.1.4.1.3375.2.4.0.190
Power Fault bay X asserted	Critical	BIGIP_LIBHAL_AOM_SENSOR_PWR_CRIT	012a0043	Power Fault bay # asserted	.1.3.6.1.4.1.3375.2.4.0.188
Power Fault bay X deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	Power Fault bay # deasserted	.1.3.6.1.4.1.3375.2.4.0.190
Power Fault switchboard asserted	Critical	BIGIP_LIBHAL_AOM_SENSOR_PWR_CRIT	012a0043	Power Fault switchboard asserted	.1.3.6.1.4.1.3375.2.4.0.188
Power Fault switchboard deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	Power Fault switchboard deasserted	.1.3.6.1.4.1.3375.2.4.0.190
Power Fault HDD x asserted	Critical	BIGIP_LIBHAL_AOM_SENSOR_PWR_CRIT	012a0043	Power Fault HDD # asserted	.1.3.6.1.4.1.3375.2.4.0.188
Power Fault HDD x deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_PWR_INFO	012a0045	Power Fault HDD # deasserted	.1.3.6.1.4.1.3375.2.4.0.190

CPU Fault sensor messages

The CPU Fault sensor is a discrete sensor indicating that a CPU fault has occurred.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
CPU Fault CATERR asserted	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	CPU Fault CATERR asserted	.1.3.6.1.4.1.3375.2.4.0.170
CPU Fault CATERR deasserted	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CPU Fault CATERR deasserted	.1.3.6.1.4.1.3375.2.4.0.172
CPU Fault THERMTRIP asserted	Emergency	BIGIP_LIBHAL_AOM_SENSOR_TEMP_EMERG	012a0032	CPU Fault THERMTRIP asserted	.1.3.6.1.4.1.3375.2.4.0.177
CPU Fault THERMTRIP deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_TEMP_INFO	012a0033	CPU Fault THERMTRIP deasserted	.1.3.6.1.4.1.3375.2.4.0.178
CPU Fault PROCHOT asserted	Critical	BIGIP_LIBHAL_AOM_SENSOR_TEMP_CRIT	012a0031	CPU Fault PROCHOT asserted	.1.3.6.1.4.1.3375.2.4.0.176
CPU Fault PROCHOT deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_TEMP_INFO	012a0033	CPU Fault PROCHOT deasserted	.1.3.6.1.4.1.3375.2.4.0.178
CPU Fault FIVR asserted	Emergency	BIGIP_LIBHAL_AOM_ALERT_EMERG	012a0026	CPU Fault FIVR asserted	.1.3.6.1.4.1.3375.2.4.0.171
CPU Fault FIVR deasserted	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CPU Fault FIVR deasserted	.1.3.6.1.4.1.3375.2.4.0.172
CPU Fault DIMM Event asserted	Warning	BIGIP_LIBHAL_AOM_ALERT_WARN	012a0022	CPU Fault DIMM asserted	.1.3.6.1.4.1.3375.2.4.0.167
CPU Fault DIMM Event deasserted	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CPU Fault DIMM deasserted	.1.3.6.1.4.1.3375.2.4.0.172
CPU Fault MEMHOT asserted	Warning	BIGIP_LIBHAL_AOM_SENSOR_TEMP_WARN	012a0028	CPU Fault MEMHOT asserted	.1.3.6.1.4.1.3375.2.4.0.173
CPU Fault MEMHOT deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_TEMP_INFO	012a0033	CPU Fault MEMHOT deasserted	.1.3.6.1.4.1.3375.2.4.0.178
CPU Fault VR Hot asserted	Warning	BIGIP_LIBHAL_AOM_SENSOR_TEMP_WARN	012a0028	CPU Fault VR Hot asserted	.1.3.6.1.4.1.3375.2.4.0.173
CPU Fault VR Hot deasserted	Info	BIGIP_LIBHAL_AOM_SENSOR_TEMP_INFO	012a0033	CPU Fault VR Hot deasserted	.1.3.6.1.4.1.3375.2.4.0.178
CPU Fault CPU IERR asserted	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	CPU Fault IERR asserted	.1.3.6.1.4.1.3375.2.4.0.170
CPU Fault CPU IERR deasserted	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CPU Fault IERR deasserted	.1.3.6.1.4.1.3375.2.4.0.172
CPU Fault CPU MCERR asserted	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	CPU Fault MCERR asserted	.1.3.6.1.4.1.3375.2.4.0.170
CPU Fault CPU MCERR deasserted	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CPU Fault MCERR deasserted	.1.3.6.1.4.1.3375.2.4.0.172

Filter sensor messages

The Filter sensor sends an event when the chassis filter needs to be replaced. This is a threshold-based sensor.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
Chassis air filter replacement due	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	Chassis air filter replacement due	.1.3.6.1.4.1.3375.2.4.0.172
Chassis air filter replacement overdue	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	Chassis air filter replacement overdue	.1.3.6.1.4.1.3375.2.4.0.170

LCD Health sensor messages

The LCD Health sensor indicates whether or not the BMC can communicate with the LCD.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
LCD Health communication OK	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	LCD Health communication OK	.1.3.6.1.4.1.3375.2.4.0.172
LCD Health communication error	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	LCD Health communication error	.1.3.6.1.4.1.3375.2.4.0.170

Module Presence sensor messages

The Module Presence sensor indicates whether or not a module is present.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
fan tray present	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	fan tray present	.1.3.6.1.4.1.3375.2.4.0.172
fan tray absent	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	fan tray absent	.1.3.6.1.4.1.3375.2.4.0.170
bay x present	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	bay # present	.1.3.6.1.4.1.3375.2.4.0.172
bay x absent	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	bay # absent	.1.3.6.1.4.1.3375.2.4.0.170
switchboard present	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	switchboard present	.1.3.6.1.4.1.3375.2.4.0.172
switchboard absent	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	switchboard absent	.1.3.6.1.4.1.3375.2.4.0.170
LCD present	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	LCD present	.1.3.6.1.4.1.3375.2.4.0.172
LCD absent	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	LCD absent	.1.3.6.1.4.1.3375.2.4.0.170
CPU 00 present	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CPU 00 present	.1.3.6.1.4.1.3375.2.4.0.172
CPU 00 absent	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	CPU 00 absent	.1.3.6.1.4.1.3375.2.4.0.170
CPU 01 present	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CPU 01 present	.1.3.6.1.4.1.3375.2.4.0.172
CPU 01 absent	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	CPU 01 absent	.1.3.6.1.4.1.3375.2.4.0.170

Leonardo Daughter Card Absent messages

A single general-purpose input/output (GPIO) on Leonardo indicates that the daughter card is absent.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
CM Daughter Card absent	Critical	BIGIP_LIBHAL_AOM_ALERT_CRIT	012a0025	CM Daughter Card absent	.1.3.6.1.4.1.3375.2.4.0.170
CM Daughter Card present	Info	BIGIP_LIBHAL_AOM_ALERT_INFO	012a0027	CM Daughter Card present	.1.3.6.1.4.1.3375.2.4.0.172

FIPS Initialization Status messages

The Federal Information Processing Standards (FIPS) Initialization Status sensor tracks the state of the Full Box FIPS Mode initialization for the host application.

Alert message	Alert level	Alert name	Error code	Log message	Trap OID
FIPS initialization error asserted	Warning	BIGIP_LIBHAL_AOM_ALERT_WARN	012a0022	FIPS initialization error asserted	.1.3.6.1.4.1.3375.2.4.0.167
FIPS initialization error deasserted	Warning	BIGIP_LIBHAL_AOM_ALERT_WARN	012a0022	FIPS initialization error deasserted	.1.3.6.1.4.1.3375.2.4.0.167

RAID Disk Failure messages

The RAID controller on a Dual-SSD system indicates that the member disk is removed from the RAID array.

Alert Message	Alert Level	Alert Name	Error code	Log message	Trap OID
RAID disk failure.	Critical	BIGIP_RAID_DISK_FAILURE	0x0	Raid#:md#: Disk failure	.1.3.6.1.4.1.3375.2.4.0.96

F5远程清除液晶屏和报警指示灯

Topic

You should consider using this procedure under the following condition:

You want to remotely clear LCD warnings and the Alarm LED.

Prerequisites

You must meet the following prerequisite to use this procedure:

You have command line access to the BIG-IP system.

Description

In some cases, you may want to remotely clear LCD warnings and the Alarm LED. Performing this action may prevent onsite personnel from discovering and reporting an old warning, or having to teach the onsite personnel how to clear the LCD. You can use the lcdwarn command line utility to control the LCD and the Alarm LED. To display its usage, run the lcdwarn command without any arguments.

Note: Starting in BIG-IP 12.1.0, you can use the tmsh show sys alert lcd command to display the list of alerts sent to the LCD front panel display.

Procedures

Clearing LCD warnings

Impact of procedure: Performing the following procedure should not have a negative impact on your system.You can use the lcdwarn command to remotely clear the LCD warnings. To do so, use the following command syntax appropriate for your BIG-IP version:

BIG-IP 12.1.5, BIG-IP 13.1.0 and later

lcdwarn -c <level>

In this command syntax, note the following:

<level> specifies the alert level to be cleared. Acceptable values include [0|1|2|3|4|5] or [warning|error|alert|critical|emergency|information]. The level can be seen under the “Priority” column when you run tmsh show sys alert from the Advanced Shell (bash) or show sys alert from within the tmsh shell:

root@(C3553740-bigip1)(cfg-sync Standalone)(Active)(/Common)(tmos)# show sys alert
—————————————————————-
Sys::LCDAlerts
Slot          Timestamp  Priority         Id         Description
—————————————————————-
0     04/22/21 02:21:07      info  0x10c0019 Unit going Active.
0     04/22/21 01:22:55      info  0x10c0019 Unit going Active.
0     04/22/21 01:16:28      info  0x10c0019 Unit going Active.
0     04/22/21 01:16:28      info  0x10c0019 Unit going Active.

For example, to clear LCD warnings with an alert level of 0, type the following command:

lcdwarn -c 0

BIG-IP 13.0.0, BIG-IP 12.1.4 and earlier

lcdwarn -c <level> <slotid>

In this command syntax, note the following:

<level> specifies the alert level to be cleared. Acceptable values include [0|1|2|3|4|5] or [warning|error|alert|critical|emergency|information].
<slotid> specifies the slot for which warnings should be cleared. Acceptable values include [0|1|2|3|4|5|6|7|8].Note: In BIG-IP 13.0.0, BIG-IP 12.1.4 and earlier, specifying any slot other than 0 is necessary only on the VIPRION platform. On a VIPRION platform, the slot ID is counted from 1, so blade 2 is 2 in the command line.

For example, to clear LCD warnings with an alert level of 0, type the following command:

lcdwarn -c 0 0

On a VIPRION system, to clear LCD warnings with an alert level of 0 for slot 2, type the following command:

lcdwarn -c 0 2

Clearing the Alarm LED

To clear the Alarm LED, you must clear all LCD warnings at all alert levels (on all slots for VIPRION systems). To do so, perform the following single command appropriate for your BIG-IP platform:

Impact of procedure: Performing the following procedure should not have a negative impact on your system.

VIPRION platforms

You can clear all LCD warnings at all alert levels on all VIPRION slots using the following single command:

BIG-IP 12.1.5, BIG-IP 13.1.0 and later

for i in 0 1 2 3 4 5; do lcdwarn -c “${i}”; done

BIG-IP 13.0.0, BIG-IP 12.1.4 and earlier

for i in 0 1 2 3 4 5; do for j in 1 2 3 4 5 6 7 8; do lcdwarn -c “${i}” “${j}”; done; done

If you run this command on a VIPRION system that has unpopulated blade slots, the system logs benign error messages to the /var/log/ltm file that appear similar to the following example:

012a0004:4: ledSet error: LopDev: sendLopCmd: Lopd status: 1 packet: action=2 obj_id=3c sub_obj=0 slot_id=2 result=2 len=0 crc=e071 payload= (error code:0x2)

You can safely ignore this message; it does not affect the traffic processing capability of the VIPRION system.

To prevent this error message on a VIPRION system with unpopulated blade slots, adjust the input values for the j variable. For example, on a VIPRION system where only blade slots 1 and 2 are populated, type the following command:

for i in 0 1 2 3 4 5; do for j in 1 2; do lcdwarn -c “${i}” “${j}”; done; done

BIG-IP iSeries platforms

For information about clearing the Alarm LED on BIG-IP iSeries platforms, refer to K24127736: Clear LCD alarms on iSeries device.

All other BIG-IP platforms (except for VIPRION platforms and BIG-IP iSeries platforms)

You can clear all LCD warnings at all alert levels on all other BIG-IP platforms (except for VIPRION platforms) using the following single command:

BIG-IP 12.1.5, BIG-IP 13.1.0 and later

for i in 0 1 2 3 4 5; do lcdwarn -c “${i}”; done

BIG-IP 13.0.0, BIG-IP 12.1.4 and earlier

for i in 0 1 2 3 4 5; do lcdwarn -c “${i}” 0; done

Note: Running this command on legacy BIG-IP platforms that are not equipped with an LCD (such as the 1000, 2400, 5100, and 5110) only clear the Alarm LED.

F5 液晶屏显示’unit going active’、’unit going standby’的含义

Topic

When a BIG-IP system transitions from active to standby state or standby to active state, a message is displayed on the LCD display indicating the amount of time that has passed since the state transition occurred.

The BIG-IP system transitioning from standby to active displays a message similar to one of the following examples:

xxS unit going active

xxM unit going active

The BIG-IP system transitioning from active to standby displays a message similar to one of the following examples:

xxS unit going standby

xxM unit going standby

In these messages, S or M indicates the units in which time is measured: S indicates seconds, M indicates minutes. The xx characters represent the number of minutes or seconds the device has been in the current state. For example, a device that transitioned to the active state 15 minutes ago displays the following message:

15M unit going active

Note: These messages may appear on standalone systems or on members of a redundant pair.

Note: To determine the cause of an unexpected state change, examine the system log files.

hdparm命令介绍

通常情况下可以使用fdisk、df等命令查看硬盘的分区情况以及当前已使用空间大小、剩余空间大小等信息。但是如果要查看硬盘的硬件信息如硬盘型号、序列号、已运行时间等信息该用什么工具查看呢？

在Linux下可以使用hdparm工具查看硬盘的相关信息或对硬盘进行测速、优化、修改硬盘相关参数设定。我主要常用这个工具来测试硬盘速度。

hdparm(hard disk parameters)
功能说明：显示与设定硬盘的参数。

该命令用来获取或设置SATA/IDE设备的参数。注意了，既可以获取也可以设置，所以使用这个命令的时候务必要小心。当然我们只是为了获取查看硬盘信息，故不会对硬盘及系统造成什么危害。通常会配合 -i 或 -I（大写的i）选项来取得硬盘的硬件信息。用法如下：

2.常用指令
a.安全擦除
hdparm –user-master u –security-set-pass PASSWORD /dev/sda 设置密码
hdparm –user-master u –security-erase PASSWORD /dev/sda 安全擦除
类似安全擦除：
hdparm –yes-i-know-what-i-am-doing –sanitize-crypto-scramle /dev/sdx
hdparm –yes-i-know-what-i-am-doing –sanitize-status /dev/sdx
b.设置硬盘容量(硬盘必须直连服务器，通过raid卡无法执行此项指令）
yum install sg3_utils.x86_64
sg_readcap /dev/sdb
hdparm -N VP –yes-i-know-what-i-am-doing /dev/sdb （VP： certain LBA blocks）
c.解锁
hdparm –user-master u –security-set-pass PASSWORD /dev/sda
hdparm –user-master u –security-unlock PASSWORD /dev/sda
hdparm –user-master u –security-disable PASSWORD /dev/sda
d.升级固件版本
hdparm –fwdownload /root/Desktop/xxx.bin –yes-i-know-what-i-am-doing –please-destroy-my-drive /dev/sdc
————————————————

hdparm -I /dev/sda

语法：hdparm [-CfghiIqtTvyYZ][-a <快取分区>][-A <0或1>][-c ][-d <0或1>][-k <0或1>][-K <0或1>][-m <分区数>][-n <0或1>][-p ][-P <分区数>][-r <0或1>][-S <时间>][-u <0或1>][-W <0或1>][-X <传输模式>][设备]

参数说明：

-a<快取分区> 设定读取文件时，预先存入块区的分区数，若不加上<快取分区>选项，则显示目前的设定。
-A<0或1> 启动或关闭读取文件时的快取功能。
-c 设定IDE32位I/O模式。
-C 检测IDE硬盘的电源管理模式。
-d<0或1> 设定磁盘的DMA模式。
-f 将内存缓冲区的数据写入硬盘，并清楚缓冲区。
-g 显示硬盘的磁轨，磁头，磁区等参数。
-h 显示帮助。
-i 显示硬盘的硬件规格信息，这些信息是在开机时由硬盘本身所提供。
-I 直接读取硬盘所提供的硬件规格信息。
-k<0或1> 重设硬盘时，保留-dmu参数的设定。
-K<0或1> 重设硬盘时，保留-APSWXZ参数的设定。
-m<磁区数> 设定硬盘多重分区存取的分区数。
-n<0或1> 忽略硬盘写入时所发生的错误。
-p 设定硬盘的PIO模式。
-P<磁区数> 设定硬盘内部快取的分区数。
-q 在执行后续的参数时，不在屏幕上显示任何信息。
-r<0或1> 设定硬盘的读写模式。
-S<时间> 设定硬盘进入省电模式前的等待时间。
-t 评估硬盘的读取效率。
-T 平谷硬盘快取的读取效率。
-u<0或1> 在硬盘存取时，允许其他中断要求同时执行。
-v 显示硬盘的相关设定。
-W<0或1> 设定硬盘的写入快取。
-X<传输模式> 设定硬盘的传输模式。
-y 使IDE硬盘进入省电模式。
-Y 使IDE硬盘进入睡眠模式。
-Z 关闭某些Seagate硬盘的自动省电功能。

hdparm常用参数使用举例：

1、显示硬盘的相关设置：
[root@oracle ~]# hdparm /dev/sda
/dev/sda:
IO_support = 0 (default 16-bit)
readonly = 0 (off)
readahead = 256 (on)
geometry = 19929［柱面数］/255［磁头数］/63［扇区数］, sectors = 320173056［总扇区数］, start = 0［起始扇区数］

2、显示硬盘的柱面、磁头、扇区数：
[root@oracle ~]# hdparm -g /dev/sda
/dev/sda:
geometry = 19929［柱面数］/255［磁头数］/63［扇区数］, sectors = 320173056［总扇区数］, start = 0［起始扇区数］

3、测试硬盘的读取速度：

[root@oracle ~]# hdparm -t /dev/xvda

/dev/xvda:
Timing buffered disk reads: 422 MB in 3.01 seconds = 140.20 MB/sec
[root@oracle ~]# hdparm -t /dev/xvda

/dev/xvda:
Timing buffered disk reads: 408 MB in 3.01 seconds = 135.59 MB/sec
[root@oracle ~]# hdparm -t /dev/xvda

/dev/xvda:
Timing buffered disk reads: 416 MB in 3.01 seconds = 138.24 MB/sec

4、测试硬盘缓存的读取速度：

[root@oracle ~]# hdparm -T /dev/xvda

/dev/xvda:
Timing cached reads: 11154 MB in 1.98 seconds = 5633.44 MB/sec
[root@oracle ~]# hdparm -T /dev/xvda

/dev/xvda:
Timing cached reads: 10064 MB in 1.98 seconds = 5077.92 MB/sec
[root@oracle ~]# hdparm -T /dev/xvda

/dev/xvda:
Timing cached reads: 10600 MB in 1.98 seconds = 5351.73 MB/sec

5、检测硬盘的电源管理模式：
[root@oracle ~]# hdparm -C /dev/sda
/dev/sda:
drive state is: standby [省电模式]

6、查询并设置硬盘多重扇区存取的扇区数，以增进硬盘的存取效率：
[root@oracle ~]# hdparm -m /dev/sda
[root@oracle ~]# hdparm -m 参数值为整数值如8 /dev/sda 附：硬盘坏道修复方法
检查：smartctl -l selftest /dev/sda
卸载：umount /dev/sda*
修复：badblocks /dev/sda

iostat命令介绍

引言：

一、概述

二、iostat用法

(1).用法：iostat [选项] [<时间间隔>] [<次数>]

(2). 命令参数：

(3).示例：

1.显示所有设备的负载情况

2.iostat -m 以M为单位显示所有信息

3.iostat -d sda 显示指定硬盘信息

4.iostat -t 报告每秒向终端读取和写入的字符数

5.iostat -d -k 1 1 查看TPS和吞吐量信息

6.iostat -d -x -k 1 1 查看设备使用率（%util）、响应时间（await）

7.iostat -c 1 2 查看cpu状态间隔1秒显示一次，总共显示2次

三、总结

引言：

iostat主要用于监控系统设备的IO负载情况，根据这个可以看出当前系统的写入量和读取量，CPU负载和磁盘负载。

一、概述

iostat 主要用于输出磁盘IO 和 CPU的统计信息。

iostat属于sysstat软件包。可以用yum install sysstat 直接安装。

二、iostat用法

(1).用法：iostat [选项] [<时间间隔>] [<次数>]

(2). 命令参数：

-c：显示CPU使用情况

-d：显示磁盘使用情况

-N：显示磁盘阵列(LVM) 信息

-n：显示NFS 使用情况

-k：以 KB 为单位显示

-m：以 M 为单位显示

-t：报告每秒向终端读取和写入的字符数和CPU的信息

-V：显示版本信息

-x：显示详细信息

-p： [磁盘] 显示磁盘和分区的情况

(3).示例：

1.显示所有设备的负载情况

cpu属性值说明：

%user：CPU处在用户模式下的时间百分比。

%nice：CPU处在带NICE值的用户模式下的时间百分比。

%system：CPU处在系统模式下的时间百分比。

%iowait：CPU等待输入输出完成时间的百分比。

%steal：管理程序维护另一个虚拟处理器时，虚拟CPU的无意识等待时间百分比。

%idle：CPU空闲时间百分比。

备注：

如果%iowait的值过高，表示硬盘存在I/O瓶颈，%idle值高，表示CPU较空闲，如果%idle值高但系统响应慢时，有可能是CPU等待分配内存，此时应加大内存容量。%idle值如果持续低于10，那么系统的CPU处理能力相对较低，表明系统中最需要解决的资源是CPU。

disk属性值说明：

磁盘名称

device:磁盘名称

tps:每秒钟发送到的I/O请求数.

Blk_read/s:每秒读取的block数.

Blk_wrtn/s:每秒写入的block数.

Blk_read:读入的block总数.

Blk_wrtn:写入的block总数.

2.iostat -m 以M为单位显示所有信息

3.iostat -d sda 显示指定硬盘信息

4.iostat -t 报告每秒向终端读取和写入的字符数

5.iostat -d -k 1 1 查看TPS和吞吐量信息

6.iostat -d -x -k 1 1 查看设备使用率（%util）、响应时间（await）

说明：

tps:每秒钟发送到的I/O请求数。

Blk_read/s:每秒读取的block数。

Blk_wrtn/s:每秒写入的block数。

Blk_read:读入的block总数。

Blk_wrtn:写入的block总数。

rrqm/s: 每秒进行 merge 的读操作数目。即 rmerge/s

wrqm/s: 每秒进行 merge 的写操作数目。即 wmerge/s

r/s: 每秒完成的读 I/O 设备次数。即 rio/s

w/s: 每秒完成的写 I/O 设备次数。即 wio/s

rkB/s: 每秒读K字节数。是 rsect/s 的一半，因为每扇区大小为512字节。

wkB/s: 每秒写K字节数。是 wsect/s 的一半。

avgrq-sz: 平均每次设备I/O操作的数据大小 (扇区)。

avgqu-sz: 平均I/O队列长度。

rsec/s: 每秒读扇区数。即 rsect/s

wsec/s: 每秒写扇区数。即 wsect/s

r_await:每个读操作平均所需的时间

不仅包括硬盘设备读操作的时间，还包括了在kernel队列中等待的时间。

w_await:每个写操作平均所需的时间

不仅包括硬盘设备写操作的时间，还包括了在kernel队列中等待的时间。

await: 平均每次设备I/O操作的等待时间 (毫秒)。

svctm: 平均每次设备I/O操作的服务时间 (毫秒)。

%util: 一秒中有百分之多少的时间用于 I/O 操作，即被io消耗的cpu百分比

备注：

如果 %util 接近 100%，说明产生的I/O请求太多，

I/O系统已经满负荷，该磁盘可能存在瓶颈。如果 svctm

比较接近 await，说明 I/O 几乎没有等待时间；如果 await

远大于 svctm，说明I/O 队列太长，io响应太慢，则需要进行必要优化。

如果avgqu-sz比较大，也表示有当量io在等待。

7.iostat -c 1 2 查看cpu状态间隔1秒显示一次，总共显示2次

iostat 1 5

间隔1秒，总共显示5次

iostat -d 2

每隔2秒,显示一次设备统计信息.

iostat -d 2 3

每隔2秒,显示一次设备统计信息.总共输出3次.

iostat -x sda sdb 2 3

每隔2秒显示一次sda, sdb两个设备的扩展统计信息,共输出3次.

iostat -p sda 2 3

每隔2秒显示一次sda及上面所有分区的统计信息,共输出3次.

三、总结

iostat是Linux中被用来监控系统的I/O设备活动情况的工具，是input/output statistics的缩写。它可以生成三种类型的报告：

CPU利用率报告

设备利用率报告

网络文件系统报告

iostat通过生成的报告来帮助管理员更好的调整系统设置来平衡各个物理磁盘之间的I/O负载。

centos7挂载本地yum源的方法

centos7虚拟机环境，光驱连接centos安装镜像。

一、创建目录挂载镜像
mkdir -p /mnt/cdrom
mount /dev/sr0 /mnt/cdrom/

二、创建本地yum源
在/etc/yum.repos.d目录下创建local.repo文件，写入内容：
[local]
name=local repo
baseurl=file:///mnt/cdrom
enable=1
gpgckeck=1
gpgkey=file:///mnt/cdrom/RPM-GPG-KEY-CentOS-7

三、将/etc/yum.repos.d目录下其余文件删除，或者移动到别的目录，使得/etc/yum.repos.d目录下只有local.repo

四、更新yum源
yum clean all
yum makecache
yum repolist all
步骤完成，可以使用yum命令安装了。

HPE 1500 iLO 4 Configuration is temporarily unavailable

Set the ilo 4 Security Override Switch on the system board to the ON position. The location of the switch is printed on a label located on the inside of the server blade hood cover. On the same maintenance switch, set switch number 3 to the ON position.
Download the Smart Update Firmware Maintenance DVD version 9.10 B and later.
Create a bootable USB key containing the contents of the Smart Update Firmware DVD.
Download the desired version of the ilo 4 firmware smart component for Linux (cp0xxxxx.scexe where xxxxx is an appropriate 5 digit number).
Copy the downloaded ilo 4 firmware to the directory /hp/swpackages on the USB key.
Put the server blade back into the enclosure and power the server blade ON. Boot to the USB key containing the Smart Update Firmware DVD and select interactive firmware update.
Use the following key sequence to exit out of the Smart Update Firmware Maintenance DVD interface. A command prompt will be displayed:
CTRL + ALT + d + b + x (Keep the CTRL and ALT keys pressed when typing d b x. )
The command prompt takes approximately 30 seconds to be displayed.
At the command prompt, navigate to the Smart Update Firmware DVD directory containing the supplemental ilo 4 firmware update by using the following command:
bash-3.1# cd /mnt/cdrom/hp/swpackages
Use the following command to unload the HPILO module:
rmmod hpilo
Use the following command to execute the ilo 4 firmware update in direct mode:
sh cp0xxxxx.scexe – – direct (This parameter requires two dash ( ) characters.).
After the ilo 4 firmware upgrade is completed, power the server blade OFF and set the ilo 4 Security Override Switch on the system board to the OFF position. On the same maintenance switch, set switch number 3 to the OFF position

HPE Integrated Lights-Out 4 (iLO 4) – HPE Active Health System (AHS) Logs and HPE OneView Profiles May Be Unavailable Causing iLO Self-Test Error 8192, Embedded Media Manager and Other Errors

On an HPE Gen8-series or HPE ProLiant Gen9-series servers with HPE Integrated Lights-Out 4 (iLO 4), the NAND flash device may not initialize or mount properly, which may cause a variety of symptoms, which are listed below:

AHS Errors

AHS Logs display a blank date when performing the following:
“Select a range of the Active Health log in days From: _______ To:_________”
AHS file system mount may fail with (I/O Error) or (No Such Device).
The HP Active Health System (AHS) Logs are unable to be downloaded. AHS Data is not available due to a filesystem error.
The iLO Diagnostic tab will display the following error message: “Embedded media manager failed initialization” or “The AHS file system mount failed with (No such device) or “The AHS file system mount failed with (I/O error)” or “Controller firmware revision 2.09.00 Could not partition embedded media device.”
Unable to download AHS log and “bb_dl_disabled” is displayed.

Embedded Media Errors

Controller Firmware Version 2.09.00 may fail to restart.
Unable to partition Embedded media device.
Embedded media manager may fail initialization.

Intelligent Provisioning Errors

Intelligent Provisioning will not execute when selecting F10.

OneView Errors

Remote Insight/Integrated Lights-Out Self-Test Error 8192.
Unable to register this HP OneView instance with iLO: There was a problem with posting a command to the iLO.
Unable to register this HP OneView instance with iLO: The iLO initialization was unable to complete.
Unable to determine if this server hardware is being managed by another management system. Received an error from iLO <ip address of iLO> with Error: Blob Store is not yet initialized. and Status: 126

SCOPE

Any HPE ProLiant Gen8-series or HPE ProLiant Gen9-series server running iLO 4.

RESOLUTION

The resolution for this issue may take several steps that need to be completed in the order specified below.

OVERVIEW

Step 1) Upgrade the iLO 4 firmware to version 2.61
Step 2) Perform a NAND format
Step 3) Check the iLO status If the iLO status is normal, then skip to Step 6 If the iLO status is still degraded, continue to Step 4
Step 4) Schedule downtime; AC power-cycle and repeat the NAND format
Step 5) Check the iLO status If the iLO status is normal, continue to
Step 6 If the iLO status is still degraded, then skip to Step 7
Step 6) Perform these final steps if the system board does not need to be replaced: Reboot the server; reinstall IP; and refresh the server in OneView (if server is managed by OneView)
Step 7) If steps 1-4 did not resolve the degraded iLO, replace the system board.

Note: The 2.61 iLO 4 firmware is a critical update. As such, HPE requires users to update to this version immediately. Install this update to take advantage of significant improvements to the write algorithm for the embedded 4 GB non-volatile flash memory (also known as the NAND). These improvements increase the NAND lifespan.

Additional considerations before performing a NAND format

A NAND format can be performed while the server is online in most cases.

Exception: For ESXi hosts booting from the Embedded SD Card – it isstrongly recommended to perform the NAND format with the ESXi OSshutdown.This recommendation also applies when updating the iLO 4firmware or resetting the iLO for ESXi hosts booting from the Embedded SDCard

A server AC power removal may be required (prior to the NAND format) in order for the NAND format to be successful. This can be accomplished for ML or DL servers by shutting down the server and disconnecting the power cables for a few seconds. For blade servers, an E-fuse can be accomplished by logging into the OA CLI and typing “reset server #” where “#” is the bay number of the blade.

An AuxPwrCycle feature was added in iLO 4 firmware version 2.55 so that the equivalent of an AC power removal can be performed remotely on a server.
Refer to Customer Notice “HPE Integrated Lights Out (iLO) 4 – RESTful Command to Allow an Auxiliary Power-Cycle Is Available in Firmware Version 2.55 (and Later)” located at the following URL:

https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-a00047494en_us

Required steps to perform after a successful NAND format

If the server is powered on when performing the NAND format, a server reboot is required after a successful NAND format. This reboot repopulates the RIS and RESTful data on the server during the next server POST with iLO 4 firmware version 2.53 or newer installed. If iLO 4 firmware version 2.50 or older is installed, connect to the iLO CLI (using putty) and use the following command “oemHP_clearRESTapistate” before rebooting the server to repopulate the RIS and RESTful data on the server.
Reinstall Intelligent Provisioning (IP) to ensure that it is working properly and reported as installed in the iLO 4 GUI “System Information – Firmware Information” page.
If using OneView: After the server is rebooted (to repopulate the RIS and RESTful data on the server), perform a server refresh in OneView; any existing errors in OneView will need to be marked as cleared after the refresh.

Detailed steps

Step 1. Upgrade the iLO 4 firmware to version 2.61. To download the firmware. The latest version of the iLO 4 firmware is available as follows:

Note: For ESXi hosts booting from the Embedded SD Card (Gen8/Gen9)- it is strongly recommended to perform this step with the ESXi OS shutdown.

Click the following link:
https://support.hpe.com/hpesc/public/home
Enter a product name (e.g., “DL380 Gen9”) in the text search field and wait for a list of products to populate. From the products displayed, identify the desired product and click on the Drivers & software icon to the right of the product.
From the Drivers & software dropdown menus on the left side of the page:
- Select the Software Type – (e.g. Firmware)
- Select the Software Sub Type – (e.g. Network)
- For further filtering if needed – Select the specific Operating System from the Operating Environment.
Select the latest release of Firmware – Lights-Out Management iLO 4 firmware version 2.61 (or later). Note: To ensure the latest version will be downloaded, click on the Revision History tab to check if a new version of the firmware/driver is available.
Click Download.

Step 2. Perform NAND format using one of the methods detailed in the customer advisory: using one of the methods detailed in the customer advisory: HPE Integrated Lights-Out 4 (iLO 4) – How to Format the NAND Used to Store AHS logs, OneView Profiles, and Intelligent Provisioning
Find the document at:

https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-a00048622en_us

Note: iLO 4 firmware version 2.44 (or later) is required to format the NAND using the following steps.

Step 3. Check the iLO status using the options available in the customer advisory: HPE Integrated Lights-Out 4 (iLO 4) – How to Format the NAND Used to Store AHS logs, OneView Profiles, and Intelligent Provisioning (find the document at the link in Step 2).

If the iLO status is normal based on the above criteria, then skip to Step 6 If the iLO status is still degraded, continue to Step 4

Step 4. If the iLO status is still degraded, perform the following steps:

a) Schedule a maintenance window
b) Shut down the server
c) Perform an E-fuse (server blade) or AC Power Pull (DL / ML series servers) d) Perform the NAND format again (refer to the instructions in Step 2 above)

Step 5. Check the iLO status (refer to the instructions in Step 3 above)
If the iLO status is normal, continue to Step 6
If the iLO status is still degraded, then skip to Step 7

Step 6. Perform these final steps after the NAND format is successful:
a) Reboot the server
b) Reinstall Intelligent Provisioning (see additional details below)
c) If the server is managed by HPE OneView, perform a server refresh to bring the server back under management.

Note: Any existing errors in OneView will need to be marked as cleared after the refresh.

Step 7. If steps 1-4 did not resolve the degraded iLO, contact HPE support to arrange a system board replacement. Follow these steps to complete the remediation:

a) Open an HPE support case to arrange for a replacement system board / arrange a maintenance window
b) During the maintenance window, shutdown the server
c) Unassign the OneView profile (if the server is under OneView management)
d) Replace the system board
e) Enter Server Model and Serial Number via RBSU
f) Update to iLO 4 firmware 2.61 (or later)
g) Check iLO status (refer to Step 3 above)
h) Reassign the OneView profile (if the server is under OneView management)

Note: Any existing errors in OneView will need to be marked as cleared after the OneView profile is applied.

If additional assistance is needed, contact HPE support and reference Advisory a00019495en_us as follows:

Click on the following URL to locate the HPE Customer Support phone number in your country:

https://h20195.www2.hpe.com/v2/Getdocument.aspx?docname=A00039121ENW .

OneView Considerations

OneView relies on the NAND to be accessible to perform many operations such as adding a new server or applying a profile. Because of this there are several things that need to be understood when dealing with this issue in an OneView environment.

OneView interaction with the NAND – OneView uses a portion of the NAND called the iLO blob store. If the blob store is not accessible, adding a server to OneView or assigning a profile to the server cannot be completed.
Impact of an inaccessible NAND when managed by OneView – If the NAND becomes inaccessible after the server was added in OneView, there are several things that can cause an outage:
1. E-fuse – If an E-fuse reset is performed, the server will not be able to be brought back under management until the NAND issue is remediated and this will cause an unexpected outage.
2. Server is removed and reinserted – This would essentially be the same as performing an E-fuse, so the same information in the E-fuse section applies here.
3. Un-assign a server profile – If a server profile is unassigned, the server profile will not be able to be reassigned until the NAND issue is remediated.
Migration from Virtual Connect to OneView – If an enclosure is migrated from Virtual Connect and a server has an inaccessible NAND, that server will not be able to be added properly until the NAND issue is remediated. If an in-service migration is attempted on a server with an inaccessible NAND, an unexpected outage will occur. Reference advisoryhttps://support.hpe.com/hpsc/doc/public/display?docId=emr_na-c05384185for more detail.
Formatting of the NAND – The NAND format wipes the iLO blob store that is used by OneView. It is important to issue a server refresh after a successful NAND format to recover the blob that OneView uses.

Reinstalling Intelligent Provisioning Considerations

There are online and offline options for reinstalling Intelligent Provisioning

Intelligent Provisioning can be reinstalled online if the server is running Windows (HPE ProLiant Gen8-series servers/HPE ProLiant Gen9-series servers only) or Linux (HPE ProLiant Gen8-series servers/HPE ProLiant Gen9-series servers). The available online Windows packages are older versions of Intelligent Provisioning.
Intelligent Provisioning must be reinstalled offline if the server is running VMware (HPE ProLiant Gen8-series servers/HPE ProLiant Gen9-series servers).
The Intelligent Provisioning software download links are provided below. Intelligent Provisioning versions
HPE ProLiant Gen8-series servers are only supported with Intelligent Provisioning 1.x
” HPE ProLiant Gen9-series servers are only supported with Intelligent Provisioning 2.x

Note: Intelligent Provisioning version 3.x is for HPE ProLiant Gen10-series servers only.

Offline method considerations:

Run the Intelligent Provisioning Restore Media to restore the Intelligent Provisioning data. Instructions to restore Intelligent Provisioning are as follows:
Instructions to create a bootable DVD with the IP image are provided on the HPE Intelligent Provisioning recovery media download site under the Installation Instructions tab.

Reference:
https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_170e0ac3cb6e4439bf313f137a#tab3

Note: The DVD can be used multiple times.

Please note the Intelligent Provisioning Recovery Media DVD may be remotely mounted using HPE Integrated Lights-Out 4 (iLO 4) Virtual Media functionality, in order to reinstall Intelligent Provisioning. Additional information is available in the HPE iLO 4 User Guide at the following URL regarding how to mount an ISO image (federated or un-federated) and perform basic virtual media operations. Reference Pages 189 and 223-237:

HPE iLO 4 User Guide:
https://support.hpe.com/hpsc/doc/public/display?docId=emr_na-c03334051-16

In addition, it is possible to write a script that utilizes HPE Integrated Lights-Out 4 (iLO 4) to reinstall Intelligent Provisioning on multiple servers. HPE Lights-Out management processors support an advanced scripting interface for group configuration and server actions. Scripts would need to be customized for the specific environment and task. Sample scripts are available for customers to reference at the following URL: HPE Lights-Out XML Scripting Sample for Windows: https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_459b8adc29c04317ad1d6a6752

Intelligent Provisioning software download links

HPE ProLiant Gen8-series servers are only supported with Intelligent Provisioning 1.x
HPE ProLiant Gen9-series servers are only supported with Intelligent Provisioning 2.x

a.) Linux RPM located at the following URLs: For HPE Gen8-series servers: http://downloads.linux.hpe.com/SDR/repo/ip/rhel/current/x86_64/gen8/

For HPE Gen9-series servers: http://downloads.linux.hpe.com/SDR/repo/ip/rhel/current/x86_64/gen9/firmware-intelligentprovisioning-ip-2.71-1.1.x86_64.rpm

b.) The Windows files are located at the following URLs:

For HPE ProLiant Gen8-series servers: Intelligent Provisioning for Microsoft Windows 64-bit Operating Systems 1.64.0.0 (2 Mar 2017) cp031302.exe https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_01b900175afc4ae68be6c87b39
For HPE ProLiant Gen9-series servers: Intelligent Provisioning for Windows x64 2.50.0.0(6 Jan 2017) cp031091.exe https://support.hpe.com/hpsc/swd/public/detail?swItemId=MTX_b031497eafd94b9ea2156cdef6

c) ISO images (e.g. for VMware) on Hewlett Packard Enterprise Support Center located at the following: https://support.hpe.com/hpesc/public/home/driverHome?sp4ts.oid=1008862660

For HPE ProLiant Gen8-series servers: Intelligent Provisioning Recovery Media, version 1.70(9 Oct 2017)
For HPE ProLiant Gen9-series servers: Intelligent Provisioning for Gen9 Servers, version 2.70(b)(28 Feb 2018)

HP DL380G9服务器开机无F10的处理办法

HP服务器的F10功能是一个安装在主板NAND空间上的独立的linux内核系统，主要功能是系统快速安装和服务器维护诊断。

如果开机自检时没有出现F10按钮的提示，有三种可能。

一、主板BIOS设置问题，可以通过恢复BIOS默认设置或者单独设置F10。
按F9进入主板BIOS设置，System Utilities——System Configuration——menucascade-separatorBIOS/
Platform Configuration (RBSU)（BIOS/平台配置）——menucascade-separatorServer Security（服务器安
全性）——menucascade-separatorIntelligent Provisioning (F10 Prompt)（F10 提示），选择Enabled启用，按F10保存即可。

二、F10系统崩溃，此时需要在HP官网下载F10镜像进行重新安装。
需要进入ilo integrated remote console控制台界面，挂载镜像，然后从F11从镜像启动。