Using the Oracle ILOM Diag Shell

At the Oracle ILOM CLI prompt, you can run HWdiag commands to check the status of a system and its components, and access HWdiag logs. Depending on the server, see Run HWdiag (CLI) to use one of the methods to access the Diag shell.

This section contains the following topics:

For information about Oracle ILOM, refer to the Oracle ILOM Documentation Library at Systems Management Documentation.

Run HWdiag (CLI)

  1. Depending on your Oracle server, do one of the following to access a diag shell:
    • For Oracle X4, X5, and X6 servers, at the Oracle ILOM CLI prompt, access the Restricted shell:

      set SESSION mode=restricted

    • For Oracle X7, X8, and X9 servers, access /SP/diag/shell:

      -> start /SP/diag/shell
      Are you sure you want to start /SP/diag/shell (y/n)? y
                                    
  2. At the diag> prompt, type diag> hwdiag parameters

    HWdiag commands use the form hwdiag main_command [subcommand ...].

    For details, see HWdiag Commands and Options.

  3. Choose any of the following ways to display help for HWdiag options and commands:
    • To display an overview of HWdiag options and main commands, type help hwdiag.

    • To display all HWdiag main commands and their subcommands, type help hwdiag -h.

    • To display the command structure with all options available by command, type help hwdiag -h -v.

    • To display help specific to a main-command, type help hwdiag -h main command.

  4. In the SP/Diag shell, type any of the following commands:
    • To display the external commands available in the SP/Diag shell, type help.

    • To display information, type echo, for example, echo $?.

    • To exit the SP/Diag shell, type exit.

    • To run hardware diagnostics, type hwdiag.

    • To list the diagnostics log directories and files, type ls.

    • To print the content of the diagnostics log files, type cat.

Using HWdiag Commands

This section describes how to use the HWdiag command line.

HWdiag commands use the form hwdiag main-command [subcommand ...].

For example:

[(flash)root@ORACLESP-1234567:~]# hwdiag led get /SYS/MB/P1/D8/SERVICE
HWdiag - Build Number 117435 (May 01 2019, 17:05:36)
         Current Date/Time: January 04 2019, 10:52:31
  LED                             VALUE
  ------------------------------------------
  /SYS/MB/P1/D8/SERVICE         : off

Many subcommands require an additional subcommand to identify the actual device or devices being acted on. This target can be an individual device or all, which represents all the valid targets for the subcommand. In the above example, hwdiag led get /SYS/MB/P1/D8/SERVICE returns the state of a single LED. If you enter hwdiag led get all, it displays the state of all the system LEDs.

Note:

Main commands and subcommands are case insensitive. However, hwdiag is not. For example, hwdiag led get all is the same as hwdiag LED GET ALL.

If you enter an incomplete command line, the HWdiag utility displays the syntax for the command and a list of valid subcommands. Use this information to re-enter the command with a complete and valid set of parameters and subcommands.

The following display shows an example.

diag> hwdiag temp
HWdiag - Build Number 81018 (January 12 2019 at 00:42:14)

Syntax: hwdiag temp ...

   get [all|<sensor>]
              - Display Temperature Sensor Reading
   info [all|<sensor>]
              - Display Temperature Sensor Information

diag> hwdiag temp get
HWdiag - Build Number 81018 (January 12 2019 at 00:42:14)

Syntax: hwdiag temp get all|<sensor>

Valid Options for Temperature :
  ALL                             /SYS/MB/T_IN_PS
  /SYS/MB/P0                      /SYS/MB/T_IN_ZONE1
  /SYS/MB/P1                      /SYS/MB/T_IN_ZONE2
  /SYS/MB/P2                      /SYS/MB/T_IN_ZONE3
  /SYS/MB/P3                      /SYS/MB/T_IN_ZONE4
  /SYS/MB/T_CORE_NET01            /SYS/MB/T_OUT_SLOT1
  /SYS/MB/T_CORE_NET23            /SYS/T_FRONT

diag> hwdiag temp get all
HWdiag - Build Number 81018 (January 12 2019 at 00:42:14)
  DEVICE                          TEMP
  -------------------------------------------
  /SYS/MB/P0                    : 64.00 margin
  /SYS/MB/P1                    : 64.00 margin
  /SYS/MB/P2                    : 63.00 margin
  /SYS/MB/P3                    : 64.00 margin
  /SYS/MB/T_CORE_NET01          : 38.75 deg C
  /SYS/MB/T_CORE_NET23          : 38.00 deg C
  /SYS/MB/T_IN_PS               : 26.75 deg C
  /SYS/MB/T_IN_ZONE1            : 30.75 deg C
  /SYS/MB/T_IN_ZONE2            : 30.75 deg C
  /SYS/MB/T_IN_ZONE3            : 29.50 deg C
  /SYS/MB/T_IN_ZONE4            : 28.25 deg C
  /SYS/MB/T_OUT_SLOT1           : 29.75 deg C
  /SYS/T_FRONT                  : 24.50 deg C

HWdiag Commands and Options

The following tables list the HWdiag utility commands and options.

Note:

Not all commands are available on all platforms. To find out which commands are available on your system, enter hwdiag -h.

HWdiag Commands

Component Action Options Description and Options

cpld

 

CPLD, FPGA tests, and utilities.

reg

all|fgpa

Dump CPLD registers.

vr_check

 

Print voltage regulator status.

log read number_of_ last_entries

 

Read the last three entries from the cpld log file.

mbus

 

(X9-8 only) Check MBUS status.

cpu

   

Display CPU information.

capid all|cpu

 

Decode CAPID values for CPU.

info all|cpu

-v

Dump CPU devices and display coding for all registers.

pirom_info all|cpu

 

Dump PIROM CPU information.

fan

   

Fan test and utilities.

get

-m

Display fan RPM.

info

-r

Display fan presence information.

gpio

   

GPIO utilities.

get all|gpio_pin

Get information about GPIO pin.

i2c

   

Test the sideband i2c topology.

scan all|bus

 

Display accessible i2c devices.

test all|bus

 

Test connectivity of platform i2c devices. This test returns a pass or fail.

io

   

IO tests and utilities.

nvme_info

-v

Display information from VPD and MI for NVMe devices.

nvme_test

-v

Check for PCIe link width and speed of NVMe drives and NVMe add-in-cards.

led

   

Get information about LEDs.

get all|led

 

Display the state of LEDs.

info all|led

 

Display information about LED registers.

mem

   

Display memory (DIMM) information.

info all|dimm_name

 

Display memory configuration.

spd all|dimm_name

-r

Display DIMM SPD information, such as size, speed, and voltage. The information displayed varies according to manufacturer.

pci

   

PCIe tests and utilities.

dump

 

Read PCIe registers.

dump <socket> <bus> <dev> <func> [std|ext]|[<offset> <count>]

  • std reads the entire space

  • ext reads the extended space

  • <offset><count> specifies a single register

info all|device

-r

Display PCIe link information for all, or for a single device.

read

 

Read the specified PCIe register.

read <socket> <bus> <dev> <func> <offset>

scan

 

Scan all PCIe devices.

status all|pci-target(s)

 

Print status of pci-target(s)

power

   

Display power information.

get

 

Display sensor readings.

get amps|volts|watts all|sensor.

info all|sensor

 

Display information about voltage reduction devices (VRD).

sensor identifies an individual sensor.

system

   

Display system information.

summary

 

Display system summary.

fabric test all| cpu

 

Test the system fabric, including QPI bus speed, PCIe link speed, and memory frequency.

info

 

Display system configuration information.

port80 number_of_last_codes

-m

Display host boot progress. Optionally, enter the number of last codes to show the last codes for port80.

rtc

 

Display the real time clock (RTC).

thermal

-m, –r

Display system thermal information, including temperatures, fan speeds, and power.

version

 

Display the version of system components.

temp

   

Display temperatures.

get all|sensor

 

Display temperature sensor readings.

info all|sensor

 

Display information about temperature sensors.

HWdiag Command Options

The following table lists the HWdiag command options.

Option Long Description

– f

force

Force execution of a command regardless of prerequisites.*

– h

help

Display help test.

– i

interactive

Prompts when you use it with a main command.*

– l

log <filename>

Enable HWdiag to start logging to <filename>.

Note: Use -t to add time stamp to logging.

– m

monitor <.1 sec>

Set monitoring interval in increments of tenths of a second (.1 second). Overrides current monitoring interval.*

– n

numberloop

Set the number of loops to run a command continually.

– p

persist

Use persistent hardware presence data for all invocations of any HWdiag command.

– q

quiterr

Exit HWdiag utility immediately after an error occurs.

– r

raw

Modify HWdiag output for easier parsing.*

– s

sampleint

Sample interval in seconds to set the number of seconds of sleep between collecting data samples. Use only with -n option.

– t

timestamp

Add time stamp to logging. Use with -l option.

– u

unit [ 1,2,4 ]

Control the output format of dump subcommands. Options are 1 byte (default), 2 byte, or 4 byte format.*

– v

verbose

Enhance the verbosity of output.*

– x

exclude

Exclude a feature.*

Note:

*Implement the HWdiag command option only on subcommands. For details on the syntax, refer to the subcommand documentation or display the help by typing help hwdiag -hv.

HWdiag Logs

In addition to running the HWdiag utility commands, you can access and display the content of the HWdiag log files. The commands ls and cat are available in the Oracle ILOM SP/Diag shell.

To read the content of the logs use the cat command. For example, type cat hwdiag/hwdiag_i2c_test.log.