background image

Appendix

 

A

 

BMC

 

Driver

 

Messages

This

 

appendix

 

explains

 

the

 

messages

 

that

 

are

 

output

 

to

 

the

 

operating

 

system

 

when

 

defects,

 

etc.

 

are

 

detected

 

in

 

the

 

BMC

 

driver.

Error

 

Messages

[ERR.]

 

xos

 

BMC

 

0001

 

-

 

internal

 

information

 

Invalid

 

parameter

 

and

 

command.

 

[CODE]

Meaning

The

 

specified

 

data

 

is

 

invalid

 

as

 

a

 

command

 

and

 

parameter.

CODE

:

 

IPMI

 

command

 

code

Action

Collect

 

investigation

 

data

 

according

 

to

 

"

5.3.1

  

Collecting

 

Information

 

for

 

Maintenance

 

Purposes

,"

 

and

 

then

 

contact

 

Fujitsu

 

Support

 

with

 

the

 

collected

 

data

 

together

 

with

 

the

 

output

 

message.

[ERR.]

 

xos

 

BMC

 

0002

 

-

 

internal

 

information

 

Invalid

 

parameter.

 

File

 

Name

 

Size:[NAME]

 

Data

 

Size:[DATA]

 

Name:[PTR1]

 

Data:[PTR2]

Meaning

A

 

specified

 

value

 

is

 

invalid

 

as

 

an

 

emergency

 

dump

 

request

 

parameter.

NAME

:

 

Filename

 

size

DATA

:

 

Data

 

size

PTR1

:

 

Pointer

 

to

 

the

 

filename

 

storage

 

area

PTR2

:

 

Pointer

 

to

 

the

 

data

 

storage

 

area

Action

Collect

 

investigation

 

data

 

according

 

to

 

"

5.3.1

  

Collecting

 

Information

 

for

 

Maintenance

 

Purposes

"

 

and

 

then

 

contact

 

Fujitsu

 

Support

 

with

 

the

 

collected

 

data

 

together

 

with

 

the

 

output

 

message.

[ERR.]

 

xos

 

BMC

 

0003

 

-

 

internal

 

information

 

Invalid

 

parameter.

 

[TYPE/(NULL)]

Meaning

The

 

specified

 

value

 

is

 

invalid

 

as

 

an

 

input

 

parameter.

TYPE

:

 

Status

 

code

Action

Collect

 

investigation

 

data

 

according

 

to

 

"

5.3.1

  

Collecting

 

Information

 

for

 

Maintenance

 

Purposes

,"

 

and

 

then

 

contact

 

Fujitsu

 

Support

 

with

 

the

 

collected

 

data

 

together

 

with

 

the

 

output

 

message.

[ERR.]

 

xos

 

BMC

 

0004

 

-

 

internal

 

information

 

Invalid

 

parameter.

 

Info:[PTR]

 

Size:[SIZE]

Meaning

A

 

specified

 

value

 

is

 

invalid

 

as

 

a

 

status

 

control

 

request

 

parameter.

PTR

:

 

Pointer

 

to

 

the

 

error

 

log

 

storage

 

area

SIZE

:

 

Error

 

log

 

size

Action

Collect

 

investigation

 

data

 

according

 

to

 

"

5.3.1

  

Collecting

 

Information

 

for

 

Maintenance

 

Purposes

"

 

and

 

then

 

contact

 

Fujitsu

 

Support

 

with

 

the

 

collected

 

data

 

together

 

with

 

the

 

output

 

message.

Appendix

 

A

 

BMC

 

Driver

 

Messages

C120-0089-03EN

65

Summary of Contents for PRIMEHPC FX1000

Page 1: ...FUJI T SUSuper co mpute r PRI M EHPC FX7 00Oper at i ng Manual FUJITSU Supercomputer PRIMEHPC FX700 Operating Manual C120 0089 03EN ...

Page 2: ...er provides an overview and information on the FX700 main unit Chapter 2 Important Information This chapter contains important information for using the FX700 correctly and safely Chapter 3 Starting Up This chapter describes the steps from installation to startup of the FX700 main unit Chapter 4 Operation This chapter describes operation of the FX700 main unit Chapter 5 Collecting Information When...

Page 3: ...ter 3 Chapter 5 Chapter 6 Updated 1 1 Overview of the FX700 Main Unit Updated 3 17 1 OS Installation Procedure and 3 17 2 OS Driver Installation Procedure Added 5 4 5 Precaution on Using the Web GUI Updated 6 1 FX700 Main Unit Specifications Revision History 1 The numbers titles of the chapters sections to which changes are made are those used in the latest version However the numbers titles of th...

Page 4: ... to keep this manual in a safe and convenient location for quick reference Fujitsu makes every effort to prevent injury to users and bystanders as well as property damage Be sure to use the product in accordance with the instructions in the manual Notes on This Product This product is designed and manufactured for use in standard applications such as office work personal devices and general indust...

Page 5: ...xchange and Foreign Trade Control Law of Japan North American Free Trade Agreement NAFTA FIPS 140 Federal Information Processing Standardization 140 U S federal standards that specify security requirements for cryptography modules NIST SP800 171 U S security standards Trade Adjustment Assistance TAA Safety Radio and Harmonics Europe Certified Standard Standard Number Safety Radio Harmonics EN IEC ...

Page 6: ...EEC Waste Electrical and Electronic Equipment Directive WEEE Directive European Parliament and Council Directive 94 62 EC of 20 December 1994 on packaging and packaging waste Export Related Europe Standard Number IATA Dangerous Goods Regulations 58th Edition 2017 Regulations on transport of lithium lithium ion batteries and electric double layer capacitors Foreign Exchange and Foreign Trade Contro...

Page 7: ...ansport of lithium lithium ion batteries and electric double layer capacitors Fundamental notices of customs law Safety Radio and Harmonics South Korea Certified Standard Standard Number Safety Radio Harmonics KCC K 60950 1 2 0 2011 12 PSU only KN32 Class A KN35 KN61000 4 2 3 4 5 6 8 11 Recycling and Disposal South Korea Standard Number Energy Saving Environmental Substances Recycling Display rule...

Page 8: ...ich case the user may be required to correct the interference at the user s own expense VCCI A The system complies with the requirements of European regulations This product is a Class A product Operation of this product in a residential area may cause radio frequency interference in which case the user will be required to correct the interference at the user s own expense Australia New Zealand Re...

Page 9: ...hall be responsible for correcting the interference caused by such unauthorized modification substitution or attachment The use of shielded I O cables is required when connecting the equipment to any optional peripheral or host device Failure to use shielded I O cables may violate FCC and ICES regulations Document Manual Code Description FUJITSU Supercomputer PRIMEHPC FX700 Operating Manual C120 0...

Page 10: ...s output by the computer and displayed on screens This font is used to indicate command output examples in boxes Shell showinfo M 2 Slot Device Status PASS Italics Indicates the name of a referenced manual See the FUJITSU Supercomputer PRIMEHPC FX700 BMC User s Guide Indicates the title of a referenced chapter section or subsection See Chapter 4 Operation Never peel off the labels Storage of Acces...

Page 11: ...marks of their respective owners Trademark indications TM R are omitted for some system and product names in this document This document shall not be reproduced or copied without the permission of the publisher All Rights Reserved Copyright FUJITSU LIMITED 2020 Safety Precautions C120 0089 03EN x ...

Page 12: ...or other property Notes on Product Handling Maintenance Modifying or Recycling the Product Disposal or Recycling of Products That Have Completed Their Life Cycle Waste must be disposed of in a professional and responsible way in accordance with environmental regulations For details please contact your nearest environmental authority or our sales representative Handling Lithium Batteries This produ...

Page 13: ...on Precautions 16 2 2 Power Voltage and Connection Precautions 17 2 3 Precautions on Handling the FX700 Main Unit 17 2 4 Environmental Protection 18 2 5 Environment Information 19 Chapter 3 Starting Up 21 3 1 Installation Procedure 21 3 2 Installation Specifications 22 3 3 Installation Environment 24 3 3 1 Dust 24 3 3 2 Corrosive Gas 24 3 3 3 Seawater Salt Spray Damage 25 3 4 Distribution Panel Cu...

Page 14: ... Connection Specifications FX700 Main Unit 49 3 15 Connecting Cables 50 3 15 1 Precautions on Connecting Disconnecting Cables 50 3 15 2 Connecting LAN Cables 50 3 15 3 Connecting the Power Cord 50 3 16 Powering On for the First Time 50 3 16 1 AC Power On 51 3 16 2 Initial BMC Settings 51 3 17 Installing the OS 52 3 17 1 OS Installation Procedure 52 3 17 2 OS Driver Installation Procedure 54 3 18 I...

Page 15: ... Console Hangs 62 5 4 3 Precaution on Using Commands 62 5 4 4 Precaution on Removing a PSU 62 5 4 5 Precaution on Using the Web GUI 63 Chapter 6 Technical Specifications 64 6 1 FX700 Main Unit Specifications 64 Appendix A BMC Driver Messages 65 Appendix B CPU MEM RAS Driver Messages 72 Contents C120 0089 03EN xiv ...

Page 16: ...Ds 10 Figure 1 15 Rear LEDs Except LAN LEDs on the Blade 11 Figure 1 16 Rear LAN LEDs on the Blade 12 Figure 1 17 BMCIFU LED ID 13 Figure 1 18 BMCIFU LAN LEDs 14 Figure 1 19 PSU LED 15 Figure 3 1 Distribution Panel Breaker Characteristics 26 Figure 3 2 Installation Area and Service Areas 27 Figure 3 3 Rack Depth 30 Figure 3 4 Rack Width 31 Figure 3 5 Support Upright Hole Shape in the Rack 32 Figur...

Page 17: ...ade 44 Figure 3 25 Installing the Dummy PSU 45 Figure 3 26 Removing the Dummy PSU 46 Figure 3 27 Installing the FANU 46 Figure 3 28 FANU Installation Completed 47 Figure 3 29 Installing the Bezel 47 Figure 3 30 Unlocking the FANU 48 Figure 3 31 Places to Hold a FANU When Removing It 48 Figure 3 32 Removing the FANU 49 Figure 4 1 Removing the Back Plates 57 Figure 4 2 Removing the Thumb Screws 58 F...

Page 18: ...nt 20 Table 3 1 Installation Specifications 22 Table 3 2 Permissible Levels of Corrosive Gases 24 Table 3 3 Distribution Panel Breaker Characteristics 25 Table 3 4 Mounting Conditions for Third Party Racks 28 Table 3 5 Power Cord Specifications 49 Table 6 1 FX700 Main Unit Specifications 64 Contents C120 0089 03EN xvii ...

Page 19: ...E is the command set architecture used by this CPU which has 48 cores and maintains performance at 3 072 TFlops operating at 2 0 GHz An HBM interface and PCI Express PCIe Gen3 16 lane controller are built in The CPU processor supports two frequencies 1 8 GHz and 2 0 GHz The main memory is High Bandwidth Memory HBM providing a high memory bandwidth of 1 024 GB s Each node is equipped with the follo...

Page 20: ...ows external views front rear top right side of the FX700 main unit Figure 1 1 Main Unit Front Figure 1 2 Main Unit Rear Figure 1 3 Main Unit Top Figure 1 4 Main Unit Right Side Chapter 1 Product Description C120 0089 03EN 1 1 Overview of the FX700 Main Unit 2 ...

Page 21: ...figuration of the FX700 Main Unit This section shows the front of the FX700 main unit Figure 1 5 Front Configuration of the FX700 Main Unit With Bezel Figure 1 6 Front Configuration of the FX700 Main Unit Without Bezel Chapter 1 Product Description C120 0089 03EN 1 1 Overview of the FX700 Main Unit 3 ...

Page 22: ... 5 PSU 00 6 PSU 01 7 PSU 02 8 BMCIF 00 1 1 3 Rear Configuration of the FX700 Main Unit This section shows the rear of the FX700 main unit Figure 1 7 Rear Configuration of the FX700 Main Unit Chapter 1 Product Description C120 0089 03EN 1 1 Overview of the FX700 Main Unit 4 ...

Page 23: ...sed for hardware status monitoring failure notification and power control 3 Node management LAN Used to connect nodes 1 1 4 LANs of the FX700 Main Unit This section shows the locations for the BMC service LAN BMC management LAN and node management LAN Figure 1 8 Rear Locations for the LANs of the FX700 Main Unit Chapter 1 Product Description C120 0089 03EN 1 1 Overview of the FX700 Main Unit 5 ...

Page 24: ...en they can be replaced By checking the LEDs maintenance workers can prevent mistakes in operation 1 2 1 Front Buttons and LEDs on the FX700 Main Unit Figure 1 9 Locations of the Front Buttons and LEDs on the FX700 Main Unit shows the locations of buttons and FANU LEDs on the front panel Figure 1 9 Locations of the Front Buttons and LEDs on the FX700 Main Unit For details see 1 2 1 1 Front Panel B...

Page 25: ...s only if all the nodes in the device are off Long press the button 4 seconds or longer to start shutdown of the operating systems on all nodes 2 BMC reset button You can reset the BMC by pressing this button Use the button for maintenance purposes when the BMC is inaccessible 1 2 1 1 Front Panel Buttons Figure 1 10 Front Panel Buttons Chapter 1 Product Description C120 0089 03EN 1 2 Buttons and L...

Page 26: ...ice contains part requiring immediate replacement Orange blinking This device contains part requiring preventive replacement 3 System power LED Off All nodes powered off On green At least 1 node powered on 4 BMC ready LED Off AC off BMC stopped On green BMC initialization completed Blinking green BMC initializing Fast blinking green BMC failed 1 2 1 2 Front Panel LEDs Figure 1 11 Front Panel LEDs ...

Page 27: ...n LED State Description 1 FANU alarm LED Off No failure On orange This FANU failed 1 2 1 3 FANU LED Figure 1 12 FANU LED Chapter 1 Product Description C120 0089 03EN 1 2 Buttons and LEDs on the FX700 Main Unit 9 ...

Page 28: ...FX700 Main Unit shows the locations of the rear LEDs on the FX700 main unit Figure 1 13 Locations of the Rear LEDs on the FX700 Main Unit For details on the LEDs see 1 2 2 1 Rear LEDs Except LAN LEDs on the Blade 1 2 2 2 Rear LAN LEDs on the Blade 1 2 2 3 BMCIFU LEDs and 1 2 2 4 PSU LED For details on the BMC service LAN node management LAN and BMC management LAN see 1 1 4 LANs of the FX700 Main U...

Page 29: ...g orange This blade contains part requiring preventive replacement 3 Identification LED Off This blade not selected as maintenance target On blue This blade selected as maintenance target Blinking blue Maintenance in progress on this blade 1 2 2 1 Rear LEDs Except LAN LEDs on the Blade Figure 1 15 Rear LEDs Except LAN LEDs on the Blade Chapter 1 Product Description C120 0089 03EN 1 2 Buttons and L...

Page 30: ...0 Mbit s Off Indicates data traffic at a transmission speed of 10 Mbit s 2 LAN link transmission LED On green A LAN connection has been established Off The LAN is not connected Green blinking LAN data is being transmitted 1 2 2 2 Rear LAN LEDs on the Blade Figure 1 16 Rear LAN LEDs on the Blade Chapter 1 Product Description C120 0089 03EN 1 2 Buttons and LEDs on the FX700 Main Unit 12 ...

Page 31: ... device not selected as maintenance target On blue This device selected as maintenance target Blinking blue Maintenance in progress on this device 1 2 2 3 BMCIFU LEDs Figure 1 17 BMCIFU LED ID Chapter 1 Product Description C120 0089 03EN 1 2 Buttons and LEDs on the FX700 Main Unit 13 ...

Page 32: ...ransmission speed of 100 Mbit s Off Indicates data traffic at a transmission speed of 10 Mbit s 2 LAN link transmission LED On green A LAN connection has been established Off The LAN is not connected Green blinking LAN data is being transmitted Figure 1 18 BMCIFU LAN LEDs Chapter 1 Product Description C120 0089 03EN 1 2 Buttons and LEDs on the FX700 Main Unit 14 ...

Page 33: ...following states Output stopped due to PSU failure No AC connection to other PSUs mounted in device or no AC input to this PSU Blinking Off AC input to PSU and output stopped On Off PSU currently operating normally 1 2 2 4 PSU LED Figure 1 19 PSU LED Chapter 1 Product Description C120 0089 03EN 1 2 Buttons and LEDs on the FX700 Main Unit 15 ...

Page 34: ...lfunctions and damage dramatically shortening the service life of the equipment Therefore measures such as installing an air cleaning system are required Also using the product in an environment exposed to dust may cause malfunctions and shorten the service life of the equipment by damaging memory media or by impeding equipment cooling Sources of corrosive gas include chemical factory areas hot sp...

Page 35: ...other equipment or anything other than its intended purpose The supplied power cord is designed to be connected to and used with the FX700 main unit and its safety has been confirmed Never use power cords from other products or for anything other than their intended purpose Otherwise fire or electric shock may result This product is also designed for an IT power system with phase to phase voltage ...

Page 36: ...llation location Do not use the FX700 main unit in proximity to devices such as cell phones that emit electromagnetic radiation Doing so may cause the FX700 main unit to malfunction The equipment passed impact tests in accordance with JIS Z 0200 and its load bearing strength has been confirmed Nonetheless take sufficient care in handling to avoid exposing the equipment to excessive shock or vibrat...

Page 37: ...e disposed of with unsorted domestic waste They can be returned free of charge to the manufacturer the dealer or an authorized agent for recycling or disposal All batteries containing pollutants are marked with a symbol a crossed out garbage can They are also marked with the chemical symbol for the heavy metal that causes them to be categorized as containing pollutants Cd Cadmium Hg Mercury Pb Lea...

Page 38: ...tion class A2 with humidity range from 20 RH to 80 RH Idle state power Watts 403 3 CMU x 1 Idle state power Watts at higher boundary temperature of declared operating condition class 598 4 CMU x 1 Maximum power Watts 2 723 Table 2 3 Critical Raw Material Content Raw Material Content Cobalt in batteries None contained Neodymium in HDDs None used Chapter 2 Important Information C120 0089 03EN 2 5 En...

Page 39: ...s on the series of FX700 system manuals see Manuals in This Series shown in Preface Remarks Separately ordered components may be delivered separately from the FX700 main unit 4 Install the rack rails on the rack See 3 8 1 Installing the Rack Rails on the Rack 5 Install the FX700 main unit in the rack See 3 8 2 Mounting the Chassis in the Rack 6 Do the wiring of the FX700 main unit See 3 15 1 Preca...

Page 40: ...llowable vibration m s2 gal Operating including standby 4 0 400 Synthetic seismic wave Stopped 10 10 0 1 000 Synthetic seismic wave Allowable dust concentration mg m3 0 15 Altitude Operating m 0 to 3 000 Stopped m 6 7 0 to 12 000 Power conditions Input voltage and phase 200 to 240 VAC 10 Single phase Frequency and variable width 50 60 Hz 3 3 Maximum power consumption W 2 723 Maximum apparent power...

Page 41: ... the recommended ambient temperature the fans may rotate at high speeds when the device is overloaded or a failure is detected 5 Condensation not allowed 6 Freezing temperatures not allowed 7 The stopped state means the device is packed and in storage 8 The actual level of noise heard varies depending on the listening position condition of mounting in the rack etc 9 The noise and sound power level...

Page 42: ...side air containing airborne particles of dust and tobacco smoke Eliminating Dust Capture airborne particles of dust etc with the air filters in air conditioners The computer room must be regularly cleaned to eliminate dust on top of and underneath the floor Be sure to clean the room in the following cases After building the computer room and before installing equipment When the computer room has ...

Page 43: ...coastal areas The installation criteria for preventing salt spray damage from sea salt particles is as follows Criteria The installation is not at sea nor within 0 5 km from the seashore An exception is an installation using air conditioners that do not take in outside air 3 4 Distribution Panel Cut Off Characteristics The characteristics of breakers in the customer s distribution panel must be co...

Page 44: ...Figure 3 1 Distribution Panel Breaker Characteristics Chapter 3 Starting Up C120 0089 03EN 3 4 Distribution Panel Cut Off Characteristics 26 ...

Page 45: ...m 3 5 Installation Area and Service Areas This section describes the installation area and service areas when the FX700 main unit is mounted in a Fujitsu 19 inch rack The installation area and service areas vary depending on the 19 inch rack used Figure 3 2 Installation Area and Service Areas Chapter 3 Starting Up C120 0089 03EN 3 5 Installation Area and Service Areas 27 ...

Page 46: ...ain unit mounted in this rack is guaranteed To safely use the product mounted in a Fujitsu 19 inch rack see the related documentation at the following URL For the Japanese market http jp fujitsu com platform server primergy peripheral rack https jp fujitsu com platform server primergy manual peri_rack html For the global market http manuals ts fujitsu com index php id 5406 5605 5606 If the product...

Page 47: ...1 mm 0 28 in The support uprights of the rack do not have threaded holes Figure 3 5 Support Upright Hole Shape in the Rack Check 9 Cable routing hole Cables can be removed from the bottom or the rear door of the rack Figure 3 3 Rack Depth Check 10 Load capacity of rack The total weight does not exceed the load capacity of the rack Note The load capacity may change when earthquake proofing measures...

Page 48: ...ription 1 Front door 2 Front support upright of rack 3 Rear support upright of rack 4 Rear door Rack Depth Conditions Figure 3 3 Rack Depth Chapter 3 Starting Up C120 0089 03EN 3 6 Rack System Requirements 30 ...

Page 49: ... upright of rack 3 Rear support upright of rack 4 Rear door 5 Bracket mounting area 6 Width for mounting brackets 7 Server width 8 Whole server Rack Width Figure 3 4 Rack Width Chapter 3 Starting Up C120 0089 03EN 3 6 Rack System Requirements 31 ...

Page 50: ...tion Specifications In particular measures need to be taken to prevent exhaust air from flowing back in through the intake vents of the equipment For example block the front of empty spaces in the rack Securing work areas for use during maintenance service areas Secure service areas for maintenance work by Fujitsu technicians Determine the service areas by referring to the service areas shown for ...

Page 51: ...ed it for transporting this product again 3 Check for damage caused during transport If any of the deliverables is damaged or does not match the invoice contact the vendor immediately 4 Check whether the delivered goods match the details printed on the invoice The product name and serial number are printed on the nameplate plate at the top of the FX700 main unit Figure 3 6 Checking the Product Nam...

Page 52: ...k The following is the procedure for mounting the chassis in the rack 3 8 1 Installing the Rack Rails on the Rack The following is the procedure for installing the rack rails on the rack 1 Prepare the necessary parts Figure 3 7 Parts for Rack Rail Installation 2 Loosen the four Phillips screws each on the left and right rails Chapter 3 Starting Up C120 0089 03EN 3 8 Mounting the Chassis in the Rac...

Page 53: ...ufactured by Fujitsu replace the pins on both the left and right rails according to the shape of the rack mounting hole Figure 3 8 Screw Positions on a Rail Figure 3 9 Before Pin Replacement Pin Diameter Φ9 2 Chapter 3 Starting Up C120 0089 03EN 3 8 Mounting the Chassis in the Rack 35 ...

Page 54: ...igure 3 10 After Pin Replacement Pin Diameter Φ6 7 3 Insert the pin at the rear of the right rail into a rack hole Figure 3 11 Rear of the Rail 4 Install the flat plate at the front of the right rail Chapter 3 Starting Up C120 0089 03EN 3 8 Mounting the Chassis in the Rack 36 ...

Page 55: ...right rail in place When using Φ6 5 pins fix it in place with countersunk head screws directly without washers depending on the size of the rack mounting hole Figure 3 13 Rail Fixing Positions 6 Tighten the four screws on the sliding part of the right rail Chapter 3 Starting Up C120 0089 03EN 3 8 Mounting the Chassis in the Rack 37 ...

Page 56: ...sis Figure 3 14 Screw Positions on the Rail 7 Repeat steps 3 to 6 to install the left rail too in the same way Figure 3 15 After the Rails are Installed 3 8 2 Mounting the Chassis in the Rack The following is the procedure for mounting the chassis in the rack Chapter 3 Starting Up C120 0089 03EN 3 8 Mounting the Chassis in the Rack 38 ...

Page 57: ... least two people are required to perform the work of mounting the chassis in the rack Thumb screws at 2 places 1 Insert the chassis along the rails into the rack Figure 3 16 Installing the Chassis 2 Fix the chassis to the rack Figure 3 17 Chassis Fixing Positions Chapter 3 Starting Up C120 0089 03EN 3 8 Mounting the Chassis in the Rack 39 ...

Page 58: ...nstalling the Blade in the Chassis 3 10 1 Installing the PSU in the Chassis 3 13 1 Installing the FANU in the Chassis and 3 11 1 Installing the Dummy Blade in the Chassis 3 9 Installing Removing the Blade 3 9 1 Installing the Blade in the Chassis The following is the procedure for installing the blade in the chassis 1 Install the blade in the chassis Chapter 3 Starting Up C120 0089 03EN 3 9 Instal...

Page 59: ... 1 Installing the Dummy Blade in the Chassis When inserting the blade be careful not to allow foreign objects such as cables to enter the chassis 3 9 2 Removing the Blade From the Chassis The following is the procedure for removing the blade from the chassis 1 While the lock is unlocked 1 pull the handle 2 to remove the blade from the chassis Chapter 3 Starting Up C120 0089 03EN 3 9 Installing Rem...

Page 60: ...moving the PSU 3 10 1 Installing the PSU in the Chassis The following is the procedure for installing the PSU in the chassis 1 Install the PSU in the chassis Figure 3 21 Installing the PSU Chapter 3 Starting Up C120 0089 03EN 3 10 Installing Removing the PSU 42 ...

Page 61: ...m the Chassis This section describes how to remove the PSU 1 Lift up the handle on the PSU halfway in the direction of the arrow 1 and push the latch 2 While pushing the latch pull out the PSU 3 Figure 3 22 Removing the PSU 3 11 Installing Removing the Dummy Blade 3 11 1 Installing the Dummy Blade in the Chassis This section describes how to install the dummy blade in the chassis 1 Push the dummy ...

Page 62: ...bjects such as cables to enter the chassis 3 11 2 Removing the Dummy Blade From the Chassis This section describes how to remove the dummy blade from the chassis 1 Unlock the lock grasp the knob and pull out the dummy blade 1 Figure 3 24 Removing the Dummy Blade Chapter 3 Starting Up C120 0089 03EN 3 11 Installing Removing the Dummy Blade 44 ...

Page 63: ...e dummy PSU in the direction of the arrow 2 Figure 3 25 Installing the Dummy PSU 2 Push in the dummy PSU until it is locked 3 12 2 Removing the Dummy PSU From the Chassis This section describes how to remove the dummy PSU from the chassis 1 While pushing the lock 1 remove the dummy PSU in the direction of the arrow 2 Chapter 3 Starting Up C120 0089 03EN 3 12 Installing Removing the Dummy PSU 45 ...

Page 64: ...U in the Chassis The following is the procedure for installing the FANU in the chassis 1 Install the FANU in the chassis Figure 3 27 Installing the FANU Remarks Confirm that the front surface of the FANU is aligned with the front panel surface Chapter 3 Starting Up C120 0089 03EN 3 13 Installing Removing the FANU 46 ...

Page 65: ...ure 3 29 Installing the Bezel 4 Insert the left and right guide pins into the chassis and fix the front panel in place with two screws 3 13 2 Removing the FANU From the Chassis This following is the procedure for removing the FANU from the chassis 1 While the first lock is unlocked 1 pull out the FANU in the direction of 2 Chapter 3 Starting Up C120 0089 03EN 3 13 Installing Removing the FANU 47 ...

Page 66: ... A Figure 3 30 Unlocking the FANU Remarks The following shows the places to hold a FANU when removing it Figure 3 31 Places to Hold a FANU When Removing It 2 While pushing the second lock 3 pull out the FANU from the chassis Chapter 3 Starting Up C120 0089 03EN 3 13 Installing Removing the FANU 48 ...

Page 67: ...4 Input Power Connection Specifications This section describes the input power connection specifications of the FX700 main unit 3 14 1 Input Power Connection Specifications FX700 Main Unit The following table shows the input power connection specifications of the FX700 main unit Remarks For the power cord connected to the device use the one that was shipped together with either the device or optio...

Page 68: ...ype is S FTP Cat5 or higher Use of unshielded or badly shielded cables may lead to increased emission of interference and or reduced fault tolerance of the device 3 15 2 Connecting LAN Cables The following is the procedure for connecting LAN cables 1 Connect the LAN cables to the LAN ports For details on the LAN ports on the FX700 main unit see 1 1 4 LANs of the FX700 Main Unit 2 Check the port LE...

Page 69: ...etails on how to log in to the BMC see 2 1 1 Login in the FUJITSU Supercomputer PRIMEHPC FX700 BMC User s Guide C120 0091EN Setting the Time To set the BMC time select Time Settings from the Configuration menu For details see 3 4 4 Time Settings in the FUJITSU Supercomputer PRIMEHPC FX700 BMC User s Guide C120 0091EN Configuring the Network To configure the network to use the BMC select Network Se...

Page 70: ...ported kernel For details on the supported OS and kernel see the following sites For the Japanese market https www fujitsu com jp products computing servers supercomputer downloads For the global market https www fujitsu com global products computing servers supercomputer documents To install the OS perform the following two procedures in the order shown Creating an Installation DVD Installing the...

Page 71: ... and Configuring a TFTP server for UEFI based clients in Performing an advanced RHEL installation on the Red Hat Inc website https access redhat com documentation en us red_hat_ enterprise_linux 8 html performing_an_advanced_rhel_installation index To configure the TFTP server set grub cfg To connect the console add the following kernel option to the grub cfg configuration file Example grub cfg 2 ...

Page 72: ... OS Driver Installation Procedure The following is the procedure for applying the OS drivers for FX700 hardware The BMC driver is required for basic operations such as powering on off the node The CPU MEM RAS driver is required for CPU and memory failure detection Driver RPM files are available online For the Japanese market https www fujitsu com jp products computing servers supercomputer downloa...

Page 73: ...ect settings Execute the following command on the FX700 node to restart the FX700 5 Check the installation status 1 BMC drivercrlf If the status of the systemd service FJSVxosbmc is active installation has been successfully completed 2 CPU MEM RAS drivercrlf If the status of the systemd service FJSVxoscpuras is active installation has been successfully completed 3 18 Installing the InfiniBand Driv...

Page 74: ...on or off the system Short press the button to power on all nodes only if all the nodes in the device are off Long press the button 4 seconds or longer to start shutdown of the operating systems on all nodes 4 1 1 Turning On Off AC Power to the FX700 Main Unit Turning On AC Power to the FX700 Main Unit The green PSU status LED blinks when the equipment is connected to the main power supply Approxi...

Page 75: ...assis 3 13 2 Removing the FANU From the Chassis and 3 11 2 Removing the Dummy Blade From the Chassis The work of removing the chassis from the rack must be done by two or more people Support bracket M5 screw x 2 4 2 Removing the Chassis 4 2 1 Removing the Chassis From the Rack The following is the procedure for removing the chassis from the rack 1 Remove the back plates Figure 4 1 Removing the Bac...

Page 76: ...Thumb screws at 2 places Figure 4 2 Removing the Thumb Screws 3 Remove the chassis from the rack Figure 4 3 Removing the Chassis Chapter 4 Operation C120 0089 03EN 4 2 Removing the Chassis 58 ...

Page 77: ...ter the system Keep the ventilation areas of the FX700 main unit clean Do not use spray cleaners including flammable types Doing so may damage the device or cause fire Wipe the FX700 main unit with a dry cloth when cleaning it If an area is particularly dirty dilute a household cleaning agent in a solution Moisten a cloth in the solution wring out the cloth thoroughly and wipe the area with the cl...

Page 78: ...ct button Type Select Partial or Full Encrypt To enable file encryption check the Enable check box Encrypt Key If the Encrypt check box is checked enter an encryption key in the field 5 The following message appears Click the OK button to start snapshot collection Figure 5 1 Message Displayed A new file is registered in Snapshot Files when collection is completed Remarks The file No corresponds to...

Page 79: ...rcomputer PRIMEHPC FX700 BMC User s Guide C120 0091EN sosreport Collect the sosreport file of the FX700 system node For the procedure in detail see the Red Hat Inc website For BMC driver related problems collect the following additional information BMC Driver var opt FJSVxos bmc log common file on the node var opt FJSVxos bmc log ipmi_message file on the node 5 4 Other Problems 5 4 1 Both the BMC ...

Page 80: ... user to the node where the hang occurred 2 Identify the PIDs of processes running on the console 3 Kill a running process 4 Kill the bash process used by the console 5 Kill the bash parent process used by the console 5 4 3 Precaution on Using Commands If an invalid character is entered a node is powered off and Node on the Web GUI screen displays Reserved for the node To clear Reserved from the d...

Page 81: ...he same PC are not supported The following message appears when multiple accesses are attempted If the message appears close the session Figure 5 2 Message From the Webpage Chapter 5 Collecting Information When Troubleshooting C120 0089 03EN 5 4 Other Problems 63 ...

Page 82: ...n for redundancy 2 1 redundancy Input voltage 200 to 240 VAC 10 single phase Input frequency 50 60 Hz 3 Hz Efficiency 80 PLUS PLATINUM FANU 7 1 redundancy Chapter 6 Technical Specifications This chapter describes the specifications of the FX main unit chassis blade PSU and FANU Note Please note that the FX700 main unit specifications may be updated without notice 6 1 FX700 Main Unit Specifications...

Page 83: ...o the data storage area Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos BMC 0003 internal information Invalid parameter TYPE NULL Meaning The specified value is invalid as an input parameter TYPE Status code Action Collect investigation data accordi...

Page 84: ...Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos BMC 0008 internal information Failed to create a log file for developers Meaning Generation of a log file for developers failed and BMC driver initialization failed Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contac...

Page 85: ...le to register the interrupt handler Meaning Registration of a BMC driver interrupt process failed Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos BMC 0015 internal information Base address is missing Address VALUE Meaning The RAM base address is an...

Page 86: ...from the user failed SIZE Size of unreadable data Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos BMC 0021 internal information BMC driver is not ready CODE Meaning The BMC driver cannot accept commands because loading is in progress or failed CODE ...

Page 87: ...aintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos BMC 0030 internal information Invalid parameter name_flag VALUE Meaning The specified value is invalid as an input parameter VALUE Set name_flag value Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Suppo...

Page 88: ...n the system by the Shutdown notification Meaning The system received a Shutdown notification and is powering off Action No action is necessary INFO xos BMC 1005 internal information Copyright c 2018 FUJITSU LIMITED All rights reserved Meaning The message displays the copyright to the BMC driver Action No action is necessary INFO xos BMC 1006 internal information BMC driver VERSION DATE Meaning Th...

Page 89: ...The current state is Busy because another command is being executed CODE IPMI command code Action No action is necessary INFO xos BMC 1012 internal information Could not accept the execution of the command under kill termination CODE Meaning The command cannot be executed because forced termination processing by kill is in progress CODE IPMI command code Action No action is necessary INFO xos BMC ...

Page 90: ...ation Failed to register GUEST SOFTWARE ERROR virq Meaning Registration of the logical interrupt number of GUEST SOFTWARE ERROR failed Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos RAS 0003 internal information GUEST SOFTWARE ERROR request_irq fai...

Page 91: ...AS 0008 internal information ITS HOST SOFTWARE ERROR detected GITS_FJ_ ITS_ERROR_STATUS DATA1 Meaning HOST SOFTWARE ERROR was detected in the ITS setting trigger DATA1 ITS error status register value Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos R...

Page 92: ... Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos RAS 0014 internal information Clearing valid bit failed ERR0STATUS DATA1 DATA2 DATA3 Meaning Clearing of a valid bit failed in the error status register DATA1 Error status register value before clearing DATA2 Error status register value after clearing an uncorrected error bit DATA3 Error sta...

Page 93: ...R register value DATA2 Error status register value Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos RAS 0019 internal information DG L2 detected esr DATA1 ERR0STATUS DATA2 Meaning The L2 cache was degraded DATA1 ESR register value DATA2 Error status ...

Page 94: ...ode or type detected when SError occurred esr DATA1 ERR0STATUS DATA2 Meaning SError occurred The error code or error type is invalid DATA1 ESR register value DATA2 Error status register value Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message ERR xos RAS 0024 ...

Page 95: ...os RAS 0028 internal information Invalid error code or type detected when memory abort occurred esr DATA1 ERR0STATUS DATA2 Meaning A memory abort occurred The error code or error type is invalid DATA1 ESR register value DATA2 Error status register value Action Collect investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the col...

Page 96: ...t investigation data according to 5 3 1 Collecting Information for Maintenance Purposes and then contact Fujitsu Support with the collected data together with the output message Information Message INFO xos RAS 0000 internal information No need to clear ERR0STATUS register Meaning The error status register does not need to be cleared Action No action is necessary Appendix B CPU MEM RAS Driver Mess...

Page 97: ......

Reviews: