Fujitsu Server M8000 User manual

Add to my manuals
546 Pages

advertisement

Fujitsu Server M8000 User manual | Manualzz
SPARC Enterprise™
M8000/M9000 Servers
Service Manual
Manual Code C120-E330-09EN
Part No. 819-7877-16
June 2010, Revision A
Copyright 2007, 2010 FUJITSU LIMITED, 1-1, Kamikodanaka 4-chome, Nakahara-ku, Kawasaki-shi, Kanagawa-ken 211-8588, Japan. All rights
reserved.
Sun Microsystems, Inc. provided technical input and review on portions of this material.
Sun Microsystems, Inc. and Fujitsu Limited each own or control intellectual property rights relating to products and technology described in
this document, and such products, technology and this document are protected by copyright laws, patents and other intellectual property laws
and international treaties. The intellectual property rights of Sun Microsystems, Inc. and Fujitsu Limited in such products, technology and this
document include, without limitation, one or more of the United States patents listed at http://www.sun.com/patents and one or more
additional patents or patent applications in the United States or other countries.
This document and the product and technology to which it pertains are distributed under licenses restricting their use, copying, distribution,
and decompilation. No part of such product or technology, or of this document, may be reproduced in any form by any means without prior
written authorization of Fujitsu Limited and Sun Microsystems, Inc., and their applicable licensors, if any. The furnishing of this document to
you does not give you any rights or licenses, express or implied, with respect to the product or technology to which it pertains, and this
document does not contain or represent any commitment of any kind on the part of Fujitsu Limited or Sun Microsystems, Inc., or any affiliate of
either of them.
This document and the product and technology described in this document may incorporate third-party intellectual property copyrighted by
and/or licensed from suppliers to Fujitsu Limited and/or Sun Microsystems, Inc., including software and font technology.
Per the terms of the GPL or LGPL, a copy of the source code governed by the GPL or LGPL, as applicable, is available upon request by the End
User. Please contact Fujitsu Limited or Sun Microsystems, Inc
This distribution may include materials developed by third parties.
Parts of the product may be derived from Berkeley BSD systems, licensed from the University of California. UNIX is a registered trademark
in the U.S. and in other countries, exclusively licensed through X/Open Company, Ltd.
Sun, Sun Microsystems, the Sun logo, Java, Netra, Solaris, Sun Ray, Answerbook2, docs.sun.com, OpenBoot, and Sun Fire are trademarks or
registered trademarks of Sun Microsystems, Inc., or its subsidiaries, in the U.S. and other countries.
Fujitsu and the Fujitsu logo are registered trademarks of Fujitsu Limited.
All SPARC trademarks are used under license and are registered trademarks of SPARC International, Inc. in the U.S. and other countries.
Products bearing SPARC trademarks are based upon architecture developed by Sun Microsystems, Inc.
SPARC64 is a trademark of SPARC International, Inc., used under license by Fujitsu Microelectronics, Inc. and Fujitsu Limited.
The OPEN LOOK and Sun™ Graphical User Interface was developed by Sun Microsystems, Inc. for its users and licensees. Sun acknowledges
the pioneering efforts of Xerox in researching and developing the concept of visual or graphical user interfaces for the computer industry. Sun
holds a non-exclusive license from Xerox to the Xerox Graphical User Interface, which license also covers Sun’s licensees who implement OPEN
LOOK GUIs and otherwise comply with Sun’s written license agreements.
United States Government Rights - Commercial use. U.S. Government users are subject to the standard government user license agreements of
Sun Microsystems, Inc. and Fujitsu Limited and the applicable provisions of the FAR and its supplements.
Disclaimer: The only warranties granted by Fujitsu Limited, Sun Microsystems, Inc. or any affiliate of either of them in connection with this
document or any product or technology described herein are those expressly set forth in the license agreement pursuant to which the product
or technology is provided. EXCEPT AS EXPRESSLY SET FORTH IN SUCH AGREEMENT, FUJITSU LIMITED, SUN MICROSYSTEMS, INC.
AND THEIR AFFILIATES MAKE NO REPRESENTATIONS OR WARRANTIES OF ANY KIND (EXPRESS OR IMPLIED) REGARDING SUCH
PRODUCT OR TECHNOLOGY OR THIS DOCUMENT, WHICH ARE ALL PROVIDED AS IS, AND ALL EXPRESS OR IMPLIED
CONDITIONS, REPRESENTATIONS AND WARRANTIES, INCLUDING WITHOUT LIMITATION ANY IMPLIED WARRANTY OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT, ARE DISCLAIMED, EXCEPT TO THE
EXTENT THAT SUCH DISCLAIMERS ARE HELD TO BE LEGALLY INVALID. Unless otherwise expressly set forth in such agreement, to the
extent allowed by applicable law, in no event shall Fujitsu Limited, Sun Microsystems, Inc. or any of their affiliates have any liability to any
third party under any legal theory for any loss of revenues or profits, loss of use or data, or business interruptions, or for any indirect, special,
incidental or consequential damages, even if advised of the possibility of such damages.
DOCUMENTATION IS PROVIDED “AS IS” AND ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND WARRANTIES,
INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR NON-INFRINGEMENT,
ARE DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD TO BE LEGALLY INVALID.
Please
Recycle
Copyright 2007, 2010 FUJITSU LIMITED, 1-1, Kamikodanaka 4-chome, Nakahara-ku, Kawasaki-shi, Kanagawa-ken 211-8588, Japon. Tous
droits réservés.
Entrée et revue tecnical fournies par Sun Microsystems, Incl sur des parties de ce matériel.
Sun Microsystems, Inc. et Fujitsu Limited détiennent et contrôlent toutes deux des droits de propriété intellectuelle relatifs aux produits et
technologies décrits dans ce document. De même, ces produits, technologies et ce document sont protégés par des lois sur le copyright, des
brevets, d’autres lois sur la propriété intellectuelle et des traités internationaux. Les droits de propriété intellectuelle de Sun Microsystems, Inc.
et Fujitsu Limited concernant ces produits, ces technologies et ce document comprennent, sans que cette liste soit exhaustive, un ou plusieurs
des brevets déposés aux États-Unis et indiqués à l’adresse http://www.sun.com/patents de même qu’un ou plusieurs brevets ou applications
brevetées supplémentaires aux États-Unis et dans d’autres pays.
Ce document, le produit et les technologies afférents sont exclusivement distribués avec des licences qui en restreignent l’utilisation, la copie,
la distribution et la décompilation. Aucune partie de ce produit, de ces technologies ou de ce document ne peut être reproduite sous quelque
forme que ce soit, par quelque moyen que ce soit, sans l’autorisation écrite préalable de Fujitsu Limited et de Sun Microsystems, Inc., et de leurs
éventuels bailleurs de licence. Ce document, bien qu’il vous ait été fourni, ne vous confère aucun droit et aucune licence, expresses ou tacites,
concernant le produit ou la technologie auxquels il se rapporte. Par ailleurs, il ne contient ni ne représente aucun engagement, de quelque type
que ce soit, de la part de Fujitsu Limited ou de Sun Microsystems, Inc., ou des sociétés affiliées.
Ce document, et le produit et les technologies qu’il décrit, peuvent inclure des droits de propriété intellectuelle de parties tierces protégés par
copyright et/ou cédés sous licence par des fournisseurs à Fujitsu Limited et/ou Sun Microsystems, Inc., y compris des logiciels et des
technologies relatives aux polices de caractères.
Par limites du GPL ou du LGPL, une copie du code source régi par le GPL ou LGPL, comme applicable, est sur demande vers la fin utilsateur
disponible; veuillez contacter Fujitsu Limted ou Sun Microsystems, Inc.
Cette distribution peut comprendre des composants développés par des tierces parties.
Des parties de ce produit pourront être dérivées des systèmes Berkeley BSD licenciés par l’Université de Californie. UNIX est une marque
déposée aux Etats-Unis et dans d’autres pays et licenciée exclusivement par X/Open Company, Ltd.
Sun, Sun Microsystems, le logo Sun, Java, Netra, Solaris, Sun Ray, Answerbook2, docs.sun.com, OpenBoot, et Sun Fire sont des marques de
fabrique ou des marques déposées de Sun Microsystems, Inc., ou ses filiales, aux Etats-Unis et dans d’autres pays.
Fujitsu et le logo Fujitsu sont des marques déposées de Fujitsu Limited.
Toutes les marques SPARC sont utilisées sous licence et sont des marques de fabrique ou des marques déposées de SPARC International, Inc.
aux Etats-Unis et dans d’autres pays. Les produits portant les marques SPARC sont basés sur une architecture développée par Sun
Microsystems, Inc.
SPARC64 est une marques déposée de SPARC International, Inc., utilisée sous le permis par Fujitsu Microelectronics, Inc. et Fujitsu Limited.
L’interface d’utilisation graphique OPEN LOOK et Sun™ a été développée par Sun Microsystems, Inc. pour ses utilisateurs et licenciés. Sun
reconnaît les efforts de pionniers de Xerox pour la recherche et le développement du concept des interfaces d’utilisation visuelle ou graphique
pour l’industrie de l’informatique. Sun détient une license non exclusive de Xerox sur l’interface d’utilisation graphique Xerox, cette licence
couvrant également les licenciés de Sun qui mettent en place l’interface d’utilisation graphique OPEN LOOK et qui, en outre, se conforment
aux licences écrites de Sun.
Droits du gouvernement américain - logiciel commercial. Les utilisateurs du gouvernement américain sont soumis aux contrats de licence
standard de Sun Microsystems, Inc. et de Fujitsu Limited ainsi qu’aux clauses applicables stipulées dans le FAR et ses suppléments.
Avis de non-responsabilité: les seules garanties octroyées par Fujitsu Limited, Sun Microsystems, Inc. ou toute société affiliée de l’une ou l’autre
entité en rapport avec ce document ou tout produit ou toute technologie décrit(e) dans les présentes correspondent aux garanties expressément
stipulées dans le contrat de licence régissant le produit ou la technologie fourni(e). SAUF MENTION CONTRAIRE EXPRESSÉMENT
STIPULÉE DANS CE CONTRAT, FUJITSU LIMITED, SUN MICROSYSTEMS, INC. ET LES SOCIÉTÉS AFFILIÉES REJETTENT TOUTE
REPRÉSENTATION OU TOUTE GARANTIE, QUELLE QU’EN SOIT LA NATURE (EXPRESSE OU IMPLICITE) CONCERNANT CE
PRODUIT, CETTE TECHNOLOGIE OU CE DOCUMENT, LESQUELS SONT FOURNIS EN L’ÉTAT. EN OUTRE, TOUTES LES CONDITIONS,
REPRÉSENTATIONS ET GARANTIES EXPRESSES OU TACITES, Y COMPRIS NOTAMMENT TOUTE GARANTIE IMPLICITE RELATIVE À
LA QUALITÉ MARCHANDE, À L’APTITUDE À UNE UTILISATION PARTICULIÈRE OU À L’ABSENCE DE CONTREFAÇON, SONT
EXCLUES, DANS LA MESURE AUTORISÉE PAR LA LOI APPLICABLE. Sauf mention contraire expressément stipulée dans ce contrat, dans
la mesure autorisée par la loi applicable, en aucun cas Fujitsu Limited, Sun Microsystems, Inc. ou l’une de leurs filiales ne sauraient être tenues
responsables envers une quelconque partie tierce, sous quelque théorie juridique que ce soit, de tout manque à gagner ou de perte de profit,
de problèmes d’utilisation ou de perte de données, ou d’interruptions d’activités, ou de tout dommage indirect, spécial, secondaire ou
consécutif, même si ces entités ont été préalablement informées d’une telle éventualité.
LA DOCUMENTATION EST FOURNIE “EN L’ETAT” ET TOUTES AUTRES CONDITIONS, DECLARATIONS ET GARANTIES EXPRESSES
OU TACITES SONT FORMELLEMENT EXCLUES, DANS LA MESURE AUTORISEE PAR LA LOI APPLICABLE, Y COMPRIS NOTAMMENT
TOUTE GARANTIE IMPLICITE RELATIVE A LA QUALITE MARCHANDE, A L’APTITUDE A UNE UTILISATION PARTICULIERE OU A
L’ABSENCE DE CONTREFACON.
Contents
Preface
1.
xvii
Safety and Tools
1.1
Symbols
1–1
1–1
1.1.1
Text Conventions
1–1
1.1.2
Prompt Notations
1–2
1.1.2.1
1–2
1.1.3
Environmental Requirements for Using This Product
1.1.4
Conventions for Alert Messages
1.1.4.1
1.2
Command syntax
Precautions
1–3
Alert Messages in the Text
Operating Environment of the Product
1.2.2
Maintenance
1.2.3
Conversion and Reuse of This Product
Tools Required for Maintenance
1.4
Antistatic Precautions
1–5
1–5
1–6
Removing Static Electricity
1.4.1.1
1–4
1–4
1.3
1.4.2
1–3
1–4
1.2.1
1.4.1
1–3
1–6
Removing Static Electricity on a CMU and an IOU
Grounding Port Connection Locations
1–8
1–11
v
2.
Product Overview and Troubleshooting
2.1
2.2
2.3
System Views
2–1
2.1.1
SPARC Enterprise M8000 Server
2.1.2
SPARC Enterprise M9000 Server (Base Cabinet)
2.1.3
SPARC Enterprise M9000 Server (Expansion Cabinet)
Labels
2–3
System Name Plate Label, Rating Label, ID Label (Japan) or EZ
Label (besides Japan), and Standard Label 2–11
2.2.2
Labels About Handling
Operator Panel
2–15
2–16
2.3.1
Operator Panel Location
2–16
2.3.2
Appearance and Operations
2.3.3
LED
2.3.4
Switch
2–17
2–18
2–20
Determining Which Diagnostics Methods To Use
2.5
Checking the Server and System Configuration
2.5.2
2.5.3
2–21
2–23
Checking the Hardware Configuration and FRU Status
2.5.1.1
2.7
2–9
2–11
2.4
2.6
2–6
2.2.1
2.5.1
vi
2–1
Checking the Hardware Configuration
2–23
2–24
Checking the Software and XSCF Firmware Configurations
2.5.2.1
Checking the Software Configuration
2.5.2.2
Checking the Firmware Configuration
Downloading the Error Log Information
Error Conditions
2–26
2.6.1
Predictive Self-Healing Tools
2.6.2
Monitoring Output
2.6.3
Messaging Output
LED Error Display
2–26
2–27
2–28
2–29
2–30
2.7.1
When target FRU is indicated by LEDs
2.7.2
When target FRU is not indicated by LEDs
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2–31
2–31
2–25
2–26
2–24
2.8
Using the Troubleshooting Commands
2.8.1
Using the showlogs Command
2.8.2
Using the fmdump Command
2.8.3
Using the fmadm Command
2.8.4
2.9
2–34
2–35
2–35
2.8.3.1
fmadm config Command
2–35
2.8.3.2
fmadm faulty Command
2–36
2.8.3.3
fmadm repair Command
2–36
Using the fmstat Command
2–37
Traditional Solaris Troubleshooting Commands
2.9.1
2.9.2
iostat Command
2–38
2.9.1.1
2–38
2.9.3
2–41
2–42
Options
2–43
2–43
2–44
Options
ps Command
2.9.6.1
2.9.7
Options
ping Command
2.9.5.1
2.9.6
2–39
netstat Command
2.9.4.1
2.9.5
Options
2–45
2–46
Options
2–46
prstat Command
2–47
2.9.7.1
2–47
Periodic Maintenance
2–37
2–39
prtconf Command
2.9.3.1
2.9.4
Options
prtdiag Command
2.9.2.1
3.
2–34
Options
3–1
3.1
Cleaning a Tape Drive Unit
3–1
3.2
Cleaning an Air Filter (Server)
3.3
Cleaning an Air Filter (I/O Unit)
3–2
3–5
Contents
vii
4.
FRU Removal Preparation
4.1
Types of Replacement Procedures
4.1.1
4.2
4–3
Disconnecting a FRU from a Domain
4–3
4.2.1.2
Disconnecting a PCI card
4–4
4.2.2
Disconnecting and Replacing a FRU
4–5
4.2.3
Configuring a FRU into a Domain
4–6
4.2.3.1
Configuring CMU/IOU
4–6
4.2.3.2
Configuring a PCI card
Confirming the Hardware
Hot Replacement
4.3.2
Confirming the Hardware
4–12
4–15
4–18
4.4.1
Powering the Server Off
4–18
4.4.2
Powering the Server On
4–19
4.4.2.1
From the Operator Panel
4.4.2.2
From the Maintenance Terminal
Confirming the Hardware
Types of Power Supply
4–19
4–20
4–20
Power-On/Off of Main Line Switch
4.5.2
4–7
4–12
Disconnecting and Replacing a FRU
Cold Replacement
4–3
4–7
4.3.1
4.5.1
viii
4–2
Disconnecting a CMU/IOU
4.4.3
4.5
4–2
4.2.1.1
4.2.4
4.4
FRU Replacement
Active Replacement
4.2.1
4.3
4–1
4–24
4–24
4.5.1.1
AC Input Power
4.5.1.2
Power System
4–24
4–26
Power-On/Off Procedures of Main Line Switch
4.5.2.1
Power-On
4–27
4.5.2.2
Power-Off
4–27
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4–26
4.5.3
6.
4–27
4.5.3.1
SPARC Enterprise M8000 Server Single-Phase Power
Feed 4–28
4.5.3.2
SPARC Enterprise M8000 Server Single-Phase and Dual
Power Feed 4–29
4.5.3.3
SPARC Enterprise M8000 Server Three-Phase Power
Feed 4–30
4.5.3.4
SPARC Enterprise M9000 Server Single-Phase Power
Feed 4–31
4.5.3.5
SPARC Enterprise M9000 Server Single-Phase and Dual
Power Feed 4–32
4.5.3.6
SPARC Enterprise M9000 Server Three-Phase Power
Feed 4–33
4.6
Emergency Switch-Off
4.7
Cable Routing of the SPARC Enterprise M8000 Server
4.7.1
5.
Main Line Switch Locations
4–34
4–34
Cable Routing When the External I/O Expansion Unit Mounted
4–34
4.7.1.1
Precautions For Cable Routing
4.7.1.2
When Three External I/O Expansion Units Mounted
4–37
4.7.1.3
For Server Using Three-phase Power Feed
4.7.1.4
When One External I/O Expansion Unit Mounted
45
Internal Components Access
4–35
How to Open and Close Doors
5.2
Corresponding Components and Doors That Can Be Accessed
5.3
How to Remove a Door
5.4
How to Remove a Side Panel
5–1
6.2
CPU Upgrade
5–2
5–3
5–4
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
Overview of the CMU
4–
5–1
5.1
6.1
4–41
6–1
6–1
6–5
Contents
ix
6.2.1
SPARC64 VII CPU Modules Added to a New Domain
6.2.2
SPARC64 VII Processors Added to an Existing Domain
7.
8.
9.
x
6–8
6.2.2.1
Preparing to Add SPARC64 VII Processors to an Existing
Domain 6–8
6.2.2.2
Adding a SPARC64 VII CPU Module to a Domain
Configured With SPARC64 VI 6–10
6.2.2.3
Upgrading a SPARC64 VI CPU Module to SPARC64 VII
on an Existing Domain 6–11
6.3
Active Replacement and Hot Replacement
6.4
Cold Replacement
6.5
CPU and DIMM Replacement
6–12
6–23
6–29
6.5.1
Replacing a CPU Module
6.5.2
Memory Module Mounting Conditions
6.5.3
6–5
6–30
6–40
6.5.2.1
Confirmation of DIMM Information
6.5.2.2
DIMM Mounting Conditions
DIMM Replacement
I/O Unit (IOU) Replacement
6–44
7–1
7.1
Overview of the IOU
7.2
Active Replacement and Hot Replacement
7.3
Cold Replacement
FAN Unit Replacement
7–2
7–8
7–17
8–1
8.1
Overview of the FAN Unit
8–2
8.2
Active Replacement and Hot Replacement
8.3
Cold Replacement
8–8
8–14
Power Supply Unit (PSU) Replacement
9–1
9.1
Overview of the PSU
9–1
9.2
Active Replacement and Hot Replacement
9.3
Cold Replacement
9–12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
9–8
6–42
6–41
10.
11.
12.
13.
14.
15.
Operator Panel Replacement
10–1
10.1
Overview of the Operator Panel
10.2
Cold Replacement
10–4
XSCF Unit Replacement
11–1
10–1
11.1
Overview of the XSCFU
11–1
11.2
Active Replacement and Hot Replacement
11.3
Cold Replacement
11–5
11–12
Hard Disk Drive (HDD) Replacement
12.1
Overview of the HDD
12.2
Active Replacement
12.3
Cold Replacement
12–1
12–1
12–3
12–5
PCI Slot Device Replacement
13–1
13.1
Overview of PCI Slot Devices
13.2
Active Replacement
13.3
Hot Replacement
13.4
Cold Replacement
13–1
13–3
13–17
13–20
CD-RW/DVD-RW Drive Unit Replacement
14–1
14.1
Overview of a CD-RW/DVD-RW Drive Unit
14.2
Active Replacement
14.3
Hot Replacement
14.4
Cold Replacement
14–1
14–5
14–9
14–11
Tape Drive Unit Replacement
15–1
15.1
Overview of the Tape Drive Unit
15.2
Active Replacement
15.3
Hot Replacement
15.4
Cold Replacement
15–1
15–5
15–9
15–11
Contents
xi
16.
17.
18.
19.
20.
21.
22.
xii
Clock Control Unit Replacement
16.1
Overview of the CLKU
16.2
Cold Replacement
16–1
16–4
Crossbar Unit Replacement
17–1
17.1
Overview of XBUs
17–1
17.2
Cold Replacement
17–4
AC Section Replacement
18–1
18.1
Overview of ACSs
18–1
18.2
Cold Replacement
18–5
DDC Replacement
16–1
19–1
19.1
Overview of the DDC
19.2
Active Replacement and Hot Replacement
19.3
Cold Replacement
19–7
Backplane Replacement
20–1
20.1
Overview of the BP
20.2
Cold Replacement
19–1
20–1
20–1
20.2.1
M8000/M9000 BPs
20.2.2
PSU BP
20–15
20.2.3
FAN BP
20–23
Sensor Unit Replacement
20–2
21–1
21.1
Overview of the SNSU
21.2
Cold Replacement
21–1
21–4
Media Backplane Replacement
22–1
22.1
Overview of the MEDBP
22–1
22.2
Cold Replacement
22–5
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
19–3
23.
24.
Switch Backplane Replacement
23.1
Overview of SWBPs
23.2
Cold Replacement
23–1
23–5
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion
Cabinet 24–1
24.1
25.
23–1
Addition and Deletion of Rack-mountable Dual Power Feed
24.1.1
Overview of RDPF
24.1.2
Addition and Deletion
24–2
24–2
24–3
24.1.2.1
Addition
24–3
24.1.2.2
Deletion
24–13
24.2
Addition and Deletion of Power Cabinet
24–13
24.3
Addition and Deletion of M9000 Expansion Cabinet
24–16
Addition and Deletion of CMU, DIMM, IOU, HDD, PCI Cards and TAPEU
25–1
25.1
25.2
Addition
25–2
25.1.1
Active Addition
25.1.2
Cold Addition
Deletion
25–2
25–3
25–4
25.2.1
Active Deletion
25.2.2
Cold Deletion
A. System Configuration
25–5
25–5
A–1
A.1
Installation Conditions
A.2
System Configuration
A–1
A–2
A.2.1
SPARC Enterprise M8000 Server
A–2
A.2.2
SPARC Enterprise M9000 Server (Base Cabinet)
A.2.3
SPARC Enterprise M9000 Server (Base Cabinet + Expansion
Cabinet) A–6
A–4
Contents
xiii
B. Components
B–1
B.1
CPU Memory Board Unit
B.2
CPU Module
B.3
Memory
B–7
B.4
I/O Unit
B–8
B.5
Hard Disk Drive
B.6
PCI Cassette
B.7
IOU Onboard Device Card
B.8
Link Card (External I/O Expansion Unit Connection Card)
B.9
Crossbar Unit
B.10
Clock Control Unit
B.11
XSCF Unit
B.12
CD-RW/DVD-RW Drive Unit
B.13
Tape Drive Unit
B.14
Operator Panel
B.15
Sensor Unit
B.16
Power Supply Unit
B.17
AC Section
B.18
FAN Unit
B.19
Power Cabinet
B.20
Rack-mountable Dual Power Feed
B.21
Backplane
B.22
DDC
B.23
PSU Backplane
B–38
B.24
FAN Backplane
B–39
B.25
Media Backplane
B–42
B.26
Switch Backplane
B–43
B–5
B–10
B–10
B–12
B–14
B–16
B–17
B–19
B–20
B–21
B–23
B–24
B–25
B–30
B–32
B–34
B–35
B–37
C. External Interface Specifications
xiv
B–4
C–1
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B–13
C.1
Serial Port
C–1
C.2
UPC Port
C–2
C.3
USB Port
C–2
C.4
Connection Diagram for Serial Cable
D. UPS Controller
C–3
D–1
D.1
Overview
D–1
D.2
Signal Cable
D.3
Configuration of Signal Lines
D.4
Power Supply Conditions
D–1
D.4.1
Input Circuit
D.4.2
Output Circuit
D.5
UPS Cable
D.6
Connections
D.7
UPC Port
D–2
D–4
D–4
D–4
D–5
D–6
D–7
E. XSCF Unit Replacement When XCP 1040 or 1041 Is in the Server
Abbreviations
Index
E–1
Abbreviations–1
Index–1
Contents
xv
xvi
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Preface
This manual a maintenance manual for the SPARC Enterprise™ M8000/M9000
servers. The manual explains basic operations and detailed replacement procedures
for field-replaceable units (FRUs), which are components that can be replaced at the
customer's site.
As a rule, Fujitsu certified service engineers on the SPARC Enterprise M8000/M9000
servers. However, another person such as the system administrator may perform
some of the simple work, under the direction of one of the service engineers. This
manual is intended for the service engineers and other persons described above.
Notes precede the chapters, sections, and paragraphs that cover the work that only
Fujitsu-certified service engineers are allowed to perform. Follow instructions for all
work.
This section explains:
■
■
■
■
■
■
“Glossary” on page xvii
“Structure and Contents of This Manual” on page xviii
“SPARC Enterprise M8000/M9000 Servers Documentation” on page xxi
“Product Handling” on page xxiv
“Limitations and Cautions” on page xxvi
“Fujitsu Welcomes Your Comments” on page xxvii
Glossary
For the terms used in the “SPARC Enterprise M8000/M9000 Servers
Documentation” on page xxi, refer to the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers Glossary.
xvii
Structure and Contents of This Manual
This manual is organized as described below:
■
PART I Basic Information for Maintenance and Troubleshooting
Provides notes on handling the SPARC Enterprise servers and rules about
operation and descriptions, and it also describes the required tools for
maintenance.
■
Chapter 1 Safety and Tools:
Provides notes on handling the SPARC Enterprise servers and rules about
operation and descriptions, and it describes the required tools for maintenance.
■
Chapter 2 Product Overview and Troubleshooting:
Provides information that is required in troubleshooting.
■
Chapter 3 Periodic Maintenance:
Explains the maintenance work that must be performed regularly regardless of
whether a problem has occurred. The actual work is limited to preventing dust
in the environment from creating pollution.
■
Chapter 4 FRU Removal Preparation:
Explains the required basic operations for replacing components.
■
PART II Maintenance
Explains how to remove and replace FRUs.
■
Chapter 5 Internal Components Access:
Explains how to access each part of the system.
■
Chapter 6 Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM:
Explains how to replace each storage device.
■
Chapter 7 I/O Unit (IOU) Replacement:
Explains the replacement procedures for an I/O unit (IOU).
■
Chapter 8 FAN Unit Replacement:
Explains the replacement procedures for a fan unit.
xviii
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
Chapter 9 Power Supply Unit (PSU) Replacement:
Explains the replacement procedures for a power supply unit (PSU).
■
Chapter 10 Operator Panel Replacement:
Explains the replacement procedures for the operator panel.
■
Chapter 11 XSCF Unit Replacement:
Explains the replacement procedures for an XSCF unit.
■
Chapter 12 Hard Disk Drive (HDD) Replacement:
Explains the replacement procedures for a hard disk drive (HDD).
■
Chapter 13 PCI Slot Device Replacement:
Explains the replacement procedures for a device mounted in a PCI slot of an
IOU.
■
Chapter 14 CD-RW/DVD-RW Drive Unit Replacement:
Explains the replacement procedures for the CD-RW/DVD-RW drive unit.
■
Chapter 15 Tape Drive Unit Replacement:
Explains the replacement procedures for the tape drive unit.
■
Chapter 16 Clock Control Unit Replacement:
Explains the replacement procedure for the clock control unit.
■
Chapter 17 Crossbar Unit Replacement:
Explains the replacement procedure for a crossbar unit.
■
Chapter 18 AC Section Replacement:
Explains the replacement procedures for a fan unit.
■
Chapter 19 DDC Replacement:
Explains the replacement procedure for the DDC.
■
Chapter 20 Backplane Replacement:
Explains the replacement procedure for a backplane.
■
Chapter 21 Sensor Unit Replacement:
Explains the replacement procedure for the sensor unit.
■
Chapter 22 Media Backplane Replacement:
Explains the replacement procedure for the media backplane.
■
Chapter 23 Switch Backplane Replacement:
Explains the replacement procedure for the switch backplane.
Preface
xix
■
Chapter 24 Addition and Deletion of a Rack-mountable Dual Power Feed
Option, Power Cabinet, and M9000 Expansion Cabinet:
Explains the replacement procedures for rack-mountable dual power
feed(RDPF).
■
Chapter 25 Addition and Deletion of CMU, DIMM, IOU, HDD, PCI Cards and
TAPEU:
Explains the procedures for adding a unit to the SPARC Enterprise
M8000/M9000 servers and deleting a unit from the SPARC Enterprise
M8000/M9000 servers.
■
Appendix A System Configuration:
Describes the installation conditions and configuration of the SPARC
Enterprise server.
■
Appendix B Components:
Provides figures showing the components that compose the SPARC Enterprise
servers.
■
Appendix C External Interface Specifications:
Describes the specifications of the connectors provided on the SPARC
Enterprise server unit.
■
Appendix D UPS Controller:
Describes the connection of UPC interface, which controls UPS
(Uninterruptible Power Supply).
■
Appendix E XSCF Unit Replacement When XCP 1040 or 1041 Is in the Server:
Provides a replacement procedure to use when the server uses an older version
of XCP firmware than is present in the replacement XSCFU.
■
Abbreviations
Provides the full spellings of abbreviations used in this manual.
■
Index
Provides keywords and corresponding reference page numbers so that the
reader can easily search for items in this manual as necessary.
xx
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
SPARC Enterprise M8000/M9000
Servers Documentation
The manuals listed below are provided for reference.
Book Titles
Manual Codes
SPARC Enterprise M8000/M9000 Servers Site Planning Guide
C120-H014
SPARC Enterprise Equipment Rack Mounting Guide
C120-H016
SPARC Enterprise M8000/M9000 Servers Getting Started Guide
C120-E323
SPARC Enterprise M8000/M9000 Servers Overview Guide
C120-E324
Important Safety Information for Hardware Systems
C120-E391
SPARC Enterprise M8000/M9000 Servers Safety and Compliance Guide
C120-E326
External I/O Expansion Unit Safety and Compliance Guide
C120-E457
SPARC Enterprise M8000/M9000 Servers Unpacking Guide
C120-E327
SPARC Enterprise M8000/M9000 Servers Installation Guide
C120-E328
SPARC Enterprise M8000/M9000 Servers Service Manual
C120-E330
External I/O Expansion Unit Installation and Service Manual
C120-E329
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI
Build Procedure
C120-E361
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers
Administration Guide
C120-E331
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
User’s Guide
C120-E332
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
Reference Manual
Go to the Web
SPARC Enterprise M4000/M5000/M8000/M9000 Servers Dynamic
Reconfiguration (DR) User’s Guide
C120-E335
SPARC Enterprise M4000/M5000/M8000/M9000 Servers Capacity on
Demand (COD) User’s Guide
C120-E336
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI
User’s Guide
C120-E360
SPARC Enterprise M8000/M9000 Servers Product Notes
Go to the Web
Preface
xxi
xxii
Book Titles
Manual Codes
External I/O Expansion Unit Product Notes
C120-E456
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers
Glossary
C120-E514
SPARC Enterprise /PRIMEQUEST Common Installation Planning
Manual
C120-H007
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
1. Manuals on the Web
The latest versions of all the SPARC Enterprise Series manuals are available at the
following websites.
Global Site
http://www.fujitsu.com/sparcenterprise/manual/
Japanese Site
http://primeserver.fujitsu.com/sparcenterprise/manual/
Note – Product Notes are available on the website only. Please check for the most
recent update on your product.
2. Documentation CD
For the Documentation CD, please contact your local sales representative.
■
SPARC Enterprise M8000/M9000 Servers Documentation CD (C120-E364)
3. Manual on the Enhanced Support Facility x.x CD-ROM disk
■
Remote maintenance service
Book Title
Manual Code
Enhanced Support Facility User's Guide for REMCS
C112-B067
4. Manual (man page) provided in the system
XSCF man page
Note – The man page can be referenced on the XSCF Shell, and it provides the same
content as the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
Reference Manual.
5. Sun Microsystems Software (for Solaris OS, etc.) Related Manuals
http://docs.sun.com
6. Information on Using the RCI function
This manual does not contain an explanation of the RCI build procedure. For
information on using the RCI function, refer to the RCI Build Procedure and RCI
User’s Guide provided on the website.
Preface
xxiii
Product Handling
Maintenance
Caution – Certain tasks in this manual should only be performed by a certified
service engineer. User must not perform these tasks. Incorrect operation of these
tasks may cause electric shock, injury, or fire.
■
Installation and reinstallation of all components, and initial settings
■
Removal of front, rear, or side covers
■
Mounting/de-mounting of optional internal devices
■
Plugging or unplugging of external interface cards
■
Maintenance and inspections (repairing, and regular diagnosis and maintenance)
Caution – The following tasks regarding this product and the optional products
provided from Fujitsu should only be performed by a certified service engineer.
Users must not perform these tasks. Incorrect operation of these tasks may cause
malfunction.
■
Unpacking optional adapters and such packages delivered to the users
■
Plugging or unplugging of external interface cards
Remodeling/Rebuilding
Caution – Any modification and/or recycling of this product and its components
may be carried out only by a certified service engineer and must not be done by the
customer under any circumstances. Otherwise, electric shock, injury or fire may
result.
xxiv
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Emission of Laser Beam (Invisible)
Caution – The main unit and high-speed optical interconnect cabinet contain
modules that generate invisible laser radiation. Laser beams are generated while the
equipment is operating, even if an optical cable is disconnected or a cover is
removed. Do not look at any light-emitting part directly or through an optical
apparatus (e.g., magnifying glass, microscope).
Preface
xxv
Limitations and Cautions
Power Control and Operator Panel Mode Switch
When you use the remote power control utilizing the RCI function or the automatic
power control system (referred to below as APCS), you can disable this remote
power control or the APCS by switching to Service mode on the operator panel.
Disabling these features ensures that you do not unintentionally switch the system
power on or off during maintenance. Note system power off with the APCS cannot
be disabled with the mode switch. Therefore, be sure to turn off automatic power
control via APCS before starting maintenance.
If you switch the mode while using the RCI or the automatic power control, the
system power is controlled as follows.
Function
Mode switch
Locked
Service
RCI
Remote power-on/power-off
operations are enabled.
Remote power-on/power-off
operations are disabled.
Automatic
power control
Automatic power-on/poweroff operations are enabled.
Automatic power-on is
disabled, but power-off
remains enabled.
To use the RCI function, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers RCI Build Procedure and the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers RCI User’s Guide which are
available on the website of manuals.
To use the APCS, see the Enhanced Support Facility User's Guide for Machine
Administration Automatic Power Control Function (Supplement Edition).
xxvi
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Fujitsu Welcomes Your Comments
If you have any comments or requests regarding this document, or if you find any
unclear statements in the document, please state your points specifically on the form
at the following URL.
For Users in U.S.A., Canada, and Mexico:
http://www.computers.us.fujitsu.com/www/support_servers.shtml?support
/servers
For Users in Other Countries:
SPARC Enterprise contact
http://www.fujitsu.com/global/contact/computing/sparce_index.html
Preface
xxvii
xxviii
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
PA RT
I
Basic Information for Maintenance and
Troubleshooting
Part I provides maintenance information, explains methods of problem analysis,
troubleshooting, and basic operations for replacing FRUs.
CHAPTER
1
Safety and Tools
This chapter provides notes on handling the high-end server descriptions and rules
about operation, and it lists the tools required for maintenance.
This information is explained in the following sections:
■
■
■
■
Section 1.1,
Section 1.2,
Section 1.3,
Section 1.4,
“Symbols” on page 1-1
“Precautions” on page 1-4
“Tools Required for Maintenance” on page 1-5
“Antistatic Precautions” on page 1-6
1.1
Symbols
1.1.1
Text Conventions
This manual uses the following fonts and symbols to express specific types of
information.
Fonts/symbols
Meaning
Example
AaBbCc123
What you type, when
contrasted with on-screen
computer output.
This font represents the
example of command input in
the frame.
XSCF> adduser jsmith
1-1
1.1.2
Fonts/symbols
Meaning
Example
AaBbCc123
The names of commands, files,
and directories; on-screen
computer output.
This font represents the
example of command input in
the frame.
XSCF> showuser -P
User Name:
jsmith
Privileges:
useradm
auditadm
Italic
Indicates the name of a
reference manual.
See the SPARC Enterprise
M3000/M4000/M5000/M8000/M
9000 Servers XSCF User’s Guide .
""
Indicates names of chapters,
sections, items, buttons, or
menus.
See Chapter 2, "Product
Overview and
Troubleshooting."
Prompt Notations
The following prompt notations are used in this manual.
1.1.2.1
Shell
Prompt Notations
XSCF
XSCF>
C shell
machine-name%
C shell super user
machine-name#
Bourne shell and Korn shell
$
Bourne shell and Korn shell super user
#
OpenBoot™ PROM
ok
Command syntax
The command syntax is as follows:
■
A variable that requires input of a value is enclosed in <>.
■
An optional element is enclosed in [ ].
■
A group of options for an optional keyword is enclosed in [ ] and delimited by |.
■
A group of options for a mandatory keyword is enclosed in {} and delimited by |.
■
The command syntax is shown in a box.
Example:
XSCF> showuser -a
1-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
1.1.3
Environmental Requirements for Using This
Product
This product is a computer that is intended to be used in a computer room. For
details on the operational environment, see the SPARC Enterprise M8000/M9000
Servers Site Planning Guide.
1.1.4
Conventions for Alert Messages
This manual uses the following conventions to show alert messages, which are
intended to prevent injury to the user or bystanders as well as property damage, and
important messages that are useful to the user.
This indicates a hazardous situation that could result in death or serious
personal injury (potential hazard) if the user does not perform the
procedure correctly.
This indicates a hazardous situation that could result in minor or moderate
personal injury if the user does not perform the procedure correctly. This
signal also indicates that damage to the product or other property may
occur if the user does not perform the procedure correctly.
This indicates information that could help the user to use the product more
effectively.
1.1.4.1
Alert Messages in the Text
An alert message in the text consists of a signal indicating an alert level followed by
an alert statement. Alert messages are indented to distinguish them from regular text
as shown in the following example. Also, a space of one line precedes and follows an
alert statement.
The tasks listed below for this product and optional product provided by Fujitsu
should be performed only by authorized service personnel.
The user must not perform these tasks. Incorrect operation of these tasks may cause
electric shock, injury, or fire.
■
■
■
■
■
Installation and reinstallation of all components
Removal of front, rear, or side covers
Mounting/unmounting of optional internal devices
Connecting/disconnecting of external interface cables
Maintenance (repair and regular diagnosis and maintenance)
Chapter 1
Safety and Tools
1-3
1.2
Precautions
The following notes must be observed in maintenance work:
1.2.1
Operating Environment of the Product
Use the SPARC Enterprise in the correct operating environment. The SPARC
Enterprise are assumed to be used in a computer room. For details of the operating
environment, see the SPARC Enterprise M8000/M9000 Servers Site Planning Guide.
1.2.2
Maintenance
The work listed below is to be performed by authorized service engineers. Persons
who are not authorized service engineers must not perform the work. Otherwise,
electric shock, injury, or fire may result.
■
■
■
■
■
Installation, transport, and initial setup of each device
Removal of the front, rear, or a side cover.
Mounting or removing internal optional components
Connecting or disconnecting an external interface cable
Maintenance (repair, regular diagnosis, and maintenance)
The work listed below is to be performed by authorized service engineers. Persons
who are not authorized service engineers must not perform the work. Otherwise, an
equipment failure may result.
1-4
■
Unpacking or installing products, such as an optional adapter, that are delivered
to the customer
■
Connecting or disconnecting an external interface cable
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
1.2.3
Conversion and Reuse of This Product
If this product is converted or a used article of this product is overhauled for the
purpose of reuse, unexpected injury on users and bystanders or damage to their
property may result.
1.3
Tools Required for Maintenance
The maintenance work described in Chapter 6 to Chapter 24 requires maintenance
software to confirm that the SPARC Enterprise and other components are operating
correctly and to collect status information and log data about the server and
components. The work for mounting, removing, or replacing a specific component
requires screwdrivers, and special tools such as an antistatic wrist strap. These items
are listed in TABLE 1-1.
TABLE 1-1
Maintenance Tools
No.
Name
Use
1
Torque wrench
[8.24 N*m (84 kgf*cm)]
Used to fix the bus bars of the power cabinet.
2
Sockets for 10 mm (M6) torque
wrench
Used to replace the BP_A in the SPARC Enterprise M8000
server.
3
Sockets for 13 mm (M8) torque
wrench
Used to fix the bus bars of the power cabinet.
4
Torque wrench extension
5
Torque screwdriver
[0.2 N*m (2.0 kgf*cm)]
Used to secure the clock cables between the cabinets if the
expansion cabinet of the SPARC Enterprise M9000 server is
mounted.
6
Slotted bit
Used to secure the clock cables between the cabinets if the
expansion cabinet of the SPARC Enterprise M9000 server is
mounted.
7
Wrist strap
For antistatic purposes
8
Conductive mat
For antistatic purposes
9
CPU module replacement tool
For mounting and removing CPU Modules (accessory)
10
SunVTS
Test program
Chapter 1
Safety and Tools
1-5
Caution – Be sure to wear an antistatic wrist strap when handling components.
Place removed components on an antistatic mat. Failure to do so may result in
serious damage or injury.
1.4
Antistatic Precautions
During normal operations, all components mounted in the server, including the
dummy (filler) units, are properly grounded through the chassis.
Prior to performing maintenance, ensure that any static electricity is discharged from
the FRUs to be inserted and the person performing the maintenance. Both must be
properly grounded.
Ensure that the procedures below are followed for proper grounding.
1.4.1
Removing Static Electricity
This section provides the information about removing static electricity.
1. Connect the earth conductor of the antistatic mat to the server grounding port.
(See FIGURE 1-7 to FIGURE 1-10)
Note – Do not use antistatic bags or packaging materials in place of a grounded
antistatic mat when handling the FRUs.
1-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 1-1
Conductive Mat
2. Connect an antistatic wrist strap clip to a server grounding port. See
Section 1.4.2, “Grounding Port Connection Locations” on page 1-11.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
Chapter 1
Safety and Tools
1-7
3. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin.
The wrist strap should be snug around the wrist so that it does not rotate.
FIGURE 1-2
Antistatic Wrist Strap Showing the Metallic Underside
metallic surface
1.4.1.1
Removing Static Electricity on a CMU and an IOU
1. Prior to mounting a new CMU or IOU, place it on the grounded antistatic mat.
1-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2. Touch the metallic chassis for 5 or more seconds with your bare hand wearing
the antistatic wrist strap. (See FIGURE 1-3 or FIGURE 1-4)
You cannot remove static electricity by touching the label.
FIGURE 1-3
Metallic Chassis (CMU)
Touch for 5 or more seconds.
FIGURE 1-4
Metallic Chassis (IOU)
Touch for 5 or more seconds.
Chapter 1
Safety and Tools
1-9
3. Touch each of the designated points on the guide blocks for 5 or more seconds
with your bare hand wearing the antistatic wrist strap. (See FIGURE 1-5 or
FIGURE 1-6)
FIGURE 1-5
Guide Block (CMU)
Touch for 5 or more seconds.
FIGURE 1-6
Guide Block (IOU)
Touch 5 or more seconds.
1-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
1.4.2
Grounding Port Connection Locations
FIGURE 1-7 to FIGURE 1-10 show the locations of the grounding port on each server.
The grounding port can be used for the grounding of the antistatic wrist strap and
the antistatic mat.
If the type of clip does not securely fit the grounding port, it can also be connected
to the grounding wire for the door. For details, see Section 5.3, “How to Remove a
Door” on page 5-3.
FIGURE 1-7
M8000 Grounding Port Connection Locations for the Wrist Strap Clip and the
Antistatic Mat (Front View)
Chapter 1
Safety and Tools
1-11
1-12
FIGURE 1-8
M8000 Grounding Port Connection Locations for the Wrist Strap Clip and the
Antistatic Mat (Rear View)
FIGURE 1-9
M9000 Grounding Port Connection Locations for the Wrist Strap Clip and the
Antistatic Mat (Front View)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 1-10
M9000 Grounding Port Connection Locations for the Wrist Strap Clip and the
Antistatic Mat (Rear View)
Chapter 1
Safety and Tools
1-13
1-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
2
Product Overview and
Troubleshooting
This chapter provides information that is required in troubleshooting.
This information is explained in the following sections:
■
■
■
■
■
■
■
■
■
2.1
Section 2.1,
Section 2.2,
Section 2.3,
Section 2.4,
Section 2.5,
Section 2.6,
Section 2.7,
Section 2.8,
Section 2.9,
“System Views” on page 2-1
“Labels” on page 2-11
“Operator Panel” on page 2-16
“Determining Which Diagnostics Methods To Use” on page 2-21
“Checking the Server and System Configuration” on page 2-23
“Error Conditions” on page 2-26
“LED Error Display” on page 2-30
“Using the Troubleshooting Commands” on page 2-34
“Traditional Solaris Troubleshooting Commands” on page 2-37
System Views
This section provides views of the high-end server. The figures can be used to locate
the component in the server to be subjected to maintenance.
In terms of its structure, the high-end server consists of a cabinet that includes
various mounted components and a front door, rear door, and side covers that
protect the mounted components. The side covers are removed when cabinets are
connected to each other or when the dual power feed option is connected to the
cabinet. The operator panel, which is mounted on the front door, is always
accessible. Each door can be locked with a key so that only the administrator can
open it.
2-1
The front and rear views of FIGURE 2-1, FIGURE 2-2, FIGURE 2-4, FIGURE 2-5, FIGURE 2-7,
and FIGURE 2-8 include names and abbreviations for field-replaceable units (FRUs).
Components that are mounted inside the system are shown FIGURE 2-3, FIGURE 2-6,
and FIGURE 2-9. The abbreviations are used in messages and the like. If multiple
FRUs of the same type are mounted, the number sign # and a sequential number is
added to their names to distinguish them from one another. Owing to the reduced
scale, certain components (FRUs) are difficult to show in the figures. Accordingly,
the layout of these components as viewed from one side is indicated in the table
connected by a lead line to the component location.
2-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2.1.1
SPARC Enterprise M8000 Server
FIGURE 2-1
Front View - M8000
PSU
DDC
XSCFU
CMU
TAPEU
DVDU
FAN_A
SNSU
FAN_B
Air Filter
Chapter 2
Product Overview and Troubleshooting
2-3
FIGURE 2-2
Rear View - M8000
ACS
FAN_B
IOU
Air Filter
2-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 2-3
Internal View - M8000
PSUBP_B
PSUBP_A
FANBP_C
BP_A
MEDBP
SWBP
FANBP_C
Chapter 2
Product Overview and Troubleshooting
2-5
2.1.2
SPARC Enterprise M9000 Server (Base Cabinet)
FIGURE 2-4
PSU
Front View - M9000 (Base Cabinet)
FAN_A
ACS
XBU
CLKU
XSCFU
IOU
TAPEU
DVDU
Air Filter
SNSU
2-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 2-5
Rear View - M9000 (Base Cabinet)
FAN_A
CMU
IOU
Air Filter
Chapter 2
Product Overview and Troubleshooting
2-7
FIGURE 2-6
Internal View - M9000 (Base Cabinet)
FANBP_B
PSUBP_A
FANBP_A
BP_B
MEDBP
SWBP
2-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2.1.3
SPARC Enterprise M9000 Server (Expansion
Cabinet)
FIGURE 2-7
Front View - M9000 (with the Expansion Cabinet)
FAN_A
PSU
ACS
cable support bracket
XBU
CLKU
XSCFU
IOU
TAPEU
DVDU
Air Filter
SNSU
Chapter 2
Product Overview and Troubleshooting
2-9
FIGURE 2-8
Rear View - M9000 (with the Expansion Cabinet)
FAN_A
CMU
IOU
Air Filter
2-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 2-9
Internal View - M9000 (with the Expansion Cabinet)
FANBP_B
PSUBP_A
FANBP_A
BP_B
MEDBP
SWBP
2.2
Labels
2.2.1
System Name Plate Label, Rating Label, ID Label
(Japan) or EZ Label (besides Japan), and Standard
Label
The important labels affixed on this server are shown in FIGURE 2-10 and FIGURE 2-11.
The actual description on the labels may differ from FIGURE 2-10 and FIGURE 2-11.
■
The system name plate label includes the model number, serial number, and
hardware version, all of which are required for maintenance and management.
■
The rating label, which is affixed near the AC power supply, includes the power
input rating for the AC power supply.
Chapter 2
Product Overview and Troubleshooting
2-11
■
The ID label or EZ label is affixed on the front door of the server, and it includes
the model name and serial number, both of which are written on the system name
plate label.
ID label (Japan)
■
EZ label (besides Japan)
The standard label is affixed near the system name plate label, and it includes the
certification standards that apply:
Safety: NRTL/C
Electrical interference: VCCI-A, FCC-A, DOC-A, and MIC
Safety and electrical interference: CE
2-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 2-10
M8000 Label Location
System Name Plate Label
Front
Rear
Standard label
Chapter 2
Product Overview and Troubleshooting
2-13
FIGURE 2-11
M9000 Label Location
System Name Plate Label
Front
Rear
Standard label
2-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2.2.2
Labels About Handling
The labels shown below, which are affixed on the high-end server, provide field
engineers with important information on component removal and mounting.
Caution – Never peel off the labels.
■
Removing and installing a CPU/memory board unit (CMU)
■
Removing a crossbar unit (XBU)
Chapter 2
Product Overview and Troubleshooting
2-15
■
2.3
Removing an I/O unit (IOU)
Operator Panel
The operator panel controls the high-end server power. The operator panel is usually
locked with a key to prevent the server from being mistakenly powered off through
an operator error during system operation.
Before starting maintenance work, ask the system administrator to unlock the
operation panel.
2.3.1
Operator Panel Location
FIGURE 2-12 indicates the location of the operator panel (OPNL) of the high-end
servers. The expansion cabinet is not equipped with the operator panel.
2-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 2-12
Operator Panel Location (at the Front of M8000)
OPNL
2.3.2
Appearance and Operations
The operator panel can be used while the front door of the server is closed. Field
engineers, and the system administrator use the operation panel to check the
operating state of the server and to perform system power operations. To check the
operating state of the server, look at the LEDs. The operating state of the server is
checked by observing the LEDs, and the power supply is operated with the POWER
switch. FIGURE 2-13 shows the appearance of the operator panel.
Chapter 2
Product Overview and Troubleshooting
2-17
FIGURE 2-13
2.3.3
Operator Panel
LED
TABLE 2-1 lists the states of the server that are displayed with the LEDs on the
operator panel.
The blinking period is one second (frequency of 1 Hz).
Besides the states listed in TABLE 2-1, the operator panel also displays various states
of the server using combinations of the three LEDs. TABLE 2-2 indicates the states that
are usually displayed in the course of operation from the power-on to power-off of
the high-end server.
TABLE 2-1
LED
State Display by the LEDs (Operator Panel)
Name
Light color
Description of function and state
POWER
Green
Indicates whether power to the SPARC Enterprise server is on.
STANDBY
XSCF
2-18
Green
Off
Indicates the power-off state.
Lit
Indicates the power-on state.
Blinking
The power-off sequence is in progress.
Indicates whether the XSCF can be powered on.
Off
Indicates that the system cannot be powered on.
Blinking
Indicates that initialization processing of the SPARC
Enterprise server is in progress after main line
switches were switched on.
Lit
Indicates that the system can be powered on.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE 2-1
LED
State Display by the LEDs (Operator Panel) (Continued)
Name
Light color
Description of function and state
CHECK
Amber
Indicates the operating status of the SPARC Enterprise server.
Off
Normal state. Otherwise, this indicates that the main
line switches were switched off or a power failure
occurred.
Blinking
(*1)
Indicates that the operator panel is the maintenance
target device.
Lit
Indicates that the server cannot be started.
Note – *1)If the maintenance target component is indicted by a blinking CHECK
LED, the LED may be called a locater.
TABLE 2-2
State Display by LED Combination (Operator Panel)
LED
POWER
XSCF
STANDBY
CHECK
Description of the state
Off
Off
Off
The main line switch is switched off.
Off
Off
On
The main line switch is switched on.
Off
Blinking
Off
The XSCF is being initialized.
Off
Blinking
On
An error occurred in the XSCF.
Off
On
Off
• The XSCF is on standby.
• The system is waiting for power-on of the air conditioning
system.
On
On
Off
• Warm-up standby processing is in progress (power-on is
delayed).
• The power-on sequence is in progress.
• The system is in operation.
Blinking
On
Off
• The power-off sequence is in progress.
• Fan termination is being delayed.
Chapter 2
Product Overview and Troubleshooting
2-19
2.3.4
Switch
The operator panel has the mode switch, which sets the operation mode, and the
POWER switch, which is used to power on and off the system.
TABLE 2-3
Switch
Switches (Operator Panel)
Name
Description of function
Mode
This key switch is used to set an operation mode for the server.
Insert the special key that is under the customer’s control, to
switch between modes.
Locked
Normal operation mode
• The system can be powered on with the POWER switch, but
it cannot be powered off with the POWER switch.
• The key can be pulled out at this key position.
Service
Mode for maintenance
• The system can be powered on and off with the POWER
switch.
• The key cannot be pulled out at this key position.
• Maintenance is performed in Service mode while the server
is stopped.
POWER
This switch is used to control the server power.
Power-on and power-off are controlled by pressing this switch
in different patterns, as described below.
Holding down for a Regardless of the mode switch state, the server (all domains) is
short time
powered on.
(less than 4 seconds) At this time, processing for waiting for facility (air
conditioners) power-on and warm-up completion is
skipped.(*1)
Holding down for a
long time in Service
mode
(4 seconds or
longer)
• If power to the server is on (at least one domain is
operating), shutdown processing is executed for all domains
before power-off processing.
• If the system is being powered on, the power-on processing
is cancelled, and the system is powered off.
• If the system is being powered off, the operation of the
POWER switch is ignored, and the power-off processing is
continued.
Note – *1)In normal operation, the server is powered on only when the computer
room environmental conditions satisfy the specified values. Then, the server
remains in the reset state until the operating system is booted.
2-20
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE 2-4
Meanings of the Mode Switch
Function
Mode Switch
Locked
Locked
Inhibition of Break Signal Reception
Enabled. Reception of the break signal
can be enabled or disabled for each
domain using setdomainmode.
Disabled
Power On/Off by power switch
Only power on is enabled
Enabled
2.4
Determining Which Diagnostics
Methods To Use
When an error occurs, a message is often displayed on the monitor. Use the
flowcharts in FIGURE 2-14 to find the correct methods for diagnosing problems.
Chapter 2
Product Overview and Troubleshooting
2-21
FIGURE 2-14
Diagnostic Method Flow Chart
Start
OS panic occurred or there’s an
error on performance
YES
e-mail sent or not by
XSCF mail function?
NO
Check OS console and XSCF
console for error information displayed
NO
Execute showlogs or fmadm
on XSCF to display the fault
information
Check /var/adm/messages
on Solaris OS
FMA message?
YES
Is there error message
on XSCF console?
NO
YES
Use fmadm ?
YES
NO
Execute fmadm to display fault
information
NO
Message ID
available?
Write down the displayed fault
information
YES
Enter Message ID in
Collect information about your
server
http://sun.com/msg/ to
refer to fault information
Contact service engineer
NO
Trouble
resolved?
YES
2-22
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
End
2.5
Checking the Server and System
Configuration
Before and after maintenance work, the state and configuration of the server and
components should be checked and the information saved. For recovery from a
problem, conditions related to the problem and the repair status must be checked.
The operating conditions must remain the same before and after maintenance.
A functioning server without any problems should not display any error conditions.
For example:
2.5.1
■
The syslog file should not display error messages.
■
* mark is not displayed in SCF shell command showhandconf
■
The administrative console should not display error messages.
■
The server processor logs should not display any error messages.
■
The Solaris™ Operating System (Solaris OS) message files should not indicate any
additional errors.
Checking the Hardware Configuration and FRU
Status
To replace a faulty component and perform the maintenance on the server it is
important to check and understand the hardware configuration of the server and the
state of each hardware component.
The hardware configuration refers to information that indicates to what layer a
component belongs in the hardware configuration.
The status of each hardware component refers to information on the condition of the
standard or optional component in the server: temperature, power supply voltage,
CPU operating conditions, and other times.
The hardware configuration and the status of each hardware component can be
checked from the maintenance terminal using XSCF Shell commands.
TABLE 2-5 lists commands for checking the hardware configuration and status. For
details, see the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
Reference Manual.
Chapter 2
Product Overview and Troubleshooting
2-23
TABLE 2-5
Commands for Checking Hardware Configuration and Status
Command
Description
showhardconf
Displays the system layer that includes a faulty component.
showstatus
Displays the status of a component. This command is used to check only a faulty
component.
showboards
Displays the use status of individual devices and resources.
showdcl
Displays domain configuration information (hardware resource information).
showfru
Displays device setting information.
ioxadm
Displays the FRU status of external I/O expansion unit as normal or abnormal.
Also some conditions can be checked based on the lit and/ or blinking state of the
component LEDs (TABLE 2-11 and TABLE 2-12).
2.5.1.1
Checking the Hardware Configuration
Login authority is required to check the hardware configuration. The following
procedure for these checks can be made from the maintenance terminal. Ask the
system administrator for necessary information, such as a password. For the detailed
procedure, see the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers
XSCF User’s Guide.
1. Log in to the XSCF.
2. Execute the showhardconf command.
XSCF> showhardconf
The showhardconf command will print the hardware configuration information
to the screen. See the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers
XSCF User’s Guide for more detailed information.
2.5.2
Checking the Software and XSCF Firmware
Configurations
The software and firmware configurations and versions affect the operation of the
server. To change the configuration or investigate a problem, check the latest
information and check for any problems in the software.
Software and firmware varies according to users.
2-24
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
The software configuration and version can be checked in the Solaris Operating
System. Refer to the Solaris OS documentation for more information.
■
The firmware configuration and versions can be checked from the maintenance
terminal using XSCF Shell commands. Refer to the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide for more detailed
information.
Check the software and firmware configuration information with assistance from the
system administrator. However, if you have received login authority from the system
administrator, the following commands can be used from the maintenance terminal
for these checks:
TABLE 2-6
Commands for Checking the Software Configuration
Command
Description
showrev(1M)
Displays information on patches applied to the system.
uname(1)
Outputs current information regarding the system to the standard output.
TABLE 2-7
Commands for Checking the XSCF Firmware Configuration
Command
Description
version(8)
XSCF Shell command that outputs the current firmware version information.
showhardconf(8)
XSCF Shell command that displays what layer of the system includes a faulty
component.
showstatus(8)
XSCF Shell command that displays the status of a component. This command is used
when only a faulty component is to be checked.
showdcl(8)
XSCF Shell command that displays the configuration information of a domain
(hardware resource information).
showfru(8)
XSCF Shell command that displays the setting information of a device.
2.5.2.1
Checking the Software Configuration
The following procedure for these checks can be made from any terminal window
terminal.
1. Execute the showrev command.
# showrev
The showrev command will print the system configuration information to the
screen.
Chapter 2
Product Overview and Troubleshooting
2-25
2.5.2.2
Checking the Firmware Configuration
Login authority is required to check the firmware configuration. The following
procedure for these checks can be made from the maintenance terminal:.
1. Log in to the XSCF.
2. Execute the version command.
XSCF> version
The version command will print the firmware version information to the screen.
See the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s
Guide for more detailed information.
2.5.3
Downloading the Error Log Information
If you want to download the error log information, use the XSCF log fetch function.
The XSCF Unit has an interface with external units so that a service engineer can
easily obtain useful maintenance information such as error logs
Connect the maintenance terminal, and use the CLI or BUI to issue a download
instruction to the maintenance terminal to download Error Log information over the
XSCF-LAN.
Note – When the XSCF unit has a redundant configuration, log in also to the
standby XSCF and obtain the log file in the same manner.
2.6
Error Conditions
This section describes error conditions and relevant corrective actions.
This work is explained in the following sections:
■
■
■
Section 2.6.1, “Predictive Self-Healing Tools” on page 2-27
Section 2.6.2, “Monitoring Output” on page 2-28
Section 2.6.3, “Messaging Output” on page 2-29
Details of the fault information, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
2-26
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
You can find more detailed descriptions of Solaris OS Predictive Self-Healing at the
website below:
http://www.sun.com/bigadmin/features/articles/selfheal.html
Predictive self-healing is an architecture and methodology for automatically
diagnosing, reporting, and handling software and hardware fault conditions. This
new technology lessens the time required to debug a hardware or software problem
and provides the administrator and technical support with detailed data about each
fault.
2.6.1
Predictive Self-Healing Tools
In Solaris OS, the fault manager runs in the background. If a failure occurs, the
system software recognizes the error and attempts to determine what hardware is
faulty. The software also takes steps to prevent that component from being used
until it has been replaced. Some of the specific activities the software takes include:
■
Receives telemetry information about problems detected by the system software
■
Diagnoses the problems
■
Initiates pro-active self-healing activities. For example, the fault manager can
disable faulty components.
The state of a FRU, group of FRUs, or part of a FRU, that has been isolated
because a fault was detected. The isolation is usually done to prevent possibly
faulty components from affecting other system components. The part that is
isolated is not always the faulty part alone; a normal part may be degraded to
isolate the faulty part. If a function required for the operation of the system is
degraded, a system failure may result.
■
When possible, causes the faulty FRU to provide an LED indication of a fault in
addition to populating the system console messages with more details
TABLE 2-8 shows a typical message generated when a fault occurs. The message
appears on your console and is recorded in the /var/adm/messages file.
Note – The message in TABLE 2-8 indicates that the fault has already been diagnosed.
Any corrective action that the system can perform has already taken place. If your
server is still running, it continues to run.
Chapter 2
Product Overview and Troubleshooting
2-27
TABLE 2-8
Predictive Self Healing Message
Output displayed
Description
Nov 1 16:30:20 dt88-292 EVENT-TIME: Tue Nov 1 16:30:20 PST 2005
EVENT-TIME: the time stamp of
the diagnosis.
Nov 1 16:30:20 dt88-292 PLATFORM: SUNW,A70, CSN: -, HOSTNAME:
dt88-292
PLATFORM: A description of the
server encountering the problem.
Nov 1 16:30:20 dt88-292 SOURCE: eft, REV: 1.13
SOURCE: Information on the
Diagnosis Engine used to
determine the fault.
Nov 1 16:30:20 dt88-292 EVENT-ID:
afc7e660-d609-4b2f-86b8-ae7c6b8d50c4
EVENT-ID: The Universally
Unique event ID for this fault.
Nov 1 16:30:20 dt88-292 DESC:
Nov 1 16:30:20 dt88-292 A problem was detected in the PCI-Express
subsystem
DESC: A basic description of the
failure.
Nov 1 16:30:20 dt88-292 Refer to http://sun.com/msg/SUN4-8000-0Y for
more information.
WEBSITE: Where to find specific
information and actions for this
fault.
Nov 1 16:30:20 dt88-292 AUTO-RESPONSE: One or more device instances AUTO-RESPONSE: What, if
may be disabled
anything, the system did to
alleviate any follow-on issues
Nov 1 16:30:20 dt88-292 IMPACT: Loss of services provided by the device
instances associated with this fault
IMPACT: A description of what
that response might have done.
Nov 1 16:30:20 dt88-292 REC-ACTION: Schedule a repair procedure to
replace the affected device. Use Nov 1 16:30:20 dt88-292 fmdump -v -u
EVENT_ID to identify the device or contact Sun for support.
REC-ACTION: A short description
of what the system administrator
should do.
2.6.2
Monitoring Output
To understand error conditions, collect monitoring output information, by using the
commands shown below.
TABLE 2-9 lists the commands for checking the monitoring output.
2-28
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE 2-9
Command
Commands for Checking the Monitoring Output
Operand
showlogs(8) console
2.6.3
Description
XSCF firmware collects console logs of console messages that were output
through the XSCF. This command collects all the console messages displayed to
users.
monitor
Logs the messages displayed in the message window of the BUI/CLI.
panic
Saves as panic logs the console logs that are logged when a reset is received after
a panic notification.
ipl
Collects the console data generated during a period from power-on of a domain
to completion of operating system startup (system running).
Messaging Output
To understand error conditions, collect messaging output information, by using the
commands shown below.
TABLE 2-10 lists the commands for checking the messaging output.
Chapter 2
Product Overview and Troubleshooting
2-29
Commands for Checking the Messaging Output
TABLE 2-10
Command
Operand
Description
showlogs
env
Collects the temperature history log. The SPARC Enterprise server environmental
temperature data and power status are collected at a 10-minute interval. The data
is stored for a maximum of six months.
power
Collects the log of power events and reset events. The target range covers the
SPARC Enterprise server, External I/O Expansion units, and UPSs.
event
Collects the message which accompanies the command or the progress of
operation such as Dynamic Reconfiguration (DR), the status of operation on the
operator panel, the event such as the shut down request to OS due to power
failure or abnormal temperature, as event log. This information is used to analyze
faults and investigate the use status of individual devices at a customer's site, and
it is kept as a maintenance work history.
error
Information on the SPARC Enterprise server hardware faults detected by the SCF,
POST/OpenBoot PROM, or ESF machine management and software monitoring
error information are logged as SCF error logs. The showlogs error command
can display with hexadecimal codes the error information stored in the SCF error
log and information on faulty components.
fmdump(1M)
fmdump(8)
Hardware and software are automatically diagnosed according to the fault
management architecture (FMA), and the diagnosis results and errors are
automatically recorded. The fmdump command can display the recorded
information. It is provided as a Solaris OS command and XSCF Shell command.
The information can be checked at the site at the specified URL by using a
displayed message ID.
Each error message logged by the predictive self-healing architecture has a code
associated with it as well as a web address that can be followed to get the most
up-to-date course of action for dealing with that error.
Refer to the Solaris OS documentation for more information on predictive
self-healing.
2.7
LED Error Display
This section explains the LEDs of each FRU that are to be checked when the relevant
FRU is replaced. Each LED can be checked after the door of a cabinet is opened.
Whether the state of the entire system is normal can be learned by checking the
operator panel (outside). When an error occurs in an individual hardware
component in the system, the LEDs of the FRU containing the hardware component
2-30
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
that has caused the error indicate that an error has occurred. The LEDs on the
operator panel (back) indicate the status of the operator as a single unit. However,
some FRUs like DIMMs do not have LEDs.
Whether a FRU without LEDs is in the normal state can be checked by executing the
XSCF Shell commands showhardconf and ioxadm from a maintenance terminal.
For details of the commands, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference Manual.
2.7.1
When target FRU is indicated by LEDs
When an error message is displayed at the system console and the cause of the error
is in hardware, a faulty FRU must be removed and replaced. Each FRU is equipped
with an LED to indicate whether an error has occurred in the FRU and an LED to
indicate whether the FRU can be removed. Most FRUs are named READY LED and
CHECK LED. In some cases, names are not indicated but the icons are always
printed or icon labels are always affixed. Such FRUs include the back of the operator
panel, XSCFUs, CMUs, XBUs, CLKUs, FANs, and HDDs.
2.7.2
When target FRU is not indicated by LEDs
For some FRUs, the READY LED and CHECK LED are not used as the names of the
LEDs that are checked at replacement. Even in such a case, the same icons as those
for the READY LED and CHECK LED are used so that the meaning of LEDs can be
understood. Even if the names of LEDs are not indicated, the icons are always
printed or icon labels are always affixed.
TABLE 2-11
LED Display That Should Be Checked When a FRU Is Replaced (Common)
LED
Display and meaning
READY
(green)
Indicates whether the unit is operating (whether it is configured into the system).
Lit
Indicates that the FRU is operating. The FRU cannot be disconnected and
removed from the system. Therefore, the FRU cannot be replaced.
Blinking
Indicates that the FRU is being configured into the system (or, for an XSCFU,
being initialized) or being disconnected from the system. However, for a PSU, it
indicates that the main line switch has been switched on.
Off
Indicates that the FRU is stopped and disconnected from the system. Therefore,
the FRU can be replaced.
Chapter 2
Product Overview and Troubleshooting
2-31
TABLE 2-11
LED Display That Should Be Checked When a FRU Is Replaced (Common) (Continued)
LED
Display and meaning
CHECK
(amber)
Indicates either that the unit contains an error or that the unit is a target device for replacement.
Lit
Indicates that an error has been detected in the hardware of the FRU. (For an
HDD, the LED is lit according to the instruction from the software or
middleware.)
Blinking (*1) Indicates that the FRU is to be replaced.
Off
Indicates that the state of the FRU is normal.
Note – *1)If the maintenance target component is indicted by a blinking CHECK
LED, the LED may be called a locater.
TABLE 2-12
Status Display of LEDs Defined Individually for Each FRU
LED
FRU
Type
Display
Meaning
XSCFU
READY
Lit (green)
Indicates that the XSCFU is in use. In this state, the
XSCFU cannot be removed (cannot be replaced).
Blinking
(green)
Indicates that the XSCFU is being initialized.
Off
Indicates that the XSCFU can be replaced.
Lit (amber)
Indicates that an error was detected in the XSCFU.
However, this LED remains on for a few minutes
immediately after power-on (until the start of
initialization). It does not indicate an error during that
time.
Blinking
(amber)
Indicates that the XSCFU is a replacement target.
Off
Indicates that the XSCFU is in the normal state.
Lit (green)
Indicates that the XSCFU is in use (active).
Off
Indicates that the XSCFU is on standby.
CHECK
ACTIVE
2-32
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE 2-12
Status Display of LEDs Defined Individually for Each FRU (Continued)
LED
FRU
Type
Display
Meaning
XSCFU and IOU
(display part for
LAN)
ACT
Lit (green)
Indicates that communication is being performed
through the Ethernet port (LAN port).
Off
Indicates that no communication is being performed
through the Ethernet port (LAN port).
Lit (amber)
Only for an IOU: Indicates that the communication
speed is 1G bps.
Lit (green)
Indicates that the communication speed is 100M bps.
Off
Indicates that the communication speed is 10M bps.
Lit (green)
Indicates that the HDD is in operation. In this state, the
HDD cannot be removed (cannot be replaced).
Blinking
(green)
Indicates that the HDD is being connected. In this state,
the HDD cannot be removed (cannot be replaced).
Off
Indicates that the HDD can be replaced.
Lit (amber)
Indicates that an error was detected in the HDD.
However, this LED remains on for a few minutes
immediately after power-on (until the start of
initialization). It does not indicate an error during that
time.
Blinking
(amber)
Indicates that the HDD is a replacement target.
Off
Indicates that the HDD is in the normal state.
Lit (green)
Indicates that power is being supplied to the PCI slot.
Off
Indicates that the PCI card in the PCI slot is stopped.
Lit (amber)
Indicates that an error occurred in the hardware of the
PCI slot.
Blinking
(amber)
Indicates that the PCI card in this PCI slot is a device to
be replaced.
Off
Indicates that the hardware of the PCI slot is normal.
LINK SPEED
HDD
READY
OK
CHECK
PCI card
(Power)
(inside an external
I/O expansion unit)
(Attention)
Chapter 2
Product Overview and Troubleshooting
2-33
TABLE 2-12
Status Display of LEDs Defined Individually for Each FRU (Continued)
LED
FRU
Type
Display
Meaning
PSU: power supply
unit
POWER
Lit (green)
Indicates that the power to the system is turned on and
being supplied.
Blinking
(green)
Indicates that power is being supplied to the PSU, but
the PSU is not turned on.
Off
Indicates that power is not being supplied to the PSU.
Lit (amber)
Indicates that an error occurred in the PSU.
Maintenance can be performed.
Off
Indicates that the PSU is normal.
Lit (amber)
Indicates that the rotational speed of the cooling fan in
the PSU is abnormal.
Off
Indicates that the rotational speed of the cooling fan in
the PSU is normal.
FAIL
PRFL
2.8
Using the Troubleshooting Commands
After the message in TABLE 2-8 is displayed, you might desire more information
about the fault. For complete information about troubleshooting commands, refer to
the Solaris OS man pages or the XSCF Shell man pages. This section describes some
details of the following commands:
■
■
■
■
2.8.1
Section 2.8.1,
Section 2.8.2,
Section 2.8.3,
Section 2.8.4,
“Using
“Using
“Using
“Using
the
the
the
the
showlogs Command” on page 2-34
fmdump Command” on page 2-35
fmadm Command” on page 2-35
fmstat Command” on page 2-37
Using the showlogs Command
The showlogs command will display the contents of a specified log in order of
timestamp starting with the oldest date. The showlogs command will display the
following logs:
■
■
■
■
■
2-34
error log
power log
event log
temperature and humidity record
monitoring message log
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
■
■
console message log
panic message log
IPL message log
XSCF> showlogs error
Dated: Mar 30 12:45:31 JST 2005 Code: 00112233-44556677-8899aabbcceeff0
Status: Alarm
Component: PSU#1,PSU#2
Msg: ACFAIL occurred (ACS=3)(FEP type = A1)
Date: Mar 30 17:45:31 JST 2005 Code: 00112233-44556677-8899aabbcceeff0
Status: Faulted
Component: PSU#1,PSU#2,*
Msg: ACFAIL occurred (ACS=3)(FEP type = A1)
XSCF>0
2.8.2
Using the fmdump Command
The fmdump command can be used to display the contents of any log files associated
with the Solaris Fault Manager.
The fmdump command produces the following output. This example assumes there
is only one fault.
# fmdump
TIME UUID SUNW-MSG-ID
Nov 02 10:04:15.4911 0ee65618-2218-4997-c0dc-b5c410ed8ec2 SUN4-8000-0Y
2.8.3
Using the fmadm Command
This section describes the use of the fmadm command.
The administrator and all service personnel can use the fmadm command. This
command can display and change the system configuration parameters managed by
the Solaris Fault Manager.
2.8.3.1
fmadm config Command
The fmadm config command outputs the version and status of the diagnostic
engine used by the server. To determine whether the latest diagnostic engine is
running, compare the version with the information on the SunSolve website.
Chapter 2
Product Overview and Troubleshooting
2-35
# fmadm config
MODULE
cpumem-diagnosis
cpumem-retire
eft
fmd-self-diagnosis
io-retire
syslog-msgs
2.8.3.2
VERSION
1.5
1.0
1.13
1.0
1.0
1.0
STATUS
active
active
active
active
active
active
DESCRIPTION
UltraSPARC-III/IV CPU/Memory Diagnosis
CPU/Memory Retire Agent
eft diagnosis engine
Fault Manager Self-Diagnosis
I/O Retire Agent
Syslog Messaging Agent
fmadm faulty Command
The fmadm faulty command can be used mainly to identify the status of faulty
components.
In the following example, the PCI card is degraded and associated with the
following UUID 49847040-ce57-e453-9adc-fe66c7c65384. Also, the "faulted"
state may be displayed.
# fmadm faulty
STATE
RESOURCE / UUID
-----------------------------------------------------degraded dev:////pci@2,600000
49847040-ce57-e453-9adc-fe66c7c65384
Note – The error information remains in the Solaris OS even when replacement of
the faulty component is completed. Identify the UUID by executing the fmadm
faulty command, and reset the error information by executing the fmadm repair
command with the UUID specified.
2.8.3.3
fmadm repair Command
The fmadm repair command can be used to reset the error information for a faulty
component in the Solaris OS.
# fmadm repair 49847040-ce57-e453-9adc-fe66c7c65384
fmadm: recorded repair to 3de29de5-6332-ec64-9b49-bacc739fe3c3
2-36
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Note – The error information remains in the Solaris OS even when replacement of
the faulty component is completed. Identify the UUID by executing the fmadm
faulty command, and reset the error information by executing the fmadm repair
command with the UUID specified.
2.8.4
Using the fmstat Command
The fmstat command can report statistics associated with the Solaris Fault
Manager. The fmstat command shows information about DE performance. In the
example below, the eft DE (also seen in the console output) has received an event
which it accepted. A case is "opened" for that event and a diagnosis is performed to
"solve" the cause for the failure.
# fmstat
module
ev_recv
cpumem-diagnosis
0
cpumem-retire
0
eft
1
fmd-self-diagnosis
0
io-retire
1
syslog-msgs
1
2.9
ev_acpt
0
0
1
0
0
0
wait svc_t
0.0
0.0
0.0
0.0
0.0 1191.8
0.0
0.0
0.0
32.4
0.0
0.5
%w
0
0
0
0
0
0
%b
0
0
0
0
0
0
open
0
0
1
0
0
0
solve
0
0
1
0
0
0
memsz
3.0
0
3.3M
0
37b
32b
bufsz
K0
0
11K
0
0
0
Traditional Solaris Troubleshooting
Commands
These superuser commands can help you determine if you have issues in your
workstation, in the network, or within another server that you are networking with.
The following commands are described in this section:
■
■
■
■
■
■
■
Section 2.9.1,
Section 2.9.2,
Section 2.9.3,
Section 2.9.4,
Section 2.9.5,
Section 2.9.6,
Section 2.9.7,
“iostat Command” on page 2-38
“prtdiag Command” on page 2-39
“prtconf Command” on page 2-41
“netstat Command” on page 2-43
“ping Command” on page 2-44
“ps Command” on page 2-46
“prstat Command” on page 2-47
Most of these commands are located in the /usr/bin or /usr/sbin directories.
Chapter 2
Product Overview and Troubleshooting
2-37
2.9.1
iostat Command
The iostat command iteratively reports terminal, drive, and tape I/O activity, as
well as CPU utilization.
2.9.1.1
Options
TABLE 2-13 describes options for the iostat command and how those options can
help troubleshoot the server.
TABLE 2-13
Options for iostat
Option
Description
How It Can Help
No option
Reports status of local I/O devices.
A quick three-line output of device
status.
-c
Reports the percentage of time the system has spent in user
mode, in system mode, waiting for I/O, and idling.
Quick report of CPU status.
-e
Displays device error summary statistics. The total errors,
hard errors, soft errors, and transport errors are displayed.
Provides a short table with
accumulated errors. Identifies
suspect I/O devices.
-E
Displays all device error statistics.
Provides information about
devices: manufacturer, model
number, serial number, size, and
errors.
-n
Displays names in descriptive format.
Descriptive format helps identify
devices.
-x
For each drive, reports extended drive statistics. The output
is in tabular form.
Similar to the -e option, but
provides rate information. This
helps identify poor performance of
internal devices and other I/O
devices across the network.
2-38
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
The following example shows output for one iostat command.
# iostat -En
c0t0d0
Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
Model: ST3120026A
Revision: 8.01 Serial No: 3JT4H4C2
Size: 120.03GB <120031641600 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 0
c0t2d0
Soft Errors: 0 Hard Errors: 0 Transport Errors: 0
Vendor: LITE-ON Product: COMBO SOHC-4832K Revision: O3K1 Serial
No:
Size: 0.00GB <0 bytes>
Media Error: 0 Device Not Ready: 0 No Device: 0 Recoverable: 0
Illegal Request: 0 Predictive Failure Analysis: 0
2.9.2
prtdiag Command
The prtdiag command displays configuration and diagnostic information. The
diagnostic information identifies any failed component.
The prtdiag command is located in the
/usr/platform/platform-name/sbin/ directory.
Note – The prtdiag command might indicate a slot number different than that
identified elsewhere in this document. This is normal.
2.9.2.1
Options
TABLE 2-14 describes options for the prtdiag command and how those options can
help troubleshooting.
TABLE 2-14
Options for prtdiag
Option
Description
How It Can Help
No option
Lists components.
Identifies CPU information, memory configuration,
mounted PCI cards, OBP version number, mode switch
status, and CPU operational mode.
-v
Verbose mode.
Provides the same information as "no option." Additionally,
lists the detail information of PCI cards.
Chapter 2
Product Overview and Troubleshooting
2-39
The following example shows output for the prtdiag command in verbose mode.
# prtdiag -v
System Configuration: Sun Microsystems
System clock frequency: 960 MHz
Memory size: 32768 Megabytes
sun4u XXXX SPARC Enterprise M8000 Server
==================================== CPUs ====================================
LSB
--00
00
00
00
CPU
Chip
---0
1
2
3
CPU
ID
---------------------------------------0,
1,
2,
3,
4,
5,
6,
7
8,
9, 10, 11, 12, 13, 14, 15
16, 17, 18, 19, 20, 21, 22, 23
24, 25, 26, 27, 28, 29, 30, 31
Run
MHz
---2640
2640
2640
2640
L2$
MB
--6.0
6.0
6.0
6.0
CPU
CPU
Impl. Mask
----- ---7
144
7
144
7
144
7
144
============================ Memory Configuration ============================
LSB
--00
00
Memory
Group
-----A
B
Available
Size
-----------------16384MB
16384MB
Memory
Status
------okay
okay
DIMM
Size
-----1024MB
1024MB
# of Mirror Interleave
DIMMs Mode
Factor
----- ------- ---------16
no
8-way
16
no
8-way
========================= IO Cards =========================
LSB
--00
00
00
00
00
00
Name
Model
-------------- -----------scsi
LSI,1064
network
N/A
network
N/A
scsi
LSI,1064
network
N/A
network
N/A
2-40
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
==================== Hardware Revisions ====================
System PROM revisions:
---------------------OBP 4.24.4 2007/11/05 10:27
=================== Environmental Status ===================
Mode switch is in UNLOCK mode
=================== System Processor Mode ===================
SPARC64-VII mode
#
2.9.3
prtconf Command
Similar to the show-devs command run at the ok prompt, the prtconf command
displays the devices that are configured.
The prtconf command identifies hardware that is recognized by the Solaris OS. If
hardware is not suspected of being bad yet software applications are having trouble
with the hardware, the prtconf command can indicate if the Solaris OS software
recognizes the hardware, and if a driver for the hardware is loaded.
Chapter 2
Product Overview and Troubleshooting
2-41
2.9.3.1
Options
TABLE 2-15 describes options for the prtconf command and how those options can
help troubleshooting.
TABLE 2-15
Options for prtconf
Option
Description
How It Can Help
No option
Displays the device tree of devices recognized by the OS.
If a hardware device is recognized,
then it is probably functioning
properly. If the message "(driver
not attached)" is displayed for the
device or for a sub-device, then the
driver for the device is corrupt or
missing.
-D
Similar to the output of no option, however the device
driver is listed.
Lists the driver needed or used by
the OS to enable the device.
-p
Similar to the output of no option, yet is abbreviated.
Reports a brief list of the devices.
-V
Displays the version and date of the OpenBoot PROM
firmware.
Provides a quick check of
firmware version.
The following example shows output for the prtconf command.
# prtconf
System Configuration: Sun Microsystems
Memory size: 32768 Megabytes
System Peripherals (Software Nodes):
sun4u
SUNW,SPARC-Enterprise
scsi_vhci, instance #0
packages (driver not attached)
SUNW,builtin-drivers (driver not attached)
deblocker (driver not attached)
disk-label (driver not attached)
terminal-emulator (driver not attached)
obp-tftp (driver not attached)
ufs-file-system (driver not attached)
chosen (driver not attached)
openprom (driver not attached)
client-services (driver not attached)
options, instance #0
aliases (driver not attached)
memory (driver not attached)
virtual-memory (driver not attached)
2-42
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
pseudo-console, instance #0
nvram (driver not attached)
pseudo-mc, instance #0
pseudo-mc, instance #1
pseudo-mc, instance #4
cmp (driver not attached)
core (driver not attached)
cpu (driver not attached)
cpu (driver not attached)
(The rest is omitted.)
2.9.4
netstat Command
The netstat command displays the network status.
2.9.4.1
Options
TABLE 2-16 describes options for the netstat command and how those options can
help troubleshooting.
TABLE 2-16
Options for netstat
Option
Description
How It Can Help
-i
Displays the interface state, including packets in/out, error
in/out, collisions, and queue.
Provides a quick overview of the
network status.
-i interval
Providing a trailing number with the -i option repeats the
netstat command every interval seconds.
Identifies intermittent or long
duration network events. By
piping netstat output to a file,
overnight activity can be viewed
all at once.
-p
Displays the media table.
Provides MAC address for hosts
on the subnet.
-r
Displays the routing table.
Provides routing information.
-n
Replaces host names with IP addresses.
Used when an address is more
useful than a host name.
Chapter 2
Product Overview and Troubleshooting
2-43
The following example shows output for the netstat -p command.
# netstat -p
Net to
Device
-----bge0
bge0
bge0
bge0
bge0
bge0
bge0
bge0
bge0
bge0
bge0
bge0
bge0
bge0
2.9.5
Media Table: IPv4
IP Address
---------------------phatair-46
ns-umpk27-02-46
moreair-46
fermpk28a-46
fermpk28as-46
kayakr
matlock
toronto2
tocknett
mpk28-lobby
efyinisedeg
froggy
d-mpk28-46-245
224.0.0.0
Mask
Flags
Phys Addr
----------------- ----- --------------255.255.255.255
08:00:20:92:4a:47
255.255.255.255
08:00:20:93:fb:99
255.255.255.255
08:00:20:8a:e5:03
255.255.255.255
00:00:0c:07:ac:2e
255.255.255.255
00:50:e2:61:d8:00
255.255.255.255
08:00:20:d1:83:c7
255.255.255.255
SP
00:03:ba:27:01:48
255.255.255.255
08:00:20:b6:15:b5
255.255.255.255
08:00:20:7c:f5:94
255.255.255.255
08:00:20:a6:d5:c8
255.255.255.255
08:00:20:8d:6a:80
255.255.255.255
08:00:20:73:70:44
255.255.255.255
00:10:60:24:0e:00
240.0.0.0
SM
01:00:5e:00:00:00
ping Command
The ping command sends ICMP ECHO_REQUEST packets to network hosts.
Depending on how the ping command is configured, the output displayed can
identify troublesome network links or nodes. The destination host is specified in the
variable hostname.
2-44
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2.9.5.1
Options
TABLE 2-17 describes options for the ping command and how those options can help
troubleshooting.
TABLE 2-17
Options for ping
Option
Description
How It Can Help
hostname
The probe packet is sent to hostname and returned.
Verifies that a host is active on the
network.
-g hostname
Forces the probe packet to route through a specified
gateway.
By identifying different routes to
the target host, those individual
routes can be tested for quality.
-i interface
Designates which interface to send and receive the probe
packet through.
Enables a simple check of
secondary network interfaces.
-n
Replaces host names with IP addresses.
Used when an address is more
beneficial than a host name.
-s
Pings continuously in one-second intervals. Ctrl-C aborts.
Upon abort, statistics are displayed.
Helps identify intermittent or
long-duration network events. By
piping ping output to a file,
activity overnight can later be
viewed at once.
-svR
Displays the route the probe packet followed in one second
intervals.
Indicates probe packet route and
number of hops. Comparing
multiple routes can identify
bottlenecks.
The following example shows output for the ping -s command.
# ping -s teddybear
PING teddybear: 56 data bytes
64 bytes from teddybear (192.146.77.140): icmp_seq=0.
64 bytes from teddybear (192.146.77.140): icmp_seq=1.
64 bytes from teddybear (192.146.77.140): icmp_seq=2.
^C
----teddybear PING Statistics---3 packets transmitted, 3 packets received, 0% packet
round-trip (ms) min/avg/max = 0/0/1
Chapter 2
time=1. ms
time=0. ms
time=0. ms
loss
Product Overview and Troubleshooting
2-45
2.9.6
ps Command
The ps command lists the status of processes. Using options and rearranging the
command output can assist in determining the resource allocation.
2.9.6.1
Options
TABLE 2-18 describes options for the ps command and how those options can help
troubleshooting.
TABLE 2-18
Options for ps
Option
Description
How It Can Help
-e
Displays information for every process.
Identifies the process ID and the
executable.
-f
Generates a full listing.
Provides the following process
information: user ID, parent
process ID, time when executed,
and the path to the executable.
-o option
Enables configurable output. The pid, pcpu, pmem, and
comm options display process ID, percent CPU consumption,
percent memory consumption, and the responsible
executable, respectively.
Provides only most important
information. Knowing the
percentage of resource
consumption helps identify
processes that are affecting
performance and might be hung.
The following example shows output for one ps command.
# ps -eo pcpu,pid,comm|sort -rn
1.4 100317 /usr/openwin/bin/Xsun
0.9 100460 dtwm
0.1 100677 ps
0.1 100600 ksh
0.1 100591 /usr/dt/bin/dtterm
0.1 100462 /usr/dt/bin/sdtperfmeter
0.1 100333 mibiisa
%CPU
PID COMMAND
0.0 100652 /bin/csh
. . .
Note – When using sort with the -r option, the column headings are printed so that
the value in the first column is equal to zero.
2-46
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2.9.7
prstat Command
The prstat utility iteratively examines all active processes and reports statistics based
on the selected output mode and sort order. The prstat command provides output
similar to the ps command.
2.9.7.1
Options
TABLE 2-19 describes options for the prstat command and how those options can
help troubleshooting.
TABLE 2-19
Options for prstat
Option
Description
How It Can Help
No option
Displays a sorted list of the top processes that are
consuming the most CPU resources. List is limited to the
height of the terminal window and the total number of
processes. Output is automatically updated every five
seconds. Ctrl-C aborts.
Output identifies process ID, user
ID, memory used, state, CPU
consumption, and command
name.
-n number
Limits output to number of lines.
Limits amount of data displayed
and identifies primary resource
consumers.
-s key
Permits sorting list by key parameter.
Useful keys are cpu (default),
time, and size.
-v
Verbose mode.
Displays additional parameters.
The following example shows output for the prstat command.
# prstat -n 5 -s size
PID
USERNAME SIZE
RSS STATE PRI NICE TIME
CPU
PROCESS/NLWP
100524 mm39236
28M
21M sleep 48
0
0:00.26 0.3% maker6X.exe/1
100317 root
28M
69M sleep 59
0
0:00.26 0.7% Xsun/1
100460 mm39236
11M 8760K sleep 59
0
0:00.03 0.0% dtwm/8
100453 mm39236 8664K 4928K sleep 48
0
0:00.00 0.0% dtsession/4
100591 mm39236 7616K 5448K sleep 49
0
0:00.02 0.1% dtterm/1
Total: 65 processes, 159 lwps, load averages: 0.03, 0.02, 0.04
Chapter 2
Product Overview and Troubleshooting
2-47
2-48
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
3
Periodic Maintenance
Caution – The information in this chapter and subsequent chapters is intended for
service engineers.
Persons other than the authorized field engineers should use this information only
for reference and should not actually perform the work.
This chapter explains the maintenance work that must be performed regularly
regardless of whether a problem has occurred. The actual work is limited to
preventing dust in the environment from creating pollution.
This information is explained in the following sections:
■
■
■
Section 3.1, “Cleaning a Tape Drive Unit” on page 3-1
Section 3.2, “Cleaning an Air Filter (Server)” on page 3-2
Section 3.3, “Cleaning an Air Filter (I/O Unit)” on page 3-5
The high-end server is equipped with air filters at the bottom of the cabinet. These
air filters filter out dust particles from the air that the fans suck in from the floor into
the cabinet. If the filters become clogged, the ventilation volume is reduced and the
temperature rises, leading to problems. Although the frequency of cleaning varies
with the operating environment, the air filters must be cleaned on a regular basis to
ensure that they do not become clogged with dust. Each I/O unit also has air filters.
Clean them at the same time that the air filters of the server are cleaned.
When the service life expiration date of an air filter has already passed, replace it by
referring to the air filter cleaning procedure.
3.1
Cleaning a Tape Drive Unit
The head in a tape drive unit must be cleaned regularly.
3-1
Each tape drive unit used for operation must be cleaned once every 24 hours of
operation. Even tape drive units not used for operation must be cleaned once every
month.
Although cleaning work can be performed in either hot or cold system maintenance
mode, the SPARC Enterprise server power must be on when a cleaning cassette is
used. The cleaning procedure is as follows.
1. If a tape cassette has been inserted in the tape drive unit, remove it from the
unit.
2. While holding the cleaning cassette with the
it into the tape drive unit slot.
mark side facing right, insert
Head cleaning begins automatically.
3. The cleaning cassette is automatically ejected when cleaning is completed.
Remove it from the slot.
4. To use the tape cassette that was removed in Step 1, reinsert it into the tape
drive unit.
5. Confirm that the tape drive unit is in the normal state.
At this point, head cleaning is finished.
If one of the following problems occurs, replace the cleaning cassette immediately:
■
The cleaning cassette is not automatically ejected within one minute after being
inserted.
■
The tape is fully wound on the take-up reel on the right side. (The cassette can no
longer be reused.)
Use only specified cleaning cassettes.
Note – Contact your sales representative for tape drive unit options on SPARC
Enterprise M8000/M9000 servers.
3.2
Cleaning an Air Filter (Server)
An air filter may be cleaned while power to the server is on. Although the air filters
must be cleaned once a year, be sure to clean them if they become visibly dirty, even
if they are not scheduled for cleaning.
A high-end server cabinet is equipped with a total of six air filters: three at the front
and three at the rear at the bottom.
3-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Note – One concern about cleaning the air filters while power to the server is on is
that dislodged dust may be sucked inside the system when the air filters are pulled
out. Therefore, gently and slowly pull them out. Complete the cleaning as quickly as
possible.
Caution – If you must use a vacuum cleaner for this work, use it outside the
computer room. Do not use it inside the computer room. Using a vacuum cleaner
inside the computer room may result in a server failure.
Because the structure and the mounting environment of air filters are the same, the
descriptions in the figures covering filter cleaning refer to, as an example, the air
filters at the front of each model.
1. Unlock and open the front and rear doors of the server. For details, see
Chapter 5.
2. Using a Phillips screwdriver, loosen the screw securing the fixing bracket of an
air filter, and turn the bracket so that it faces downward.
Chapter 3
Periodic Maintenance
3-3
FIGURE 3-1
Removing Air Filters (Example for the M8000)
Fixing bracket (x3)
3. Pull out all of the air filters.
4. Use a cleaner to remove dust from the air filters. Attach a brush to the tip of the
cleaner, and clean both sides of the filters.
5. Restore each air filter to its original location and orientation, which means the
knob is on the side closest to you and the arrow on the label points up (the
latticework faces upward).
6. When this restoring work is completed for all the air filters, turn the fixing
brackets of the air filters until they face upward, and then tighten the screws
firmly with the Phillips screwdriver. Finally, close the front and rear doors of
the SPARC Enterprise server.
Removal of Air Filters
This filter cleaning procedure applies to both high-end servers.
3-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 3-2
Removing of Air Filters (Example Using the M9000 Base Cabinet)
Fixing bracket (x3)
3.3
Cleaning an Air Filter (I/O Unit)
Each I/O unit has two air filters. Clean them at the same time that the air filters of
the server are cleaned.
Caution – If you must use a vacuum cleaner for this work, use it outside the
computer room. Do not use it inside the computer room. Using a vacuum cleaner
inside the computer room may result in a server failure.
The cleaning procedure is as follows.
1. Loosen the screws securing the filter cover, and remove the filter cover.
2. Pull out the air filter from the filter cover.
Chapter 3
Periodic Maintenance
3-5
FIGURE 3-3
Removing of Air Filters (I/O Unit)
3. Use a vacuum cleaner to remove dust from the air filter.
4. After the cleaning is completed, follow the removal procedure in reverse order
to mount it.
3-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
4
FRU Removal Preparation
This chapter explains the required basic operations for replacing components, in the
following sections:
■
■
■
■
■
■
■
Section 4.1,
Section 4.2,
Section 4.3,
Section 4.4,
Section 4.5,
Section 4.6,
Section 4.7,
“Types of Replacement Procedures” on page 4-2
“Active Replacement” on page 4-3
“Hot Replacement” on page 4-12
“Cold Replacement” on page 4-18
“Power-On/Off of Main Line Switch” on page 4-24
“Emergency Switch-Off” on page 4-34
“Cable Routing of the SPARC Enterprise M8000 Server” on page 4-34
When actually performing the work of replacing a component, use the operator
panel and the maintenance terminal by referring to the operator panel display,
maintenance terminal display, and the LED display of the component.
Depending on the target component, the server must be powered off or a domain
must be stopped.
For the LED display of each component, see Section 2.7, “LED Error Display” on
page 2-30. Three replacement types are defined for judging whether power-off of the
server or stopping a domain is necessary: active replacement, hot replacement, and
cold replacement. See Part II, Maintenance. For information on the swapping types
of each component, see Appendix B.
Note – Some of the XSCF functions have restrictions on their use. Register the
necessary login privileges for each field engineer in advance. Field engineers cannot
use functions that have not been registered for them. The system administrator sets
and changes the users and their privileges. For details, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
Power-on and power-off of the server and emergency power-off are explained in the
last part of this chapter.
■
Section 4.5, “Power-On/Off of Main Line Switch” on page 4-24
4-1
■
Section 4.6, “Emergency Switch-Off” on page 4-34
4.1
Types of Replacement Procedures
4.1.1
FRU Replacement
The three types of replacement procedures explained below are supported for FRU
replacement. Choose the most suitable replacement procedure according to the
customer's system environment.
■
Active replacement
A target FRU is operated while the Solaris OS of the domain to which the FRU
belongs is operating. The target FRU is operated by using Solaris OS commands or
XSCF commands. Because the power supply unit (PSU) and fan unit (FAN) do not
belong to any domain, they are operated by using XSCF commands regardless of the
operating state of the Solaris OS.
Note – The procedure for disconnecting a hard disk drive from the domain depends
on whether disk mirroring software or similar support software is active. For
details, see the related individual software manuals.
■
Hot replacement
A target FRU is operated while the domain to which the FRU belongs is stopped.
Depending on the target FRU, there are two cases as follows:
■
■
Operated with XSCF commands.
■
Operated directly, not by using XSCF commands.
Cold replacement
After all the domains are stopped and then the server is powered off, a FRU is
operated.
Note – Do not operate a target FRU while the OpenBoot PROM is running (the ok
prompt is displayed). After stopping the relevant domain (power-off) or starting the
Solaris OS, operate the target FRU.
4-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.2
Active Replacement
In active replacement the Solaris OS must be configured to allow the component to
be replaced. Active replacement has four stages:
■
■
■
■
Section 4.2.1,
Section 4.2.2,
Section 4.2.3,
Section 4.2.4,
“Disconnecting a FRU from a Domain” on page 4-3
“Disconnecting and Replacing a FRU” on page 4-5
“Configuring a FRU into a Domain” on page 4-6
“Confirming the Hardware” on page 4-7
Note – If the hard disk drive is the boot device, it is necessary to replace it according
to the cold replacement procedure. However, if the disk mirroring software or other
support software can disconnect the relevant boot disk from Solaris OS, active
replacement can be performed.
4.2.1
Disconnecting a FRU from a Domain
4.2.1.1
Disconnecting a CMU/IOU
Perform the following procedure to disconnect a CMU or IOU when the Solaris OS is
operating:
1. Checking resources
Check the resources that are connected to a CMU or IOU to be disconnected, and
verify that the system is not affected when it is disconnected.
2. Disconnecting from the domain
To disconnect the CMU or IOU from the domain, enter the following command
from the terminal that is connected to the XSCF:
XSCF> deleteboard
01-0
The system administrator permission is required for executing this command.
Chapter 4
FRU Removal Preparation
4-3
4.2.1.2
Disconnecting a PCI card
Caution – Before you remove the PCI cassette, make sure that there is no activity on
the card in the cassette.
Caution – In the PCI cassette part, when removing cables such as LAN cable, if
your finger can't reach the latch lock of the connector, press the latch with a flathead
screwdriver to remove the cable. Forcing your finger into the clearance can cause
damage to the PCI card.
1. From the Solaris OS use the cfgadm command to get the component status:
# cfgadm
AP_ID
...
iou#0-pci#1
iou#0-pci#2
iou#0-pci#3
iou#0-pci#4
Type
Receptacle
Occupant
Condition
unknown
unknown
etherne/hp
fibre/hp
empty
empty
connected
connected
unconfigured
unconfigured
configured
configured
unknown
unknown
ok
ok
AP_ID is comprised of the IOU number (iou#0 or iou#1) and the PCI cassette slot
number (pci#1, pci#2, pci#3, pci#4)
2. Use the cfgadm command to unconfigure the component from the hardware:
# cfgadm -c unconfigure AP_ID
where AP_ID is the IOU and PCI card as shown in the output of cfgadm.
3. Use the cfgadm command to stop supplying power to the component:
# cfgadm -c disconnect AP_ID
where AP_ID is the IOU and PCI card as shown in the output of cfgadm.
4. Use the cfgadm command to confirm the component from the domain is now
disconnected and unconfigured:
# cfgadm
AP_ID
...
iou#0-pci#1
iou#0-pci#2
iou#0-pci#3
iou#0-pci#4
4-4
Type
Receptacle
Occupant
Condition
unknown
unknown
etherne/hp
fibre/hp
empty
empty
disconnected
connected
unconfigured
unconfigured
unconfigured
configured
unknown
unknown
ok
ok
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.2.2
Disconnecting and Replacing a FRU
1. From the XSCF Shell prompt, use the replacefru command:
XSCF> replacefru
---------------------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
---------------------------------------------------------------Select [1-5|c:cancel] :
Note – DDC_A is displayed only for the M8000.
The command is menu-driven. The example continues using a FAN unit.
Select [1,2|c:cancel] :1
--------------------------------------------------Maintenance/Replacement Menu
Please select a FAN to be replaced.
No. FRU
Status
--- --------------- -----------------1. FAN_A#0
Normal
2. FAN_A#1
Normal
3. FAN_A#2
Normal
4. FAN_A#3
Normal
--------------------------------------------------Select [1-4|b:back] :1
You are about to replace FAN_A#0.
Do you want to continue?[r:replace|c:cancel] :r
Please confirm the Check LED is blinking.
If this is the case, please replace FAN_A#0.
After replacement has been completed, please select[f:finish] :f
The replacefru command will automatically test the status of the component after
the disconnecting off and replace has finished.
Chapter 4
FRU Removal Preparation
4-5
Diagnostic tests for FAN_A#0 have started.
[This operation may take up to 2 minute(s)]
(progress scale reported in seconds)
0..... 30..... 60..... 90.....done
-------------------------------------------------Maintenance/Replacement Menu
Status of the replaced unit.
FRU
Status
------------- -------FAN_A#0
Normal
-------------------------------------------------The replacement of FAN_A#0 has completed, normally.[f:finish] :f
-------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. FAN
(Fan Unit)
2. PSU
(Power Supply Unit)
-------------------------------------------------Select [1,2|c:cancel] : C
XSCF>
When the tests are complete the program will return to the original menu. Select
cancel to return to the XSCF Shell prompt.
Note – The display may vary depending on the XCP version.
4.2.3
Configuring a FRU into a Domain
4.2.3.1
Configuring CMU/IOU
Perform the following procedure to configure a CMU or IOU when the Solaris OS is
operating:
4-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
1. Configuring into a domain
To configure CMU or IOU into the domain, enter the following command from
the terminal that is connected to the XSCF:
XSCF> addboard -c configure
-d 0 00-0
The system administrator permission is required for executing this command.
2. Checking resources
Log in to the XSCF and execute the showboard command to verify that the target
CMU or IOU has been correctly configured.
4.2.3.2
Configuring a PCI card
1. From the Solaris OS, use the cfgadm command to configure and connect the
component to the domain:
# cfgadm -c configure AP_ID
where AP_ID is the IOU and PCI card as shown in the output of cfgadm.
2. Use the cfgadm command to confirm the component is now connected and
configured:
# cfgadm
AP_ID
..
iou#0-pci#1
iou#0-pci#2
iou#0-pci#3
iou#0-pci#4
4.2.4
Type
unknown
unknown
etherne/hp
fibre/hp
Receptacle
Occupant
Condition
empty
empty
connected
connected
unconfigured
unconfigured
configured
configured
unknown
unknown
ok
ok
Confirming the Hardware
1. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for a replacement CMU, CPU, or DIMM, it is operating
normally.
Chapter 4
FRU Removal Preparation
4-7
Note – The display may vary depending on the XCP version.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
4-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
XSCF> showhardconf
SPARC Enterprise M9000;
+ Serial:PA30601004; Operator_Panel_Switch:Locked;
+ Power_Supply_System:Single-1Phase; Ex:Single-1Phase; SCF-ID:XSCF#0;
+ System_Power:On; System_Phase:Cabinet Power On;
Domain#0 Domain_Status:Running;
CMU#0 Status:Normal; Ver:0101h; Serial:PP0616B579 ;
+ FRU-Part-Number:CA06629-D001 A4
;
+ Memory_Size:128 GB;
+ Type:A
CPUM#0-CHIP#0 Status:Normal; Ver:0101h; Serial:PP091505ZY ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#1-CHIP#0 Status:Normal; Ver:0101h; Serial:PP091505ZW ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#2-CHIP#0 Status:Normal; Ver:0101h; Serial:PP0915060H ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#3-CHIP#0 Status:Normal; Ver:0101h; Serial:PP09150603 ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
MEM#00A Status:Normal;
+ Code:7f7ffe00000000004aEBE41RE4ABHA-5C-E 3020-22211d88;
+ Type:4B; Size:4 GB;
MEM#00B Status:Normal;
+ Code:7f7ffe00000000004aEBE41RE4ABHA-5C-E 3020-2a002a55;
+ Type:4B; Size:4 GB;
:
MEM#33A Status:Normal;
+ Code:ce0000000000000001M3 93T5168AZ0-CD5 3041-741a8ea1;
+ Type:4B; Size:4 GB;
MEM#33B Status:Normal;
+ Code:ce0000000000000001M3 93T5168AZ0-CD5 3041-741a8ed3;
+ Type:4B; Size:4 GB;
:
CMU#2 Status:Normal; Ver:4201h; Serial:PP0618K472 ;
+ FRU-Part-Number:CA06620-D003 A0
/371-4617-01
;
+ Memory_Size:32 GB;
+ Type:B
CPUM#0-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0608J517 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
Chapter 4
FRU Removal Preparation
;
;
;
;
;
4-9
Output for the showhardconf command (Continued)
+ Core:4; Strand:2;
CPUM#1-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0620P552 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
;
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#2-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0631Q396 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
;
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#3-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0629H443 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
;
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
MEM#00A Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221d6855;
+ Type:1A; Size:1 GB;
MEM#00B Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221fcdb7;
+ Type:1A; Size:1 GB;
:
MEM#33A Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221d678b;
+ Type:1A; Size:1 GB;
MEM#33B Status:Normal;
+ Code:2cffffffffffffff0818HTF12872Y-53EB3 0300-69aedd7a;
+ Type:1A; Size:1 GB;
:
IOU#0 Status:Normal; Ver:0101h; Serial:PP072102UN ;
+ FRU-Part-Number:CA06620-D102 B1
/371-2217-02
;
PCI#0 Name_Property:pci; Card_Type:IOUA;
+ Serial:PP0611T826 ;
+ FRU-Part-Number:CA21126-B20X 002AB
IOU#1 Status:Normal; Ver:0101h; Serial:PP072102UM ;
+ FRU-Part-Number:CA06620-D102 B1
/371-2217-02
;
PCI#0 Name_Property:pci; Card_Type:IOUA;
+ Serial:PP0611T825 ;
+ FRU-Part-Number:CA21126-B20X 002AB
:
XSCFU_B#0 Status:Normal,Active; Ver:0201h; Serial:PP080600DW ;
+ FRU-Part-Number:CA06620-D342 C0
/371-2228-02
;
:
XBU_B#0 Status:Normal; Ver:0201h; Serial:PP0641X324 ;
+ FRU-Part-Number:CA06620-D301 A6
;
:
CLKU_B#0 Status:Normal; Ver:0101h; Serial:PP0542M679 ;
+ FRU-Part-Number:CA06629-D042 A1
;
:
4-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Output for the showhardconf command (Continued)
OPNL#0 Status:Normal; Ver:0101h; Serial:PP06058246 ;
+ FRU-Part-Number:CA06629-D061 A1
PSU#0 Status:Normal; Serial:;
+ FRU-Part-Number:CA01022-0690;
+ Power_Status:On;
:
FANBP_A#0 Status:Normal; Ver:0101h; Serial:PP0607D266 ;
+ FRU-Part-Number:CA21123-B54X 003AC
FAN_A#0 Status:Normal; Serial:PA0605B287;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FAN_A#15 Status:Normal; Serial:PA0605B303;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FANBP_B#0 Status:Normal; Ver:0201h; Serial:PP0607D270 ;
+ FRU-Part-Number:CA21123-B55X 003AC
FAN_A#4 Status:Normal; Serial:PA0605B297;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FAN_A#9 Status:Normal; Serial:PA0605B300;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
SWBP#0 Status:Normal; Ver:0101h; Serial:PP0607E759 ;
+ FRU-Part-Number:CA20394-B16X 001AA
MEDBP#0 Status:Normal; Ver:0101h; Serial:PP06058497 ;
+ FRU-Part-Number:CA20394-B17X 002AB
;
;
;
;
;
;
;
;
;
2. Verify the status LEDs on the PCI cassette.
The POWER LED should be lit and the CHECK LED should not be lit.
Note – If an XSCF failover occurs while executing the addfru, deletefru or
replacefru command, the addition, removal or replacement of FRUs may not have
been completed successfully. In such case, log in again to the active XSCF to confirm
that the addition, removal or replacement of FRUs was successfully completed. If the
addition, removal or replacement of the FRUs was not completed successfully,
re-execute the following commands in the manner described:
- For addfru command execution, execute the deletefru command for the
relevant FRU, and then re-execute the addfru command.
- For deletefru command execution, re-execute the deletefru command for the
relevant FRU.
- For replacefru command execution, re-execute the replacefru command for
the relevant FRU.
Chapter 4
FRU Removal Preparation
4-11
4.3
Hot Replacement
In hot replacement the Solaris OS does not need to be configured to allow the
component to be replaced. The domain might need to be stopped and restarted in
the case of CD-RW/DVD-RW drive unit or tape drive units.
4.3.1
■
Section 4.3.1, “Disconnecting and Replacing a FRU” on page 4-12
■
Section 4.3.2, “Confirming the Hardware” on page 4-15
Disconnecting and Replacing a FRU
1. From the XSCF Shell prompt, specify the domain, and execute the poweroff
command.
Example: In case you specify Domain ID 0
XSCF> poweroff -d 0
2. From the XSCF Shell prompt, execute the replacefru command:
XSCF> replacefru
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
--------------------------------------------------Select [1-5|c:cancel] :
Note – DDC_A is displayed only for the M8000.
4-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
The command is menu-driven. The example continues using a FAN unit.
Select [1,2|c:cancel] :1
--------------------------------------------------Maintenance/Replacement Menu
Please select a FAN to be replaced.
No. FRU
Status
--- --------------- -----------------1. FAN_A#0
Normal
2. FAN_A#1
Normal
3. FAN_A#2
Normal
4. FAN_A#3
Normal
--------------------------------------------------Select [1-4|b:back] :1
You are about to replace FAN_A#0.
Do you want to continue?[r:replace|c:cancel] :r
Please confirm the Check LED is blinking.
If this is the case, please replace FAN_A#0.
After replacement has been completed, please select[f:finish] :f
The replacefru command will automatically test the status of the component after
the disconnecting off and replace has finished.
Chapter 4
FRU Removal Preparation
4-13
Diagnostic tests for FAN_A#0 have started.
[This operation may take up to 2 minute(s)]
(progress scale reported in seconds)
0..... 30..... 60..... 90.....done
--------------------------------------------------Maintenance/Replacement Menu
Status of the replaced unit.
FRU
Status
-------------------FAN_A#0
Normal
--------------------------------------------------The replacement of FAN_A#0 has completed, normally.[f:finish] :f
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. FAN
(Fan Unit)
2. PSU
(Power Supply Unit)
--------------------------------------------------Select [1,2|c:cancel] : C
XSCF>
When the tests are complete the program will return to the original menu. Select
cancel to return to the XSCF Shell prompt.
Note – The display may vary depending on the XCP version.
4-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.3.2
Confirming the Hardware
1. Use the showhardconf command to confirm the new component has been
installed:
XSCF> showhardconf
SPARC Enterprise M9000;
+ Serial:PA30601004; Operator_Panel_Switch:Locked;
+ Power_Supply_System:Single-1Phase; Ex:Single-1Phase; SCF-ID:XSCF#0;
+ System_Power:On; System_Phase:Cabinet Power On;
Domain#0 Domain_Status:Running;
CMU#0 Status:Normal; Ver:0101h; Serial:PP0616B579 ;
+ FRU-Part-Number:CA06629-D001 A4
;
+ Memory_Size:128 GB;
+ Type:A
CPUM#0-CHIP#0 Status:Normal; Ver:0101h; Serial:PP091505ZY ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#1-CHIP#0 Status:Normal; Ver:0101h; Serial:PP091505ZW ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#2-CHIP#0 Status:Normal; Ver:0101h; Serial:PP0915060H ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#3-CHIP#0 Status:Normal; Ver:0101h; Serial:PP09150603 ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
MEM#00A Status:Normal;
+ Code:7f7ffe00000000004aEBE41RE4ABHA-5C-E 3020-22211d88;
+ Type:4B; Size:4 GB;
MEM#00B Status:Normal;
+ Code:7f7ffe00000000004aEBE41RE4ABHA-5C-E 3020-2a002a55;
+ Type:4B; Size:4 GB;
:
MEM#33A Status:Normal;
+ Code:ce0000000000000001M3 93T5168AZ0-CD5 3041-741a8ea1;
+ Type:4B; Size:4 GB;
MEM#33B Status:Normal;
+ Code:ce0000000000000001M3 93T5168AZ0-CD5 3041-741a8ed3;
+ Type:4B; Size:4 GB;
:
CMU#2 Status:Normal; Ver:4201h; Serial:PP0618K472 ;
+ FRU-Part-Number:CA06620-D003 A0
/371-4617-01
;
Chapter 4
;
;
;
;
FRU Removal Preparation
4-15
Output for the showhardconf command (Continued)
+ Memory_Size:32 GB;
+ Type:B
CPUM#0-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0608J517 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#1-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0620P552 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#2-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0631Q396 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#3-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0629H443 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
MEM#00A Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221d6855;
+ Type:1A; Size:1 GB;
MEM#00B Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221fcdb7;
+ Type:1A; Size:1 GB;
:
MEM#33A Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221d678b;
+ Type:1A; Size:1 GB;
MEM#33B Status:Normal;
+ Code:2cffffffffffffff0818HTF12872Y-53EB3 0300-69aedd7a;
+ Type:1A; Size:1 GB;
:
IOU#0 Status:Normal; Ver:0101h; Serial:PP072102UN ;
+ FRU-Part-Number:CA06620-D102 B1
/371-2217-02
;
PCI#0 Name_Property:pci; Card_Type:IOUA;
+ Serial:PP0611T826 ;
+ FRU-Part-Number:CA21126-B20X 002AB
IOU#1 Status:Normal; Ver:0101h; Serial:PP072102UM ;
+ FRU-Part-Number:CA06620-D102 B1
/371-2217-02
;
PCI#0 Name_Property:pci; Card_Type:IOUA;
+ Serial:PP0611T825 ;
+ FRU-Part-Number:CA21126-B20X 002AB
:
XSCFU_B#0 Status:Normal,Active; Ver:0201h; Serial:PP080600DW ;
+ FRU-Part-Number:CA06620-D342 C0
/371-2228-02
;
:
4-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
;
;
;
;
Output for the showhardconf command (Continued)
XBU_B#0 Status:Normal; Ver:0201h; Serial:PP0641X324 ;
+ FRU-Part-Number:CA06620-D301 A6
:
CLKU_B#0 Status:Normal; Ver:0101h; Serial:PP0542M679 ;
+ FRU-Part-Number:CA06629-D042 A1
:
OPNL#0 Status:Normal; Ver:0101h; Serial:PP06058246 ;
+ FRU-Part-Number:CA06629-D061 A1
PSU#0 Status:Normal; Serial:;
+ FRU-Part-Number:CA01022-0690;
+ Power_Status:On;
:
FANBP_A#0 Status:Normal; Ver:0101h; Serial:PP0607D266 ;
+ FRU-Part-Number:CA21123-B54X 003AC
FAN_A#0 Status:Normal; Serial:PA0605B287;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FAN_A#15 Status:Normal; Serial:PA0605B303;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FANBP_B#0 Status:Normal; Ver:0201h; Serial:PP0607D270 ;
+ FRU-Part-Number:CA21123-B55X 003AC
FAN_A#4 Status:Normal; Serial:PA0605B297;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FAN_A#9 Status:Normal; Serial:PA0605B300;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
SWBP#0 Status:Normal; Ver:0101h; Serial:PP0607E759 ;
+ FRU-Part-Number:CA20394-B16X 001AA
MEDBP#0 Status:Normal; Ver:0101h; Serial:PP06058497 ;
+ FRU-Part-Number:CA20394-B17X 002AB
;
;
;
;
;
;
;
;
;
;
;
Refer to the showhardconf man page for more information.
2. Verify the indicator LEDs on the FRU.
Refer to TABLE 2-11 and TABLE 2-12 for LED status.
Note – For notes to observe when executing the addfru, deletefru or
replacefru command, see Section 4.2.4, “Confirming the Hardware” on page 4-7.
Chapter 4
FRU Removal Preparation
4-17
4.4
Cold Replacement
In cold replacement all business operations are stopped. Cold replacement is the act
of powering off the server and disconnecting input power. This is normally required
for safety when the inside of the server is accessed. The work of cold replacement
consists of the following three steps:
4.4.1
■
Section 4.4.1, “Powering the Server Off” on page 4-18
■
Section 4.4.2, “Powering the Server On” on page 4-19
■
Section 4.4.3, “Confirming the Hardware” on page 4-20
Powering the Server Off
Two power-off procedures are available.
From the Operator Panel
Follow the power-off procedure below.
1. Confirm that the mode switch on the operator panel is set to the Service
position.
2. Confirm that the POWER LED and CHECK LED on the operator panel are lit
and the XSCF LED on the operator panel is off.
3. When the XSCF LED is lit, see Section 2.7, “LED Error Display” on page 2-30 to
identify the cause and give preference to these functions.
4. Press and hold down the POWER switch on the operator panel for at least four
seconds.
5. Check the POWER LED. When it turns off, the power-off procedure is
completed.
If the state of any of the LEDs is different from the above, see Section 2.7, “LED
Error Display” on page 2-30
From the Maintenance Terminal
Follow the power-off procedure below using the XSCF maintenance functions.
1. Notify users that the server is going down.
4-18
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2. Back up the system files and data as necessary.
3. Log into the XSCF Shell and type the poweroff command
XSCF> poweroff -a
The following activities occur when the poweroff command is used:
■
The Solaris OS shuts down cleanly.
■
The server powers off to Standby mode (the XSCF Unit and one fan will still have
power).
For details of the command, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
4. Switch off all main line switches of the AC section.
Caution – There is an electrical hazard if the power cords are not disconnected. All
power cords must be disconnected to completely remove power from the server.
4.4.2
Powering the Server On
Two power-on procedures are available.
4.4.2.1
From the Operator Panel
Follow the power-on procedure below.
1. Confirm that the mode switch on the operator panel is set to the Service
position.
2. Confirm that the POWER LED and CHECK LED on the operator panel are off
and the XSCF LED on the operator panel is lit.
3. If the XSCF LED is off, switch on all the system main line switches.
4. Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
5. Check the POWER LED.
When it becomes lit, the power-on procedure is completed.
If the state of any of the LEDs is different from the above, see Section 2.7, “LED
Error Display” on page 2-30.
Chapter 4
FRU Removal Preparation
4-19
4.4.2.2
From the Maintenance Terminal
Follow the power-on procedure below using the XSCF maintenance functions.
1. Make sure that the server has enough power supply units to run the desired
configuration.
2. Switch on all main line switches of the AC section.
3. Make sure the XSCF Unit Ready LED is lit.
4. Log into the XSCF Shell and type the poweron command.
XSCF> poweron
Refer to the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
User’s Guide for details.
5. After a delay the following activities occur:
■
The operator panel Active LED will light.
■
The system will execute the power-on self-test (POST).
Then, the server is completely powered on.
6. Turn the keyswitch on the operator panel to the Locked position.
4.4.3
Confirming the Hardware
1. Use the showhardconf command to confirm the new component has been
installed:
4-20
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
XSCF> showhardconf
SPARC Enterprise M9000;
+ Serial:PA30601004; Operator_Panel_Switch:Locked;
+ Power_Supply_System:Single-1Phase; Ex:Single-1Phase; SCF-ID:XSCF#0;
+ System_Power:On; System_Phase:Cabinet Power On;
Domain#0 Domain_Status:Running;
CMU#0 Status:Normal; Ver:0101h; Serial:PP0616B579 ;
+ FRU-Part-Number:CA06629-D001 A4
;
+ Memory_Size:128 GB;
+ Type:A
CPUM#0-CHIP#0 Status:Normal; Ver:0101h; Serial:PP091505ZY ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#1-CHIP#0 Status:Normal; Ver:0101h; Serial:PP091505ZW ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#2-CHIP#0 Status:Normal; Ver:0101h; Serial:PP0915060H ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
CPUM#3-CHIP#0 Status:Normal; Ver:0101h; Serial:PP09150603 ;
+ FRU-Part-Number:CA06620-D021 A3
+ Freq:2.280 GHz; Type:16;
+ Core:2; Strand:2;
MEM#00A Status:Normal;
+ Code:7f7ffe00000000004aEBE41RE4ABHA-5C-E 3020-22211d88;
+ Type:4B; Size:4 GB;
MEM#00B Status:Normal;
+ Code:7f7ffe00000000004aEBE41RE4ABHA-5C-E 3020-2a002a55;
+ Type:4B; Size:4 GB;
:
MEM#33A Status:Normal;
+ Code:ce0000000000000001M3 93T5168AZ0-CD5 3041-741a8ea1;
+ Type:4B; Size:4 GB;
MEM#33B Status:Normal;
+ Code:ce0000000000000001M3 93T5168AZ0-CD5 3041-741a8ed3;
+ Type:4B; Size:4 GB;
:
CMU#2 Status:Normal; Ver:4201h; Serial:PP0618K472 ;
+ FRU-Part-Number:CA06620-D003 A0
/371-4617-01
;
+ Memory_Size:32 GB;
+ Type:B
Chapter 4
;
;
;
;
FRU Removal Preparation
4-21
Output for the showhardconf command (Continued)
CPUM#0-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0608J517 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#1-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0620P552 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#2-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0631Q396 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
CPUM#3-CHIP#0 Status:Normal; Ver:0901h; Serial:PP0629H443 ;
+ FRU-Part-Number:CA06620-D051 A0
/371-4616-01
+ Freq:2.880 GHz; Type:32;
+ Core:4; Strand:2;
MEM#00A Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221d6855;
+ Type:1A; Size:1 GB;
MEM#00B Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221fcdb7;
+ Type:1A; Size:1 GB;
:
MEM#33A Status:Normal;
+ Code:7f7ffe00000000004aEBE10RD4AGFA-5C-E 3020-221d678b;
+ Type:1A; Size:1 GB;
MEM#33B Status:Normal;
+ Code:2cffffffffffffff0818HTF12872Y-53EB3 0300-69aedd7a;
+ Type:1A; Size:1 GB;
:
IOU#0 Status:Normal; Ver:0101h; Serial:PP072102UN ;
+ FRU-Part-Number:CA06620-D102 B1
/371-2217-02
;
PCI#0 Name_Property:pci; Card_Type:IOUA;
+ Serial:PP0611T826 ;
+ FRU-Part-Number:CA21126-B20X 002AB
IOU#1 Status:Normal; Ver:0101h; Serial:PP072102UM ;
+ FRU-Part-Number:CA06620-D102 B1
/371-2217-02
;
PCI#0 Name_Property:pci; Card_Type:IOUA;
+ Serial:PP0611T825 ;
+ FRU-Part-Number:CA21126-B20X 002AB
:
XSCFU_B#0 Status:Normal,Active; Ver:0201h; Serial:PP080600DW ;
+ FRU-Part-Number:CA06620-D342 C0
/371-2228-02
;
:
4-22
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
;
;
;
;
Output for the showhardconf command (Continued)
XBU_B#0 Status:Normal; Ver:0201h; Serial:PP0641X324 ;
+ FRU-Part-Number:CA06620-D301 A6
:
CLKU_B#0 Status:Normal; Ver:0101h; Serial:PP0542M679 ;
+ FRU-Part-Number:CA06629-D042 A1
:
OPNL#0 Status:Normal; Ver:0101h; Serial:PP06058246 ;
+ FRU-Part-Number:CA06629-D061 A1
PSU#0 Status:Normal; Serial:;
+ FRU-Part-Number:CA01022-0690;
+ Power_Status:On;
:
FANBP_A#0 Status:Normal; Ver:0101h; Serial:PP0607D266 ;
+ FRU-Part-Number:CA21123-B54X 003AC
FAN_A#0 Status:Normal; Serial:PA0605B287;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FAN_A#15 Status:Normal; Serial:PA0605B303;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FANBP_B#0 Status:Normal; Ver:0201h; Serial:PP0607D270 ;
+ FRU-Part-Number:CA21123-B55X 003AC
FAN_A#4 Status:Normal; Serial:PA0605B297;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
FAN_A#9 Status:Normal; Serial:PA0605B300;
+ FRU-Part-Number:CA06501-D023 A2
/371-2222-00
:
SWBP#0 Status:Normal; Ver:0101h; Serial:PP0607E759 ;
+ FRU-Part-Number:CA20394-B16X 001AA
MEDBP#0 Status:Normal; Ver:0101h; Serial:PP06058497 ;
+ FRU-Part-Number:CA20394-B17X 002AB
;
;
;
;
;
;
;
;
;
;
;
Refer to the showhardconf man page for more information.
2. Verify the state of the status LEDs on the FRU.
Refer to TABLE 2-11 and TABLE 2-12 for LED status.
Chapter 4
FRU Removal Preparation
4-23
4.5
Power-On/Off of Main Line Switch
4.5.1
Types of Power Supply
TABLE 4-1 shows how the components of the server vary depending on the type of
power supply. For details on the electrical specifications, see the SPARC Enterprise
M8000/M9000 Servers Site Planning Guide.
TABLE 4-1
Types of Power Supply and the Components of the Server
Types of power supply
Reference diagram
Model
AC input power
Single/Dual power
Components of the server
M8000
Single-phase
Single power
Single-phase power feed option
See FIGURE 4-4.
Single-phase
Dual power
Rack-mountable dual power feed option
See FIGURE 4-5.
Three-phase
Dual power
Power cabinet + three-phase power feed
option
See FIGURE 4-6.
Single-phase
Single power
Single-phase power feed option
See FIGURE 4-7.
Single-phase
Dual power
Power cabinet + single-phase power feed
option
See FIGURE 4-8.
Three-phase
Dual power
Power cabinet + three-phase power feed
option
See FIGURE 4-9.
M9000
Note – When you choose three-phase power feed option, the dual power feed
option is always included.
4.5.1.1
AC Input Power
There are two types of AC input power: single-phase and three-phase.
FIGURE 4-1 and FIGURE 4-2 show how the power supplied from the external AC power
supply (distribution panel) runs through in the server. The power input from the
distribution panel is supplied to the main line switch, the switch (in case of
three-phase power feed), and the power supply unit.
4-24
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 4-1
Internal Connection of Power Supply (Single-Phase Power Feed)
Distribution panel
Main line switch
AC
ACS
(ON/OFF)
(ON/OFF)
Power supply unit
PSU
PSU
PSU
FIGURE 4-2
Internal Connection of Power Supply (Three-Phase Power Feed)
Distribution panel
Main line switch
Switch
Power supply unit
PSU
AC
ACS
ACS_D
(ON/OFF)
(ON/OFF)
(ON/OFF)
ACS_D
PSU
ACS_D
PSU
ACS_D
PSU
To use three-phase power feed, you need the three-phase power feed option and the
power cabinet for mounting the power supply.
In case of three-phase power feed, AC section is not located in the SPARC Enterprise
server cabinet. The AC section for the three-phase power feed is located in the power
cabinet.
Two types of options are available for three-phase power feed, the star connection
and the delta connection: the star connection option has two sorts of connections,
connection for the neutral line and connections for phases, and the delta connection
option has connections for phases. For details, see the SPARC Enterprise
M8000/M9000 Servers Site Planning Guide.
Chapter 4
FRU Removal Preparation
4-25
4.5.1.2
Power System
There are two types of power system: single power and dual power.
FIGURE 4-3 shows the conceptual diagram of power system.
FIGURE 4-3
Power System (Single Power / Dual Power)
Dual power feed
Single power feed
Cabinet
Cabinet
PSU
PSU
PSU
ACS
ACS
ACS
AC
AC
AC
Distribution panel
Distribution panel
Distribution panel
The dual power feed option receives power from two external AC power sources
that are independent of each other, and it duplicates the input power system.
The dual power feed can be used with either single-phase or three-phase power
feed.
4.5.2
Power-On/Off Procedures of Main Line Switch
On the server with expansion cabinet, when you turn on or turn off the main line
switch, do not fail to follow the order described below.
4-26
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.5.2.1
Power-On
1. Turn on all the main line switches of the expansion cabinet.
If an optional power cabinet is connected, also turn on all the main line switches
of the power cabinet.
2. Turn on all the main line switches of the base cabinet.
If an optional power cabinet is connected, also turn on all the main line switches
of the power cabinet.
4.5.2.2
Power-Off
1. Turn off all the main line switches of the base cabinet.
If an optional power cabinet is connected, also turn off all the main line switches
of the power cabinet. (Note)
2. Turn off all the main line switches of the expansion cabinet.
Note – Please wait at least 30 seconds before turning on the system power that you
turned off, by using the main line switch or the circuit breakers on the distribution
panel.
4.5.3
Main Line Switch Locations
This section describes the locations of main line switch for each type of power
supply. (See FIGURE 4-4 through FIGURE 4-9)
A power cabinet is required when using the dual power feed option. However, in
the case of the SPARC Enterprise M8000 server single-phase and dual power feed,
the power cabinet is not required because the rack-mountable dual power feed
option is mounted in the SPARC Enterprise M8000 server cabinet (See FIGURE 4-5).
In the case of the SPARC Enterprise M9000 server single-phase and dual power feed,
main line switches are mounted in both the SPARC Enterprise server cabinet and the
power cabinet (See FIGURE 4-8).
In the case of the three-phase dual power feed, the main line switch is located in the
power cabinet (See FIGURE 4-6 and FIGURE 4-9).
Chapter 4
FRU Removal Preparation
4-27
4.5.3.1
SPARC Enterprise M8000 Server Single-Phase Power Feed
FIGURE 4-4
Main Line Switch Locations (Rear of the M8000)
M8000 single-phase power feed:
-Main line switch (x3)
-ACS_A
4-28
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.5.3.2
SPARC Enterprise M8000 Server Single-Phase and Dual
Power Feed
FIGURE 4-5
Main Line Switch Locations (Rear of the M8000)
M8000 single-phase and
dual power feed:
M8000 single-phase and
dual power feed:
-Main line switch (x3)
-Main line switch (x3)
-ACS_C
-ACS_A
Chapter 4
FRU Removal Preparation
4-29
4.5.3.3
SPARC Enterprise M8000 Server Three-Phase Power Feed
FIGURE 4-6
Main Line Switch Locations (Power Cabinet)
M8000 three-phase power feed:
M8000 three-phase power feed:
-Switch (x4)
-Switch (x5)
-ACS_D
-ACS_D
Front
M8000 three-phase power feed
(for dual power feed option):
-Main line switch (x1)
-ACSTPH_A (delta connection)
or ACSTPH_B (star connection)
Rear
M8000 three-phase power feed
(for base power feed):
-Main line switch (x1)
-ACSTPH_A (delta connection)
or ACSTPH_B (star connection)
Note – Same type of switch is used for main line switches in base cabinet and power
cabinet on a system with three-phase dual power feed.
4-30
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.5.3.4
SPARC Enterprise M9000 Server Single-Phase Power Feed
FIGURE 4-7
Main Line Switch Locations (Front of the M9000)
M9000 single-phase power feed:
-Main line switch (x5)
-ACS_B
Chapter 4
FRU Removal Preparation
4-31
4.5.3.5
FIGURE 4-8
SPARC Enterprise M9000 Server Single-Phase and Dual
Power Feed
Main Line Switch Locations (Front of the Power Cabinet + M9000)
M9000 single-phase and
dual power feed:
-Main line switch (x5)
-ACS_B
M9000 single-phase and
dual power feed:
-Main line switch (x5)
-ACS_B
4-32
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.5.3.6
FIGURE 4-9
SPARC Enterprise M9000 Server Three-Phase Power Feed
Main Line Switch Locations (Power Cabinet)
M9000 three-phase power feed:
M9000 three-phase power feed:
-Switch (x5)
-Switch (x5)
-ACS_D
-ACS_D
Front
M9000 three-phase power feed
(for dual power feed option):
-Main line switch (x1)
-ACSTPH_A (delta connection)
or ACSTPH_B (star connection)
Rear
M9000 three-phase power feed
(for base power feed):
-Main line switch (x1)
-ACSTPH_A (delta connection)
or ACSTPH_B (star connection)
Note – Same type of switch is used for main line switches in base cabinet and power
cabinet on a system with three-phase dual power feed.
Chapter 4
FRU Removal Preparation
4-33
4.6
Emergency Switch-Off
In emergency situations (such as smoke or flames are coming from the server), you
need to immediately stop the use and turn off the power supply.
Caution – If smoke or flames are coming from the server, switch off all main line
switches immediately, or use the customer's distribution panel to immediately stop
the power supply. Regardless of the job at hand, remember to give top priority to fire
prevention.
4.7
Cable Routing of the SPARC Enterprise
M8000 Server
In the rack space for the SPARC Enterprise M8000 server, when a 19-inch unit such
as the external I/O expansion unit mounted besides the rack-mountable dual power
feed, the cables connected to the unit must be properly routed.
This section provides an example of the cable routing when the external I/O
expansion unit mounted. Perform the cable routing properly with reference to this
section while following the precautions.
To do maintenance on the I/O Unit or the FAN unit, you need the cable routing in
the same manner. See Chapter 7 and Chapter 8.
4.7.1
Cable Routing When the External I/O Expansion
Unit Mounted
FIGURE 4-10 shows the example of the external I/O expansion units mounted in the
rack space. Up to 3 external I/O expansion units can be mounted in the rack space
for the SPARC Enterprise M8000 server. Follow the precautions below when routing
the cables.
4-34
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 4-10
4.7.1.1
Example of the External I/O Expansion Units Mounted in the Rack Space for
the SPARC Enterprise M8000 Server (front)
Precautions For Cable Routing
Note – In principle, run LAN cables to the external I/O expansion units along the
route on the right side of the rack (right route). If the total number of LAN cables
exceeds 72, place only 72 of the LAN cables along the right route, and place the
remaining LAN cables along the route on the left side of the rack (left route).
Chapter 4
FRU Removal Preparation
4-35
Note – The SPARC Enterprise M8000 server comes with 22 short hook-and-loop
fasteners and 4 long hook-and-loop fasteners for cables. Use the type of
hook-and-loop fastener appropriate to the situation.
Note – To do maintenance on the IOU or the FAN unit of the SPARC Enterprise
M8000 server, push aside the cables.
Caution – Do not place any cable over a PCI cassette at the front of an IOU in the
SPARC Enterprise M8000 server. (Adjust the positioning to divide and bundle the
cables.)
Caution – Take care to prevent the cables from blocking the ventilation slits of the
SPARC Enterprise M8000 server. (Fujitsu recommends running the cables above the
handle of the FAN unit.)
Caution – Keep optical fiber cables as far apart from other cables as possible to
prevent the optical fiber cables from being covered by the other cables.
Caution – Fasten the bundles of cables with the attached hook-and-loop fastener to
the SPARC Enterprise M8000 server, while avoiding the following locations which
may affect the system operation:
- The handle or the lever of the AC section
- The handle of the FAN unit
- The handle or the lever of the IOU
- The lever of the PCI cassette
Caution – When bending a cable, keep its bending radius within the permissible
bending radius.
Caution – Take sufficient care not to touch the main line switch when laying cables.
4-36
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.7.1.2
When Three External I/O Expansion Units Mounted
This section provides an example of the cable routing procedure applied when three
external I/O expansion units are mounted in the rack space for the SPARC
Enterprise M8000 server.
Cable Routing Procedure
The cable routing procedure explained below is an example of the recommended
procedure.
This section explains the recommended cable routing procedure in an example using
a configuration in which three external I/O expansion units are mounted and 34
LAN cables are connected to each external I/O expansion unit.
When actually routing cables, perform the work properly with reference to this
procedure while following the precautions for cable routing.
Hook-and-loop Fasteners Used
This procedure specifies only the short and long hook-and-loop fasteners
recommended for use. During actual cable routing, use the type of hook-and-loop
fastener appropriate for the number of cables to be bound.
1. Run 72 LAN cables (of the cables connected to the external I/O expansion units)
along the right route.
Chapter 4
FRU Removal Preparation
4-37
2. Run the remaining cables along the left route.
FIGURE 4-11
Cable Routing When Three External I/O Expansion Units Mounted
(M8000, rear view)
Left route
Right route
Part A
Ventilation slits
Part B
Part C
4-38
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
3. Bind the cables connected to each external I/O expansion unit with a short
hook-and-loop fastener. (See (1) in FIGURE 4-12.)
Caution – Keep optical fiber cables as far apart from other cables as possible to
prevent the optical fiber cables from being covered by the other cables.
4. Bind the bundles of cables with a long hook-and-loop fastener at location (2)
shown in FIGURE 4-12 while taking care not to touch the main line switch of the
ACS.
5. Fasten the bundles of cables with a long hook-and-loop fastener to the cable
holder at location (3) shown in FIGURE 4-12 while taking care not to touch the
main line switch of the ACS.
Chapter 4
FRU Removal Preparation
4-39
FIGURE 4-12
4-40
Magnified View of Part A (M8000, rear view)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.7.1.3
For Server Using Three-phase Power Feed
Note – The server using three-phase power feed does not have any cable holder. In
this case, fasten the cables by running the hook-and-loop fastener through the slit
between the bracket at the cable holder mounting location and a column in the
cabinet. (FIGURE 4-13 shows the fastening location on the left route.)
FIGURE 4-13
Cable Routing of the Server Using Three-Phase Power Feed
Location for Attaching a Hook-and-loop Fastener
Column of cabinet
Cable holder
Attached Hook-and-loop Fastener
Hook-and-loop fastener
mounting bracket
6. Fasten the cables running along the left route to the cable holder with a long
hook-and-loop fastener. (See (4) in FIGURE 4-14.)
7. Allow 12 cables of all the cables running along the right route to hang along
the right frame. (See (5) in FIGURE 4-14.)
8. Run the remaining cables from the right side to the left side while taking care
to prevent the cables from blocking the ventilation slits. (See (6) in FIGURE 4-14.)
Note – Fujitsu recommends running the cables above the handle of the FAN unit.
9. Fasten 24 cables with a short hook-and-loop fastener to the handle of the fan
shelf. (See (7) in FIGURE 4-14.)
Chapter 4
FRU Removal Preparation
4-41
10. Divide the remaining cables into three routes (about 12 cables per route), bind
the cables on each route with a short hook-and-loop fastener, and allow them to
hang from the fastener. (See (8) in FIGURE 4-14.)
Caution – Adjust the positioning of the routes appropriately so that the cables do
not cover a PCI cassette at the front of an IOU.
FIGURE 4-14
4-42
Magnified View of Part B (M8000, rear view)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
11. Fasten the cables running along the left route to the cable holder with a short
hook-and-loop fastener. (See (9) in FIGURE 4-15.)
12. Bind each remaining bundle of hanging cables with a short hook-and-loop
fastener. (See (10) in FIGURE 4-15.)
13. Also bind each of these bundles of cables with a short hook-and-loop fastener
at a location near the floor panel. (See (11) in FIGURE 4-15.)
Binding the cables near the floor panel facilitates cable handling through floor
panel openings.
FIGURE 4-15
Magnified View of Part C (M8000, rear view)
Chapter 4
FRU Removal Preparation
4-43
FIGURE 4-16
4-44
Completed Cable Routing with Three External I/O Expansion Units Mounted
(M8000, rear view)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
4.7.1.4
When One External I/O Expansion Unit Mounted
This section provides an example of the cable routing procedure applied when one
external I/O expansion unit is mounted in the rack space for the SPARC Enterprise
M8000 server.
Cable Routing Procedure
The cable routing procedure explained below is an example of the recommended
procedure.
This section explains the recommended cable routing procedure in an example using
a configuration in which one external I/O expansion unit is mounted and 34 LAN
cables are connected to the external I/O expansion unit.
When actually routing cables, perform the work properly with reference to this
procedure while following the precautions for cable routing.
Hook-and-loop Fasteners Used
This procedure specifies only the short and long hook-and-loop fasteners
recommended for use. During actual cable routing, use the type of hook-and-loop
fastener appropriate for the number of cables to be bound.
1. Run all the cables connected to the external I/O expansion unit along the right
route.
Chapter 4
FRU Removal Preparation
4-45
FIGURE 4-17
Cable Routing When One External I/O Expansion Unit Mounted (M8000, rear view)
Part A
Ventilation
slits
Part B
4-46
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2. Bind the cables from the external I/O expansion unit with a short
hook-and-loop fastener while taking care not to touch the main line switch of
the ACS. (See (1) in FIGURE 4-18.)
3. Fasten the cables with a short hook-and-loop fastener to the cable holder at
location (2) shown in FIGURE 4-18.
FIGURE 4-18
Magnified View of Part A (M8000, rear view)
4. Fasten the cables with a short hook-and-loop fastener to the cable holder at
location (3) shown in FIGURE 4-19.
5. Also bind the cables with a short hook-and-loop fastener at a location near the
floor panel. (See (4) in FIGURE 4-19.)
Binding the cables near the floor panel facilitates cable handling through floor
panel openings.
Chapter 4
FRU Removal Preparation
4-47
FIGURE 4-19
4-48
Magnified View of Part B (M8000, rear view)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 4-20
Completed Cable Routing with One External I/O Expansion Unit Mounted
(M8000, rear view)
Chapter 4
FRU Removal Preparation
4-49
4-50
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
PA RT
II
Maintenance
Part II explains how to remove and replace FRUs.
The chapters are organized according to the FRU mounting locations.
Chapter 6 starts with an overview of FRUs, followed by explanations of the FRU
removal and replacement procedures.
CHAPTER
5
Internal Components Access
This chapter explains how to access each unit of the server, in the following sections:
■
■
■
■
5.1
Section 5.1,
Section 5.2,
page 5-2
Section 5.3,
Section 5.4,
“How to Open and Close Doors” on page 5-1
“Corresponding Components and Doors That Can Be Accessed” on
“How to Remove a Door” on page 5-3
“How to Remove a Side Panel” on page 5-4
How to Open and Close Doors
The high-end server is equipped with doors at the front and rear. The operator panel
can be operated and checked when the doors are closed, but, to replace the operator
panel or to access or maintain other components, the doors must be opened.
However, the backplane cannot be accessed by only opening a door. It cannot be
accessed unless some units and working parts connected to it are removed.
As shown in FIGURE 5-1, each door can be opened and closed by inserting a key into
the handle lock and turning the handle. Borrow the key from the system
administrator.
Note – There are two types of key: one for the doors of the server and the other for
the operator panel. Both are managed by the system administrator.
The detailed procedure below explains how to open a door. The same procedure
applies to both the front and rear doors.
1. Insert the key into the lock above the handle, and turn it clockwise 90 degrees.
5-1
2. Push the projection down below the handle.
The part below the handle projects outward.
3. Turn the handle counterclockwise and pull it out to open the door.
To close the door, reverse the above steps for opening it.
FIGURE 5-1
5.2
How to Open a Door (Front Door as an Example)
Corresponding Components and Doors
That Can Be Accessed
TABLE 5-1 indicates which door should be opened to access a certain component.
Components are represented by their abbreviated names. For their formal names, see
TABLE B-1 in Appendix B.
5-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE 5-1
Model
Relationship Between Components and Doors That Can Be Accessed
Component (abbreviation) maintained through front
door
Component (abbreviation) maintained through rear
door
M8000
CMU, XSCFU, TAPEU, DVDU, DDC_A, PSU, IOU, ACS_A, and FAN_B (2-FAN #2 to #7)
FAN_A (3-FAN), FAN_B (2-FAN #0 and #1),
OPNL, SNSU, and RDPF
M9000
CMU, IOU (odd ID), and FAN_A (3-FAN #4
PSU, IOU (even ID), XSCFU_B, XSCFU_C
to #15)
(expansion cabinet), CLKU_B, XBU_B,
TAPEU, DVDU, PSU, ACS_B, FAN_A (3-FAN
#0 to #3), OPNL, and SNSU
5.3
How to Remove a Door
If it is absolutely necessary to remove a door during maintenance, refer to
FIGURE 5-2 and remove the grounding wire from the fixing bracket at the upper
part of the door and the hinge parts. Then, lift the door to disengage it. Keep the
removed door in a safe place.
The detailed procedure below explains how to remove a door. Most of the steps
are the same when applied to the front or rear door.
1. Remove the operator panel from the front door. This step is not required for the
rear door.
a. Unhook the cable holders to free the cables of the operator panel.
b. Using a Phillips screwdriver, loosen the two screws securing the operator
panel, and remove the operator panel from the front door.
c. Affix the operator panel to some convenient location on the cabinet.
Note – If the operator panel need not be removed for maintenance, you can remove
the cables from the sensor unit side.
2. Remove the screw securing the upper hinge bracket on the right, and remove
the grounding wire as well.
3. While supporting the door with one hand, loosen the screw securing the upper
hinge bracket on the left.
4. Lift the door, and remove it from the lower hinge bracket.
Chapter 5
Internal Components Access
5-3
5. Keep the door in a safe place.
To reinstall the door, reverse the above steps for removing it.
FIGURE 5-2
How to Remove a Door (Front Door)
Operator panel
Cable holder
Sensor unit
5.4
How to Remove a Side Panel
To install an optional power supply cabinet or to connect an expansion cabinet to the
M9000 base cabinet, the side panel of the cabinet must be removed.
The side panel hangs from a bracket on the top of the cabinet, and the lower part of
the side panel is secured with a screw or screws.
The structures of side panels are the same even for side panels of different sizes. A
large side panel is secured with two screws, and a small one is secured with one
screw.
5-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
The procedure for removing the side panel is follows.
1. Using a Phillips screwdriver, completely loosen the one or two screws at the
lower part of the side panel.
2. Lift the side panel, and remove it from the cabinet.
3. Keep the side panel in a safe place.
To reinstall the side panel, reverse the above steps for removing it.
FIGURE 5-3
How to Remove a Side Panel
Chapter 5
Internal Components Access
5-5
5-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
6
Replacement of CPU/Memory
Board Unit (CMU), CPU, and DIMM
This chapter explains the procedures for replacing the CPU/Memory board unit
(CMU) and the CPUs and DIMMs mounted in the CMU. It covers the following:
■
■
■
■
Section 6.1,
Section 6.3,
Section 6.4,
Section 6.5,
“Overview of the CMU” on page 6-1
“Active Replacement and Hot Replacement” on page 6-12
“Cold Replacement” on page 6-23
“CPU and DIMM Replacement” on page 6-29
There are three methods for replacing a CMU: active replacement, hot replacement,
and cold replacement. For the definition of each method, see Section 4.1, “Types of
Replacement Procedures” on page 4-2.
The section on swapping methods explains how to replace a CMU. For details on
swapping only a CPU or DIMM, see also Section 6.5, “CPU and DIMM
Replacement” on page 6-29.
Section 6.5, “CPU and DIMM Replacement” on page 6-29 also covers the memory
installation conditions when adding more DIMMs.
6.1
Overview of the CMU
This section provides an overview of the CMU and shows CMU mounting locations.
Each CMU has 4 CPU slots and 32 DIMM slots. When a CMU is replaced, the
mounted parts (CPUs and DIMMs) must be remounted in the new CMU. These
mounted parts must be remounted at the same locations in the new CMU as their
original pre-swap mounting locations. In addition, the flash memory of each CMU
(FMEM) stores the POST/OpenBoot™ PROM firmware. The firmware is
automatically restored to its pre-swap version.
6-1
For the version matching of firmware, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
FIGURE 6-1, FIGURE 6-2, and FIGURE 6-3 show the mounting locations of CMUs in the
high-end server cabinets.
FIGURE 6-1
CMU Mounting Locations (Front of the M8000)
CMU#3
CMU#2
CMU#1
CMU#0
6-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 6-2
CMU Mounting Locations (Rear of the M9000 Base Cabinet)
CMU#7
CMU#6
CMU#5
CMU#4
CMU#3
CMU#2
CMU#1
CMU#0
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-3
CMU Mounting Locations (Rear of M9000 with the Expansion Cabinet)
FIGURE 6-3
CMU#15
CMU#14
CMU#13
CMU#12
CMU#11
CMU#10
CMU#9
CMU#8
CMU#7
CMU#6
CMU#5
CMU#4
CMU#3
CMU#2
CMU#1
CMU#0
TABLE 6-1 lists the abbreviated names and component numbers of the CMUs.
TABLE 6-1
Abbreviated Names and Component Numbers of the CMUs
Model name
Abbreviated names and numbers
M8000
CMU#0 to 3
M9000 base cabinet
CMU#0 to 7
M9000 expansion cabinet
CMU#8 to 15
6-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
6.2
CPU Upgrade
This section describes the procedure to mount the SPARC64 VII processor on the
SPARC Enterprise M8000/M9000 server.
Caution – The SPARC64 VII processors are supported by the XCP firmware
(beginning with XCP 1070) and Solaris software. For more information about these
minimum software and firmware requirements, see the latest version of the Product
Notes (no earlier than the XCP 1070 edition) for your server.
Caution – You must complete the upgrades to the XCP firmware and to Solaris
before inserting SPARC 64 VII processors into the chassis. When they have the
minimum requirements of the XCP firmware and Solaris OS , it is not necessary to
update them.
For more information about configuring combinations of processors in domains,
refer to Section 2.2.13, "Domain Mode Configuration," in the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide. In particular, see the
section "SPARC64 VI and SPARC64 VII Processors and CPU Operational Modes."
6.2.1
SPARC64 VII CPU Modules Added to a New
Domain
Note – If you want to install Solaris 10 8/07 on the new domain, you must install
from a patched image on the installation server. (See Step 18.)
1. Log in to the XSCF on an account with platform administrative privileges.
2. Use the showstatus(8) command to confirm that a component in Faulted or
Deconfigured status doesn't exist.
XSCF> showstatus
3. Change the key position on the operator panel from Locked to Service.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-5
4. Collect an XSCF snapshot to archive system status prior to upgrade.
This will be help in case any problem occurred in this procedure.
XSCF> snapshot -t user@host:directory
5. Update the XCP version.
For the firmware updating procedures, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
6. After updating the XCP, reset the XSCF.
XSCF> rebootxscf
7. After resetting the XSCF, log in to the XSCF again.
8. Mount the CPU module (CPUM) on the CMU for add-on.
For the procedure, see the description about the CPU module installation in
Section 6.5.1, “Replacing a CPU Module” on page 6-30.
9. Mount the CMU, described in Step 8, on the server.
a. Executing the addfru(8) command and select “CMU/IOU” from the
maintenance menu.
XSCF> addfru
b. Perform the operation according to the instructions displayed in the
maintenance menu, referring to Section 6.3, “Active Replacement and Hot
Replacement” on page 6-12.
Note – Do not fail to execute the diagnosis of newly mounted CMU in the
maintenance menu of addfru(8) command.
10. Confirm that the mounted CPU module has been recognized by the server, and
the error indicator asterisk (*) is not displayed.
XSCF> showhardconf -M
11. Confirm no abnormality occurred by using showlogs error -v and
showstatus(8) commands.
XSCF> showlogs error -v
XSCF> showstatus
12. Move the key position on the operator panel from Service to Locked.
6-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
13. Sets the following for the CMU.
■
Set up XSB.
■
Set up the domain
■
Set up the CPU operational mode of the domain.
For each setting, see the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Use the setdomainmode(8) command, disable the autoboot function of the
domain.
Refer to the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF
User’s Guide and the setdomainmode(8) man page for more information.
15. Power on the domains.
XSCF> poweron -d domain_id
16. Confirm that the target domain has been correctly started.
XSCF> showlogs power
17. Confirm no abnormality occurred by using showlogs error -v and
showstatus(8) commands.
XSCF> showlogs error -v
XSCF> showstatus
18. Install a version of Solaris OS that supports SPARC64 VII processors.
Note – The SPARC64 VII processors are supported by the XCP firmware (beginning
with XCP 1070) and Solaris software. For more information about these minimum
software and firmware requirements, see the latest version of the Product Notes (no
earlier than the XCP 1070 edition) for your server.
For information about network-based installations, refer to Solaris 10 8/07
Installation Guide: Network-Based Installations (part 820-0177).
19. Use the setdomainmode(8) command to enable the autoboot function of the
domain.
The autoboot function is applied by a domain reboot. For more information, refer
to the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s
Guide and the setdomainmode(8) man page.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-7
6.2.2
SPARC64 VII Processors Added to an Existing
Domain
Adding SPARC64 VII Processors to an existing domain is a two step process. First
you must prepare the system (see Section 6.2.2.1, “Preparing to Add SPARC64 VII
Processors to an Existing Domain” on page 6-8) and then you must install the
Processors using the instructions that correspond to your installation scenario.
6.2.2.1
Preparing to Add SPARC64 VII Processors to an Existing
Domain
1. If necessary, upgrade to a version of Solaris OS that supports SPARC64 VII
processors.
Note – The SPARC64 VII processors are supported by the XCP firmware (beginning
with XCP 1070) and Solaris software. For more information about these minimum
software and firmware requirements, see the latest version of the Product Notes (no
earlier than the XCP 1070 edition) for your server.
2. Log in to the XSCF on an account with platform administrative privileges.
3. Use the showstatus(8) command to confirm that a component in Faulted or
Deconfigured status doesn't exist.
XSCF> showstatus
4. Turn off the power to a domain assigned the CMU of which you upgrade or add
the CPU.
XSCF> poweroff -d domain_id
5. Confirm that the target domain stopped.
XSCF> showlogs power
6. Change the key position on the operator panel from Locked to Service.
7. Collect an XSCF snapshot to archive system status prior to upgrade.
This will be help in case any problem occurred in this procedure.
XSCF> snapshot -t user@host:directory
6-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
8. Update the XCP version.
For the firmware updating procedures, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
9. After updating the XCP, reset the XSCF.
XSCF> rebootxscf
10. After resetting the XSCF, log in to the XSCF again.
11. Power on the target domain, and apply OpenBoot PROM firmware.
XSCF> poweron -d domain_id
It stops at ok prompt. It is not necessary to start Solaris OS.
12. Check the updated OpenBoot PROM version.
The OpenBoot PROM version of XCP 1082 is 02.09.0000.
XSCF> version -c cmu -v
DomainID
DomainID
DomainID
DomainID
...
DomainID
00
01
02
03
XSB#00-0
XSB#00-1
XSB#00-2
XSB#00-3
XSB#01-0
XSB#01-1
XSB#01-2
XSB#01-3
...
:
:
:
:
:
:
:
:
:
:
:
:
02.03.0000
02.03.0000
02.09.0000
02.09.0000
15 : 02.03.0000
02.03.0000
02.03.0000
02.03.0000
02.03.0000
02.03.0000
02.03.0000
02.03.0000
02.03.0000
(Current),
(Current),
(Current),
(Current),
(Reserve),
(Reserve),
(Reserve),
(Reserve),
02.02.0000
02.02.0000
02.02.0000
02.02.0000
02.09.0000
02.09.0000
02.09.0000
02.09.0000
(Reserve)
(Reserve)
(Reserve)
(Reserve)
(Current)
(Current)
(Current)
(Current)
13. Turn off the power to the target domain.
XSCF> poweroff -d domain_id
14. Continue with the appropriate installation procedure:
■
If you are adding a new SPARC64 VII-equipped CPU module to a domain
configured with SPARC64 VI processors, continue with Section 6.2.2.2, “Adding a
SPARC64 VII CPU Module to a Domain Configured With SPARC64 VI” on
page 6-10.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-9
■
6.2.2.2
If you are upgrading an existing SPARC64 VI CPU module in an existing domain
to SPARC64 VII processors, continue with Section 6.2.2.3, “Upgrading a SPARC64
VI CPU Module to SPARC64 VII on an Existing Domain” on page 6-11.
Adding a SPARC64 VII CPU Module to a Domain Configured
With SPARC64 VI
This procedure must be preceded by Section 6.2.2.1, “Preparing to Add SPARC64 VII
Processors to an Existing Domain” on page 6-8. If you have not completed that
procedure, do so before continuing.
1. Install the CMU in the server.
■
For adding the CPU, operate by hot replacement, referring to Section 6.3, “Active
Replacement and Hot Replacement” on page 6-12.
Note – Do not fail to execute the diagnosis of newly mounted CMU in the
maintenance menu of replacefru(8) command.
2. Log in to the XSCF on an account with platform administrative privileges.
3. Confirm that the mounted CPU module has been recognized by the server, and
the error indicator asterisk (*) is not displayed.
XSCF> showhardconf -M
4. Confirm no abnormality occurred by using showlogs error -v and
showstatus(8) commands.
XSCF> showlogs error -v
XSCF> showstatus
5. Move the key position on the operator panel from Service to Locked.
6. In case of addition, sets the following for the CMU.
■
Set up XSB.
■
Set up the LSB.
■
Add the XSB to the domain.
Refer to Chapter 2, "Setting Up XSCF," in the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide for information
about these settings.
7. Power on the target domains.
XSCF> poweron -d domain_id
6-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
8. Confirm that the target domain has been correctly started.
XSCF> showlogs power
9. Confirm no abnormality occurred by using showlogs error -v and
showstatus(8) commands.
XSCF> showlogs error -v
XSCF> showstatus
6.2.2.3
Upgrading a SPARC64 VI CPU Module to SPARC64 VII on an
Existing Domain
This procedure must be proceeded by Section 6.2.2.1, “Preparing to Add SPARC64
VII Processors to an Existing Domain” on page 6-8. If you have not completed that
procedure please do so before continuing.
1. Log in to the XSCF on an account with platform administrative privileges.
2. Replace the SPARC64 VI CPU module with the SPARC64 VII CPU module.
■
For replacing the CPU, operate by hot replacement, referring to Section 6.3,
“Active Replacement and Hot Replacement” on page 6-12.
3. Confirm that the mounted CPU module has been recognized by the server, and
the error indicator asterisk (*) is not displayed.
XSCF> showhardconf -M
4. Confirm no abnormality occurred by using showlogs error -v and
showstatus(8) commands.
XSCF> showlogs error -v
XSCF> showstatus
5. Move the key position on the operator panel from Service to Locked.
6. Set up and confirm the CPU operational mode of the domain.
For more information, refer to Chapter 2, "Setting Up XSCF," in the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
7. Power on the target domains.
XSCF> poweron -d domain_id
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-11
8. Confirm that the target domain has been correctly started.
XSCF> showlogs power
9. Confirm no abnormality occurred by using showlogs error -v and
showstatus(8) commands.
XSCF> showlogs error -v
XSCF> showstatus
6.3
Active Replacement and Hot
Replacement
The active replacement procedure and hot replacement procedure are almost the
same. This section explains both procedures together. Each step that is different
between these procedures is distinguished by "Active replacement" or "Hot
replacement" as a header. The main difference between active replacement and hot
replacement is as follows:
■
Active replacement: A CMU is disconnected from or configured into a domain by
dynamic reconfiguration (DR).
There are restrictions on the DR operation depending on whether the Solaris OS
operates in the SPARC64™ VII enhanced mode or in the SPARC64™ VI
compatible mode. For DR operation, see the SPARC Enterprise
M4000/M5000/M8000/M9000 Servers Dynamic Reconfiguration (DR) User’s Guide.
■
Hot replacement: A domain must be stopped and restarted.
The procedure for active replacement and hot replacement is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching CMU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Check whether the CMU to be replaced is logically divided and what domains
are affected by the replacement.
a. From the console that is connected to XSCF, log in to XSCF, and execute the
showfru command. Then, check whether the CMU is logically divided.
6-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
b. If it is logically divided, execute the showdcl command to check whether
the CMU to be replaced and another CMU comprise a domain.
c. Execute the showboards command to check the use states of the devices and
resources in the CMU.
For details on how to log in to XSCF and commands, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
For details of commands, see also the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference Manual, or the man
page.
3. Prepare the CMU that is to be replaced.
■
Active replacement:
From the terminal connected to XSCF, execute the deleteboard command to
dynamically disconnect the CMU to be replaced from all the domains that have
been recognized in Step 2 as including the CMU. (The CMU status can be
checked by executing the showboards command.)
To execute this command, you need the privilege of platform administrator
(platadm) or domain administrator (domainadm). For details, see Section 4.2.1,
“Disconnecting a FRU from a Domain” on page 4-3, and the SPARC Enterprise
M4000/M5000/M8000/M9000 Servers Dynamic Reconfiguration (DR) User’s Guide.
For details on how to log in to the XSCF, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
■
Hot replacement:
From the terminal connected to the XSCF, select all the domains that have been
recognized in Step 2 as including the CMU to be replaced, and execute the
poweroff -d command to shut down the domains. (The CMU status can be
checked by executing the showboards command.)
Domain administrator authority is required for executing this command. For
details on how to log in to the XSCF, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-13
4. Use the maintenance menu to select the CMU to be replaced. (The maintenance
menu is used up to Step 20.)
The fieldeng permission is required for executing this command.
a. From the console that is connected to XSCF, log in to XSCF, and execute the
replacefru command.
As shown in FIGURE 6-4, the initial page of the maintenance menu is displayed.
FIGURE 6-4
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
--------------------------------------------------Select [1-5|c:cancel] :
Note – DDC_A is displayed only for the M8000.
b. From the replacement parts list in the maintenance menu, select 1,
"CMU/IOU."
c. Then, perform the operation according to the instructions displayed in the
maintenance menu.
Check the states of the LEDs (POWER LED: off; CHECK LED: blinking) on the
CMU to be replaced, and begin the actual replacement work when a message
instructing the user to replace the target CMU is displayed. Leave the
maintenance menu as is until Step 17.
5. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
6-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
6. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
7. Remove the CMU to be replaced.
a. Using a Phillips screwdriver, turn the knob screws securing the upper and
lower eject/lock levers 90 to 180 degrees to unlock the levers. (Turn the knob
screw while lightly pulling the lever by one hand.)
b. Unlock the eject/lock levers to pull the CMU off the backplane.
c. While grasping the handle at the front of the CMU, pull the CMU out
partially until the [NEAR TO END] label comes into view.
d. Raise the handle at the upper part of the CMU.
e. While grasping the front and upper handles of the CMU, pull it out slowly
from the slot.
Note – To upgrade the CMU, remove the dummy (filler) unit before mounting the
CMU to be added. Since the same levers are used on the dummy (filler) unit, please
operate the eject/lock levers in the same manner.
Caution – A CMU with mounted components in the maximum configuration
weighs about 22 kg.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-15
FIGURE 6-5
Removing the CMU (Front of the M8000)
3
1
2
1
6-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 6-6
Removing the CMU (Rear of the M9000 Base Cabinet)
3
1
2
1
8. Place the removed CMU with the cover and its label facing upward on an
antistatic mat.
9. Remove the cover of the CMU.
a. Pull the two latches on the connector side of the CMU to unlock the latches.
b. Slide the cover of the CMU toward the connector side to release it from the
guide pins (two pins on each side).
c. Raise the cover of the CMU to remove it.
10. Remove the DIMM duct cover. (See FIGURE 6-7)
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-17
11. Replace parts.
In CMU replacement:
Remove all the CPUs and DIMMs mounted in the removed CMU, and remount
them in the replacement CMU.
Caution – Be sure to remount the CPUs and DIMMs in the same locations in the
replacement CMU as their mounting locations in the removed CMU.
For CPU replacement:
Replace each swap target CPU mounted in the removed CMU with a replacement
CPU. For the replacement procedure, see Section 6.5, “CPU and DIMM
Replacement” on page 6-29.
6-18
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
For DIMM replacement:
Replace each swap target DIMM mounted in the removed CMU with a replacement
DIMM. For the replacement procedure, see Section 6.5, “CPU and DIMM
Replacement” on page 6-29.
FIGURE 6-7
CPU and DIMM Locations
Duct cover
DIMM (x32)
CPU (x4)
12. Place the DIMM duct cover back at its original location. Insert the L-shaped
projections on the CMU frame into the two positioning slits of the DIMM duct
cover.
13. Follow the removal instructions in Step 9 in reverse order to attach the cover of
the CMU, and press the latch pins to lock the latches.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-19
14. Place the replacement CMU on the antistatic mat, and touch the metallic chassis
for five or more seconds with your bare hand wearing the antistatic wrist strap,
to remove static electricity.
Caution – To mount a new CMU, ensure that the board is placed on a grounded
antistatic mat. Then, touch it with your bare hand while wearing a properly
connected antistatic wrist strap. This will remove any static electricity before
installing it in the server. Take care not to damage the connector on the CMU edge.
FIGURE 6-8
Metallic Chassis (CMU)
Touch for 5 or more seconds.
6-20
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
15. Touch each of the designated points on the guide blocks for 5 or more seconds
with your bare hand wearing the antistatic wrist strap.
FIGURE 6-9
Guide Block (CMU)
Touch for 5 or more seconds.
16. Mount the replacement CMU by following the removal instructions in Step 7 in
reverse order. Align the CMU with the slot guides, insert it carefully, and
secure it firmly.
Push the eject/lock levers as far as they will go, and then push the knob screws
on the upper and lower eject/lock levers to lock them in position. When the knob
screws remain in position, the CMU is securely mounted.
Note – Mount the dummy (filler) unit in the same manner as CMU. Since the same
levers are used on the dummy (filler) unit, please operate the eject/lock levers in the
same manner.
Caution – Do not forcibly push the CMU when inserting it, even if it is not moving
smoothly. If the CMU is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
17. Go back to the terminal that is connected to XSCF, and perform the input
operation according to the instructions of the maintenance menu that is
displayed after the replacement.
a. Because the displayed messages are different between active replacement
and hot replacement, perform the operation according to the instructions for
each case. When the display of the maintenance menu returns to the initial
page, the operation is completed.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-21
b. Exit the maintenance menu.
18. Configure the CMU into the system.
■
Active replacement:
Use the addboard command to dynamically configure the relevant CMU into
a domain. Perform this operation for all the domains that have been recognized
in Step 2 as including the CMU to be replaced.
For details, see Section 4.2.2, “Disconnecting and Replacing a FRU” on
page 4-5, and the SPARC Enterprise M4000/M5000/M8000/M9000 Servers
Dynamic Reconfiguration (DR) User’s Guide.
■
Hot replacement:
From the console that is connected to XSCF, execute the poweron -d
command of XSCF to turn on power to the domain.
When the power to the domain is turned on, version synchronization of
OpenBoot PROM (OBP) is automatically performed. When version
synchronization is completed, the power-on self test (POST) is performed.
When ok prompt displayed on the console of relevant domain, POST is
completed.
19. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for a replacement CMU, CPU, or DIMM, it is operating
normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
20. Close the door on the server, and lock it.
6-22
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
6.4
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching CMU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-23
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the CMU to be replaced. (See FIGURE 6-5 or FIGURE 6-6)
a. Using a Phillips screwdriver, turn the knob screws securing the upper and
lower eject/lock levers 90 to 180 degrees to unlock the levers. (Turn the knob
screw while lightly pulling the lever by one hand.)
b. Unlock the eject/lock levers to pull the CMU off the backplane.
c. While grasping the handle at the front of the CMU, pull out the CMU
partially until the [NEAR TO END] label comes into view, and then stop
pulling.
d. Raise the handle at the upper part of the CMU.
6-24
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
e. While grasping the front and upper handles of the CMU, pull it out slowly
from the slot.
Note – To upgrade the CMU, remove the dummy (filler) unit before mounting the
CMU to be added. Since the same levers are used on the dummy (filler) unit, please
operate the eject/lock levers in the same manner.
Caution – A CMU with mounted components in the maximum configuration
weighs about 22 kg.
9. Place the removed CMU with the cover facing upward on an antistatic mat.
10. Remove the cover of the CMU.
a. Pull the two latches on the connector side of the CMU to unlock the latches.
b. Slide the cover of the CMU toward the connector side to release it from the
guide pins (two pins on each side).
c. Raise the cover of the CMU to remove it.
11. Remove the DIMM duct cover.
12. Replace parts.
In CMU replacement:
Remove all the CPUs and DIMMs mounted in the removed CMU, and remount
them in the replacement CMU.
Note – Be sure to remount the CPUs and DIMMs in the same locations in the
replacement CMU as their mounting locations in the swap target CMU.
For CPU replacement:
Replace each swap target CPU mounted in the removed CMU with a replacement
CPU. For the replacement procedure, see Section 6.5, “CPU and DIMM
Replacement” on page 6-29.
For DIMM replacement:
Replace each swap target DIMM mounted in the removed CMU with a replacement
DIMM. For the replacement procedure, see Section 6.5, “CPU and DIMM
Replacement” on page 6-29.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-25
13. Place the DIMM duct cover back at its original location. Insert the L-shaped
projections on the CMU frame into the two positioning slits of the DIMM duct
cover.
14. Follow the removal instructions in Step 10 in reverse order to attach the cover of
the CMU, and press the latch pins to lock the latches.
15. Place the replacement CMU on the antistatic mat, and touch the metallic chassis
for five or more seconds with your bare hand wearing the antistatic wrist strap,
to remove static electricity.
Caution – To mount a new CMU, ensure that the board is placed on a grounded
antistatic mat. Then, touch it with your bare hand while wearing a properly
connected antistatic wrist strap. This will remove any static electricity before
installing it in the server. Take care not to damage the connector on the FRU edge.
FIGURE 6-10
Metallic Chassis (CMU)
Touch for 5 or more seconds.
6-26
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
16. Touch each of the designated points on the guide blocks for 5 or more seconds
with your bare hand wearing the antistatic wrist strap.
FIGURE 6-11
Guide Block (CMU)
Touch for 5 or more seconds.
17. Mount the replacement CMU by following the removal instructions in Step 8 in
reverse order. Align the CMU with the slot guides, insert it carefully, and
secure it firmly.
Push the eject/lock levers as far as they will go, and then push the knob screws
on the upper and lower eject/lock levers to lock them in position. When the knob
screws remain in position, the CMU is securely mounted.
Note – Mount the dummy (filler) unit in the same manner as CMU. Since the same
levers are used on the dummy (filler) unit, please operate the eject/lock levers in the
same manner.
Caution – Do not forcibly push the CMU when inserting it, even if it is not moving
smoothly. If the CMU is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
18. Switch on all main line switches for the AC section (ACS) in the power supply
system.
19. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-27
20. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From the console that is connected to XSCF, execute the
poweron -d <domain-id> command of XSCF for the domain of the swap
target new CMU.
Power to all the domains is turned on. Firmware version synchronization of the
target domain is automatically performed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
21. Using the XSCF version command, confirm that the firmware version of the
target domain is the same in the replacement CMU.
For details of firmware version synchronization, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
22. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for a replacement CMU, CPU, or DIMM, it is operating
normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
23. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
24. Close the door on the server, and lock it.
6-28
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
6.5
CPU and DIMM Replacement
When replacing a CPU module or DIMM, replace it with one of the same type. For
details of the mounting locations indicated in messages, see FIGURE 6-12.
FIGURE 6-12
CPU Module and DIMM (MEM) Mounting Locations
Front side
CPUM#3
CPUM#2
CPUM#1
CPUM#0
Chapter 6
MEM#30A
MEM#32A
MEM#30B
MEM#32B
MEM#31A
MEM#33A
MEM#31B
MEM#33B
MEM#21B
MEM#23B
MEM#21A
MEM#23A
MEM#20B
MEM#22B
MEM#20A
MEM#22A
MEM#10A
MEM#12A
MEM#10B
MEM#12B
MEM#11A
MEM#13A
MEM#11B
MEM#13B
MEM#01B
MEM#03B
MEM#01A
MEM#03A
MEM#00B
MEM#02B
MEM#00A
MEM#02A
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-29
6.5.1
Replacing a CPU Module
To replace the CPU module, use the CPU Module insertion/extraction tool. Side A is
for insertion, and side B is for extraction. The CPU Module insertion/expansion tool
is stored inside the server. In the SPARC Enterprise M8000 server, the tool can be
found on the left-side wall when the front door of the server is open. In the SPARC
Enterprise M9000 server base cabinet, the tool can be found on the upper part of the
unit when the rear door is open.
In the SPARC Enterprise M9000 server expansion cabinet, there is no
insertion/extraction tool.
FIGURE 6-13
CPU Module Insertion/Extraction Tool
B
A
6-30
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
1. Remove the CPU Module.
a. Insert the tips on the extraction side of the CPU Module insertion/extraction
tool (B) under the CPU Module guide pins. Insert the shorter guide pin into
the groove of the guide lock of the CMU.
FIGURE 6-14
Set CPU Module Insertion/Extraction Tool
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-31
FIGURE 6-15
CMU Guide Lock (Enlarged)
Caution – When using the CPU Module insertion/extraction tool, make sure that it
is set in position and used in the manner shown in the figure. If the CPU Module
insertion/extraction tool is set and used in a different direction from the figure,
serious damage may result.
b. Tilt the CPU Module insertion/extraction tool to the DIMM side, and pull up
the CPU Module to disengage the CMU connector.
6-32
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 6-16
Lever Up the CPU Module
Caution – When tilting the CPU Module insertion/extraction tool toward the
DIMM, do not tilt it to such an extent that the CPU Module is raised beyond its limit
height.
Caution – Be sure not to remove the tool before removing the CPU Module.
Otherwise, the CMU may be damaged.
c. Slide the CPU Module upward at an inclined angle to remove it from the
guide lock.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-33
FIGURE 6-17
Remove the CPU Module
d. Remove the CPU Module insertion/extraction tool.
6-34
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2. Attach the replacement CPU Module.
a. Slide and insert the guide pin on the DDC side of the CPU Module along the
groove of the CMU guide lock.
FIGURE 6-18
Set CPU Module on the Guide Lock (1)
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-35
FIGURE 6-19
6-36
CMU Guide Lock (Enlarged)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
b. Fit the guide pin on the heat sink side of the CPU Module into the groove of
the CMU guide lock.
FIGURE 6-20
Set CPU Module on the Guide Lock (2)
Note – Each CPU Module has two guide pins and two CMU guide locks on the
DDC side. Perform the work in such a way that both guide pins and both guide
locks are mounted correctly.
c. Move the CPU Module back and forth to confirm that the pins are in the
grooves.
d. Fit the shorter guide pin on the insertion side of the CPU Module
insertion/extraction tool (A) into the CMU guide lock.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-37
FIGURE 6-21
6-38
Set CPU Module Insertion/Extraction Tool
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 6-22
CMU Guide Lock (Enlarged)
Caution – When using the CPU Module insertion/extraction tool, make sure that it
is set in position and used in the manner shown in the figure. If the CPU Module
insertion/extraction tool is set and used in a different direction from the figure,
serious damage may result.
e. Raise the CPU Module insertion/extraction tool on the CPU Module side
until the tool is standing upright, perpendicular to the ground.
The CPU Module and CMU connectors are coupled, and the CPU Module is
mounted.
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-39
FIGURE 6-23
Mounting the CPU Module
Caution – Each CPU Module has two CMU guide locks. Perform the work in such
a way that both guide locks are correctly set. Raise the insertion/extraction tool until
it touches the stopper and is perpendicular to the ground. The connector is securely
engaged when the insertion/extraction tool is raised vertically and it comes into
contact with the stopper. Do not forcibly raise the insertion/extraction tool further.
f. Reverse Step 2 to Step e to remove the CPU Module insertion/extraction tool.
Caution – After using the CPUM insertion/extraction tool, be sure to return it to
the storage location in the server.
6.5.2
Memory Module Mounting Conditions
This section describes memory module (DIMM) mounting conditions.
6-40
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
The CMU has 16 DIMMs as standard. The number of DIMMs that can be mounted
on the CMU is 16 or 32, in sets of 16 DIMMs. Therefore, 16 DIMMs can be added to
the standard product.
Caution – To replace or install a DIMM, do not fail to confirm the DIMM
information and comply with the conditions to mount the memory.
6.5.2.1
Confirmation of DIMM Information
Confirm the DIMM information (size/rank) in the following way.
■
Execute the showhardconf(8) command on XSCFU.
The Type field shows the DIMM size and the DIMM rank.
XSCF> showhardconf
...
CMU#1 Status:Normal; Ver:0101h; Serial:PP0642Z470 ;
+ FRU-Part-Number:CA06620-D001 A8 ;
+ Memory_Size:64 GB;
...
MEM#00A Status:Normal;
+ Code:2cffffffffffffff0836HTF25672Y-53EB1 0100-d409da25;
+ Type:2B; Size:2 GB;
MEM#00B Status:Normal;
+ Code:2cffffffffffffff0836HTF25672Y-53EB1 0100-d409da1e;
+ Type:2B; Size:2 GB;
MEM#01A Status:Normal;
+ Code:2cffffffffffffff0836HTF25672Y-53EB1 0100-d409da22;
+ Type:2B; Size:2 GB;
MEM#01B Status:Normal;
+ Code:2cffffffffffffff0836HTF25672Y-53EB1 0100-d409da23;
+ Type:2B; Size:2 GB;
...
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-41
FIGURE 6-24 shows an example of how to read the DIMM information displayed in
the Type field.
FIGURE 6-24
Example of Reading DIMM Information
MEM#00A Status:Normal;
Memory slot No.
+ Code:2cffffffffffffff0836HTF25672Y-53EB1 0100-d409da25;
+ Type:2B; Size:2 GB;
DIMM size
1: 1 GB
2: 2 GB
4: 4 GB
...
DIMM rank
A: 1 rank
B: 2 rank
6.5.2.2
DIMM Mounting Conditions
■
■
■
■
■
Memory modules are added in sets of 16 DIMMs.
The DIMMs mounted in the MEM#xxA slot are defined as Group A. The DIMMs
mounted in the MEM#xxB slot are defined as Group B. The standard memory
modules are included in Group A.
The memory size of Group A is equal to or greater than that of Group B.
Group B need not always include memory modules.
Mount memory modules of the same size and rank in each group. Memory
modules of different sizes or ranks cannot be mounted in a single group.
Note – When you upgrade using larger capacity memory than the capacity of
Group-A, move the memory mounted in Group-A to Group-B, and then mount the
upgrade memory to Group-A.
6-42
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 6-25
Memory Mounting Conditions
DIMM
MEM#30A
CPUM#3 MEM#30B
MEM#31A
MEM#31B
MEM#21B
CPUM#2 MEM#21A
MEM#20B
MEM#20A
MEM#10A
CPUM#1 MEM#10B
MEM#11A
MEM#11B
MEM#01B
CPUM#0 MEM#01A
MEM#00B
MEM#00A
Memory modules
mounted as standard
MEM#32A
MEM#32B
MEM#33A
MEM#33B
MEM#23B
MEM#23A
MEM#22B
MEM#22A
MEM#12A
MEM#12B
MEM#13A
MEM#13B
MEM#03B
MEM#03A
MEM#02B
MEM#02A
Added
memory modules
Conceptual diagram of
CPU and memory module locations
Chapter 6
Mounting locations of
CPUs and memory modules
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-43
6.5.3
DIMM Replacement
1. Open outward the tabs of the connector in which the DIMM to be replaced is
mounted, and pull the DIMM out vertically.
FIGURE 6-26
Removing the DIMM
Note – While grasping both ends of the DIMM, pull it out vertically.
6-44
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2. Perform the above operation in reverse order to mount the DIMM.
Note – To mount DIMM, position the notch of DIMM to the corresponding part of
the connector.
FIGURE 6-27
Notch of DIMM
Chapter 6
Replacement of CPU/Memory Board Unit (CMU), CPU, and DIMM
6-45
6-46
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
7
I/O Unit (IOU) Replacement
This chapter explains the replacement procedures for the I/O unit (IOU). It covers
the following:
■
■
■
Section 7.1, “Overview of the IOU” on page 7-2
Section 7.2, “Active Replacement and Hot Replacement” on page 7-8
Section 7.3, “Cold Replacement” on page 7-17
There are three methods for replacing an IOU: active replacement, hot replacement
and cold replacement. For the definition of each type of method, see Section 4.1,
“Types of Replacement Procedures” on page 4-2.
In case of the SPARC Enterprise M8000 server, when you do maintenance on the I/O
unit, the cables connected to the unit which mounted in the rack space for the
SPARC Enterprise M8000 server must be properly routed. For details, see Section 4.7,
“Cable Routing of the SPARC Enterprise M8000 Server” on page 4-34.
Note – To do maintenance on the IOU or the FAN unit of the SPARC Enterprise
M8000 server, push aside the cables.
Note – When mounting an IOU, also mount the CPU/memory board (CMU) whose
component number represents the same mounting location as the IOU.
Example: Mount IOU#1 together with CMU#1.
Note – In hot replacement of an IOU, the CMU whose component number
represents the same mounting location as the IOU must be disconnected from the
domain. Neither the DR operation for only an IOU nor hot replacement of only an
IOU are possible.
7-1
Caution – Adjust the positioning of the routes appropriately so that the cables do
not cover a PCI cassette at the front of an IOU in the SPARC Enterprise M8000
server.
7.1
Overview of the IOU
This section provides an overview of IOUs and describes their mounting locations.
Each IOU consists of one I/O board (IOB), eight PCI slots, and up to four hard disk
drives (HDD). A PCI card or link card is mounted in a PCI slot by using a PCI
cassette. For slots in which no PCI card or link card is mounted, a filler panel may
occupy the slot. When an IOU is replaced, these mounted components (PCI cassettes,
PCI cards, HDDs, and other components) must be remounted. These mounted parts
must be remounted at the same locations in the new IOU as their original pre-swap
mounting locations.
7-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 7-1, FIGURE 7-2, FIGURE 7-3, FIGURE 7-4, and FIGURE 7-5 indicate the mounting
locations of IOUs.
FIGURE 7-1
IOU Mounting Locations (Rear of the M8000)
IOU#3
IOU#2
IOU#1
IOU#0
Chapter 7
I/O Unit (IOU) Replacement
7-3
FIGURE 7-2
IOU mounting Locations (Front of the M9000 Base Cabinet)
IOU#0
IOU#2
IOU#4
IOU#6
7-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 7-3
IOU Mounting Locations (Rear of the M9000 Base Cabinet)
IOU#7
IOU#5
IOU#3
IOU#1
Chapter 7
I/O Unit (IOU) Replacement
7-5
FIGURE 7-4
IOU Mounting Locations (Front of M9000 with the Expansion Cabinet)
IOU#0
IOU#2
IOU#4
IOU#6
IOU#8
IOU#10
IOU#11
IOU#12
7-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 7-5
IOU Mounting Locations (Rear of M9000 with the Expansion Cabinet)
IOU#15
IOU#13
IOU#11
IOU#9
IOU#7
IOU#5
IOU#3
IOU#1
TABLE 7-1 lists the abbreviated names and component numbers of the IOUs.
TABLE 7-1
Abbreviated Names and Component Numbers of IOUs
Model name
Abbreviated names and numbers
M8000
IOU#0 to 3
M9000 base cabinet
IOU#0, 2, 4, and 6 (front)
IOU#1, 3, 5, and 7 (rear)
M9000 expansion cabinet
IOU#8, 10, 12, and 14 (front)
IOU#9, 11, 13, and 15 (rear)
Chapter 7
I/O Unit (IOU) Replacement
7-7
7.2
Active Replacement and Hot
Replacement
The active replacement procedure and hot replacement procedure are almost the
same. This section explains both procedures together. Each step that is different
between these procedures is distinguished by "Active replacement" or "Hot
replacement" as a header. The main difference between active replacement and hot
replacement is as follows:
■
Active replacement: The CMU of an IOU is disconnected from or configured into
a domain by dynamic reconfiguration (DR).
■
Hot replacement: A domain must be stopped and restarted.
The procedure for active replacement and hot replacement is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching IOU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Check whether the CMU that is connected to the IOU to be replaced (the target
CMU) is logically divided and what domains are affected by the replacement.
a. Log in to the XSCF from a terminal connected to the XSCF, and execute the
showfru command to check whether the target CMU is logically divided.
b. If it is logically divided, execute the showdcl command to check whether
the target CMU and another CMU comprise a domain.
c. Execute the showboards command to check the use states of the devices and
resources in the CMU.
For details on how to log in to the XSCF and of commands, see the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide. For
details of commands, see also the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference Manual, or the man
page.
7-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
3. Prepare the IOU that is to be replaced.
■
Active replacement:
From the terminal connected to the XSCF, execute the deleteboard command
to dynamically disconnect the target CMU from all the domains that have been
recognized in Step 2 as including the target CMU. (The IOU status can be
checked by executing the showboards command.)
To execute this command, you need the privilege of platform administrator
(platadm) or domain administrator (domainadm). For details, see
Section 4.2.1, “Disconnecting a FRU from a Domain” on page 4-3, and the
SPARC Enterprise M4000/M5000/M8000/M9000 Servers Dynamic Reconfiguration
(DR) User’s Guide . For details on how to log in to XSCF, see the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
■
Hot replacement:
From the terminal connected to the XSCF, select all the domains that have been
recognized in Step 2 as including the CMU to be replaced, and execute the
poweroff -d command to shut down the domains. (The IOU status can be
checked by executing the showboards command.)
Domain administrator authority is required for executing this command. For
details on how to log in to XSCF, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
4. Use the maintenance menu to select the replacement target IOU. (The
maintenance menu is used up to Step 14.)
The fieldeng permission is required for executing this command.
a. From the console that is connected to XSCF, log in to XSCF, and execute the
replacefru command.As shown in FIGURE 7-6, the initial page of the
maintenance menu is displayed.
FIGURE 7-6
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
--------------------------------------------------Select [1-5|c:cancel] :
Chapter 7
I/O Unit (IOU) Replacement
7-9
Note – DDC_A is displayed only for the M8000.
b. From the replacement parts list in the maintenance menu, select 1,
"CMU/IOU."
c. From here, perform the operation according to the guidance displayed in the
maintenance menu.
Check the states of LEDs (POWER LED: off; CHECK LED: blinking) on the
IOU to be replaced, and begin the actual replacement work when a message
instructing the user to replace the target IOU is displayed. Leave the
maintenance menu as is until Step 13.
5. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
6. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
7. Remove the IOU to be replaced.
a. Using a Phillips screwdriver, turn the knob screws securing the upper and
lower eject/lock levers 90 to 180 degrees to unlock the levers. (Turn the knob
screw while lightly pulling the lever by one hand.)
b. Unlock the eject/lock levers to pull the IOU off the backplane.
c. While grasping the handle at the front of the IOU, pull the IOU out partially
until the [NEAR TO END] label comes into view.
d. Raise the handle at the upper part of the IOU.
e. While grasping the front and upper handles of the IOU, pull it out slowly
from the slot.
Note – To upgrade the IOU, remove the dummy (filler) unit before mounting the
IOU to be added. Since the same levers are used on the dummy (filler) unit, please
operate the eject/lock levers in the same manner.
7-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Caution – An IOU with all options mounted weighs about 21 kg.
Removing the IOU (Rear of M8000)
FIGURE 7-7
3
1
2
1
Chapter 7
I/O Unit (IOU) Replacement
7-11
FIGURE 7-8
Removing the IOU (Front of M9000)
3
1
2
1
7-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 7-9
Removing the IOU (Rear of M9000)
3
1
2
1
8. Place the removed IOU on an antistatic mat.
9. Dismount the HDD (or filler panel) and the PCI cards on the detached IOU,
and remount them all on the replacing IOU. As for the PCI cassette, which
serves as the housing for PCI card, use the one which is mounted on the
replacing IOU. For the remounting procedures, see Chapter 12 and Chapter 13.
Note – Be sure to remount the PCI cards, filler boards, and HDDs in the same
locations in the replacement IOU as their mounting locations in the removed IOU.
Note – When mounting PCI cards and HDDs, insert them all the way in so that the
connectors are fully engaged.
Chapter 7
I/O Unit (IOU) Replacement
7-13
FIGURE 7-10
Locations of PCI Slots and HDDs
HDD#0
HDD#1
HDD#2
HDD#3
PCI#4
PCI#5
PCI#6
PCI#7
PCI#0
PCI#1
PCI#2
PCI#3
10. Place the replacement IOU on the antistatic mat, and touch the metallic chassis
for five or more seconds with your bare hand wearing the antistatic wrist strap,
to remove static electricity.
Caution – To mount a new CMU, ensure that the board is placed on a grounded
antistatic mat. Then, touch it with your bare hand while wearing a properly
connected antistatic wrist strap. This will remove any static electricity before
installing it in the server. Take care not to damage the connector on the CMU edge.
7-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 7-11
Metallic Chassis (IOU)
Touch for 5 or more seconds.
11. Touch each of the designated points on the guide blocks for 5 or more seconds
with your bare hand wearing the antistatic wrist strap.
FIGURE 7-12
Guide Block (IOU)
Touch 5 or more seconds.
Chapter 7
I/O Unit (IOU) Replacement
7-15
12. Mount the replacement IOU by following the removal instructions in Step 7 in
reverse order. Align the IOU with the slot guides, insert it carefully, and secure
it firmly.
Push the eject/lock levers as far as they will go, and then push the knob screws
on the upper and lower eject/lock levers to lock them in position. When the knob
screws remain in position, the IOU is securely mounted.
Note – Mount the dummy (filler) unit in the same manner as IOU. Since the same
levers are used on the dummy (filler) unit, please operate the eject/lock levers in the
same manner.
Note – In hot addition of an IOU, mount the CMU and IOU in the proper location.
Otherwise, the added IOU may not operate.
Caution – Do not forcibly push the IOU when inserting it, even if it is not moving
smoothly. If the IOU is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
13. Go back to the terminal that is connected to XSCF, and perform the input
operation according to the instructions of the maintenance menu that is
displayed after the replacement.
a. Because the displayed messages are different between active replacement
and hot replacement, perform the operation according to the instructions for
each case. When the display of the maintenance menu returns to the initial
page, the operation is completed.
b. Exit the maintenance menu.
14. Configure the IOU into the system.
■
Active replacement:
Configure dynamically the IOU into a domain by using the addboard
command. Do this operation for all the domains that have been recognized in
Step 2 as including the target CMU.
For details, see Section 4.2.2, “Disconnecting and Replacing a FRU” on
page 4-5, and the SPARC Enterprise M4000/M5000/M8000/M9000 Servers
Dynamic Reconfiguration (DR) User’s Guide.
7-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
Hot replacement:
From the console that is connected to XSCF, execute the poweron -d
command of XSCF to turn on power to the domain.
When the power to the domain is turned on, the power-on self test (POST) is
performed. When ok prompt displayed on the console of relevant domain,
POST is completed.
15. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for a replacement IOU, CPU, or DIMM, it is operating
normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
16. Close the door on the server, and lock it.
7.3
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching IOU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
Chapter 7
I/O Unit (IOU) Replacement
7-17
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
7-18
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
8. Remove the IOU to be replaced. (See FIGURE 7-7 - FIGURE 7-9)
a. Using a Phillips screwdriver, turn the knob screws securing the upper and
lower eject/lock levers 90 to 180 degrees to unlock the levers. (Turn the knob
screw while lightly pulling the lever by one hand.)
b. Unlock the eject/lock levers to pull the IOU off the backplane.
c. While grasping the handle at the front of the IOU, pull the IOU out partially
until the [NEAR TO END] label comes into view.
d. Raise the handle at the upper part of the IOU.
e. While grasping the front and upper handles of the IOU, pull it out slowly
from the slot.
Note – To upgrade the IOU, remove the dummy (filler) unit before mounting the
IOU to be added. Since the same levers are used on the dummy (filler) unit, please
operate the eject/lock levers in the same manner.
Caution – An IOU with mounted components in the maximum configuration
weighs about 21 kg.
9. Place the removed IOU on an antistatic mat.
10. Dismount the HDD (or filler panel) and the PCI cards on the detached IOU,
and remount them all on the replacing IOU. As for the PCI cassette, which
serves as the housing for PCI card, use the one which is mounted on the
replacing IOU. For the remounting procedures, see Chapter 12 and Chapter 13.
Note – Be sure to remount the PCI cards, filler boards, and HDDs in the same
locations in the replacement IOU as their mounting locations in the removed IOU.
Note – When mounting PCI cards and HDDs, fully insert them so that the
connectors are completely engaged.
Chapter 7
I/O Unit (IOU) Replacement
7-19
11. Place the replacement IOU on the antistatic mat, and touch the metallic chassis
for five or more seconds with your bare hand wearing the antistatic wrist strap,
to remove static electricity.
Caution – To mount a new CMU, ensure that the board is placed on a grounded
antistatic mat. Then, touch it with your bare hand while wearing a properly
connected antistatic wrist strap. This will remove any static electricity before
installing it in the server. Take care not to damage the connector on the FRU edge.
FIGURE 7-13
Metallic Chassis (IOU)
Touch for 5 or more seconds.
7-20
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
12. Touch each of the designated points on the guide blocks for 5 or more seconds
with your bare hand wearing the antistatic wrist strap.
FIGURE 7-14
Guide Block (IOU)
Touch 5 or more seconds.
13. Mount the replacement IOU by following the removal instructions in Step 8 in
reverse order. Align the IOU with the slot guides, insert it carefully, and secure
it firmly.
Push the eject/lock levers as far as they will go, and then push the knob screws
on the upper and lower eject/lock levers to lock them in position. When the knob
screws remain in position, the IOU is securely mounted.
Note – Mount the dummy (filler) unit in the same manner as IOU. Since the same
levers are used on the dummy (filler) unit, please operate the eject/lock levers in the
same manner.
Caution – Do not forcibly push the IOU when inserting it, even if it is not moving
smoothly. If the IOU is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
Note – In addition of an IOU, mount the CMU and IOU in the proper location.
Otherwise, the added IOU may not operate after power-on.
Chapter 7
I/O Unit (IOU) Replacement
7-21
14. Switch on all main line switches for the AC section (ACS) in the power supply
system.
15. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
16. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From the console that is connected to XSCF, execute the
poweron -d <domain-id> command of XSCF for the domain of the
replacement target new IOU.
Power to the specified domains is turned on.
When the power to the domain is turned on, the power-on self test (POST) is
performed. When ok prompt displayed on the console of relevant domain, POST
is completed. For details of the power-on operation, see Section 4.4.2, “Powering
the Server On” on page 4-19, and the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
17. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement IOU (including PCI cards and HDDs),
it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
18. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
19. Close the door on the server, and lock it.
7-22
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
8
FAN Unit Replacement
This chapter explains the replacement procedures for the FAN unit (FANU). It covers
the following:
■
■
■
Section 8.1, “Overview of the FAN Unit” on page 8-2
Section 8.2, “Active Replacement and Hot Replacement” on page 8-8
Section 8.3, “Cold Replacement” on page 8-14
There are three methods for replacing a FAN unit: active replacement , hot
replacement, and cold replacement. For the definition of each type of method, see
Section 4.1, “Types of Replacement Procedures” on page 4-2.
In case of the SPARC Enterprise M8000 server, when you do maintenance on the
FAN_B on the rear side, the cables connected to the unit which mounted in the rack
space for the SPARC Enterprise M8000 server must be properly routed. For details,
see Section 4.7, “Cable Routing of the SPARC Enterprise M8000 Server” on
page 4-34.
Note – To do maintenance on the IOU or the FAN unit of the SPARC Enterprise
M8000 server, push aside the cables.
Caution – Take care to prevent the cables from blocking the ventilation slits of the
SPARC Enterprise M8000 server. (Fujitsu recommends running the cables above the
handle of the FAN unit.)
8-1
8.1
Overview of the FAN Unit
This section provides an overview of FAN unit and describes their mounting
locations.
Each FAN unit consists of two or three cooling fans, and two or more FAN unit are
used for cooling components in a cabinet. There are two types of FAN unit: FAN_A
(3-FAN unit), which is used commonly with the high-end server, and FAN_B (2-FAN
unit), which is used only with the M8000. FAN units have an N+1 redundancy
configuration, and active maintenance can be performed on them.
FIGURE 8-1, FIGURE 8-2, FIGURE 8-3, FIGURE 8-4, FIGURE 8-5, and FIGURE 8-6 indicate the
mounting locations of FAN unit in the high-end servers.
FIGURE 8-1
FAN Mounting Locations (Front of the M8000)
FAN_B#1
FAN_B#0
FAN_A#0-#3
8-2
#1
#3
#0
#2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 8-2
FAN Mounting Locations (Rear of the M8000)
FAN_B#2-#7
#7
#5
#3
#6
#4
#2
Chapter 8
FAN Unit Replacement
8-3
FIGURE 8-3
FAN Mounting Locations (Front of the M9000 Base Cabinet)
FAN_A#0-#3
8-4
#1
#3
#0
#2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 8-4
FAN Mounting Locations (Rear of the M9000 Base Cabinet)
FAN_A#4-#15
#15 #12
#9
#6
#14 #11
#8
#5
#13 #10
#7
#4
Chapter 8
FAN Unit Replacement
8-5
FIGURE 8-5
FAN Mounting Locations (Front of M9000 with the Expansion Cabinet)
FAN_A#0-#3
#1
#3
#0
#2
FAN_A#20-#23
#21 #23
#20 #22
base cabinet
Expansion cabinet
8-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 8-6
FAN Mounting Locations (Rear of M9000 with the Expansion Cabinet)
FAN_A#4-#15
#15 #12
#9
#6
#14 #11
#8
#5
#13 #10
#7
#4
FAN_A#24-#35
#35 #32 #29 #26
#34 #31 #28 #25
#33 #30 #27 #24
base cabinet
Expansion cabinet
TABLE 8-1 lists the abbreviated names and component numbers of the FAN unit.
TABLE 8-1
Abbreviated Names and Component Numbers of the FAN Unit
Model name
Abbreviated names and numbers
M8000
FAN_A#0 to 3 (3-FAN unit, front)
FAN_B#0, 1 (2-FAN unit, front)
FAN_B#2 to 7 (2-FAN unit, rear)
M9000 base cabinet
FAN_A#0 to 3 (3-FAN unit, front)
FAN_A#4 to 15 (3-FAN unit, rear)
M9000 expansion cabinet
FAN_A#20 to 23 (3-FAN unit, front)
FAN_A#24 to 35 (3-FAN unit, rear)
Chapter 8
FAN Unit Replacement
8-7
8.2
Active Replacement and Hot
Replacement
Caution – Each cabinet contains multiple FAN units so that the cooling
specifications are satisfied even when one individual fan inside the FAN units fails.
When replacing multiple FAN units, replace one at a time.
The active replacement procedure and hot replacement procedure for the FAN units
are the same.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Use the maintenance menu to select a swap target FAN unit. (The maintenance
menu is used up to Step 9.)
The fieldeng permission is required for executing this command.
a. From the console that is connected to XSCF, log in to XSCF, and execute the
replacefru command.
As shown in FIGURE 8-7, the initial page of the maintenance menu is displayed.
8-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 8-7
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
--------------------------------------------------Select [1-5|c:cancel] :
Note – DDC_A is displayed only for the M8000.
b. From the replacement parts list on the maintenance menu, select 2, "FAN."
c. From here, perform the operation according to the instructions displayed in
the maintenance menu.
Check the states of LEDs (POWER LED: off; CHECK LED: blinking) on the
FAN unit to be replaced, and begin the actual replacement work when a
message instructing the user to replace the target FAN unit is displayed. Leave
the maintenance menu as is until Step 8.
3. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
4. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
5. Using a Phillips screwdriver, loosen the knob screw of the fixing bracket of the
FAN unit to release the bracket.
Chapter 8
FAN Unit Replacement
8-9
6. While grasping the handle of the FAN unit to be replaced, pull it out slowly
from the slot.
FIGURE 8-8
Removing the FAN Unit (Front of M8000)
1
2
8-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 8-9
Removing the FAN Unit (Rear of M8000)
1
2
Chapter 8
FAN Unit Replacement
8-11
FIGURE 8-10
Removing the FAN Unit (Front of M9000)
1
2
8-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 8-11
Removing the FAN Unit (Rear of M9000)
1
2
7. Place the removed FAN unit on an antistatic mat.
8. Mount the replacement FAN unit by following the removal instructions in
Step 5 and Step 6 in reverse order. Align the FAN unit with the slot guides,
insert it carefully, and secure it firmly.
Caution – Each cabinet contains multiple FAN unit so that the cooling
specifications are satisfied even if one FAN unit fails. When replacing a FAN unit,
however, replace it as quickly as possible.
Caution – Do not forcibly push the FAN unit when inserting it, even if it is not
moving smoothly. If the FAN unit is forcibly inserted despite the presence of any
obstruction in a slot or any problem with a connector pin, serious damage may
result.
Chapter 8
FAN Unit Replacement
8-13
9. Go back to the terminal that is connected to XSCF, and perform the input
operation according to the instructions of the maintenance menu that is
displayed after the replacement.
a. Perform operation according to instructions of messages. When the display
of the maintenance menu returns to the initial page, the operation is
completed.
b. Exit the maintenance menu.
10. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement FAN, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
11. Close the door on the server, and lock it.
8.3
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
8-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Using a Phillips screwdriver, loosen the knob screw of the fixing bracket of the
FAN unit to release the bracket. (See FIGURE 8-11)
Chapter 8
FAN Unit Replacement
8-15
9. While grasping the handle of the FAN unit to be replaced, pull it out slowly
from the slot.
10. Place the removed FAN unit on an antistatic mat.
11. Mount the replacement FAN unit by following the removal instructions in
Step 8 and Step 9 in reverse order. Align the FAN unit with the slot guides,
insert it carefully, and secure it firmly.
Caution – Do not forcibly push the FAN unit when inserting it, even if it is not
moving smoothly. If the FAN unit is forcibly inserted despite the presence of any
obstruction in a slot or any problem with a connector pin, serious damage may
result.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement FAN, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
8-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
16. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
17. Close the door on the server, and lock it.
Chapter 8
FAN Unit Replacement
8-17
8-18
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
9
Power Supply Unit (PSU)
Replacement
This chapter explains the replacement procedures for the power supply unit (PSU).
It covers the following:
■
■
■
Section 9.1, “Overview of the PSU” on page 9-1
Section 9.2, “Active Replacement and Hot Replacement” on page 9-8
Section 9.3, “Cold Replacement” on page 9-12
There are three methods for replacing a PSU: active replacement , hot replacement,
and cold replacement. For the definition of each type of method, see Section 4.1,
“Types of Replacement Procedures” on page 4-2.
The same power supply unit is mounted on the rack-mountable dual power feed
option or power cabinet (see Chapter 24). The orientation of the mounted power
supply unit may vary by 90 degrees, but the replacement procedure is the same. See
the procedure described in subsequent sections.
9.1
Overview of the PSU
This section provides an overview of the PSU and describes its mounting locations.
A PSU receives alternating power, and it distributes 48 V and 12 V of direct current
to each component. PSUs have an N+1 redundancy configuration, and active
maintenance can be performed on them.
FIGURE 9-1 and FIGURE 9-2 show the PSU locations inside the M8000 server. FIGURE 9-3
and FIGURE 9-4 show the PSU locations inside the M9000 server (base cabinet).
FIGURE 9-5 and FIGURE 9-6 show the PSU locations inside the M9000 server (with an
expansion cabinet).
9-1
For details of the types and locations of power supply, see Section 4.5,
“Power-On/Off of Main Line Switch” on page 4-24.
FIGURE 9-1
PSU Mounting Locations (Front of the M8000 + Power Cabinet)
Power cabinet
PSU#0-#8
#8
#7
Power supply unit
#6
#5
PSU#10-#19
#19
#18
#17
#16
#15
#14
#13
#12
#11
#10
9-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
#4
#3
#2
#1
#0
FIGURE 9-2
PSU Mounting Locations (Rear of M8000 + Power Cabinet)
Power cabinet
Chapter 9
Power Supply Unit (PSU) Replacement
9-3
FIGURE 9-3
PSU Mounting Locations (Front of the M9000 Base Cabinet + Power Cabinet)
Power cabinet
PSU#0-#15
Power supply unit
PSU#20-#34
#2
#5
#8
#11 #14
#1
#4
#7
#10 #13
#0
#3
#6
#9
#34
#33
#32
#31
#30
#29
#28
#27
#26
#25
#24
#23
#22
#21
#20
9-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
#12
FIGURE 9-4
PSU Mounting Locations (Rear of the M9000 Base Cabinet + Power Cabinet)
Power cabinet
Chapter 9
Power Supply Unit (PSU) Replacement
9-5
FIGURE 9-5
PSU Mounting Locations (Front of M9000 with an Expansion Cabinet + Power Cabinet)
PSU#20-#34
#22 #25 #28 #31 #34
#21 #24 #27 #30 #33
#20 #23 #26 #29 #32
Power cabinet
Power cabinet
PSU#0-#14
#2
#5
#8
#11 #14
#1
#4
#7
#10 #13
#0
#3
#6
#9
#12
PSU#40-#54
PSU#60-#74
#54
#74
#53
#73
#52
#72
#51
#71
#50
#70
#49
#69
#48
#68
#47
#46
9-6
#67
Base cabinet
Expansion cabinet
#66
#45
#65
#44
#64
#43
#63
#42
#62
#41
#61
#40
#60
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 9-6
PSU Mounting Locations (Rear of M9000 with an Expansion Cabinet + Power Cabinet)
Power cabinet
Power cabinet
Base cabinet
Expansion cabinet
TABLE 9-1 lists the abbreviated names and component numbers of the PSUs.
TABLE 9-1
Abbreviated Names with Component Numbers of PSUs
Model name
Abbreviated names and numbers
Single power feed
Dual power feed
M8000
PSU#0 to 8
PSU#10 to 19
M9000 base cabinet
PSU#0 to 14
PSU#20 to 34
M9000 expansion cabinet
PSU#20 to 34
PSU#40 to 74
Chapter 9
Power Supply Unit (PSU) Replacement
9-7
9.2
Active Replacement and Hot
Replacement
Caution – Each cabinet contains multiple PSUs so that the power output
specifications are satisfied even if one PSU fails. However, avoid continuing
operation for a long period with a faulty PSU left in the cabinet.
The active replacement procedure and hot replacement procedure for the PSUs are
the same.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Display the maintenance menu, and select the swap target PSU. (The
maintenance menu is used up to Step 9.)
The fieldeng permission is required for executing this command.
a. From the console that is connected to XSCF, log in to XSCF, and execute the
replacefru command.
As shown in FIGURE 9-7, the initial page of the maintenance menu is displayed.
9-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 9-7
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
--------------------------------------------------Select [1-5|c:cancel] :
Note – DDC_A is displayed only for the M8000.
b. From the replacement parts list on the maintenance menu, select 3, "PSU."
c. From here, perform the operation according to the instructions displayed in
the maintenance menu.
Check the states of LEDs (POWER LED: off; CHECK LED: blinking) on the
PSU to be replaced, and begin the actual replacement work when a message
instructing the user to replace the target PSU is displayed. Leave the
maintenance menu as is until Step 7.
3. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
Chapter 9
Power Supply Unit (PSU) Replacement
9-9
4. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
5. Using a Phillips screwdriver, loosen the knob screw of the fixing bracket of the
PSU to release the bracket.
6. While pushing the lock lever (green) of the PSU to be replaced, raise the handle
to about 45 degrees, and pull out the PSU slowly from the slot.
FIGURE 9-8
Removing the PSU (Front of M8000)
3
2
1
9-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 9-9
Removing the PSU (Front of the M9000 Base Cabinet)
3
1
2
7. Place the removed PSU on an antistatic mat.
8. Mount the replacement PSU by following the removal instructions in Step 5
and Step 6 in reverse order. Align the PSU with the slot guides, insert it
carefully, and secure it firmly.
Caution – Each cabinet contains multiple PSUs so that the power output
specifications are satisfied even if one PSU fails. However, please refrain from the
long time operation as one PSU remains failed.
Caution – Do not forcibly push the PSU when inserting it, even if it is not moving
smoothly. If the PSU is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
9. Go back to the terminal that is connected to XSCF, and perform the input
operation according to the instructions on the maintenance menu that is
displayed after the replacement.
Chapter 9
Power Supply Unit (PSU) Replacement
9-11
a. Perform operation according to instructions of messages. When the display
of the maintenance menu returns to the initial page, the operation is
completed.
b. Exit the maintenance menu.
10. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement PSU, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
11. Close the door on the server, and lock it.
9.3
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
9-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
Chapter 9
Power Supply Unit (PSU) Replacement
9-13
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Using a Phillips screwdriver, loosen the knob screw of the fixing bracket of the
PSU to release the bracket, and open it 180 degrees. (See FIGURE 9-8 or
FIGURE 9-9)
9. While pushing the lock lever (green) of the PSU to be replaced, raise the handle
to about 45 degrees, and pull out the PSU slowly from the slot.
10. Place the removed PSU on an antistatic mat.
11. Mount the replacement PSU by following the removal instructions in Step 8
and Step 9 in reverse order. Align the PSU with the slot guides, insert it
carefully, and secure it firmly.
Caution – Do not forcibly push the PSU when inserting it, even if it is not moving
smoothly. If the PSU is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to the XSCF, execute the poweron -a
command of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
9-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
15. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement PSU, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
16. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
17. Close the door on the server, and lock it.
Chapter 9
Power Supply Unit (PSU) Replacement
9-15
9-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
10
Operator Panel Replacement
This chapter explains the replacement procedures for the operator panel (OPNL). It
covers the following:
■
■
Section 10.1, “Overview of the Operator Panel” on page 10-1
Section 10.2, “Cold Replacement” on page 10-4
Cold replacement is the only swapping method that can be used for the OPNL. For
the definition of each type of swapping method, see Section 4.1, “Types of
Replacement Procedures” on page 4-2.
Note – Operation on the operator panel requires a different key from the key for the
doors of the server. Both are managed by the system administrator.
Note – If you replace the XSCF unit and the operator panel simultaneously, system
will not operate normally. Execute the showhardconf command or the
showstatus command to confirm that the component replaced earlier is operating
normally, before replacing the subsequent FRU.
10.1
Overview of the Operator Panel
This section provides an overview of the OPNL and describes its mounting location.
The OPNL is used for the operation and condition display of the server. Active
maintenance is not supported. The OPNL is mounted on the front cover.
10-1
FIGURE 10-1
OPNL Mounting Location (Front of the M8000)
OPNL
10-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 10-2
OPNL Mounting Location (Front of the M9000 Base Cabinet)
OPNL
Chapter 10
Operator Panel Replacement
10-3
FIGURE 10-3
OPNL Mounting Location (Front of M9000 with the Expansion Cabinet)
OPNL
Base cabinet
Expansion cabinet
TABLE 10-1 lists the abbreviated name of the OPNL.
TABLE 10-1
10.2
Abbreviated Name of the Operator Panel
Model name
Abbreviated name
M8000/M9000 Operator Panel
OPNL
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
10-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the OPNL from the Locked position to the
Service position.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
Chapter 10
Operator Panel Replacement
10-5
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Change the mode switch setting on the OPNL to the Locked position and pull
out the key.
The key cannot be pulled out at the Service position.
9. Remove the cable from the connector of the OPNL.
10. Using a Phillips screwdriver, loosen the two knob screws securing the OPNL,
and remove the OPNL from the front cover.
FIGURE 10-4
10-6
Removing the OPNL
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
11. Place the removed OPNL on an antistatic mat.
12. Mount the replacement OPNL by following the removal instructions in Step 8
to Step 10 in reverse order.
Caution – Do not forcibly push the cable in when inserting it. If there is problem
with the connector pin, serious damage may result.
13. Insert the OPNL key, and then set the key to the Service position.
14. Switch on all main line switches for the AC section (ACS) in the power supply
system.
15. Confirm that the XSCF STANDBY LED (green) on the OPNL remains lit. If it is
blinking, wait until it remains lit.
16. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
17. Confirm that the replaced components are normal by using the showhardconf
or showstatus command of XSCF.
Unless "*" is displayed for the replacement OPNL, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
Chapter 10
Operator Panel Replacement
10-7
18. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
19. Close the door on the server, and lock it.
10-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
11
XSCF Unit Replacement
This chapter explains the replacement procedures for an eXtended System Control
Facility unit (XSCFU). It covers the following:
■
■
■
Section 11.1, “Overview of the XSCFU” on page 11-1
Section 11.2, “Active Replacement and Hot Replacement” on page 11-5
Section 11.3, “Cold Replacement” on page 11-12
There are three methods for replacing an XSCFU: active replacement, hot
replacement, and cold replacement. For the definition of each type of method, see
Section 4.1, “Types of Replacement Procedures” on page 4-2.
Note – If you replace the XSCF unit and the operator panel simultaneously, system
will not operate normally. Execute the showhardconf command or the
showstatus command to confirm that the component replaced earlier is
operating normally, before replacing the subsequent FRU.
11.1
Overview of the XSCFU
This section provides an overview of the XSCFU and describes their mounting
locations.
The XSCFU controls and manages the server hardware. There are two types of
XSCFU: the XSCFU_B, which is used for both high-end servers, and the XSCFU_C,
which is used for the expansion cabinet of the M9000. The XSCFU is duplicated, and
active maintenance can be performed on it.
FIGURE 11-1, FIGURE 11-2, and FIGURE 11-3 show the mounting locations of XSCFUs.
11-1
FIGURE 11-1
XSCFU Mounting Locations (Front of the M8000)
XSCFU_B#0
XSCFU_B#1
11-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 11-2
XSCFU Mounting Locations (Front of the M9000 Base Cabinet)
XSCFU_B#1
XSCFU_B#0
Chapter 11
XSCF Unit Replacement
11-3
FIGURE 11-3
XSCFU Mounting Locations (Front of M9000 with the Expansion Cabinet)
XSCFU_C#1
XSCFU_C#0
XSCFU_B#1
XSCFU_B#0
TABLE 11-1 lists the abbreviated names and component numbers of the XSCFUs.
TABLE 11-1
Abbreviated Names and Component Numbers of XSCFUs
Model name
Abbreviated names and numbers
M8000
XSCFU_B#0, 1
M9000 base cabinet
XSCFU_B#0, 1
M9000 expansion cabinet
XSCFU_C#0, 1
11-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
11.2
Active Replacement and Hot
Replacement
Note – To replace the XSCF Unit, you must consider the version of the XCP that is
currently in the server.
To determine the firmware version of the server, type:
XSCF> version -c xcp
For more information, refer to the version (8) man page or the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference Manual.
If the version of XCP in your server is XCP 1041 or earlier, see Appendix E. If the
server has XCP1050 or later, follow the procedure described below.
The active replacement procedure and hot replacement procedure for the XSCFU are
the same.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. When the Active LED (green) of the XSCFU to be replaced is lit, log in to the
XSCF from a terminal connected to the XSCF, and then execute the switchscf
command for switching to Standby.
XSCF> switchscf -t Standby
Chapter 11
XSCF Unit Replacement
11-5
3. Use the maintenance menu to select the XSCFU to be replaced. (The
maintenance menu is used up to Step 10.)
The fieldeng permission is required for executing this command.
a. From the console that is connected to the Active XSCF, log in to the XSCF,
and execute the replacefru command.
As shown in FIGURE 11-4, the initial page of the maintenance menu is displayed.
FIGURE 11-4
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
--------------------------------------------------Select [1-5|c:cancel] :
Note – DDC_A is displayed only for the M8000
b. From the replacement parts list on the maintenance menu, select 4, "XSCFU."
c. From here, perform the operation according to the guidance displayed in the
maintenance menu.
Check the states of LEDs (POWER LED: off; CHECK LED: blinking) on the
XSCFU to be replaced, and begin the actual replacement work when a message
instructing the user to replace the target XSCFU is displayed. Leave the
maintenance menu as is until Step 6.
4. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
11-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
5. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
6. For the M9000 (with the expansion cabinet), disconnect the cables from all
connectors of the XSCFU to be replaced.
7. Remove the XSCFU to be replaced.
a. If the SPARC Enterprise M9000 server expansion cabinet is installed, slide
the cable support bracket at the front of the XSCF unit up or down by one
hole, and then remove the cable connected to the XSCF unit.
b. Using a Phillips screwdriver, turn the knob screws securing the upper and
lower eject/lock levers (in the SPARC Enterprise M8000 server) or left and
right eject/lock levers (in the SPARC Enterprise M9000 server) 90 to 180
degrees to unlock the levers. (Turn the knob screw while lightly pulling the
lever.)
c. Unlock the eject/lock levers to pull the XSCFU off the backplane.
d. While grasping the eject/lock levers, pull out the XSCFU slowly from the
slot.
Chapter 11
XSCF Unit Replacement
11-7
FIGURE 11-5
Removing the XSCFU (Front of the M8000)
1
2
1
11-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 11-6
Removing the XSCFU (Front of the M9000 Base Cabinet)
1
2
1
8. Place the removed XSCFU on an antistatic mat.
9. Mount the replacement XSCFU by following the removal instructions in Step 6
and Step 7 in reverse order. Align the XSCFU with the slot guides, insert it
carefully, and secure it firmly.
Push the levers as far as they will go, and then push the knob screws on the
eject/lock levers to lock them in position. When the knob screws remain in
position, the XSCF unit is securely mounted.
Chapter 11
XSCF Unit Replacement
11-9
Note – When your system is the SPARC Enterprise M9000 server with Expansion
Cabinet, pay attention to the XSCF message or the error log information which
pointing out the suspect unit.
* When it is "XSCFU_C," replace XSCFU_C.
* When it is "XSCFU_B," you must always once extract both XSCFU_B and
XSCFU_C.
For insertion, insert XSCFU_C first, and then XSCFU_B.
Caution – Do not forcibly push the XSCFU in when inserting it. If the XSCFU is
forcibly inserted despite the presence of any obstruction in a slot or any problems
with the connector pin, serious damage may result.
Caution – Do not forcibly push the cable connector when inserting it. If the cable
connector is forcibly inserted, the cable connector might be damaged.
10. Go back to the terminal that is connected to XSCF, and perform the input
operation according to the guidance of the maintenance menu that is displayed
after the replacement.
a. Perform operation according to instructions of messages. When the display
of the maintenance menu returns to the initial page, the operation is
completed.
b. Exit the maintenance menu.
(In case communication with XSCF is lost)
During execution of the maintenance command (such as the replacefru
command) in the active XSCF unit, in case communication with XSCF is lost
for some reason such as LAN disconnection, re-execute the command in the
manner described below:
i. Reconnect to XSCF.
In case multiple maintenance commands executed, the XSCF becomes
locked status and the message as follows appears.
Unable to perform maintenance commands at this time.
Another user is currently executing a maintenance command.
When the above message displayed, execute the unlockmaintenance
command to forcibly release the XSCF locked status.
11-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
XSCF> unlockmaintenance
For details of the unlockmaintenance command, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide, the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
ii. Execute the showhardconf command.
And, confirm that the currently active XSCF unit is the one, which you are
going to replace with using replacefru command.
iii. After confirming that the system did not switch the other XSCF unit as
active, re-execute the replacefru command.
(a)When you can execute the replacefru command
Follow the replacefru command to replace the unit.
(b)In case communication with XSCF is lost after you physically replaced
the target XSCF
The new XSCF unit is physically mounted; however, the mounting process
is not yet finished completely. When the replacefru command showed a
message which prompts the replacement of XSCF unit, once extract the
inserted XSCF unit which has undergone the replacement, and then reinsert
it.
Note – If an attempt to re-execute the replacefru command from the active
XSCF fails, or if an XSCF failover occurred and the XSCF units switched, both XSCF
units will require cold replacement. For information on cold replacement, see
Section 11.3, “Cold Replacement” on page 11-12.
11. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement XSCFU, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
12. Confirm that the version of the replaced XSCF unit is the same. If the version
number is different, upgrade the firmware again to match up the version
numbers.
Chapter 11
XSCF Unit Replacement 11-11
For the version matching of firmware, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
13. Close the door on the server, and lock it.
11.3
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold down the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
11-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. For the M9000 (with the expansion cabinet), disconnect the cables from the
connectors of the XSCFU to be replaced. Disconnect any other cables, such as a
LAN cable, as necessary.
9. Remove the XSCFU to be replaced. (See FIGURE 11-5 or FIGURE 11-6)
a. Using a Phillips screwdriver, turn the knob screws securing the upper and
lower eject/lock levers (in the SPARC Enterprise M8000 server) or left and
right eject/lock levers (in the SPARC Enterprise M9000 server) 90 to 180
degrees to unlock the levers. (Turn the knob screw while lightly pulling the
lever by one hand.)
b. Unlock the eject/lock levers to pull the XSCFU off the backplane.
c. While grasping the eject/lock levers, pull out the XSCFU slowly from the
slot.
10. Place the removed XSCFU on an antistatic mat.
Chapter 11
XSCF Unit Replacement 11-13
11. Mount the replacement XSCFU by following the removal instructions in Step 8
and Step 9 in reverse order. Align the XSCFU with the slot guides, insert it
carefully, and secure it firmly.
Push the levers as far as they will go, and then push the knob screws on the
eject/lock levers to lock them in position. When the knob screws remain in
position, the XSCF unit is securely mounted.
Caution – Do not forcibly push the XSCFU in when inserting it. If the XSCFU is
forcibly inserted despite the presence of any obstruction in a slot or any problems
with the connector pin, serious damage may result.
Caution – Do not forcibly push the cable in when inserting it. If the cable is forcibly
inserted despite the presence of any obstruction in a slot or any problems with the
connector pin, serious damage may result.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
14. Update the XSCF firmware.
In case the version of the firmware on the replaced XSCFU is different from the
version of the original, it is necessary to match the version of the firmware.
For the version matching of firmware, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide.
15. Set the XSCF clock.
From the terminal connected to XSCF, log in to XSCF and use setdate -s
MMDDhhmmCCYY.ss command to set the current time.
For details of the setdate command, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide, the SPARC
Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference Manual, or
the man page.
11-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
16. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
17. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement XSCFU, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
18. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
19. Close the door on the server, and lock it.
Chapter 11
XSCF Unit Replacement 11-15
11-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
12
Hard Disk Drive (HDD)
Replacement
This chapter explains the replacement procedures for the hard disk drive (HDD). It
covers the following:
■
Section 12.1, “Overview of the HDD” on page 12-1
■
Section 12.2, “Active Replacement” on page 12-3
■
Section 12.3, “Cold Replacement” on page 12-5
There are two methods for replacing a HDD: active replacement and cold
replacement. For the definition of each type of method, see Section 4.1, “Types of
Replacement Procedures” on page 4-2.
12.1
Overview of the HDD
This section provides an overview of the HDD and shows its mounting locations.
HDDs are mounted in the four HDD slots in an IOU to expand the storage capacity
of the system. In each HDD slot, a 2.5-inch SAS interface HDD can be mounted. A
filler panel may be mounted in an HDD slot. Active maintenance can be performed
on an HDD.
FIGURE 12-1 indicates the HDD mounting locations in an IOU cabinet. For
information on the IOU mounting locations, see FIGURE 7-1 to FIGURE 7-5 in
Chapter 7.
12-1
FIGURE 12-1
HDD Mounting Locations
HDD#0
HDD#1
HDD#2
HDD#3
TABLE 12-1 lists the abbreviated name and numbers of HDDs in an IOU.
TABLE 12-1
12-2
Abbreviated Name and Numbers of Hard Disk Drives
Component name
Abbreviated name and number
Hard disk drive
HDD#0 to 3
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
12.2
Active Replacement
The procedure for active replacement is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Prepare for the replacement of the target HDD.
From the OS, execute the cfgadm -a command to verify Ap_Id.
Stop the use of HDD by the application and disconnect the HDD from the OS. The
procedure of disconnecting the HDD from the OS varies depending on if you are
using the disk mirroring software or other supporting software. When you are not
using such kind of software, you can use the cfgadm command to disconnect the
target from the OS. When you are using the above software, see the manual of
that software for how to disconnect the disk.
The READY LED (green) of the HDD is turned off.
3. Blink the CHECK LED of the HDD to be replaced.
From the OS, execute the cfgadm -x led=fault,mode=blink <Ap_Id>
command.
4. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
5. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
Chapter 12
Hard Disk Drive (HDD) Replacement
12-3
6. Remove the HDD unit to be replaced.
a. Press the square push button on the HDD unit to be replaced.
The eject/lock lever protrudes at an angle of about 45 degrees.
b. Pull the eject/lock lever to an angle of about 90 degrees to release the lock.
c. While grasping the eject/lock lever, pull out the HDD unit slowly from the
slot.
FIGURE 12-2
Removing the HDD
2
1
7. Place the removed HDD on an antistatic mat.
12-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
8. Mount the replacement HDD by following the removal instructions in Step 6 in
reverse order. Align the HDD with the slot guides, insert it carefully, and
secure it firmly.
Caution – Do not forcibly push the HDD when inserting it, even if it is not moving
smoothly. If the HDD is forcibly inserted despite the presence of any problem with a
connector pin, serious damage may result.
Do not insert an HDD into a slot while the eject/lock lever is closed. If an HDD is
inserted in this condition, it is stopped halfway and becomes difficult to pull out.
9. Configure the HDD into the system.
From the OS, execute the cfgadm -x led=fault,mode=off <Ap_Id>
command to turn off the CHECK LED.
10. Configure the HDD into the OS.
The procedure of configuring the HDD into the OS varies depending on whether
or not using the disk mirroring software or other supporting software. When you
are not using such kind of software, you can use the cfgadm command to
configure the target into the OS. After configured, conduct the necessary
operation such as format, as required. When you are using the above software,
see the manual of that software for how to configure the disk.
The READY LED (green) of the HDD is turned on.
11. Close the door on the server, and lock it.
12.3
Cold Replacement
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Chapter 12
Hard Disk Drive (HDD) Replacement
12-5
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold down the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that shutdown processing is completed by confirming that the READY
LED (green) on the HDD is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
12-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
8. Remove the HDD unit to be replaced. (See FIGURE 12-2)
a. Press the square push button on the HDD unit to be replaced.
The eject/lock lever protrudes at an angle of about 45 degrees.
b. Pull the eject/lock lever to an angle of about 90 degrees to release the lock.
c. While grasping the eject/lock lever, pull out the HDD unit slowly from the
slot.
9. Place the removed HDD on an antistatic mat.
10. Mount the replacement HDD by following the removal instructions in Step 8 in
reverse order. Align the HDD with the slot guides, insert it carefully, and
secure it firmly.
Caution – Do not forcibly push the HDD when inserting it, even if it is not moving
smoothly. If the HDD is forcibly inserted despite the presence of any problem with a
connector pin, serious damage may result.
Do not insert an HDD into a slot while the eject/lock lever is closed. If an HDD is
inserted in this condition, it is stopped halfway and becomes difficult to pull out.
11. Switch on all main line switches for the AC section (ACS) in the power supply
system.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Start the domain as necessary.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
Chapter 12
Hard Disk Drive (HDD) Replacement
12-7
16. Close the door on the server, and lock it.
12-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
13
PCI Slot Device Replacement
This chapter explains the replacement procedures for devices that are mounted in
the PCI slots of an IOU. It covers the following:
■ Section 13.1, “Overview of PCI Slot Devices” on page 13-1
■ Section 13.2, “Active Replacement” on page 13-3
■ Section 13.3, “Hot Replacement” on page 13-17
■ Section 13.4, “Cold Replacement” on page 13-20
There are three methods for replacing PCI slot devices: active replacement, hot
replacement and cold replacement. For the definition of each type of method, see
Chapter 4
13.1
Overview of PCI Slot Devices
This section provides an overview of devices that are inserted in the PCI slots of an
IOU, and describes their mounting locations.
Each IOU has eight PCI slots. Every PCI slot is equipped with a PCI cassette
(PCICS). Any of the devices below can be mounted in a PCI cassette, and the system
functions can be expanded with these devices.
■
PCI card
The cards that can be mounted are PCI-Express short cards.
The PHP function enables active maintenance of PCI cards.
■
IOU onboard device card (IOUA)
A card required for mounting a built in CD-RW/DVD-RW drive unit or tape
drive unit
The PHP function does not enable active maintenance of IOUAs. Perform either
hot maintenance or cold maintenance.
13-1
■
Link card
If PCI cards mounted in all the IOUs are not sufficient in number, this card serves
as a card for connecting an external I/O expansion unit that is used for mounting
additional PCI cards.
The PHP function enables active maintenance of link cards.
A filler panel may be mounted in a PCICS.
FIGURE 13-1 indicates the PCI slot locations in an IOU cabinet. For information on the
IOU mounting locations, see FIGURE 7-1 to FIGURE 7-5 in Chapter 7.
FIGURE 13-1
PCI Slot Locations
PCI#4
PCI#5
PCI#6
PCI#7
PCI#0
PCI#1
PCI#2
PCI#3
13-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE 13-1 shows abbreviated name and numbers of PCI slots in an IOU.
TABLE 13-1
Abbreviated Name and Numbers of PCI Slots
Slot name
Abbreviated name and numbers
PCI slot
PCI#0 to 7
For the following devices, replacement procedures have the same steps except that
LEDs to be checked, eject levers, and names used for status display are different.
■
PCI cassette (PCICS)
■
PCI card
■
IOU onboard device card (IOUA)
■
Link card
Note – The steps of the replacement procedures for I/O boxes and link cards
mounted in external I/O expansion units, PCI cards, I/O boards, and power supply
units are basically the same as those described in this manual. For details, see the
External I/O Expansion Unit Installation and Service Manual.
The following explains the replacement procedure for a PCI card, which is a typical
example.
13.2
Active Replacement
The PHP function enables active replacement of PCI cassettes (PCICSs) and link
card. Active replacement cannot be performed for IOU onboard device cards
(IOUAs).
The active replacement procedure is as follows.
1. Unlock and open the appropriate door on the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
Chapter 13
PCI Slot Device Replacement
13-3
2. From the OS, execute the cfgadm -a command to verify the Ap_Id of the PCI
card.
Example: If the PCI card in PCI slot 3 of IOU#0 is replaced
# cfgadm -a
Ap_Id
iou#0_pci#3
Type
Receptacle Occupant
Condition
:
pci_pci/hp connected configured ok
The AP_Id (iou#0_pci#3) verified here is referred to as PCI_ApId, and the
subsequent explanations of cfgadm operations for the PCI card use the term
PCI_ApId.
3. From the OS, execute the cfgadm -c unconfigure <PCI_ApId> command to
unconfigure the replacement target PCI card from the driver.
# cfgadm -c unconfigure iou#0_pci#3
4. From the OS, execute the cfgadm -c disconnect <PCI_ApId> command to
stop supplying power to the replacement target PCI card.
# cfgadm -c disconnect iou#0_pci#3
5. From the OS, execute the cfgadm -a command to confirm that the status of the
PCI card has changed from "configured" to "unconfigured," and from
"connected" to "disconnected."
# cfgadm -a
Ap_Id
Type
iou#0_PCI#3
:
unknown
Receptacle Occupant Condition
disconnected
unconfigured
unknown
6. Blink the ATTENTION LED (amber) of the PCI card to be replaced.
# cfgadm -x led=attn, mode=blink iou#0_pci#3
7. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
13-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
8. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
9. Remove the PCI cassette of the PCI card to be replaced.
Caution – To disconnect the link cable, which is a cable connected between the link
card mounted on the server and external I/O expansion unit, pull the latch lock of
the connector. If you forcibly pull the connector or the cable, the cable may be
damaged.
Caution – When removing cables such as LAN cable, if your finger can't reach the
latch lock of the connector, press the latch with a flathead screwdriver to remove the
cable. Forcing your finger into the clearance can cause damage to the PCI card.
a. Using a Phillips screwdriver, loosen the green screw securing the lever.
b. While grasping the upper part of the PCI cassette, pull the lever until it is
stopped, and continue pulling it to pull out the PCI cassette.
c. Place the removed PCI cassette on an antistatic mat.
Chapter 13
PCI Slot Device Replacement
13-5
FIGURE 13-2
Removing the PCI Cassette
2
1
10. Remove the PCI card from the PCI cassette. (Perform this work on the
conductive mat.)
a. While grasping the lock bracket, turn it in the direction of the arrow and
push the lever a little. Then, push in the lever completely. (See FIGURE 13-3)
Once you start moving the lever, you need not continue holding the lock
bracket.
b. Loosen the two screws on the back of the PCI cassette, slide out the latch,
and temporarily secure it with screws. (See FIGURE 13-4)
Caution – Do not loosen the screws at the bottom because they are used as a
reference position.
c. Slide the PCI card in the direction of the arrow to remove it. (See FIGURE 13-5)
13-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 13-3
Pushing In the Lever
Lever
Lock bracket
Chapter 13
PCI Slot Device Replacement
13-7
FIGURE 13-4
13-8
Loosening Screws of Latches
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 13-5
Removing the PCI Card
Chapter 13
PCI Slot Device Replacement
13-9
11. Mount the replacement PCI card on the PCI cassette. (Perform this work on the
conductive mat.)
FIGURE 13-6 shows card positioning post and card positioning tab.
FIGURE 13-6
Card positioning post and card positioning tab
1
2
TABLE 13-2
Card positioning post and card positioning tab
Item
Description
1
Card positioning post
2
Card positioning tab
The post (item 1 in FIGURE 13-6) is seated in the notch of the bracket of the PCI card.
If the post is not properly seated in the notch, the mounting bracket of the card may
be bent and the card may end up being affixed in an inclined position on the PCI
cassette. If a card is affixed in an inclined position, it will not be in proper electric
contact with the socket on the IOU.
The tab (item 2 in FIGURE 13-6) fits in the notch on the lower part of the PCI card.
This tab is helpful in positioning the card when it is mounted on the PCI cassette
(however, some types of cards do not have the notch).
13-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Note – When the PCI cassette is removed from the IOU, the tab raises the front of
the card from the card connector.
a. Move the PCI card in the direction of the arrow, and mount it on the PCI
cassette by inserting the part projecting out at the bottom of the bracket of
the PCI card into the hole at the bottom of the front panel of the PCI cassette
and inserting the pin of the PCI cassette into the oval hole at the top of the
bracket of the PCI card.
FIGURE 13-7
Mounting the PCI Card
b. Push the PCI card until the notched part on the base plate of the PCI card
touches the card-positioning tab from the PCI cassette.
Note – When fixing the PCI card in position after aligning it with the card
positioning tab, do so while pulling the lever about 2 cm from the frame.
Chapter 13
PCI Slot Device Replacement 13-11
c. Align the PCI card with the center part of the lower latch of the cassette.
Then, while pressing the PCI card against the center part of each of the
upper latches of the cassette, tighten the fixing screws in the sequence
shown in FIGURE 13-8 to fix the card in position..
Note – To ensure that the PCI card is properly mounted, you need to confirm that
the PCI card base is properly centered at the latches and securely fastened in
position.
FIGURE 13-8
13-12
Sequence of Fixing the PCI Card in Position
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Caution – If the card is short, move the screw at the bottom of the cassette as
required. In such cases, place the latches as close as possible to the far end of the
card. (FIGURE 13-9 reference)
To shift the location of the bottom screw for the PCI card, first secure the card at the
lower latch (part a in FIGURE 13-8). Then, tighten the fixing screws in the sequence of
1-2 shown in FIGURE 13-8 to fix the card in position.
FIGURE 13-9
Latch Positions at the Bottom of the Cassette
12. Mount the PCI cassette.
a. Pull the lever of the PCI cassette until the PCI cassette is pulled out
completely.
b. Align the PCI cassette with the groove of the PCI slot, and insert the PCI
cassette.
c. Press the upper part of the cassette until the cassette is inserted in the same
position as the other cassettes.
d. Push in the lever.
Chapter 13
PCI Slot Device Replacement 13-13
e. Using a Phillips screwdriver, tighten the green screw.
Be sure to stop tightening the screw when the bottom of the screw head is flush
with the projecting part of the handle. Otherwise, over tightening the screw
may cause the projecting part of the handle to break. (See FIGURE 13-10.)
FIGURE 13-10
Tightening the Screw
Screw head
Projecting part of the handle
Caution – Align the PCI card with the slot guides, insert it carefully, and secure it
firmly.
Caution – Do not forcibly push the PCI card when inserting it, even if it is not
moving smoothly. If the PCI card is forcibly inserted despite the presence of any
obstruction in the slot or any problem with a connector pin, serious damage may
result.
13-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
If the lever of a PCI cassette cannot be inserted:
a. When the lever of the PCI cassette is pushed in but it stops halfway without
reaching the far end, do not forcibly push it in any further. Instead, pull it
out once and then insert it again.
FIGURE 13-11
Reinserting the Lever
Chapter 13
PCI Slot Device Replacement 13-15
b. If the lever stops halfway even when the PCI cassette is inserted again, pull
the PCI cassette out from the IOU, and then check whether the PCI card is
set in the normal position (see a, b, and c of Step 11). Then, reinsert the PCI
cassette.
Caution – Do not forcibly push the lever of the PCI cassette when the lever is
pushed but stops halfway. In this case, restore the lever to its original position, and
then reinsert it. If the lever stops halfway even when the PCI cassette is inserted
again, it is possible that the PCI card is not set in the normal position (see a, b, and c
of Step 11). Pull the PCI cassette out from the IOU, verify the position of the PCI
card, and then reinsert the PCI cassette.
13. From the OS, execute the cfgadm -c connect <PCI_ApId> command to
supply power to the PCI card.
# cfgadm -c connect iou#0_pci#3
14. From the OS, execute the cfgadm -c configure <PCI_ApId> command to
configure the PCI card into the domain.
# cfgadm -c configure iou#0_pci#3
15. From the OS, execute the cfgadm -a command to confirm that the status of the
PCI card has changed from "unconfigured" to "configured."
# cfgadm -a
Ap_Id
iou#0_pci#3
Type
Receptacle Occupant
Condition
:
pci_pci/hp connected configured ok
16. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement XSCFU, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
17. Close the door on the server and lock it.
13-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
13.3
Hot Replacement
The procedure for hot replacement is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Turn off power to the relevant domain.
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -d command.
The shutdown processing is executed for the relevant domain, and then the
power-off processing is executed.
3. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the PCI card is off.
4. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
5. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
Chapter 13
PCI Slot Device Replacement 13-17
6. Remove the PCI cassette of the PCI card to be replaced. (See FIGURE 13-2)
Caution – To disconnect the link cable, which is a cable connected between the link
card mounted on the server and external I/O expansion unit, pull the latch lock of
the connector. If you forcibly pull the connector or the cable, the cable may be
damaged.
Caution – When removing cables such as LAN cable, if your finger can't reach the
latch lock of the connector, press the latch with a flathead screwdriver to remove the
cable. Forcing your finger into the clearance can cause damage to the PCI card.
a. Using a Phillips screwdriver, loosen the green screw securing the lever.
b. While grasping the upper part of the PCI cassette, pull the lever until it is
stopped, and continue pulling it to pull out the PCI cassette.
c. Place the removed PCI cassette on an antistatic mat.
7. Remove the PCI card from the PCI cassette. (Perform the work on the
conductive mat.)
a. While grasping the lock bracket, turn it in the direction of the arrow and
push the lever a little. Then, push in the lever completely. (See FIGURE 13-3)
Once you start moving the lever, the lock bracket need not be held.
b. Loosen the two screws at the rear of the PCI cassette, slide out the latch, and
secure it with screws. (See FIGURE 13-4)
Caution – Do not loosen the screws at the bottom because they are used as a
reference position.
c. Slide the PCI card in the direction of the arrow to remove it. (See FIGURE 13-5)
8. Mount the replacement PCI card on the PCI cassette. (Perform the work on the
conductive mat.) (See FIGURE 13-6 and TABLE 13-2)
a. Move the PCI card in the direction of the arrow, and mount it on the PCI
cassette by inserting the part projecting out at the bottom of the bracket of
the PCI card into the hole at the bottom of the front panel of the PCI cassette
and inserting the pin of the PCI cassette into the oval hole at the top of the
bracket of the PCI card. (See FIGURE 13-7)
b. Push the PCI card until the notched part on the base plate of the PCI card
touches the card-positioning tab from the PCI cassette.
13-18
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
c. Set the screws in the center part of each of the two latches, and then tighten
the screws while pressing them against the PCI card.
Note – When affixing the PCI card after aligning its position with the
card-positioning tab, do so while pulling the lever about 2 cm from the frame.
Caution – If the card is short, move the screw at the bottom of the cassette as
required. In such cases, place the latches as close as possible to the far end of the
card. (See FIGURE 13-9)
9. Mount the PCI cassette.
a. Pull the lever of the PCI cassette until the PCI cassette is pulled out
completely.
b. Align the PCI cassette with the groove of the PCI slot, and insert the PCI
cassette.
c. Press the upper part of the cassette until the cassette is inserted in the same
position as the other cassettes.
d. Push in the lever.
e. Using a Phillips screwdriver, tighten the green screw.
Caution – Align the PCI card with the slot guides, insert it carefully, and secure it
firmly.
Caution – Do not forcibly push the PCI card when inserting it, even if it is not
moving smoothly. If the PCI card is forcibly inserted despite the presence of any
obstruction in a slot or any problem with a connector pin, serious damage may
result.
■
If the lever of a PCI cassette cannot be inserted:
a. When the lever of the PCI cassette is pushed in but it stops halfway without
reaching the far end, do not forcibly push it in any further. Instead, pull it
out once and then insert it again. (See FIGURE 13-11)
Chapter 13
PCI Slot Device Replacement 13-19
b. If the lever stops halfway even when the PCI cassette is inserted again, pull
the PCI cassette out from the IOU, and then check whether the PCI card is
set in the normal position (see a, b and c of Step 8). Then, reinsert the PCI
cassette.
Caution – Do not forcibly push the lever of the PCI cassette when the lever is
pushed but stops halfway. In this case, restore the lever to its original position, and
then reinsert it. If the lever stops halfway even when the PCI cassette is inserted
again, it is possible that the PCI card is not set in the normal position (see a, b and c
of Step 7). Pull the PCI cassette out from the IOU, check the position of the PCI card,
and then reinsert it.
10. Turn on power to the relevant domain.
From a console that is connected to XSCF, execute the poweron -d command of
XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
11. Close the door on the server, and lock it.
13.4
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
13-20
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
2. Change the mode switch setting on the OPNL from the Locked position to the
Service position.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the PCI card is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
Chapter 13
PCI Slot Device Replacement 13-21
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the PCI cassette of the PCI card to be replaced.
Caution – To disconnect the link cable, which is a cable connected between the link
card mounted on the server and external I/O expansion unit, pull the latch lock of
the connector. If you forcibly pull the connector or the cable, the cable may be
damaged.
Caution – When removing cables such as LAN cable, if your finger can't reach the
latch lock of the connector, press the latch with a flathead screwdriver to remove the
cable. Forcing your finger into the clearance can cause damage to the PCI card.
a. Using a Phillips screwdriver, loosen the green screw securing the lever.
b. While grasping the upper part of the PCI cassette, pull the lever until it is
stopped, and continue pulling it to pull out the PCI cassette.
c. Place the removed PCI cassette on an antistatic mat.
9. Remove the PCI card from the PCI cassette. (Perform the work on the
conductive mat.)
a. While grasping the lock bracket, turn it in the direction of the arrow and
push the lever a little. Then, push in the lever completely. (See FIGURE 13-3)
Once you start moving the lever, the lock bracket need not be held.
b. Loosen the two screws at the rear of the PCI cassette, slide out the latch, and
secure it with screws. (See FIGURE 13-4)
Note – Do not loosen the screws at the bottom because they are used as a reference
position.
c. Slide the PCI card in the direction of the arrow to remove it. (See FIGURE 13-5)
13-22
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
10. Mount the replacement PCI card on the PCI cassette. (Perform the work on the
conductive mat.)
a. Move the PCI card in the direction of the arrow, and mount it on the PCI
cassette by inserting the part projecting out at the bottom of the bracket of
the PCI card into the hole at the bottom of the front panel of the PCI cassette
and inserting the pin of the PCI cassette into the oval hole at the top of the
bracket of the PCI card.
b. Push the PCI card until the notched part on the base plate of the PCI card
touches the part projecting from the PCI cassette.
c. Push in the two latches on the PCI card, and tighten the screws.
Note – If the card is short, move the screw at the bottom of the cassette as required.
In such cases, place the latches as close as possible to the far end of the card. (See
FIGURE 13-9)
11. Mount the PCI cassette.
a. Pull the lever of the PCI cassette until the PCI cassette is pulled out
completely
b. Align the PCI cassette with the groove of the PCI slot, and insert the PCI
cassette.
c. Press the upper part of the cassette until the cassette is inserted in the same
position as the other cassettes.
d. Push in the lever.
e. Using a Phillips screwdriver, tighten the green screw.
Caution – Align the PCI card with the slot guides, insert it carefully, and secure it
firmly.
Caution – Do not forcibly push the PCI card when inserting it, even if it is not
moving smoothly. If the PCI card is forcibly inserted despite the presence of any
obstruction in a slot or any problem with a connector pin, serious damage may
result.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
Chapter 13
PCI Slot Device Replacement 13-23
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
13-24
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
14
CD-RW/DVD-RW Drive Unit
Replacement
This chapter explains the replacement procedures for a CD-RW/DVD-RW drive unit
(DVDU). It covers the following:
■
■
■
■
Section 14.1,
Section 14.2,
Section 14.3,
Section 14.4,
“Overview of a CD-RW/DVD-RW Drive Unit” on page 14-1
“Active Replacement” on page 14-5
“Hot Replacement” on page 14-9
“Cold Replacement” on page 14-11
There are three methods for replacing a CD-RW/DVD-RW drive unit: active
replacement, hot replacement, and cold replacement. For the definition of each type
of method, see Chapter 4.
14.1
Overview of a CD-RW/DVD-RW Drive
Unit
This section provides an overview of CD-RW/DVD-RW drive units and describes
their mounting locations.
A CD-RW/DVD-RW drive unit is a device for an optical disk that can be used for
high-capacity data recording in digital form. One CD-RW/DVD-RW drive unit is
mounted in each of the M8000, M9000 (base cabinet), and SPARC Enterprise M9000
server (expansion cabinet) model. Active replacement and hot replacement can be
performed on a CD-RW/DVD-RW drive unit.
FIGURE 14-1, FIGURE 14-2, and FIGURE 14-3 show the mounting location of the
CD-RW/DVD-RW drive unit.
14-1
FIGURE 14-1
CD-RW/DVD-RW Drive Unit Mounting Location (Front of the M8000)
DVDU
14-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 14-2
CD-RW/DVD-RW Drive Unit Mounting Location (Front of the M9000 Base
Cabinet)
DVDU
Chapter 14
CD-RW/DVD-RW Drive Unit Replacement
14-3
FIGURE 14-3
CD-RW/DVD-RW Drive Unit Mounting Location (Front of M9000 with the Expansion Cabinet)
DVDU
DVDU
Base cabinet
Expansion cabinet
TABLE 14-1 shows the abbreviated name of the CD-RW/DVD-RW drive unit.
TABLE 14-1
14-4
Abbreviated Name of the CD-RW/DVD-RW Drive Unit
Part name
Abbreviated name
CD-RW/DVD-RW drive unit
DVDU
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
14.2
Active Replacement
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Stop all connection to the target CD-RW/DVD-RW drive unit.
3. From the OS, execute the cfgadm -a command to verify the Ap_Id of
CD-RW/DVD-RW drive unit.
Example: If the CD-RW/DVD-RW drive unit is c0t4d0,
Ap_Id
c0::dsk/c0t4d0
Type
CD-ROM
Receptacle Occupant
:
connected configured
Condition
unknown
The Ap_Id which verified here is referred to as DVD_ApId, and the subsequent
cfgadm operations to CD-RW/DVD-RW drive unit use DVD_ApId.
4. From the OS, execute the cfgadm -c unconfigure DVD_Ap_Id command to
unconfigure the CD-RW/DVD-RW drive unit from the domain.
# cfgadm -c unconfigure c0::dsk/c0t4d0
5. Stop the connection to a tape drive unit if it happens to be on the same
controller as the target CD-RW/DVD-RW drive unit.
6. From the OS, execute the cfgadm -a command to verify the Ap_Id of tape drive
unit.
Example: If the tape drive unit is rmt/0,
# cfgadm -a
Ap_Id
Type Receptacle Occupant Condition
:
c0::rmt/0 tape connected configured unknown
The Ap_Id which verified here is referred to as TAPE_ApId, and the subsequent
cfgadm operations to tape drive unit use TAPE_ApId.
Chapter 14
CD-RW/DVD-RW Drive Unit Replacement
14-5
7. From the OS, execute the cfgadm -c unconfigure TAPE_Ap_Id command to
unconfigure the tape drive unit from the domain.
# cfgadm -c unconfigure c0::rmt/0
8. From the OS, execute the cfgadm -a command to verify that the status of
CD-RW/DVD-RW drive unit and tape drive unit has changed from
"configured" to "unconfigured."
# cfgadm -a
Ap_Id
c0::dsk/c0t4d0
c0::rmt/0
Type
Receptacle Occupant
Condition
:
CD-ROM connected unconfigured unknown
tape
connected unconfigured unknown
:
9. Detach the target CD-RW/DVD-RW drive unit from the port.
From the terminal that is connected to XSCF, log in to XSCF, and execute the
cfgdevice -c detach -p port no. command.
10. Confirm that the target CD-RW/DVD-RW drive unit is unconfigured from the
domain.
From XSCF, execute cfgdevice -l to confirm that the relevant port is not
displayed in the current connection.
11. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
12. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
14-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
13. Remove the CD-RW/DVD-RW drive unit to be replaced.
a. Using a Phillips screwdriver, loosen the two knob screws securing the
CD-RW/DVD-RW drive unit to be replaced.
b. While grasping the knob screws, pull out the CD-RW/DVD-RW drive unit
slowly from the slot.
FIGURE 14-4
Removing the CD-RW/DVD-RW Drive Unit (Front of the M8000)
1
2
Chapter 14
CD-RW/DVD-RW Drive Unit Replacement
14-7
FIGURE 14-5
Removing the CD-RW/DVD-RW Drive Unit (Front of the M9000)
1
2
14. Place the removed CD-RW/DVD-RW drive unit on an antistatic mat.
15. Mount the replacement CD-RW/DVD-RW drive unit by following the removal
instructions in Step 13 in reverse order. Align the CD-RW/DVD-RW drive unit
with the slot guides, insert it carefully, and secure it firmly.
16. Attach the CD-RW/DVD-RW drive unit to the port.
Execute the cfgdevice -c attach -p port no. command of XSCF.
17. Confirm that the replaced CD-RW/DVD-RW drive unit is configured to the
domain.
From XSCF, execute the cfgdevice -l command to verify that the specified Port
number is displayed in the current connection.
18. From the OS, execute the cfgadm -c configure DVD_Ap_Id command to
configure the CD-RW/DVD-RW drive unit to the domain.
# cfgadm -c configure c0::dsk/c0t4d0
14-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
19. From the OS, execute the cfgadm -c configure TAPE_Ap_Id command to
configure the tape drive unit to the domain.
# cfgadm -c configure c0::rmt/0
20. From the OS, execute the cfgadm -a command to verify that the status of
CD-RW/DVD-RW drive unit and the tape drive unit has changed from
"unconfigured" to "configured."
# cfgadm -a
Ap_Id
Type
Receptacle Occupant
:
c0::dsk/c0t4d0 CD-ROM connected configured
c0::rmt/0
tape
connected configured
:
Condition
unknown
unknown
21. Close the door on the server, and lock it.
14.3
Hot Replacement
The procedure is as follows.
1. Unlock and open the appropriate door on the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Stop all connection to the target CD-RW/DVD-RW drive unit.
3. Turn off power to the relevant domain.
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -d command.
The OS shutdown processing is executed for the relevant domain, and then the
power-off processing is executed.
4. Detach the target CD-RW/DVD-RW drive unit from the port.
From a terminal that is connected to XSCF, log in to XSCF, and execute the
cfgdevice -c detach -p <port no.> command.
Chapter 14
CD-RW/DVD-RW Drive Unit Replacement
14-9
5. Confirm that the target CD-RW/DVD-RW drive unit is unconfigured from the
domain.
From XSCF, execute cfgdevice -l to confirm that the relevant port is not displayed
in the current connection.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the CD-RW/DVD-RW drive unit to be replaced. (See FIGURE 14-4 or
FIGURE 14-5)
a. Using a Phillips screwdriver, loosen the two knob screws securing the
CD-RW/DVD-RW drive unit.
b. While grasping the knob screws, pull out the CD-RW/DVD-RW drive unit
slowly from the slot.
9. Place the removed CD-RW/DVD-RW drive unit on an antistatic mat.
10. Mount the replacement CD-RW/DVD-RW drive unit by following the removal
instructions in Step 8 in reverse order.
Caution – Do not forcibly push the CD-RW/DVD-RW drive unit in when inserting
it. If the CD-RW/DVD-RW drive unit is forcibly inserted despite the presence of any
obstruction in a slot or any problems with the connector pin, serious damage may
result.
11. Attach the CD-RW/DVD-RW drive unit to the port.
Execute the cfgdevice -c attach -p <port no.> command of XSCF.
12. Confirm that the replacement CD-RW/DVD-RW drive unit is configured to the
domain.
From XSCF, execute the cfgdevice -l command to confirm that the specified port
number is displayed in the current connection.
14-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
13. Turn on the power to the relevant domain.
From a console that is connected to XSCF, execute the poweron -d command of
XSCF.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Start up domains as necessary.
15. Close the door on the server and lock it.
14.4
Cold Replacement
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the OPNL from the Locked position to the
Service position.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
Chapter 14
CD-RW/DVD-RW Drive Unit Replacement 14-11
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the CD-RW/DVD-RW drive unit to be replaced. (See FIGURE 14-4 or
FIGURE 14-5)
a. Using a Phillips screwdriver, loosen the two knob screws securing the
CD-RW/DVD-RW drive unit to be replaced.
b. While grasping the knob screws, pull out the CD-RW/DVD-RW drive unit
slowly from the slot.
9. Place the removed CD-RW/DVD-RW drive unit on an antistatic mat.
14-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
10. Mount the replacement CD-RW/DVD-RW drive unit by following the removal
instructions in Step 8 in reverse order.
Caution – Do not forcibly push the CD-RW/DVD-RW drive unit in when inserting
it. If the CD-RW/DVD-RW drive unit is forcibly inserted despite the presence of any
obstruction in a slot or any problems with the connector pin, serious damage may
result.
11. Switch on all main line switches for the AC section (ACS) in the power supply
system.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command of
XSCF.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Start the domain as necessary.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
Chapter 14
CD-RW/DVD-RW Drive Unit Replacement 14-13
14-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
15
Tape Drive Unit Replacement
This chapter explains the replacement procedure for the tape drive unit (TAPEU). It
covers the following:
■
■
■
■
Section 15.1,
Section 15.2,
Section 15.3,
Section 15.4,
“Overview of the Tape Drive Unit” on page 15-1
“Active Replacement” on page 15-5
“Hot Replacement” on page 15-9
“Cold Replacement” on page 15-11
There are three methods for replacing a tape drive unit: active replacement, hot
replacement, and cold replacement. For the definition of each type of method, see
Section 4.1, “Types of Replacement Procedures” on page 4-2.
Note – Contact your sales representative for tape drive unit options on SPARC
Enterprise M8000/M9000 servers.
15.1
Overview of the Tape Drive Unit
This section provides an overview of tape drive unit s and describes their mounting
locations. Since the tape was originally called a digital audio tape, it is sometimes
referred to as a DAT.
One tape drive unit can be mounted in each cabinet. Active replacement and hot
replacement can be performed on a tape drive unit.
FIGURE 15-1, FIGURE 15-2, and FIGURE 15-3 show the mounting locations of tape drive
unit.
15-1
FIGURE 15-1
Tape Drive Unit Mounting Locations (Front of the M8000)
TAPEU
15-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 15-2
Tape Drive Unit Mounting Locations (Front of the M9000 Base Cabinet)
TAPEU
Chapter 15
Tape Drive Unit Replacement
15-3
FIGURE 15-3
Tape Drive Unit Mounting Locations (Front of M9000 with the Expansion Cabinet)
TAPEU
TAPEU
Base cabinet
Expansion cabinet
TABLE 15-1 shows the abbreviated name of the tape drive unit.
TABLE 15-1
15-4
Abbreviated Name of the Tape Drive Unit
Part name
Abbreviated name
Tape drive unit
TAPEU
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
15.2
Active Replacement
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Stop all connection to the target tape drive unit.
3. From the OS, execute the cfgadm -a command to verify the Ap_Id of tape drive
unit.
Example: If tape drive unit is rmt/0,
# cfgadm -a
Ap_Id
Type
Receptacle Occupant Condition
:
c0::rmt/0 tape connected configured unknown
The Ap_Id which verified here is referred to as TAPE_ApId, and the subsequent
cfgadm operations to tape drive unit use TAPE_ApId.
4. From the OS, execute the cfgadm -c unconfigure TAPE_Ap_Id command to
unconfigure the tape drive unit from the domain.
# cfgadm -c unconfigure c0::rmt/0
5. Stop the connection to the CD-RW/DVD-RW drive unit on the same controller
as the target tape drive unit.
6. From the OS, execute the cfgadm -a command to verify the Ap_Id of
CD-RW/DVD-RW drive unit.
Example: If CD-RW/DVD-RW drive unit is c0t4d0,
# cfgadm -a
Ap_Id
Type
Receptacle
Occupant
Condition
:
c0::dsk/c0t4d0
CD-ROM
connected
configured
unknown
The Ap_Id which verified here is referred to as DVD_ApId, and the subsequent
cfgadm operations to CD-RW/DVD-RW drive unit uses DVD_ApId.
Chapter 15
Tape Drive Unit Replacement
15-5
7. From the OS, execute the cfgadm -c unconfigure DVD_Ap_Id command to
unconfigure the CD-RW/DVD-RW drive unit from the domain.
# cfgadm -c unconfigure c0::dsk/c0t4d0
8. From the OS, execute the cfgadm -a command to verify that the status of tape
drive unit and CD-RW/DVD-RW drive unit has changed from "configured" to
"unconfigured."
# cfgadm -a
Ap_Id
c0::dsk/c0t4d0
c0::rmt/0
Type
Receptacle Occupant
:
CD-ROM connected unconfigured
tape
connected unconfigured
:
Condition
unknown
unknown
9. Detach the target tape drive unit from the port.
From the terminal that is connected to XSCF, log in to XSCF, and execute the
cfgdevice -c detach -p port no. command.
10. Confirm that the target tape drive unit is unconfigured from the domain.
From XSCF, execute cfgdevice -l to confirm that the relevant port is not
displayed in the current connection.
11. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
12. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
15-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
13. Remove the tape drive unit to be replaced.
a. Using a Phillips screwdriver, loosen the two knob screws securing the tape
drive unit to be replaced.
b. While grasping the knob screws, pull out the tape drive unit slowly from the
slot.
FIGURE 15-4
Removing the Tape Drive Unit (Front of the M8000)
1
2
Chapter 15
Tape Drive Unit Replacement
15-7
FIGURE 15-5
Removing the Tape Drive Unit (Front of the M9000)
1
2
14. Place the removed tape drive unit on an antistatic mat.
15. Mount the replacement tape drive unit by following the removal instructions in
Step 13 in reverse order. Align the tape drive unit with the slot guides, insert it
carefully, and secure it firmly.
16. Attach the tape drive unit to the port.
Execute the cfgdevice -c attach -p port no. command of XSCF.
17. Confirm that the replaced tape drive unit is configured to the domain.
From XSCF, execute the cfgdevice -l command to verify that the specified Port
number is displayed in the current connection.
18. From the OS, execute the cfgadm -c configure TAPE_Ap_Id command to
configure the tape drive unit to the domain.
# cfgadm -c configure c0::rmt/0
19. From the OS, execute the cfgadm -c configure DVD_Ap_Id command to
configure the CD-RW/DVD-RW drive unit to the domain.
# cfgadm -c configure c0::dsk/c0t4d0
15-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
20. From the OS, execute the cfgadm -a command to verify that the status of tape
drive unit and the CD-RW/DVD-RW drive unit has changed from
"unconfigured" to "configured."
# cfgadm -a
Ap_Id
Type
Receptacle Occupant
Condition
:
c0::dsk/c0t4d0 CD-ROM connected unconfigured unknown
c0::rmt/0
tape
connected unconfigured unknown
:
21. Close the door on the server, and lock it.
15.3
Hot Replacement
The procedure is as follows.
1. Unlock and open the appropriate door on the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Stop all connection to the target tape drive unit.
3. Turn off power to the relevant domain.
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -d command.
The OS shutdown processing is executed for the relevant domain, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Detach the target tape drive unit from the port.
From a terminal that is connected to XSCF, log in to XSCF, and execute the
cfgdevice -c detach -p <port no.> command.
6. Confirm that the target tape drive unit is unconfigured from the domain.
From XSCF, execute cfgdevice -l to confirm that the relevant port is not
displayed in the current connection.
Chapter 15
Tape Drive Unit Replacement
15-9
7. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
8. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
9. Remove the tape drive unit to be replaced. (See FIGURE 15-4 or FIGURE 15-5)
a. Using a Phillips screwdriver, loosen the two knob screws securing the tape
drive unit.
b. While grasping the knob screws, pull out the tape drive unit slowly from the
slot.
10. Place the removed tape drive unit on an antistatic mat.
11. Mount the replacement tape drive unit by following the removal instructions in
Step 9 in reverse order.
Caution – Do not forcibly push the tape drive unit in when inserting it. If the tape
drive unit is forcibly inserted despite the presence of any obstruction in a slot or any
problems with the connector pin, serious damage may result.
12. Attach the tape drive unit to the port.
Execute the cfgdevice -c attach -p <port no.> command of XSCF.
13. Confirm that the replacement tape drive unit is configured to the domain.
From XSCF, execute the cfgdevice -l command to confirm that the specified
port number is displayed in the current connection.
14. Turn on the power to the relevant domain.
From a console that is connected to XSCF, execute the poweron -d command of
XSCF.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Start up domains as necessary.
16. Close the door on the server and lock it.
15-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
15.4
Cold Replacement
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
Chapter 15
Tape Drive Unit Replacement 15-11
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the tape drive unit to be replaced. (See FIGURE 15-4 or FIGURE 15-5)
a. Using a Phillips screwdriver, loosen the two knob screws securing the tape
drive unit to be replaced.
b. While grasping the knob screws, pull out the tape drive unit slowly from the
slot.
9. Place the removed tape drive unit on an antistatic mat.
15-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
10. Mount the replacement tape drive unit by following the removal instructions in
Step 8 in reverse order.
Caution – Do not forcibly push the tape drive unit in when inserting it. If the tape
drive unit is forcibly inserted despite the presence of any obstruction in a slot or any
problems with the connector pin, serious damage may result.
11. Switch on all main line switches for the AC section (ACS) in the power supply
system.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Start the domain as necessary.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
Chapter 15
Tape Drive Unit Replacement 15-13
15-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
16
Clock Control Unit Replacement
This chapter explains the replacement procedure for the clock control unit (CLKU).
It covers the following:
■
■
Section 16.1, “Overview of the CLKU” on page 16-1
Section 16.2, “Cold Replacement” on page 16-4
Cold replacement is the only swapping method that can be used for the CLKU. For
the definition of the swapping method, see Section 4.1, “Types of Replacement
Procedures” on page 4-2.
16.1
Overview of the CLKU
This section provides an overview of the CLKU and describes its mounting
locations.
The CLKU is mounted only in the M9000, and it supplies clock signals to CPU
memory board units (CMUs) and crossbar units (XBUs). CLKUs are duplicated, but
active maintenance cannot be performed on them. For the M9000 with the expansion
cabinet, there are cables available for connecting the clocks in the base cabinet to
those in the extension cabinet.
FIGURE 16-1 and FIGURE 16-2 show the CKLU mounting locations in the cabinets of the
M9000 (base cabinet) and M9000 (with the expansion cabinet), respectively.
16-1
FIGURE 16-1
CLKU Mounting Locations (Front of the M9000 Base Cabinet)
CLKU_B#0
CLKU_B#1
16-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 16-2
CLKU Mounting Locations (Front of M9000 with the Expansion Cabinet)
CLKU_B#0
CLKU_B#1
Base cabinet
CLKU_B#2
CLKU_B#3
Expansion cabinet
TABLE 16-1 lists the abbreviated names and component numbers of CLKUs.
TABLE 16-1
Abbreviated Names and Numbers of CLKUs
Model name
Abbreviated names and numbers
M9000 base cabinet
CLKU_B #0, 1
M9000 expansion cabinet
CLKU_B #2, 3
Chapter 16
Clock Control Unit Replacement
16-3
16.2
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
16-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
Chapter 16
Clock Control Unit Replacement
16-5
8. Remove the CLKU to be replaced. (See FIGURE 16-3)
a. If the M9000 expansion cabinet is installed, slide the cable support bracket at
the front of the clock control unit up or down by one hole, and then use a
flathead screwdriver to remove the cable connected to the clock control unit.
(four cables from the base unit and two from the expansion unit).
b. Using a Phillips screwdriver, turn the knob screws securing the left and right
eject/lock levers 90 to 180 degrees to unlock the levers. (Turn the knob screw
while lightly pulling the lever by one hand.)
Caution – Move the cable support bracket of the crossbar unit up by one hole for
the work.
c. Unlock the eject/lock levers to pull the CLKU off the backplane.
d. While grasping the eject/lock levers, pull out the CLKU slowly from the slot.
FIGURE 16-3
Removing the CLKU (Front of the M9000 Base Cabinet)
2
1
1
9. Place the removed CLKU on an antistatic mat.
16-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
10. Mount the replacement CLKU by following the removal instructions in Step 8
in reverse order. Align the CLKU with the slot guides, insert it carefully, and
secure it firmly.
Push the eject/lock levers as far as they will go, and then push the knob screws
on the right and left eject/lock levers to lock them in position. If the knob screws
remain in position, the clock control unit is securely mounted.
Note – Use a torque screwdriver to secure the clock cable connector with the torque
of 0.2N*m (2.0kgf*cm).
Caution – If you are unable to obtain a torque screwdriver, finger tighten the clock
cable connectors. Do not secure them with a regular screwdriver.
Caution – Do not forcibly push the CLKU when inserting it, even if it is not moving
smoothly. If the CLKU is forcibly inserted despite the presence of any obstruction in
a slot or any problem with a connector pin, serious damage may result.
11. Switch on all main line switches for the AC section (ACS) in the power supply
system.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
Chapter 16
Clock Control Unit Replacement
16-7
14. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement CLKU, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
16-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
17
Crossbar Unit Replacement
This chapter explains the replacement procedure for crossbar units (XBUs). It covers
the following:
■
■
Section 17.1, “Overview of XBUs” on page 17-1
Section 17.2, “Cold Replacement” on page 17-4
As a replacement method for XBUs, only the cold replacement method can be used.
For the definition of the replacement method, see Section 4.1, “Types of Replacement
Procedures” on page 4-2.
17.1
Overview of XBUs
This section provides an overview of XBUs and describes their mounting locations.
XBUs are mounted only in the M9000. XBUs logically switch between connections to
CPU memory board units (CMUs) and I/O units (IOUs). Active maintenance cannot
be performed on XBUs even during degraded operation involving the XBUs.
FIGURE 17-1 and FIGURE 17-2 show the mounting locations of XBUs in the cabinets of
the M9000 (base cabinet), and M9000 (with the expansion cabinet), respectively.
17-1
FIGURE 17-1
XBU Mounting Locations (Front of the M9000 Base Cabinet)
XBU_B#0-#7
#7
#6
#5
#4
#3
#2
#1
#0
17-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 17-2
XBU Mounting Locations (Front of M9000 with the Expansion Cabinet)
XBU_B#0-#7
XBU_B#8-#15
#7
#15
#6
#14
#5
#13
#4
#12
#3
#11
#2
#10
#1
#9
#0
#8
Base cabinet
Expansion cabinet
TABLE 17-1 lists the abbreviated names and numbers of the XBU.
TABLE 17-1
Abbreviated Names and Numbers of XBUs
Model name
Abbreviated names and numbers
M9000 base cabinet
XBU_B #0 to 7
M9000 expansion cabinet
XBU_B #8 to 15
Chapter 17
Crossbar Unit Replacement
17-3
17.2
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
17-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the XBU to be replaced. (See FIGURE 17-3)
a. If the M9000 expansion cabinet is installed, slide both the top and bottom
cable support brackets at the front of the replacement target crossbar unit up
or down by two holes, remove the clock cables connecting the crossbar units
in the base and the expansion cabinets, and then use a flathead screwdriver
to remove the data cables.
b. Using a Phillips screwdriver, turn the knob screws securing the left and right
eject/lock levers 90 to 180 degrees to unlock the levers. (Turn the knob screw
while lightly pulling the lever by one hand.)
c. Unlock the eject/lock levers to pull the XBU off the backplane.
d. While grasping the eject/lock levers, pull out the XBU slowly from the slot.
Caution – Do not hang or yank the cable. Especially do not while the one side is
connected.
Caution – Tighten or remove the cable screws evenly on both sides.
Chapter 17
Crossbar Unit Replacement
17-5
FIGURE 17-3
Removing the XBU (Front of the M9000 Base Cabinet)
2
1
1
9. Place the removed crossbar unit on an antistatic mat, and then install the
replacement crossbar unit.
10. While grasping the eject/lock levers, properly align the crossbar unit with the
slot guide, and then gently insert it.
Caution – If you are unable to obtain a torque screwdriver, finger tighten the clock
cable connectors. Do not secure them with a regular screwdriver.
Caution – Do not forcibly push the crossbar unit in when inserting it. If the crossbar
unit is forcibly inserted despite the presence of any obstruction in a slot or any
problems with the connector pin, serious damage may result.
11. Lock the eject/lock levers to affix the crossbar unit to the back plane.
17-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
12. Push the eject/lock levers as far as they will go, and then push the knob screws
on the right and left eject/lock levers to lock them in position. If the knob
screws remain in position, the crossbar unit is securely mounted.
13. If a SPARC Enterprise M9000 server expansion cabinet is installed, connect
cables for connecting the base cabinet to the expansion cabinet crossbar unit.
a. To connect the cables, hold the cable connector covers and insert the
connectors perpendicularly with respect to the front plate of the XB unit
until they come into contact with the front plate.
b. Then, while ensuring that the cable weight does not cause the cable
connectors to become slanted, fix the data cable connectors in position with
a standard screwdriver while supporting the cable connector with one hand.
c. Use a torque screwdriver to secure the clock cable connector with the torque
of 0.2N*m (2.0kgf*cm).
Note – After fixing the cable connectors in position, confirm that the connector
housing is securely affixed to the XB unit, and that it is not slanted sideways,
upward or downward. If the cable connectors are affixed slanted, the affixed part
will become unstable, leading to communication failure.
Caution – Do not hang or yank the cable. Especially do not while the one side is
connected.
Caution – Tighten or remove the cable screws evenly on both sides.
Caution – If you are unable to obtain a torque screwdriver, finger tighten the clock
cable connectors. Do not secure them with a regular screwdriver.
Caution – Do not secure the clock cables to the cable folder along with the data
cables, and lay the clock cables on the data cables.
14. Return the top and bottom cable support brackets (in front of the replacement
crossbar unit) to their original positions in their respective upper and lower
holes to securely affix them.
Note – Visually check from the opening in the upper cover to ensure that the two
printed circuit boards do not come into contact with the cable.
Chapter 17
Crossbar Unit Replacement
17-7
15. Switch on all main line switches for the AC section (ACS) in the power supply
system.
16. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
17. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
18. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement XBU, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
19. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
20. Close the door on the server, and lock it.
17-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
18
AC Section Replacement
This chapter explains the replacement procedures for AC sections (ACSs). It covers
the following:
■
■
Section 18.1, “Overview of ACSs” on page 18-1
Section 18.2, “Cold Replacement” on page 18-5
As a replacement method for ACSs, only the cold replacement method can be used
for the sensor units. For the definition of the replacement method, see Section 4.1,
“Types of Replacement Procedures” on page 4-2.
On the rack-mountable dual power feed option and power cabinet (see Chapter 24),
the replacement procedure differs partially depending on the mounted AC section.
For details on types of AC section, see B.17.
18.1
■
The same replacement procedure is used for all single-phase AC sections (ACS_A,
ACS_B, and ACS_C). See the replacement procedure for the AC section for the
SPARC Enterprise M8000 server or SPARC Enterprise M9000 server.
■
The replacement procedure used for the AC section for the dual power feed
option (ACS_D) is the same as for the single-phase AC section, except for the fact
that the ACS_D does not include a power cord. See the replacement procedure for
the AC section for the SPARC Enterprise M8000 server or SPARC Enterprise
M9000 server.
■
The connector part of the power cable for the three-phase AC sections
(ACSTPH_A, ACSTPH_B, ACSTPH_C, and ACSTPH_D) is a bolt pin. See the
replacement procedure for the three-phase AC section.
Overview of ACSs
This section provides an overview of ACSs and shows their mounting locations.
18-1
An ACS distributes AC input power from external sources to power supply units
(PSUs). If the dual power feed option is not installed, a main line switch is mounted
in an ACS. An ACS can be duplicated, but active replacement cannot be performed
even in such a case.
FIGURE 18-1, FIGURE 18-2, and FIGURE 18-3 show the mounting locations of CMUs.
FIGURE 18-1
ACSs Mounting Locations (Rear of the M8000)
ACS_A
18-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 18-2
ACS Mounting Location (Front of the M9000 Base Cabinet)
ACS_B
Chapter 18
AC Section Replacement
18-3
FIGURE 18-3
ACS Mounting Locations (Front of M9000 with the Expansion Cabinet)
ACS_B#0
ACS_B#1
Base cabinet
Expansion cabinet
TABLE 18-1 lists the abbreviated names and component numbers of the ACSs.
TABLE 18-1
18-4
Abbreviated Names and Numbers of ACSs
Model name
Abbreviated names and numbers
M8000
ACS_A
M9000 with the base cabinet
ACS_B
M9000 with the expansion cabinet (base
cabinet side)
ACS_B#0
M9000 with the expansion cabinet
(expansion cabinet side)
ACS_B#1
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
18.2
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
Chapter 18
AC Section Replacement
18-5
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the ACS to be replaced. This section explains the ACS used for
single-phase power feed in the standard configuration.
M8000 (See FIGURE 18-4):
a. Using a Phillips screwdriver, loosen the thumb screw to remove the
connector protecting plate.
b. Loosen the connector retaining screws to disconnect three power cords.
c. Using a Phillips screwdriver, loosen the two thumb screws in the upper part
of the ACS.
Pull up the two levers to the open position.
d. Move the eject lever left to the open position.
e. While grasping the handle, pull out the ACS carefully from the slot.
M9000 (See FIGURE 18-5):
18-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
a. Loosen the cable support plate retaining screw, and remove the plate by
sliding it to the left.
b. Using a Phillips screwdriver, loosen the two thumb screws to remove the
connector protecting plate.
c. Using a Phillips screwdriver, loosen the connector retaining screws to
disconnect five power cords.
d. Move the eject lever up to the open position.
e. Using a Phillips screwdriver, loosen the two thumb screws in the upper part
of the ACS.
Pull up the two levers to the open position.
f. While grasping the handle, pull out the ACS carefully from the slot.
For three-phase AC section:
a. Unlock and open the door of the power cabinet.
b. Loosen the four screws on the ACS cover, and remove the ACS cover. (See
FIGURE 18-6)
c. Remove the power cables.
Caution – When removing power cables, switch off the main line switches. The
power cord must be removed by a qualified worker. Ask a system administrator.
d. Loosen the knob screws on the AC section, and pull out the AC section
while lifting the two levers.
e. With the AC section pulled out partially, grasp the front and top handles and
remove the AC section.
Chapter 18
AC Section Replacement
18-7
FIGURE 18-4
1
Removing the AC Section (Rear of the M8000)
1
2
18-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 18-5
1
Removing the AC Section (Front of the M9000)
1
2
Chapter 18
AC Section Replacement
18-9
FIGURE 18-6
Removing the Three-phase AC Section (Power Cabinet)
9. Place the removed AC section on an antistatic mat.
10. Mount the replacement AC section by following the removal instructions in
Step 8 in reverse order. Align the AC section with the slot guides, insert it
carefully, press the lever to the closed position, and secure it firmly.
Caution – Do not forcibly push the ACS when inserting it, even if it is not moving
smoothly. If the ACS is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
11. Switch on all main line switches for the AC section (ACS) in the power supply
system.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
18-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement ACS, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
Chapter 18
AC Section Replacement 18-11
18-12
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
19
DDC Replacement
This chapter explains the replacement procedures for DDCs (DC-to-DC Converters).
It covers the following:
■
■
■
Section 19.1, “Overview of the DDC” on page 19-1
Section 19.2, “Active Replacement and Hot Replacement” on page 19-3
Section 19.3, “Cold Replacement” on page 19-7
There are three methods for replacing DDCs: active replacement, hot replacement
and cold replacement. For the definition of the replacement method, see Section 4.1,
“Types of Replacement Procedures” on page 4-2.
19.1
Overview of the DDC
This section provides an overview of the DDCs that are mounted on the back plane
of the M8000 (BP_A), and describes their mounting locations.
A DDC supplies power to the back plane. DDCs are duplicated, and active
maintenance can be performed on them.
FIGURE 19-1 indicates the DDC mounting location in the M8000.
19-1
FIGURE 19-1
Mounting Locations of the DDCs (Front of the M8000)
DDC_A#0
DDC_A#1
TABLE 19-1 lists the abbreviated names and numbers of the DDCs.
TABLE 19-1
19-2
Abbreviated Names and Numbers of DDCs
Model name
Abbreviated names and numbers
M8000
DDC_A#0, 1
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
19.2
Active Replacement and Hot
Replacement
The active replacement procedure and hot replacement procedure for the DDC are
the same.
The procedure is as follows.
1. Unlock and open the appropriate door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Using the maintenance menu, select the DDC to be replaced. (Use the
maintenance menu for Step 8.)
The fieldeng permission is required for executing this command.
a. Log in to the XSCF from the terminal connected to the XSCF, and execute the
replacefru command.
The initial page of the maintenance menu is displayed as shown in FIGURE 19-2.
FIGURE 19-2
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Replacement Menu
Please select a type of FRU to be replaced.
1. CMU/IOU(CPU Memory Board Unit/IO Unit)
2. FAN(Fan Unit)
3. PSU(Power Supply Unit)
4. XSCFU(Extended System Control Facility Unit)
5. DDC_A(DDC for BP_A)
--------------------------------------------------Select [1-5|c:cancel] :
Note – DDC_A is displayed only for the M8000.
b. Select "5. DDC" from the replacement parts list in the maintenance menu.
Chapter 19
DDC Replacement
19-3
c. From here, perform operations according to the guidance displayed in the
maintenance menu.
Check the states of LEDs (POWER LED: off; CHECK LED: blinking) on the
DDC to be replaced, and begin the actual replacement work when a message
instructing the user to replace the target DDC is displayed. Leave the
maintenance menu as is until Step 7.
3. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
4. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
19-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
5. Remove the DDC to be replaced.
a. Using a Phillips screwdriver, turn the knob screw securing the bottom
eject/lock lever 90 to 180 degrees to unlock the lever. (Turn the knob screw
while lightly pulling the lever with one hand.)
b. Unlock the eject/lock levers to pull the DDC off the backplane.
c. While grasping the eject/lock levers, pull out the DDC slowly from the slot.
FIGURE 19-3
Removing the DDC (Front of the M8000)
1
2
1
6. Place the removed DDC on an antistatic mat.
Chapter 19
DDC Replacement
19-5
7. Mount the replacement DDC by following the removal instructions in Step 5
and Step 6 in reverse order. Align the DDC with the slot guides, insert it
carefully, and secure it firmly.
Push the bottom eject/lock lever as far as it will go, and then push the knob screw
on the lever to lock it in position. If the knob screws remain in position, the DDC
is securely mounted.
Caution – Do not forcibly push the DDC when inserting it, even if it is not moving
smoothly. If the DDC is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
8. Go back to the terminal that is connected to the XSCF, and enter data according
to the guidance of the maintenance menu displayed after replacement.
a. Perform this operation according to instructions in messages. The operation
is completed when the display of the maintenance menu returns to the
initial page.
b. Exit the maintenance menu.
9. Using the XSCF command showhardconf or showstatus, confirm that each
replacement component is normal.
Unless "*" is displayed for the replacement DDC, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
10. Close and lock the door of the server.
19-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
19.3
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Caution – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
Chapter 19
DDC Replacement
19-7
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the DDC to be replaced. (See FIGURE 19-3)
a. Using a Phillips screwdriver, turn the knob screw securing the bottom
eject/lock lever 90 to 180 degrees to unlock the lever. (Turn the knob screw
while lightly pulling the lever with one hand.)
b. Unlock the eject/lock levers to pull the DDC off the backplane.
c. While grasping the eject/lock levers, pull out the DDC slowly from the slot.
9. Place the removed DDC on an antistatic mat.
10. Mount the replacement DDC by following the removal instructions in Step 8 in
reverse order. Align the DDC with the slot guides, insert it carefully, and secure
it firmly.
Push the eject/lock levers as far as they will go, and then push the knob screws
on the right and left eject/lock levers to lock them in position. If the knob screws
remain in position, the DDC is securely mounted.
19-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Caution – Do not forcibly push the DDC when inserting it, even if it is not moving
smoothly. If the DDC is forcibly inserted despite the presence of any obstruction in a
slot or any problem with a connector pin, serious damage may result.
11. Switch on all main line switches for the AC section (ACS) in the power supply
system.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is not displayed for the replacement ACS, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
Chapter 19
DDC Replacement
19-9
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
19-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
20
Backplane Replacement
This chapter explains the replacement procedure for backplanes (BPs). It covers the
following:
■
■
Section 20.1, “Overview of the BP” on page 20-1
Section 20.2, “Cold Replacement” on page 20-1
As a replacement method for BPs, only the cold replacement method can be used.
For the definition of the replacement method, see Section 4.1, “Types of Replacement
Procedures” on page 4-2.
20.1
Overview of the BP
This section provides an overview of the BPs and shows their mounting locations.
A BP has connectors for interconnecting replaceable units in a cabinet. The following
BPs are available:
20.2
■
BP_A: used for interconnecting the CMUs and IOUs of the M8000
■
BP_B: used for interconnecting the CMUs, IOUs, XBUs of the M9000
■
PSUBP: used for interconnecting power supply units (PSUs)
■
FANBP: used for interconnecting FAN units
Cold Replacement
BPs are mounted in a cabinet, and it is necessary to remove many other components
before it can be accessed. Consequently, active maintenance cannot be performed on
BPs. Cold replacement must be used when replacing them.
20-1
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
20.2.1
M8000/M9000 BPs
FIGURE 20-1, FIGURE 20-2, and FIGURE 20-3 show the mounting locations of BPs of
M8000, BP for M9000 (base cabinet), and M9000 (with the expansion cabinet),
respectively.
FIGURE 20-1
BP Mounting Locations (Rear of the M8000)
BP_A
20-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-2
BP Mounting Locations (Front of the M9000 Base Cabinet)
BP_B
Chapter 20
Backplane Replacement
20-3
FIGURE 20-3
BP Mounting Locations (Front of M9000 with the Expansion Cabinet)
BP_B#0
BP_B#1
Base cabinet
Expansion cabinet
TABLE 20-1 lists the abbreviated names and numbers of the BPs.
TABLE 20-1
Abbreviated Names and Numbers of BPs
Model name
Abbreviated names and numbers
M8000
BP_A
M9000 with the base cabinet
BP_B
M9000 with the expansion cabinet (base cabinet side)
BP_B#0
M9000 with the expansion cabinet (expansion cabinet side)
BP_B#1
20-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
The cold replacement procedure is as follows:
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
For details on how to log in to the XSCF, see the SPARC Enterprise
M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide. Domain
administrator authority is required for executing this command.
The OS shutdown processing is executed for all the domains, and then the poweroff processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
Chapter 20
Backplane Replacement
20-5
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
The XSCF STANDBY LED (green) on the operator panel goes out.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Release the following components from the BP to be replaced or remove them
completely from the cabinet so that you can access the BP.
■
M8000 (See FIGURE 20-6):
Perform the following operations from the front of the unit:
a. Release all mounted CMUs from the BP. (See Chapter 6)
b. Pull two DDCs off the BP. (See Chapter 19)
c. Pull two XSCF units off the BP. (See Chapter 11)
20-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Perform the following operations from the rear of the unit (See FIGURE 20-4):
a. Remove the grounding wire on one side at the top of the rear door so that the
door can be opened at least 180 degrees.
b. Remove all FAN units on the I/O unit. (See Chapter 8) Mount the fixing
brackets of the FAN unit at their original locations, and secure them
temporarily.
c. Using a Phillips screwdriver, loosen the two thumb screws securing the duct
on the fan shelf. While grasping the handle, remove the duct.
d. Using a Phillips screwdriver, loosen the four thumb screws on the left and
right sides of the fan shelf.
e. While grasping the handle of the fan shelf, pull it out partially. Move your
hands to the indentations on the both sides of the fan shelf, and then pull
out the fan shelf carefully from the cabinet.
f. Remove all mounted I/O units (IOUs) including filler panels. (See Chapter 7)
g. Using a Phillips screwdriver, loosen the two knob screws on both sides of
the IOU shelf and the two screws on the innermost part of the shelf.
h. While grasping the frame of the IOU shelf, carefully pull out the IOU shelf
partially, then switch to grasping the shelf by the indentations on its side,
and remove it from the cabinet.
i. Using a Phillips screwdriver, loosen the two retaining screws of the Lshaped box on the left side in the cabinet to remove the box.
j. Using a Phillips screwdriver, loosen the retaining screws on the left and
right sides to remove the horizontal IOU-fan connector panel on the far side
in the cabinet. Disconnect the cables as well. Also remove the connector on
the right side.
Chapter 20
Backplane Replacement
20-7
■
M9000 (See FIGURE 20-7):
Perform the following operations from the rear of the unit:
a. Pull all mounted CMUs off the BP. (See Chapter 6)
b. Pull all mounted I/O units (IOUs) off the BP. (See Chapter 7)
Perform the following operations from the front of the unit (See FIGURE 20-5):
a. Remove all mounted crossbar units (XBUs).
For a model with the expansion cabinet, remove all cables and cable support
brackets as well. Disconnect the cables completely from both cabinets. (See
Chapter 17)
b. Remove two clock control units (CLKUs).
For a model with the expansion cabinet, disconnect the cables as well.
Disconnect the cables completely from both cabinets. (See Chapter 11)
c. Remove two XSCF units.
For a model with the expansion cabinet, disconnect the cables as well.
Disconnect the cables completely from both cabinets. (See Chapter 17)
d. Loosen the four screws securing the XBU shelf. While grasping the handle,
remove the shelf from the cabinet.
e. Remove all mounted I/O units (IOUs) including filler panels. (See Chapter 7)
f. Loosen the four screws securing the IOU shelf. While grasping the handle,
pull the shelf out partially. Then, move your hands to the top frame of the
IOU shelf, and remove the shelf from the cabinet.
20-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-4
Preparation for Accessing BP_A (Rear of the M8000)
Chapter 20
Backplane Replacement
20-9
FIGURE 20-5
Preparation for Accessing BP_B (Front of M9000)
9. Remove the BP to be replaced.
■
M8000 (See FIGURE 20-6):
a. Using a torque wrench (box driver size: M6), remove the four bolts that hold
the part shaped like an inverted L at the top edge of the bus bar of the BP.
b. While grasping the lock of the connector, remove every cable connected to
the BP.
Note – Be sure to press both lock release buttons inward and pull the connector
when removing the SAS cables (lock release buttons of the connector are yellow).
20-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
c. Arrange the cables so that the connectors do not get caught in anything
while the BP is being removed.
d. Using a Phillips screwdriver, loosen the eight thumb screws securing the BP.
The BP does not fall because it is suspended from the two guide rails inside the
upper part of the cabinet.
e. While grasping the two handles of the BP, remove the BP by moving it
carefully along the guide rails.
■
M9000 (See FIGURE 20-7):
a. Using a torque wrench, remove the four hexagonal bolts on the bus bar of
the BP.
b. With the connectors unlocked, disconnect all cables from the BP.
Note – Be sure to press both lock release buttons inward and pull the connector
when removing the SAS cables (lock release buttons of the connector are yellow).
c. Arrange the cables so that the connectors do not get caught on anything
while the BP is being removed.
d. Using a Phillips screwdriver, loosen the six thumb screws securing the BP.
The BP does not fall because it is mounted on the two guide rails inside the
lower part of the cabinet.
e. While grasping the upper parts of the BP frame, lower it toward you, and
place it horizontally on the guide rails.
f. While grasping the left and right sides of the BP frame, remove the BP by
moving it carefully along the guide rails.
Chapter 20
Backplane Replacement
20-11
FIGURE 20-6
20-12
Removing the BP (Rear of the M8000)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-7
Removing the BP (Front of the M9000)
10. Place the removed BP on an antistatic mat.
11. Mount the replacement BP by following the removal instructions in Step 8 and
Step 9 in reverse order.
Caution – For tightening the bus bar, choose a torque depending on the bolt size.
* For M8 bolts, use a torque of 8.24 N*m (84 kgf*cm).
* For M6 bolts, use a torque of 3.73 N*m (38 kgf*cm).
Caution – Remove the cover which is attached for shipment. Be sure not to hit
connector parts of the BP against anything. Otherwise, the connection pins may be
seriously damaged.
Caution – Be sure not to hit the connector parts on anything when mounting the
fan shelf. The connector on the edge of the base plate of the switch BP is not used to
connect the cable of the BP of the M8000. Be aware of this when connecting cables.
Chapter 20
Backplane Replacement
20-13
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
Domain administrator authority is required for executing this command.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Return the mode switch on the operator panel to its original position.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Using the XSCF command showhardconf or showstatus, confirm that each
replacement component is normal.
Unless "*" is displayed for the replacement BP_A or BP_B, it is operating
normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
17. Close the door on the server, and lock it.
20-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
20.2.2
PSU BP
FIGURE 20-8, FIGURE 20-9, and FIGURE 20-10 show the mounting locations of PSUBPs of
M8000, M9000 (base cabinet), and M9000 (with the expansion cabinet), respectively.
FIGURE 20-8
PSUBP Mounting Locations (Front of the M8000)
PSUBP_B
PSUBP_A
Chapter 20
Backplane Replacement
20-15
FIGURE 20-9
PSUBP Mounting Locations (Front of the M9000 Base Cabinet)
PSUBP_A#0
PSUBP_A#1
PSUBP_A#2
20-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-10
PSUBP Mounting Locations (Front of M9000 with the Expansion Cabinet)
PSUBP_A#3
PSUBP_A#0
PSUBP_A#4
PSUBP_A#1
PSUBP_A#5
PSUBP_A#2
Base cabinet
Expansion cabinet
TABLE 20-2 lists the abbreviated names and component numbers of the PSUBPs.
TABLE 20-2
Abbreviated Names and Numbers of PSU BPs
Model name
Abbreviated names and numbers
M8000
PSUBP_A and PSUBP_B
M9000 base cabinet
PSUBP_A#0 - 2
M9000 expansion cabinet
PSUBP_A#3 - 5
Chapter 20
Backplane Replacement
20-17
The cold replacement procedure is as follows:
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the poweroff processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
20-18
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove following components so that you can access the PSU BP to be
replaced.
■
M8000 (See FIGURE 20-11):
a. Remove all nine PSUs. (See Chapter 9)
b. Return the fixing brackets of the PSU at their original locations, and secure
them with screws temporarily.
c. Using a Phillips screwdriver, loosen the three thumb screws on the right side
of the PSU shelf.
d. While grasping the handle on the lower right side of the PSU shelf, pull out
the shelf carefully.
■
M9000 (See FIGURE 20-12):
a. Remove all 15 PSUs. (See Chapter 9)
b. Mount the fixing brackets of the PSU at their original locations, and secure
them with screws temporarily.
c. Using a Phillips screwdriver, loosen the two thumb screws in the upper part
of the PSU shelf.
d. While grasping the handle at the upper part of the PSU shelf, pull out the
shelf carefully.
9. Remove the PSU backplane (PSUBP) to be replaced. Although two types of BPs
are used, the replacement procedures are the same, with the only difference
being the numbers of connectors they have (PSUBP_A: 5 connectors; PSUBP_B:
4 connectors).
a. Using a Phillips screwdriver, loosen the two thumb screws in the PSU BP.
b. Pull out the PSUBP carefully.
Chapter 20
Backplane Replacement
20-19
FIGURE 20-11
Removing the PSUBP (Front of the M8000)
PSUBP_B
PSUBP_A
20-20
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-12
Removing the PSUBP (Front of the M9000)
10. Place the removed BP on an antistatic mat.
11. Mount the replacement BP by following the removal instructions in Step 8 and
Step 9 in reverse order. Align the holes by the side of the thumb screws of the
PSUBP with the two guide pins in the cabinet.
Note – To mount a FAN BP on a fan shelf that has a guide pin, align the FAN BP
with the guide pin during mounting.
Caution – Be sure not to hit connector parts of the BP against anything. Otherwise,
the connection pins may be seriously damaged.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
Chapter 20
Backplane Replacement
20-21
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Restore the original mode switch setting on the operator panel.
20-22
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
20.2.3
FAN BP
FIGURE 20-13, FIGURE 20-14, and FIGURE 20-15 show the mounting locations of FAN BPs
of M8000, M9000 (base cabinet), and M9000 (with the expansion cabinet),
respectively.
FIGURE 20-13
FANBP Mounting Locations (M8000)
FANBP_C#0
FANBP_C#1
Front
Rear
Chapter 20
Backplane Replacement
20-23
FIGURE 20-14
FANBP Mounting Locations (Rear of the M9000 Base Cabinet)
FANBP_B
FANBP_A
20-24
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-15
FANBP Mounting Locations (Rear of M9000 with the Expansion Cabinet)
FANBP_A#1
FANBP_B#1
FANBP_A#0
FANBP_B#0
Expansion cabinet
Base cabinet
TABLE 20-3 lists the abbreviated names and component numbers of the FANBPs.
TABLE 20-3
Abbreviated Names and Component Numbers of the FANBPs
Model name
Abbreviated names and numbers
M8000
FANBP_C#0 (front), FANBP_C#1 (rear)
M9000 with the base cabinet
FANBP_A, FANBP_B
M9000 with the expansion cabinet (base cabinet side)
FANBP_A#0, FANBP_B#0
M9000 with the expansion cabinet (expansion cabinet side)
FANBP_A#1, FANBP_B#1
Chapter 20
Backplane Replacement
20-25
The cold replacement procedure is as follows:
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Caution – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the poweroff processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
20-26
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Release the following components from the FAN backplane (FANBP) to be
replaced or remove them completely from the cabinet so that you can access the
FANBP. The FANBP is mounted at the rear of the fan shelf.
■
M8000 (See FIGURE 20-11):
For FANBP_C#0, perform the following operations from the front of the unit:
a. Remove all six FAN units. (See Chapter 8) After the FAN units are removed,
return the fan fixing brackets at their original locations, and secure them
temporarily.
b. Using a Phillips screwdriver, loosen the thumb screws on the left and right
sides of the fan shelf.
c. While grasping the handle at the lower center of the fan shelf, pull out the
fan shelf carefully.
For FANBP_C #1, perform the following operations from the rear of the unit:
a. Remove all six FAN units. (See Chapter 8) After the FAN units are removed,
mount the fan fixing brackets at their original locations, and secure them
temporarily.
b. Using a Phillips screwdriver, loosen the two thumb screws on each the left
and right sides of the fan shelf.
c. While grasping the handles on the left and right sides of the fan shelf, pull
out the shelf carefully.
■
M9000 (See FIGURE 20-12):
For FANBP_A, perform the following operations from the front and rear of the
unit:
a. Pull the four FAN units at the front off the BP. (See Chapter 8)
b. Go to the rear of the unit.
Chapter 20
Backplane Replacement
20-27
c. Remove all six FAN units on the left side. (See Chapter 8) After the FAN
units are removed, mount the fan fixing brackets at their original locations,
and secure them temporarily.
d. Using a Phillips screwdriver, loosen the two thumb screws in the upper part
of the fan shelf.
e. While grasping the handle at the upper part of the fan shelf, pull out the
shelf carefully.
For FANBP_B, perform the following operations from the rear of the unit:
a. Remove all six FAN units on the right side. (See Chapter 8)
b. Using a Phillips screwdriver, loosen the two thumb screws in the upper part
of the fan shelf.
c. While grasping the handle at the upper part of the fan shelf, pull out the
shelf carefully.
9. Remove the FANBP to be replaced from the fan shelf. Although three types of
FANBPs and three types of fan shelves are used, the replacement procedures
for the BP are the same.
a. Using a Phillips screwdriver, loosen two thumb screws on each the left and
right sides of the FANBP (FANBP_A or FANBP_B in the M9000) or three
thumb screws in each the upper and lower parts of the FANBP (FANBP_C in
the M8000).
b. Remove the FANBP.
20-28
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-16
Removing the FANBP (Rear of the M8000)
Chapter 20
Backplane Replacement
20-29
FIGURE 20-17
20-30
Removing the FANBP (Front of M8000)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 20-18
Removing the FANBP (Rear of M9000)
10. Place the removed BP on an antistatic mat.
11. Mount the replacement BP by following the removal instructions in Step 8 and
Step 9 in reverse order.
Note – To mount a FAN BP on a fan shelf that has a guide pin, align the FAN BP
with the guide pin during mounting.
Caution – Be sure not to hit connector parts of the BP against anything. Otherwise,
the connection pins may be seriously damaged.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
Chapter 20
Backplane Replacement
20-31
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement FANBP_A or FANBP_B, it is operating
normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
16. Restore the original mode switch setting on the operator panel.
20-32
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
21
Sensor Unit Replacement
This chapter explains the replacement procedure for the sensor unit (SNSU). It
covers the following:
■
■
Section 21.1, “Overview of the SNSU” on page 21-1
Section 21.2, “Cold Replacement” on page 21-4
As a replacement method for the SNSU, only the cold replacement method can be
used. For the definition of the replacement method, see Section 4.1, “Types of
Replacement Procedures” on page 4-2.
21.1
Overview of the SNSU
This section provides an overview of the SNSU and describes its mounting locations.
The sensor unit monitors the temperature and humidity of air drawn in by the
server. The sensor unit consists of only a detecting element, and the controller block
is built in to the operator panel. Active maintenance cannot be performed on the
SNSU.
FIGURE 21-1, FIGURE 21-2, and FIGURE 21-3 show the mounting locations of the SNSUs
of SPARC Enterprise M8000 server, SPARC Enterprise M9000 server (base cabinet),
and the base cabinet of SPARC Enterprise M9000 server with the expansion cabinet
respectively.
21-1
FIGURE 21-1
Mounting Location of the SNSU (Front of the M8000)
SNSU
SNSU
21-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 21-2
SNSU Mounting Location (Front of the M9000 Base Cabinet)
SNSU
Chapter 21
Sensor Unit Replacement
21-3
FIGURE 21-3
SNSU Mounting Location (Front of M9000 with the Expansion Cabinet)
SNSU
Base cabinet
Expansion cabinet
TABLE 21-1 lists the abbreviated names of the SNSUs.
TABLE 21-1
21.2
Abbreviated Names of the SNSU
Part name
Abbreviated name
Sensor unit
SNSU
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
21-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
Chapter 21
Sensor Unit Replacement
21-5
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
21-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
8. Remove the SNSU to be replaced. (See FIGURE 21-4 or FIGURE 21-5)
a. Using a Phillips screwdriver, loosen the two thumb screws in the SNSU.
b. While grasping the thumb screws, pull out the SNSU carefully from the slot.
FIGURE 21-4
Removing the SNSU (Front of the M8000)
1
2
Chapter 21
Sensor Unit Replacement
21-7
FIGURE 21-5
Removing the SNSU (Front of the M9000)
2
1
9. Place the removed SNSU on an antistatic mat.
10. Mount the replacement SNSU by following the removal instructions in Step 8
in reverse order.
Caution – Do not forcibly push the SNSU in when inserting it. If the SNSU is
forcibly inserted despite the presence of any obstruction in a slot or any problems
with the connector pin, serious damage may result.
11. Switch on all main line switches for the AC section (ACS) in the power supply
system.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
21-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
15. Close the door on the server, and lock it.
Chapter 21
Sensor Unit Replacement
21-9
21-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
22
Media Backplane Replacement
This chapter explains the replacement procedure for media backplanes (MEDBPs). It
covers the following:
■
■
Section 22.1, “Overview of the MEDBP” on page 22-1
Section 22.2, “Cold Replacement” on page 22-5
As a replacement method for MEDBPs, only the cold replacement method can be
used. For the definition of the replacement method, see Section 4.1, “Types of
Replacement Procedures” on page 4-2.
22.1
Overview of the MEDBP
This section provides an overview of MEDBPs and describes their mounting
locations.
A MEDBP provides the connectors for mounting a tape drive unit (TAPEU) and
CD-RW/DVD-RW drive unit (DVDU) in a cabinet. These connectors serve as sockets
on the cabinet side for connecting a tape drive unit and CD-RW/DVD-RW drive unit
to the cabinet. A MEDBP is further connected to a switch backplane, through which
it connects the tape drive unit or CD-RW/DVD-RW drive unit to the IOU. Active
maintenance cannot be performed on a MEDBP.
FIGURE 22-1, FIGURE 22-2, and FIGURE 22-3 show the mounting locations of the MEDBP
of M8000, M9000 (base cabinet), and M9000 (with the expansion cabinet),
respectively.
22-1
FIGURE 22-1
MEDBP Mounting Location (Rear of the M8000)
MEDBP
22-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 22-2
MEDBP Mounting Location (Front of the M9000 Base Cabinet)
MEDBP
Chapter 22
Media Backplane Replacement
22-3
FIGURE 22-3
MEDBP Mounting Locations (Front of M9000 with the Expansion Cabinet)
Base cabinet
MEDBP#0
Expansion cabinet
MEDBP#1
TABLE 22-1 lists the abbreviated names and numbers of the MEDBPs.
Abbreviated Names and Numbers of MEDBPs
TABLE 22-1
Model name
Abbreviated names and numbers
M8000
MEDBP
M9000 with the base cabinet
MEDBP
M9000 with the expansion cabinet (base cabinet side)
MEDBP#0
M9000 with the expansion cabinet (expansion cabinet side)
MEDBP#1
22-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
22.2
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
Chapter 22
Media Backplane Replacement
22-5
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the following components so that you can access the MEDBP to be
replaced.
M8000 (See FIGURE 22-4):
a. Remove the tape drive unit. (See Chapter 15)
b. Remove the CD-RW/DVD-RW drive unit. (See Chapter 14)
c. Using a Phillips screwdriver, loosen the two retaining screws to remove the
partition box located between the CD-RW/DVD-RW drive unit and XSCFU.
d. Remove the two XSCFUs. (See Chapter 11) If any cable is connected, remove
it.
22-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
M9000 (See FIGURE 22-5):
a. Remove the tape drive unit. (See Chapter 15)
b. Remove the CD-RW/DVD-RW drive unit. (See Chapter 14)
c. Remove I/O units #0, #2, and #4 (IOU#0, IOU#2, and IOU#4) on the right side
of the CD-RW/DVD-RW drive unit. A filler panel may be mounted
depending on the option configuration. (See Chapter 7)
9. Remove the MEDBP to be replaced.
a. Using a Phillips screwdriver, loosen the two thumb screws securing the
MEDBP.
b. Slide the MEDBP to the right.
c. Disconnect the two cables connected to the rear of the MEDBP.
Note – Be sure to press both lock release buttons inward and pull the connector
when removing the SAS cable (lock release buttons of the connector are yellow).
d. Along the guide, and remove MEDBP carefully
Chapter 22
Media Backplane Replacement
22-7
FIGURE 22-4
22-8
Removing the MEDBP (Front of the M8000)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 22-5
Removing the MEDBP (Front of the M9000)
10. Place the removed MEDBP on an antistatic mat.
11. Mount the replacement MEDBP by following the removal instructions in Step 8
and Step 9 in reverse order.
Caution – Do not forcibly push the MEDBP in when inserting it. If the MEDBP is
forcibly inserted despite the presence of any obstruction in a slot or any problems
with the connector pin, serious damage may result.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
Chapter 22
Media Backplane Replacement
22-9
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
22-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
23
Switch Backplane Replacement
This chapter explains the replacement procedure for switch backplanes (SWBPs). It
covers the following:
■
■
Section 23.1, “Overview of SWBPs” on page 23-1
Section 23.2, “Cold Replacement” on page 23-5
As a replacement method for SWBPs, only cold replacement can be used. For the
definition of the replacement method, see Section 4.1, “Types of Replacement
Procedures” on page 4-2.
23.1
Overview of SWBPs
This section provides an overview of SWBPs and describes their mounting locations.
A SWBP is connected to a tape drive unit (TAPEU) and CD-RW/DVD-RW drive unit
(DVDU) to an I/O unit (IOU) through the media backplane. An expander that
performs the switching of serial attached SCSI (SAS) interfaces is mounted on a
SWBP. Active maintenance cannot be performed on an SWBP.
FIGURE 23-1, FIGURE 23-2, and FIGURE 23-3 show the mounting locations of SWBPs of
M8000, M9000 (base cabinet), and M9000 (with the expansion cabinet), respectively.
23-1
FIGURE 23-1
SWBP Mounting Location (Rear of the M8000)
SWBP
23-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 23-2
SWBP Mounting Location (Front of the M9000 Base Cabinet)
SWBP
Chapter 23
Switch Backplane Replacement
23-3
FIGURE 23-3
SWBP Mounting Locations (Front of M9000 with the Expansion Cabinet)
Base cabinet
SWBP#0
Expansion cabinet
SWBP#1
TABLE 23-1 lists the abbreviated names and numbers of the SWBPs.
Abbreviated Names and Numbers of SWBPs
TABLE 23-1
Model name
Abbreviated names and numbers
M8000
SWBP
M9000 with the base cabinet
SWBP
M9000 with the expansion cabinet (base cabinet side)
SWBP#0
M9000 with the expansion cabinet (expansion cabinet side)
SWBP#1
23-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
23.2
Cold Replacement
In cold replacement, none of the replacement work is performed with the
maintenance menu from a terminal connected to XSCF.
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for the AC section (ACS) in the power supply
system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Note – If the optional three-phase 200 VAC input power supply unit is installed, the
main line switches are located on the three-phase ACS.
Chapter 23
Switch Backplane Replacement
23-5
Caution – If the dual power feed option is installed, switch off the main line
switches for both power supply units.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Remove the following parts so that you can access the SWBP to be replaced.
M8000 (See FIGURE 23-4):
Remove all I/O units (IOUs). (See Chapter 7)
M9000 (See FIGURE 23-5):
Remove all I/O units (IOUs) on the operator panel side. (See Chapter 7)
9. Remove the SWBP to be replaced. (See FIGURE 23-4 or FIGURE 23-5).
a. Disconnect five (SPARC Enterprise M9000 server) or three (SPARC
Enterprise M8000 server) signal cables and two power cords from the BP
while unlocking the connectors.
Note – Be sure to press both lock release buttons inward and pull the connector
when removing the SAS cables (lock release buttons of the connector are yellow).
b. Using a Phillips screwdriver, loosen the two screws securing the SWBP.
c. Pull out the SWBP carefully.
23-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 23-4
Removing the SWBP (Rear of the M8000)
Chapter 23
Switch Backplane Replacement
23-7
FIGURE 23-5
Removing the SWBP (Front of the M9000)
10. Place the removed SWBP on an antistatic mat.
11. Mount the replacement SWBP by following the removal instructions in Step 8
and Step 9 in reverse order.
Caution – Do not forcibly push the SWBP in when inserting it. If the SWBP is
forcibly inserted despite the presence of any obstruction in a slot or any problems
with the connector pin, serious damage may result.
The connector on the left edge of the base plate of the switch BP is not used for the
SPARC Enterprise M8000 server. Be aware of this when connecting cables.
12. Switch on all main line switches for the AC section (ACS) in the power supply
system.
13. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
23-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
14. Start (turn on power) all domains together. Take either of the following two
actions:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
15. Restore the original mode switch setting on the operator panel.
Note – When the maintenance work is completed, be sure to confirm the mode
switch setting and return the switch to Locked.
16. Close the door on the server, and lock it.
Chapter 23
Switch Backplane Replacement
23-9
23-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
24
Addition and Deletion of a RDPF
Option, Power Cabinet, and M9000
Expansion Cabinet
This chapter explains the procedures for adding and deleting a rack-mountable dual
power feed option (RDPF), a power cabinet, and an M9000 expansion cabinet, all of
which are server options. In the chapter, deletion of such an option means that the
work of removing it involves moving the server.
These options are not FRUs. Normally, the system configuration has been
determined prior to shipment, so the options are not intended to be added or deleted
by a user or field engineer. However, if the system configuration is changed or the
system is transported to another location, addition or deletion of one of the options
may be required. The sections below explain the following:
■
Section 24.1, “Addition and Deletion of Rack-mountable Dual Power Feed” on
page 24-2
■
Section 24.2, “Addition and Deletion of Power Cabinet” on page 24-13
■
Section 24.3, “Addition and Deletion of M9000 Expansion Cabinet” on page 24-16
For the basic installation procedures, see the SPARC Enterprise M8000/M9000 Servers
Installation Guide. No section in the SPARC Enterprise M8000/M9000 Servers
Installation Guide explains this addition and deletion work. Interpret an installation
procedure as an addition procedure and a removal procedure as a deletion
procedure.
24-1
24.1
Addition and Deletion of
Rack-mountable Dual Power Feed
This section explains the procedure for replacing a rack-mountable dual power feed
unit (RDPF). The sections below explain the following:
■
■
Section 24.1.1, “Overview of RDPF” on page 24-2
Section 24.1.2, “Addition and Deletion” on page 24-3
Turn off the server power before adding or deleting a rack-mountable dual power
feed option.
24.1.1
Overview of RDPF
This section provides an overview of rack-mountable dual power feed and describes
the mounting location.
Rack-mountable dual power feed is the M8000 dedicated option. It is mounted on
equipment rack of M8000. In case that M8000 uses the single-phase AC power input,
rack-mountable dual power feed enables it to receive the supply of power from the
separate independent external power supply and assures the power output to the
servers in time of errors of external power. A rack-mountable dual power feed
option is never a target of hot replacement because it is not a FRU.
FIGURE 24-1 shows the mounting location of rack-mountable dual power feed of
M8000.
24-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 24-1
RDPF Mounting Location (M8000)
RDPF
Front
Rear
TABLE 24-1 lists the abbreviated name of rack-mountable dual power feed of M8000.
TABLE 24-1
Abbreviated Name of Rack-mountable Dual Power Feed
Component name
Abbreviated name
Rack-mountable dual power feed
RDPF
24.1.2
Addition and Deletion
24.1.2.1
Addition
Note – When you order the rack-mountable dual power feed, it is attached with the
rack-mountable kit (bus bar, fixing bracket, screws etc.) to mount the main body of
rack-mountable dual power feed to M8000.
Chapter 24
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet
24-3
The procedure is as follows.
1. Unlock and open a door of the server.
For details, see Chapter 5.
Caution – Before touching FRU or the dummy (filler) unit, see Section 1.4,
“Antistatic Precautions” on page 1-6 and be sure to wear the antistatic wrist strap.
Failure to do so might result in serious damage to operating domains.
2. Change the mode switch setting on the operator panel from Locked to Service.
Note – Unless the mode switch is set to Locked when the maintenance work is
completed, the error notification function is disabled. Be sure to restore and confirm
the switch setting when the work is completed.
3. Stop (turn off power) all domains together. Take either of the following two
actions:
■
Press and hold the POWER switch on the operator panel for at least four
seconds.
■
From a console that is connected to XSCF, log in to XSCF, and execute the
poweroff -a command.
The OS shutdown processing is executed for all the domains, and then the
power-off processing is executed.
4. Confirm that the power-off processing is completed by confirming that the
POWER LED (green) on the operator panel is off.
5. Switch off all main line switches for AC section (ACS) of power supply system.
For details, see Section 4.5.3, “Main Line Switch Locations” on page 4-27.
Caution – After stopping all the domains, be sure to switch off the main line
switches in the power supply system to completely shut off the supply of power.
Disconnect the power cord from the customer's distribution panel.
6. Connect the clip of the antistatic wrist strap and of the antistatic mat to the
server grounding port. See Section 1.4, “Antistatic Precautions” on page 1-6.
24-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
7. Ensure that the metallic underside of the wrist strap is in direct contact with
your skin. The wrist strap should be snug around the wrist so that it does not
rotate. See Section 1.4, “Antistatic Precautions” on page 1-6.
Caution – Before handling FRUs, be sure to connect the clip of the antistatic wrist
strap and of the antistatic mat to the server grounding port, and attach the band of
the wrist strap to one of your wrists. Place the FRUs on the grounded antistatic mat.
Failure to take these antistatic measures might result in serious damage.
8. Place the additional rack-mountable dual power feed on an antistatic mat.
9. Mount the rack-mountable dual power feed.
a. Remove the six dummy panels from the bottom of the equipment rack of the
M8000.
b. Remove the cover panel (together with four screws) of the M8000, and mount
the two bus bars (L shape) from the installation kit by using two bolts.
(Torque: 8.24 N*m (84 kgf*cm))
c. Remove the connector panel (together with a thumb screw) of the M8000,
and mount it (with a thumb screw) on the mounting bracket from the
installation kit.
d. Mount the mounting bracket on the M8000 (with four screws, at the
mounting location of the covering panel).
e. Mount the cage nuts from the installation kit on the two poles (two for each
pole) at the rear of the equipment rack of the M8000.
Chapter 24
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet
24-5
FIGURE 24-2
Mounting Locations of cage Nuts
14
14
2
4
f. Mount the dual power feed unit at the front of the equipment rack of the
M8000.
24-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
g. Mount the dual power feed unit (with four screws) on the two poles at the
front of the equipment rack of the M8000.
Note – When securing the rack-mountable dual power feed unit on the equipment
rack, keep it positioned close to the PSUs of the M8000.
FIGURE 24-3
Mounting the Screws of Rack-mountable Dual Power Feed (Rear of M8000)
screws (three on both sides)
h. Remove the right screw securing the connector unit at the rear of the dual
power feed unit, loosen the left screw, slide the connector unit to the right,
and remove it.
i. Mount the two brackets (mound shaped) on the two poles (with two screws
for each pole) at the rear of the equipment rack of the SPARC Enterprise
M8000 server.
j. Secure the rear of the dual power feed unit on the two brackets (with three
screws each).
k. Connect the bus bars that were mounted in Step b to the bus bar at the rear
of the dual power feed unit by using the two bus bars (Z shape) from the
installation kit. Secure them with bolts. (Torque: 8.24 N*m (84 kgf*cm))
l. Connect the connectors of the four cables protruding from the rear of the
dual power feed unit to the DC1 connector panel.
Chapter 24
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet
24-7
FIGURE 24-4
Mounting the Cable (Rear of M8000)
connector panel
bus bar
conductive-springattached bracket
m. Mount the bracket with a conductive spring (with two screws) at the bottom
rear of the dual power feed unit.
n. Mount the connector unit that was removed in Step h at its original location.
24-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 24-5
Mounting the Connector Unit (Rear of M8000)
o. Mount the two brackets (with two screws each) for securing the ACS on the
side panel at the rear of the equipment rack of the M8000. For their
mounting locations, use the second and third holes from the bottom, and
secure them.
p. From the rear of the server, mount the ACS mounting frame, and secure it at
the rear of the server (with three screws each on both sides). Secure the two
brackets for securing the ACS (one screw each) and the dual power feed unit
at the rear of the server (with three screws on each side).
q. Secure the ACS mounting frame and the two brackets mounted in Step o
with screws (one screw each).
Chapter 24
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet
24-9
FIGURE 24-6
Mounting the Frame for ACS Mount (Rear of M8000)
r. Using two thumb screws, mount the AC section (ACS) on the frame for
mounting the ACS. For details, see Chapter 18.
24-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 24-7
Removing the ACS (Rear of M8000)
s. Mount nine power supply units (PSUs) at the front of the dual power feed
unit. For details, see Chapter 9.
Chapter 24
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet 24-11
FIGURE 24-8
Mounting the PSU (Front of the M8000)
screws
10. Connect power cords to the added rack-mountable dual power feed unit.
11. Switch on all main line switches for AC section (ACS) of both power supply
systems.
12. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
13. Start (turn on power) all domains together. Take either of the following two
actions:
■
24-12
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
From a console that is connected to XSCF, execute the poweron -a command
of XSCF.
The power to all the domains is turned on, and then the power-on self test (POST)
is executed. When ok prompt displayed on the console of relevant domain, POST
is completed.
For details of the power-on operation, see Section 4.4.2, “Powering the Server On”
on page 4-19, and the SPARC Enterprise M3000/M4000/M5000/M8000/M9000
Servers XSCF User’s Guide.
14. Confirm that the replacement components are normal by using the
showhardconf or showstatus command of XSCF.
Unless "*" is displayed for the replacement RDPF, it is operating normally.
For details of the showhardconf command and showstatus command, see the
SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF User’s Guide,
the SPARC Enterprise M3000/M4000/M5000/M8000/M9000 Servers XSCF Reference
Manual, or the man page.
Note – The showstatus command displays information on degraded components.
15. Restore the original mode switch setting on the operator panel.
Note – Unless the mode switch is set to Locked when the maintenance work is
completed, the error notification function is disabled. Be sure to restore and confirm
the switch setting when the work is completed.
16. Close the door on the server, and lock it.
24.1.2.2
Deletion
Follow the addition instructions in reverse order.
24.2
Addition and Deletion of Power Cabinet
The three procedures listed below can be used. Regardless of the procedure used, see
the power cabinet installation procedure explained in the SPARC Enterprise
M8000/M9000 Servers Installation Guide.
■
Chapter 24
Adding or deleting a power cabinet for the SPARC Enterprise M8000 server with
three-phase dual power feed configured
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet 24-13
■
Adding or deleting a power cabinet for the SPARC Enterprise M9000 server base
cabinet with single-phase or three-phase dual power feed configured
■
Adding or deleting a power cabinet for the SPARC Enterprise M9000 server
expansion cabinet with single-phase or three-phase dual power feed configured
FIGURE 24-9, FIGURE 24-10, and FIGURE 24-11 show installation diagrams for each case.
FIGURE 24-9
M8000 + Power Cabinet
Power cabinet
24-14
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 24-10
M9000 Base Cabinet + Power Cabinet
Power cabinet
Chapter 24
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet 24-15
FIGURE 24-11
M9000 Expansion Cabinet + Power Cabinet
Power cabinet
Power cabinet
24.3
Addition and Deletion of M9000
Expansion Cabinet
The procedure listed below can be used. For details, see the power cabinet
installation procedure explained in the SPARC Enterprise M8000/M9000 Servers
Installation Guide.
■
Adding or deleting an M9000 expansion cabinet for the M9000 base cabinet
FIGURE 24-12 shows an installation diagram of the cabinets.
24-16
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 24-12
M9000 Base Cabinet + M9000 Expansion Cabinet
M9000 expansion cabinet
Chapter 24
Addition and Deletion of a RDPF Option, Power Cabinet, and M9000 Expansion Cabinet 24-17
24-18
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
25
Addition and Deletion of CMU,
DIMM, IOU, HDD, PCI Cards and
TAPEU
This chapter explains the procedures for installing new FRUs and removing existing
FRUs. These procedures are called addition and deletion.
The following topics are covered in this chapter.
■
■
Section 25.1, “Addition” on page 25-2
Section 25.2, “Deletion” on page 25-4
All FRUs can be replaced, as shown in the chapters preceding , but addition and
deletion are limited to the following FRUs.
Since the details for adding and deleting FRUs are the same as some details for
replacing FRUs, this chapter explains only the main points. For more details, refer to
the appropriate chapter on FRU replacement.
These FRUs can be added:
■
■
■
■
■
■
CPU/Memory board unit (CMU)
DIMM
I/O unit (IOU)
Hard Disk Drive (HDD)
PCI card (PCI)
Tape Drive Unit (TAPEU)
These FRUs can be deleted:
■
■
■
CPU/Memory board unit (CMU)
I/O unit (IOU)
PCI card (PCI)
25-1
Some FRUs can be actively added or actively deleted:
■
■
■
■
Active addition or removal of a CMU or IOU is possible using the DR function.
Maintenance or addition of a DIMM module involves removing or installing a
CMU.
Active addition or removal of PCI cards (including link cards) is possible using
the PHP function.
Active addition of TAPEUs is possible
Caution – For deletion, a filler panel (TABLE 25-1) must be installed in the empty
slot. Without filler panels, the system might overheat.
TABLE 25-1
25.1
Filler Panels
Slot location
Filler panel name
CMU
CMU filler panel
IOU
IOU filler panel
PCI
PCI filler panel
Addition
Some steps for adding a FRU are the same as for replacing a FRU. For more details,
refer to the appropriate chapter on FRU replacement.
25.1.1
Active Addition
First, select a unit to be added as described below.
1. Log in to the XSCF.
2. Execute the addition command (addfru).
XSCF> addfru
3. When the following addition menu is displayed, select the type of FRU to be
added.
25-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE 25-1
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Addition Menu
Please select a type of FRU to be added.
1. CMU/IOU
(CPU Memory Board Unit/IO Unit)
2. FAN
(Fan Unit)
3. PSU
(Power Supply Unit)
--------------------------------------------------Select [1-3|c:cancel]:
Note – DDC_A is displayed only for the M8000.
Perform the operation according to the instructions displayed in the addition
menu.
Note – In hot addition of an IOU, mount the CMU and IOU in the proper location.
Otherwise, the added IOU may not operate.
25.1.2
Cold Addition
In cold addition the main line switches of the system are turned off and the addition
menu is not available. Perform the addition according to the following procedure.
1. Set the mode switch on the operator panel to Service.
2. Turn off power of all domains together.
Press and hold the POWER switch on the operator panel for at least four seconds
or execute the poweroff -a command from XSCF.
3. Switch off all main line switches for the AC section (ACS) in the power supply
system.
Chapter 25
Addition and Deletion of CMU, DIMM, IOU, HDD, PCI Cards and TAPEU
25-3
4. Add the FRUs.
Some steps for adding a FRU are the same as for replacing a FRU. For more
details about the addition procedure, refer to the appropriate chapter on FRU
replacement.
Note – In addition of an IOU, mount the CMU and IOU in the proper location.
Otherwise, the added IOU may not operate after power-on.
5. Switch on all main line switches for the AC section (ACS) in the power supply
system.
6. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
7. Start (turn on power to) all domains together using one of these steps:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
Execute the poweron -a command from XSCF.
8. Confirm that the addition components are normal by using the showhardconf
or showstatus command of XSCF.
XSCF> showhardconf
If "*" is not displayed for the addition FRU, it is operating normally.
Note – The showhardconf command cannot check the PCI card, Hard Disk Drive,
and tape drive unit. To check the Hard Disk Drive and tape drive unit, use the
probe-scsi-all command from the ok prompt. To check the PCI card, use the
show-devs command from the ok prompt.
9. When CMUs or IOUs are added, perform the SB settings and domain settings
for each CMU and IOU.
Ask a system administrator or a domain administrator for assistance.
10. Set the mode switch on the operator panel to Locked.
25.2
Deletion
Some steps for deleting a FRU are the same as for replacing a FRU. For more details,
refer to the appropriate chapter on FRU replacement.
25-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
25.2.1
Active Deletion
First, select a unit to be deleted as described below.
1. Log in to the XSCF.
2. Execute the deletion command (deletefru).
XSCF> deletefru
3. When the following deletion menu is displayed, select the type of FRU to be
deleted.
FIGURE 25-2
Initial Page of the Maintenance Menu
--------------------------------------------------Maintenance/Removal (Deletion) Menu
Please select a type of FRU to be removed (deleted).
1. CMU/IOU
(CPU Memory Board Unit/IO Unit)
--------------------------------------------------Select [1|c:cancel]:
Note – DDC_A is displayed only for the M8000.
4. Perform the operation according to the instructions displayed in the deletion
menu.
Caution – For deletion, a filler panel (TABLE 25-1) must be installed in the empty
slot. Without filler panels, the system might overheat.
25.2.2
Cold Deletion
In cold deletion the main line switches of the system are turned off and the deletion
menu is not available. Perform the deletion according to the following steps.
1. Set the mode switch on the operator panel to Service.
2. Turn off power of all domains together.
Press and hold the POWER switch on the operator panel for at least four seconds
or execute the poweroff -a command from XSCF.
Chapter 25
Addition and Deletion of CMU, DIMM, IOU, HDD, PCI Cards and TAPEU
25-5
3. Switch off all main line switches for the AC section (ACS) in the power supply
system.
4. Delete the FRUs.
The procedures are the same for removing of the replacement for the appropriate
FRU. Some steps for deleting a FRU are the same as for replacing a FRU. For more
details about the deletion procedure, refer to the appropriate chapter on FRU
replacement.
Caution – For deletion, a filler panel (TABLE 25-1) must be installed in the empty
slot. Without filler panels, the system might overheat.
5. Switch on all main line switches for the AC section (ACS) in the power supply
system.
6. Confirm that the XSCF STANDBY LED (green) on the operator panel remains
lit. If it is blinking, wait until it remains lit.
7. Start (turn on power to) all domains together using one of these steps:
■
Press and hold down (for less than four seconds) the POWER switch on the
operator panel.
■
Execute the poweron -a command from XSCF.
8. Confirm configuration information of the system by using the showhardconf
command of XSCF.
XSCF> showhardconf
If the deleted FRU is not displayed, it is operating normally.
Note – The showhardconf command cannot check the PCI card, Hard Disk Drive,
and tape drive unit. To check the Hard Disk Drive and tape drive unit, use the
probe-scsi-all command from the ok prompt. To check the PCI card, use the
show-devs command from the ok prompt.
9. When CMUs or IOUs are deleted, cancel the SB settings and domain settings
for each CMU and IOU.
Ask a system administrator or a domain administrator for assistance.
10. Set the mode switch on the operator panel to Locked.
25-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
APPENDIX
A
System Configuration
This appendix describes the installation conditions and configuration of the SPARC
Enterprise server.
A.1
Installation Conditions
The basic component of a domain configuration is the system board. The system
board consists of only a single CPU/Memory Board Unit (CMU), or it consists of one
CMU and one I/O Unit (IOU).
The system board mounting conditions are as follows:
■
CPU Module (CPUM)
Mount either two sets of two CPUMs, or one set of four identical CPUMs on the
CMU.
In the SPARC Enterprise M8000/M9000 server, the SPARC64 VI processor and
SPARC64 VII processor can be mounted together.
The following four CPUM combinations are supported as system board
configurations:
(1)
(2)
(3)
(4)
CPU0-SPARC64 VI
CPU1-SPARC64 VI
CPU2-SPARC64 VI
CPU2-SPARC64 VI
CPU0-SPARC64 VI
CPU1-SPARC64 VI
CPU2-SPARC64 VII
CPU3-SPARC64 VII
CPU0-SPARC64 VII
CPU1-SPARC64 VII
CPU2-SPARC64 VI
CPU3-SPARC64 VI
CPU0-SPARC64 VII
CPU1-SPARC64 VII
CPU2-SPARC64 VII
CPU3-SPARC64 VII
A-1
Be sure to use the same type of CPUM in each pair (CPUM#0 and CPUM#1, or
CPUM#2 and CPUM#3).
■
Memory Module (DIMM)
Mount DIMMs in sets of 16 DIMMs on the CMU.
Mount memory modules of the same size and rank in each group.
■
I/O Unit (IOU)
When mounting an IOU, also mount the CMU whose component number
represents the same mounting location as the IOU.
Example: Be sure to mount IOU#1 together with CMU#1.
A.2
System Configuration
TABLE A-1 to TABLE A-3 outline the system configuration. These tables do not include
some types of back plane.
A.2.1
SPARC Enterprise M8000 Server
This model has one cabinet.
Up to four CMUs (16 CPU chips / 32 cores) and four IOUs can be mounted on this
model. A total of 32 core expansion is possible. The primary units such as the CMU,
IOU, power supply unit, and FAN unit are common among the SPARC Enterprise
servers. The model is a floor-stand type, with an equipment rack space provided at
the top. (32 cores for the SPARC64 VI processors, 64 cores for the SPARC64 VII
processors)
A-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE A-1
System Configuration for SPARC Enterprise M8000 Server
Unit
Number
Cabinet
1
CMU
1 to 4
CPU module
Comments
2 or 4
• In case of SPARC64 VI processor
2.28 GHz, L2 cache 5M bytes
2.4 GHz, L2 cache 6M bytes
Two CPU cores per CPU chip
Two threads per CPU core (four threads per CPU
module)
Maximum of 64 threads per system
• In case of SPARC64 VII processor
2.52 GHz, L2 cache 6M bytes
2.88 GHz, L2 cache 6M bytes
Four CPU cores per CPU chip
Two threads per CPU core (eight threads per CPU
module)
Maximum of 128 threads per system
Memory module (DIMM)
16 or 32
1 rank:DIMM-1GB, DIMM-2GB, DIMM-4GB
2 rank:DIMM-2GB, DIMM-4GB, DIMM-8GB
In case of DIMM-2GB, the maximum is
256 GB/system.
1 to 4
IOU
PCI cassette (card feeding adaptor)
8
At maximum, 32 cassettes per system
Slot#0, 2, 4, 6: PCI-EX short cards or IOU onboard
device cards
Slot#1, 3, 5, 7: PCI-EX short cards or link cards
Appendix A
System Configuration
A-3
TABLE A-1
System Configuration for SPARC Enterprise M8000 Server (Continued)
Unit
Number
HDD
0 to 4
Comments
HDD #0, 1: Slot#0 must have an IOU onboard device
card.
HDD #2, 3: Slot#4 must have an IOU onboard device
card.
XSCFU
2
duplicated configuration
FAN unit
12
N+1 fans redundant
configuration
Power supply unit
9
N+1 redundant configuration
AC section
1
in case of single-phase power
feed
Operator panel
1
CD-RW/DVD-RW drive unit
1
Tape drive unit
1
Option
BP power supply (DDC)
2
duplicated configuration
Other option
External I/O expansion unit
Single-phase and dual power feed option
Power cabinet
(three-phase power feed option: includes 2 units of
AC section)
A.2.2
SPARC Enterprise M9000 Server (Base Cabinet)
The SPARC Enterprise M9000 server is available in two models.
■
This section describes the base cabinet.
■
For a description of the base cabinet with an expansion cabinet, see Section A.2.3,
“SPARC Enterprise M9000 Server (Base Cabinet + Expansion Cabinet)” on
page A-6.
A maximum of eight CMUs (32 CPU chips / 64 cores) and eight IOUs can be
mounted in the base cabinet. A total of 64 core expansion is possible. The primary
units such as the CMU, IOU, power supply unit, and FAN unit are common among
the SPARC Enterprise servers. This is a floor-stand type. (64 cores for the SPARC64
VI processors, 128 cores for the SPARC64 VII processors)
A-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE A-2
System Configuration for SPARC Enterprise M9000 Server
Unit
Number
Base cabinet
1
CMU
1 to 8
CPU module
Comments
2 or 4
• In case of SPARC64 VI processor
2.28 GHz, L2 cache 5M bytes
2.4 GHz, L2 cache 6M bytes
Two CPU cores per CPU chip
Two threads per CPU core (four threads per CPU
module)
Maximum of 128 threads per system
• In case of SPARC64 VII processor
2.52 GHz, L2 cache 6M bytes
2.88 GHz, L2 cache 6M bytes
Four CPU cores per CPU chip
Two threads per CPU core (eight threads per CPU
module)
Maximum of 256 threads per system
Memory module (DIMM)
16 or 32
1 rank:DIMM-1GB, DIMM-2GB, DIMM-4GB
2 rank:DIMM-2GB, DIMM-4GB, DIMM-8GB
In case of DIMM-2GB, the maximum is 512
GB/system.
1 to 8
IOU
PCI cassette (card feeding adaptor)
8
At maximum, 64 cassettes per system
Slot#0, 2, 4, 6: PCI-EX short cards or IOU onboard
device cards
Slot#1, 3, 5, 7: PCI-EX short cards or link cards
HDD
0 to 4
HDD #0, 1: Slot#0 must have an IOU onboard device
card.
Appendix A
System Configuration
A-5
TABLE A-2
System Configuration for SPARC Enterprise M9000 Server (Continued)
Unit
Number
Comments
HDD #2, 3: Slot#4 must have an IOU onboard device
card.
XBU
8
XSCFU
2
duplicated configuration
2
duplicated configuration
16
N+1 fans redundant
configuration
15 or
30 x 2
N+1 redundant configuration
AC section
1
in case of single-phase power
feed
Operator panel
1
CD-RW/DVD-RW drive unit
1
Tape drive unit
1
CLKU
FAN unit
*
Power supply unit *
15 in single power, 30 in dual power (for each power
feed)
Option
Other options
External I/O expansion unit
Single-phase and dual power feed option
Power cabinet
(three-phase power feed option: includes 2 units of
AC section)
* Even if the number of mounted CMUs or IOUs is small, the server is shipped with a full set of FAN units and power supply units installed.
A.2.3
SPARC Enterprise M9000 Server (Base Cabinet +
Expansion Cabinet)
This configuration includes the base cabinet and an expansion cabinet.
A-6
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
A maximum of 16 CMUs (64 CPU chips / 128 cores) and 16 IOUs can be mounted in
this model. A total of 128 core expansion is possible. The primary units such as the
CMU, IOU, power supply unit, and FAN unit are common among the SPARC
Enterprise servers. This is a floor-stand type. (128 cores for the SPARC64 VI
processors, 256 cores for the SPARC64 VII processors)
TABLE A-3
System Configuration for M9000 with Expansion Cabinet
Unit
Number
Comments
Cabinet
2
Connected by cables
CMU
2 to 16
1 to 8 per cabinet
CPU module
2 or 4
• In case of SPARC64 VI processor
2.28 GHz, L2 cache 5M bytes
2.4 GHz, L2 cache 6M bytes
Two CPU cores per CPU chip
Two threads per CPU core (four threads per CPU
module)
Maximum of 256 threads per system
• In case of SPARC64 VII processor:
2.52 GHz, L2 cache 6M bytes
2.88 GHz, L2 cache 6M bytes
Four CPU cores per CPU chip
Two threads per CPU core (eight threads per CPU
module)
Maximum of 512 threads per system
Memory module (DIMM)
16 or 32
1 rank:DIMM-1GB, DIMM-2GB, DIMM-4GB
2 rank:DIMM-2GB, DIMM-4GB, DIMM-8GB
In case of DIMM-2GB, the maximum is 1024
GB/system.
2 to 16
IOU
1 to 8 per cabinet
PCI cassette (card feeding adaptor)
8
At maximum, 128 cassettes per system
Appendix A
System Configuration
A-7
TABLE A-3
System Configuration for M9000 with Expansion Cabinet (Continued)
Unit
Number
Comments
Slot#0, 2, 4, 6: PCI-EX short cards or IOU onboard
device cards
Slot#1, 3, 5, 7: PCI-EX short cards or external I/O
expansion unit connection cards
HDD
0 to 4
HDD #0, 1: Slot#0 must have an IOU onboard device
card.
HDD #2, 3: Slot#4 must have an IOU onboard device
card.
16
XBU
8 per cabinet
2 per cabinet duplicated
configuration
XSCFU
with base cabinet having only the relaying function
CLKU
2 per cabinet duplicated
configuration
FAN unit *
32
N+1 fans redundant
configuration
30
N+1 redundant
configuration
2
in case of
single-phase power
feed
Operator panel
1
Mounted only on
Base Cabinet
CD-RW/DVD-RW drive unit
2
16 in base cabinet, 16 in expansion cabinet
Power supply unit *
15 in base cabinet, 15 in expansion cabinet
AC section
1 in base cabinet, 1 in expansion cabinet
1 in base cabinet, 1 in expansion cabinet
2
Tape drive unit
1 per cabinet
Other options
External I/O expansion unit
A-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
option
TABLE A-3
System Configuration for M9000 with Expansion Cabinet (Continued)
Unit
Number
Comments
Single-phase and dual power feed option
Power cabinet
(three-phase power feed option: includes 2 units of
AC section)
* Even if the number of mounted CMUs or IOUs is small, the server is shipped with a full set of FAN units and power supply units
installed.
Appendix A
System Configuration
A-9
A-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
APPENDIX
B
Components
This Appendix describes each component making up SPARC Enterprise.
For the most up-to-date information, contact a certified service engineer.
TABLE B-1 gives the name and abbreviation of each component, and whether active
replacement and hot replacement can be applied to each, together with the chapter
and the section where the replacement procedure is explained.
A circle symbol ( ) means a component is replaceable. A dash (—) means a
component is not replaceable.
TABLE B-1
FRU Components of High-end Server Models
Name:
CPU memory board unit: See B.1
• CPU module: See B.2
• Memory: See B.3
I/O unit: See B.4
Abbreviation
Active
replacement
Hot
replacement
Replacement procedure
CMU
(*1)
Chapter 6
CPUM
(*1)
Chapter 6
In case of SPARC64 VI
CPUM
(*1)
Chapter 6
In case of SPARC64 VII
MEM
(*1)
Chapter 6
IOU
(*1)
Chapter 7
• Hard disk drive: See B.5
HDD
Chapter 12
• PCI cassette: See B.6
PCICS
(*2)
Chapter 13
• IOU Onboard Device Card: See B.7
IOUA
— (*3)
Chapter 13
• External I/O expansion unit
connection card: See B.8
PCIB-EX
(*2)
Chapter 13
Crossbar Unit: See B.9
XBU_B
—
—
Chapter 17
Clock Control Unit: See B.10
CLKU_B
—
—
Chapter 16
B-1
TABLE B-1
FRU Components of High-end Server Models (Continued)
Name:
Abbreviation
Active
replacement
Hot
replacement
Replacement procedure
XSCF unit (M8000/M9000 Servers for base XSCFU_B
cabinet) B.11
Chapter 11
XSCF Unit (For M9000 Server Expansion
cabinet) B.11
XSCFU_C
Chapter 11
CD-RW/DVD-RW drive unit: See B.12
DVDU
Chapter 14
Tape drive unit: See B.13
TAPEU
Chapter 15
Operator Panel: See B.14
OPNL
—
—
Chapter 10
Sensor Unit: See B.15
SNSU
—
—
Chapter 21
Power supply unit: See B.16
PSU
AC section (For M8000 Server) B.17
ACS_A
ACSTPH-A
(*4)
—
Chapter 9
—
Chapter 18
Single-phase
Single-phase dual
power feed
Three-phase dual power
feed (delta connection)
Three-phase dual power
feed (star connection)
AC section (For M9000 Server) B.17
ACS_B
ACSTPH-B
—
—
Chapter 18
Single-phase
Three-phase dual power
feed (delta connection)
Three-phase dual power
feed (star connection)
AC section (For dual power feed option)
B.17
ACS_D
FAN unit (3 FAN): See B.18
FAN_A
(*5)
Chapter 8
FAN unit (2 FAN): See B.18
FAN_B
(*5)
Chapter 8
M8000 Backplane B.21
BP_A
• DC-DC Converter: See B.22
—
—
—
—
DDC_A
Chapter 18
Option
Section 20.2.1
Chapter 19
M9000 Backplane: See B.21
BP_B
—
—
Section 20.2.1
PSU Backplane (For M9000 Server) B.23
PSUBP_A
—
—
Section 20.2.2
PSU Backplane (For M8000 Server) B.23
PSUBP_B
—
—
Section 20.2.2
B-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE B-1
FRU Components of High-end Server Models (Continued)
Name:
Abbreviation
Active
replacement
Hot
replacement
Replacement procedure
FAN Backplane (For M9000 Server) B.24
FANBP_A
—
—
Section 20.2.3
FAN Backplane (For M9000 Server) B.24
FANBP_B
—
—
Section 20.2.3
FAN Backplane (For M8000 Server) B.24
FANBP_C
—
—
Section 20.2.3
Media Backplane: See B.25
MEDBP
—
—
Chapter 22
Switch Backplane: See B.26
SWBP
—
—
Chapter 23
Rack-mountable dual power feed option:
See B.20
RDPF
—
—
Chapter 24
Dual power feed option: See B.19
—
—
—
Three-phase power feed option: See B.19
—
—
—
Note – *1 It uses the DR function. It is available in the configuration in which all the
XSB on the target FRU can be detached by DR function.
Note – *2 It uses the PHP function. Depending on the PCI card usage, it requires the
multi-path setting.
Note – *3 Active replacement of an IOU onboard device card (IOUA) is not possible,
but active replacement of an I/O unit (IOU) as a whole is possible.
Note – *4 In case of the unit error, replace one unit at a time. When N+1 redundancy
is lost, cold replacement will be necessary.
Note – *5 In case of the unit error, replace one unit at a time.
Appendix B
Components
B-3
B.1
CPU Memory Board Unit
The CPU memory board unit (CMU) consists of a CPU memory board (CMB) with a
DC-DC converter mounted, a memory, and a CPU module (CPUM). It can be
combined with an I/O unit to configure a system (domain).
FIGURE B-1 shows the CMU.
FIGURE B-1
CPU Memory Board Unit (CMU)
The CMU has the following characteristics.
B-4
■
Active maintenance of CMUs is possible using the DR function.
■
In CMU maintenance, you can replace the target in units of CMU as a whole, CPU
module, or memory. To perform active maintenance, you need to pull out the
CMU from the system before replacing the target.
■
A CMU can be installed or removed using the DR function during system
operation.
■
A dedicated LSI is mounted for system control (SC) and memory access control
(MAC).
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.2
■
For the bus connecting SC and SC as well as SC and crossbar unit, a high-speed
link is used.
■
For the bus connecting SC and MAC as well as SC and CPU, a high-speed link is
used.
■
For the memory, DIMM is used.
■
The CMU has a label bearing the printed component number and manufacture's
serial number.
CPU Module
CPU module consists of CPU and DDC. Up to four CPU modules can be mounted
on a CMU. There are four types of CPU modules.
FIGURE B-2 shows the CPU module.
FIGURE B-2
CPU Module
The CPU module has the following characteristics.
■
Active maintenance of CPU modules is possible using the DR function.
■
To perform active maintenance, you need to pull out the CMU from the system
before replacing the target CPU module.
Appendix B
Components
B-5
B-6
■
A dedicated LSI for CPU is mounted.
■
One CPU chip is mounted on a CPU module. In case of SPARC 64 VI processor,
one CPU chip has two CPU cores, each of which has two threads. This means four
threads per CPU module.
In case of SPARC 64 VII processor, one CPU chip has four CPU cores, each of
which has two threads. This means eight threads per CPU module.
A thread is a component of a CPU that can execute threads. When viewed from
Solaris OS, one thread is seen as one CPU.
■
The performance difference between four types of CPU module is as follows.
■
In case of SPARC64 VI processor:
2.28 GHz, L2 cache 5M bytes/2.4 GHz, L2 cache 6M bytes
■
In case of SPARC64 VII processor:
2.52 GHz, L2 cache 6M bytes/2.88 GHz, L2 cache 6M bytes
■
For the bus connecting CPU and SC, a high-speed link is used.
■
The CPU module has a label bearing the printed component number and
manufacture's serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.3
Memory
The type of memory used is DIMM.
FIGURE B-3 shows the memory.
FIGURE B-3
Memory
The memory has the following characteristics.
■
Active maintenance of memory is possible using the DR function.
■
Maintenance or installation of a DIMM module involves removing or installing a
CMU.
■
The category of the DIMM as specified by capacity and type is as follows:
■
■
DIMM-1GB
■
DIMM-2GB
■
DIMM-4GB
■
DIMM-8GB
Depending on the DIMMs used, the label format of component numbers and
serial numbers differs.
Appendix B
Components
B-7
B.4
I/O Unit
An I/O Unit (IOU) consists of an I/O board (IOB) with two types of control LSIs and
a DDC mounted, an HDD, and a PCI cassette. An IOU can be combined with a CMU
to configure a system (domain).
FIGURE B-4 shows the IOU.
FIGURE B-4
I/O Unit (IOU)
The IOU has the following characteristics.
B-8
■
Active maintenance of IOUs is possible using the DR function.
■
Active replacement of a failed IOU during system operation is possible using the
DR function.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
An IOU can be added or removed using the DR function during system
operation.
■
Active maintenance of PCI slots is possible using the PHP function.
■
The IOU has eight PCI slots with PCI cassettes mounted. PCI cards can be
mounted on PCI cassettes. Up to four HDDs (2.5 inch with SAS interface) can be
mounted on an IOU.
■
Up to eight PCI-Express short cards can be mounted. If an IOU onboard device
card (IOUA) is mounted instead of a PCI-Express short card, such units as the
internal hard disk drive (HDD), internal CD-RW/DVD-RW drive unit, and tape
drive unit can be connected. Also, if a link card is mounted, the external I/O
expansion unit can be cable-connected to the SPARC Enterprise servers, enabling
more PCI cards to be mounted.
■
Even-number slots (#0, #2, #4, #6): PCI-Express short card or IOUA
Only an IOUA card in slot #0 or #4 can control up to two internal HDDs.
■
Odd-number slots (#1, #3, #5, #7): PCI-Express short card or link card
■
A dedicated LSI is used for the system control.
■
For the bus connecting the control portion and crossbar unit, a high-speed link is
used.
■
PCI-Express interface is being used.
■
The IOU has a label bearing the printed component number and manufacture's
serial number.
Appendix B
Components
B-9
B.5
Hard Disk Drive
The hard disk drive (HDD) is mounted on an IOU. Up to four HDDs can be
mounted on an IOU.
FIGURE B-5 shows the HDD.
FIGURE B-5
Hard Disk Drive (HDD)
The HDD has the following characteristics.
B.6
■
Active replacement of a failed HDD during system operation is possible.
■
Active installation of HDDs during system operation is possible.
■
It allows mounting of a 2.5 inch HDD with an SAS interface.
■
Depending on the HDDs used, the label format of component numbers and serial
numbers differs.
PCI Cassette
The PCI cassette is an adapter used to mount different types of PCI cards in an IOU.
A cassette for PCI Express Short card (PCICS) is also available.
B-10
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE B-6 shows the PCICS.
FIGURE B-6
PCI Express Short card cassette (PCICS)
The PCI cassette has the following characteristics.
■
Active maintenance of PCICSs is possible using the PHP function.
■
Active installation or removal of PCI cards (including link cards) is possible using
the PHP function.
■
A PCICS can have a short card with the PCI-Express interface mounted.
■
A PCICS has a label bearing the printed component number and manufacturer's
serial number.
Appendix B
Components
B-11
B.7
IOU Onboard Device Card
The IOU onboard device card (IOUA) is an option for mounting CD-RW/DVD-RW
drive units or tape drive units in SPARC Enterprise server cabinets. Connect to the
CD-RW/DVD-RW drive unit or tape drive unit is controlled by IOUA. The IOUA to
control these devices must be mounted on PCI slot 0 or 4 of the IOU.
FIGURE B-7 shows the IOUA.
FIGURE B-7
IOU Onboard Device Card (IOUA)
The IOUA has the following characteristics.
B-12
■
Hot replacement of an IOUA is possible.
■
SAS interface is not supported.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.8
Link Card (External I/O Expansion Unit
Connection Card)
The external I/O expansion unit connection card includes two link cards. One link
card is mounted on the host server. The other link card is mounted on the I/O boat.
The cards are supplied with a cable for connection.
FIGURE B-8 shows the PCIB-EX.
FIGURE B-8
External I/O Expansion Unit Connection Card (PCIB-EX)
The PCIB-EX has the following characteristics.
■
Active maintenance of a PCIB_EX is possible using the PHP function.
■
Active installation or removal of PCI cards (including link cards) is possible using
the PHP function.
■
The maximum cable length for a card is 25 m.
Appendix B
Components
B-13
B.9
Crossbar Unit
A crossbar unit (XBU: formal abbreviation is XBU_B) is mounted only in the SPARC
Enterprise M9000 server. The XBU consists of a connector and a crossbar switch
(XB). The connector is used to physically connect a CMU and IOU, and the XB
switches logical connections and controls data transfer between the CMU and IOU.
Up to eight XBUs can be mounted in a cabinet.
FIGURE B-9 shows the XBU_B.
FIGURE B-9
Crossbar Unit (XBU_B)
The XBU has the following characteristics.
B-14
■
Only cold replacement is available for the XBU.
■
In the SPARC Enterprise M9000 server (with an expansion cabinet) model, control
over the expansion cabinet is done by connecting the XBUs of both base cabinet
and expansion cabinet with cables.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
The XBU has a normal mode and a degraded mode. In normal mode, all eight sets
operate; and in degraded mode in which a failure has occurred, the four sets
containing the failed set are disconnected, and operation continues with the
remaining four normal sets. If an XB fails, the system is shut down to disconnect
the four sets including the failed XB, and it is then automatically recovered.
■
A dedicated LSI chip is used for the crossbar switch (XB).
■
For the bus connecting CMU and XBU, a high-speed link is used.
■
For the bus connecting the IOU and XBU and the bus connecting XBUs of the base
cabinet and expansion cabinet, a high-speed link is being used.
■
The XBU has a label bearing the printed component number and manufacture's
serial number.
Appendix B
Components
B-15
B.10
Clock Control Unit
A clock control unit (CLKU: formal abbreviation is CLKU_B) is mounted only in the
SPARC Enterprise M9000 server, and it supplies clock signals to a CMU and crossbar
unit (XBU). Furthermore, in the SPARC Enterprise M9000 server with the expansion
cabinet, both CLKUs in the base cabinet and expansion cabinet are connected to
enable each to supply clock signals to the other cabinet.
In the M8000, two LSI chips for clock signals instead of the CLKU are mounted on
the backplane.
FIGURE B-10 shows the CLKU_B.
FIGURE B-10
Clock Control Unit (CLKU_B)
The CLKU has the following characteristics.
B-16
■
Only cold replacement is available for CLKU_B.
■
A dedicated LSI chip is used for clock control.
■
CMU and XBU are supplied with the standard clocks, respectively.
■
The CLKU has a label bearing the printed component number and manufacture's
serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.11
XSCF Unit
XSCF unit (XSCFU: eXtended System Control Facility unit) controls the system.
Two types of XSCFU are available: the XSCFU_B for the high-end servers and
XSCFU_C for the expansion cabinet of the M9000. XSCFU_B has a connector for
cable connection to XSCFU_C provided for expansion cabinet use. XSCFU_C has no
functions for controlling the system, but it provides an interface for passing the
status of each unit in the expansion cabinet to XSCFU_B.
FIGURE B-11 and FIGURE B-12 show the XSCFU respectively.
FIGURE B-11
eXtended System Control Facility Unit (XSCFU_B)
Appendix B
Components
B-17
FIGURE B-12
eXtended System Control Facility Unit (XSCFU_C)
XSCFU has the following characteristics.
B-18
■
The XSCFU has a duplicated configuration as the standard specification, allowing
hot replacement of a failed XSCFU during system operation.
■
XSCFU is the central control unit of the system and has the following functions:
■
It controls each unit.
■
It works as the console of each domain by using the TTY communication
protocol.
■
It communicates with each domain by using the command-line interface.
■
It manages the device information.
■
It supports the browser-based user interface (BUI) function called XSCF Web.
■
It allows connection to maintenance terminals.
■
It supports remote maintenance services.
■
It allows interlocking control for power supply and cluster control by using a
remote cabinet interface (RCI).
■
The XSCFU has a label bearing the printed component number and manufacture's
serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.12
CD-RW/DVD-RW Drive Unit
The CD-RW/DVD-RW drive unit (DVDU) can be mounted on each cabinet. One for
the M8000/M9000 with the base cabinet, and up to two for the M9000 with an
expansion cabinet can be mounted. The CD-RW/DVD-RW drive unit is controlled
by the optional IOU onboard device card (IOUA).
The CD-RW/DVD-RW drive unit is shown in FIGURE B-13.
FIGURE B-13
CD-RW/DVD-RW Drive Unit (DVDU)
The CD-RW/DVD-RW drive unit has the following characteristics.
■
Active replacement or hot replacement of CD-RW/DVD-RW drive units is
possible.
■
CD-RW/DVD-RW drive unit is mounted with the slim type CD-RW/DVD-RW
Unit by default.
■
XSCF command can switch the domain which can use the CD-RW/DVD-RW
drive unit.
■
The CD-RW/DVD-RW drive unit has a label bearing the printed component
number and manufacture's serial number.
Appendix B
Components
B-19
B.13
Tape Drive Unit
Tape drive unit (TAPEU) is optional and can be mounted on each cabinet. One for
theM8000 /M9000 with the base cabinet, and up to two for the M9000 with an
expansion cabinet can be mounted. TAPEU is controlled by the optional IOU
onboard device card (IOUA).
The TAPEU is shown in FIGURE B-14.
FIGURE B-14
Tape Drive Unit (TAPEU)
Tape drive unit has the following characteristics.
B-20
■
Active replacement or hot replacement of tape drive units is possible.
■
Active installation of tape drive units during system operation is possible.
■
Tape drive unit supports DAT72 (SATA interface).
■
By changing the switch setting on the tape drive unit, it can be used by each
domain.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
B.14
The tape drive unit has a label bearing the printed component number and
manufacture's serial number.
Operator Panel
The operator panel (OPNL) allows the operation of the high-end server and the
display of condition. The OPNL is mounted on the front door. With the M9000, the
OPNL is mounted only on the base cabinet.
The front and rear views of the OPNL are shown in FIGURE B-15 and FIGURE B-16
respectively.
FIGURE B-15
Front View of the Operator Panel (OPNL)
Appendix B
Components
B-21
FIGURE B-16
Rear View of the Operator Panel (OPNL)
OPNL has the following characteristics.
B-22
■
Only cold replacement is available for the OPNL.
■
The following switches are mounted on the OPNL to enable operation of the
server.
■
POWER switch to turn on the power.
■
Mode switch to control the POWER switch
■
The LED on the front of the OPNL displays the condition of the server. The LED
of the rear of the OPNL displays the status of OPNL itself.
■
On OPNL, EEPROM is mounted to store the device identification information and
the user setting information. These sets of information are also stored in XSCFU.
■
The OPNL has a label bearing the printed component number and manufacture's
serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.15
Sensor Unit
The sensor unit (SNSU) monitors the intake-air temperature and humidity. In the
M9000, this unit is mounted only on a base cabinet. When the output value of the
sensor exceeds the reference value, XSCF issues an alarm to the system.
FIGURE B-17 shows the SNSU.
FIGURE B-17
Sensor Unit (SNSU)
SNSU has the following characteristics.
■
Only cold replacement is available for the SNSU.
■
Only the temperature sensor and the humidity sensor are mounted.
■
Data processing such as A/D conversion is performed with the operator panel
(OPNL).
■
The SNSU has a label bearing the printed component number and manufacture's
serial number.
Appendix B
Components
B-23
B.16
Power Supply Unit
The power supply unit (PSU) receives AC 200 to 240 V, and supplies each unit of the
SPARC Enterprise with DC48 V and 12 VL. When all CMUs and IOUs are mounted
without the dual power feed option, nine units for the M8000, 15 units for the M9000
(base cabinet only) and 30 units for the M9000 (with expansion cabinet) will be
mounted. The number of mounted units also vary with the system configuration.
The PSU is shown in FIGURE B-18.
FIGURE B-18
Power Supply Unit (PSU)
The PSU has the following characteristics.
B-24
■
Because PSUs are normally operating in redundancy, active maintenance of a
failed PSU during system operation is possible.
■
The internal power factor correction (PFC) circuit secures the stable output and
suppresses the harmonic current that adversely affects the environment.
■
The current balancing function enables the concurrent operation of a large
number of PSUs, allowing the redundant configurations.
■
The PSU has a label bearing the printed component number and manufacture's
serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.17
AC Section
AC section (ACS) is positioned at the input side of the power supply unit (PSU), to
protect the system from electromagnetic interference (EMI).
There are two types of ACS, ACS for M8000 and ACS for M9000. Both single-phase
AC and three-phase AC types are available for both types.
ACS consists of a terminal strip for power input, a main line switch, and so on.
Among the ACS, ACS_D is a unit for supplying the internal current to the threephase power feed cabinet, which doesn't have a main line switch and the external
connection.
FIGURE B-19, FIGURE B-20, FIGURE B-21, FIGURE B-22, FIGURE B-23, FIGURE B-24,
FIGURE B-25, and FIGURE B-26 show the ACS.
■
The ACS_A is a single-phase power feed unit for the SPARC Enterprise M8000
server. This unit uses three power cords to receive power and distribute AC
power to nine PSUs.
FIGURE B-19
AC Section (ACS_A)
Appendix B
Components
B-25
■
The ACS_B is a single-phase power feed unit for the SPARC Enterprise M9000
server and the SPARC Enterprise M9000 server single-phase dual power feed
cabinet (option). This unit uses five power cords to receive power and distribute
AC power to 15 PSUs.
FIGURE B-20
■
The ACS_C is a single-phase power feed unit for the SPARC Enterprise M8000
server single-phase dual power feed cabinet (option). This unit uses three power
cords to receive power and distribute AC power to nine PSUs.
FIGURE B-21
■
B-26
AC Section (ACS_B)
AC Section (ACS_C)
The ACSTPH_A is a unit that receives three-phase (delta) power for the SPARC
Enterprise M8000 server three-phase dual power feed cabinet and switches
between three-phase and single-phase. This unit uses two power cables to receive
power, switch from three-phase to single-phase current internally, and distribute
AC power to 18 PSUs.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE B-22
■
The ACSTPH_B is a unit that receives three-phase (star) power for the SPARC
Enterprise M8000 server three-phase dual power feed cabinet and switches
between three-phase and single-phase. This unit uses two power cables to receive
power, switch from three-phase to single-phase current internally, and distribute
AC power to 18 PSUs.
FIGURE B-23
■
AC Section (ACSTPH_A)
AC Section (ACSTPH_B)
The ACSTPH_C is a unit that receives three-phase (delta) power for the SPARC
Enterprise M9000 server three-phase dual power feed cabinet and switches
between three-phase and single-phase. This unit uses two power cables to receive
power, switch from three-phase to single-phase current internally, and distribute
AC power to 30 PSUs.
Appendix B
Components
B-27
FIGURE B-24
■
The ACSTPH_D is a unit that receives three-phase (star) power for the SPARC
Enterprise M9000 server three-phase dual power feed cabinet and switches
between three-phase and single-phase. This unit uses two power cables to receive
power, switch from three-phase to single-phase current internally, and distribute
AC power to 30 PSUs.
FIGURE B-25
B-28
AC Section (ACSTPH_C)
AC Section (ACSTPH_D)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
The ACS_D is an internal current power feed unit for the SPARC Enterprise
M8000 server/SPARC Enterprise M9000 server three-phase dual power feed
cabinet. This unit supplies single-phase current to the server. It does not use
external cables.
FIGURE B-26
AC Section (ACS_D)
ACS has the following characteristics.
■
Only cold replacement is available for the ACS.
Appendix B
Components
B-29
B.18
FAN Unit
The FAN unit consists of two to three cabinet cooling FAN units. Two types of FAN
unit are available: the FAN_A type, which is the common FAN unit for SPARC
Enterprise, and the FAN_B type, which is the a dedicated FAN unit for the M8000.
FIGURE B-27 and FIGURE B-28 show the FAN unit.
FIGURE B-27
B-30
FAN Unit (FAN_A)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
FIGURE B-28
FAN Unit (FAN_B)
The FAN unit has the following characteristics.
■
Active maintenance of a failed FAN unit during system operation is possible.
■
The FAN units are controlled by XSCF.
■
Even if a FAN unit fails during system operation, the redundant FAN unit
configuration ensures that system operation can continue.
■
The FAN unit has a label bearing the printed component number and
manufacturer's serial number.
Appendix B
Components
B-31
B.19
Power Cabinet
A dual power feeder (DPF) and a three-phase AC power feeder are mounted on the
power cabinet. All of these are optional.
The dual power feeder receives its power supply from two independent external
power sources, ensuring that, even if a failure occurs in one source, power is
supplied to the system from the other source that in normal state. It supports both
single-phase and three-phase AC.
The three-phase AC power feeder is required when the input power is three-phase
AC.
B-32
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
The power cabinet is shown in FIGURE B-29.
FIGURE B-29
Power Cabinet
The power cabinet has the following characteristics.
■
Only cold replacement is available for the power cabinet.
■
It provides the mounting locations for all components of the dual power feeder
(DPF) and the three-phase AC power feeder. There are no exclusive conditions for
mounting.
■
The power cabinet has a label bearing the printed component number and
manufacture's serial number.
Appendix B
Components
B-33
B.20
Rack-mountable Dual Power Feed
Rack-mountable Dual Power Feed (RDPF) is a dedicated option for M8000.
The Rack-mountable Dual Power Feed receives its power supply from two
independent external power sources, ensuring that, even if a failure occurs in one
source, power is supplied to the system from the other source that in normal state. It
is applicable only for single-phase AC input. For three-phase AC power input, a
power cabinet is required.
The Rack-mountable Dual Power Feed is shown in FIGURE B-30.
FIGURE B-30
Rack-mountable Dual Power Feed (RDPF)
The Rack-mountable Dual Power Feed has the following characteristics.
B-34
■
Only cold replacement is available for the rack-mountable dual power feed.
■
It provides the mounting locations for all components of the dual power feeder
(DPF) and the three-phase AC power feeder. There are no exclusive conditions for
mounting.
■
The Rack-mountable Dual Power Feed has a label bearing the printed component
number and manufacture's serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.21
Backplane
The backplane (BP) is a printed circuit board (PCB) that has connectors and PC
board used to interconnect the cabinet-mounted units. There are two types of BP for
SPARC Enterprise available: namely, the BP_A for the M8000 and the BP_B for the
M9000. The BP_A makes connections among CMU, IOU, XSCFU and DDC_A (See
next section). The BP_B makes connections among CMU, IOU, XSCFU, CLKU and
XBU.
The BP is shown in FIGURE B-31 and FIGURE B-32.
FIGURE B-31
Backplane (BP_A)
Appendix B
Components
B-35
FIGURE B-32
Backplane (BP_B)
The BP has the following characteristics.
B-36
■
Only cold replacement is available for the BP.
■
Since there are no crossbar units for the M8000, such electronic devices as LSI for
crossbar switches (XB) are mounted on BP_A.
■
BP_B, which consists of connectors and mechanical components only, has no
electronic devices.
■
The BP for SPARC Enterprise has a label bearing the printed component number
and manufacture's serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.22
DDC
DC-DC converter (DDC_A) supplies DC power to electronic components mounted
on the backplane (BP_A) for M8000. DDC_A receives 48 VDC and 12 VL and outputs
1.0 VDC, 1.8 VDC, and 3.3 VL.
The DDC_A is shown in FIGURE B-33.
FIGURE B-33
DC-DC Converter (DDC_A)
DDC_A has the following characteristics.
■
Because DDC_As are normally operating in redundancy, active maintenance or
hot maintenance of a failed DDC_A during system operation is possible.
■
The current balancing function enables concurrent redundant operations.
■
The DDC_A has a label bearing the printed component number and
manufacture's serial number
Appendix B
Components
B-37
B.23
PSU Backplane
PSU backplane (PSUBP) is a PCB that has a connector for interconnecting multiple
power supply units (PSU) concurrently operated. There are two types of PSUBP,
namely the PSUBP_A commonly used for all SPARC Enterprise and the PSUBP_B
for the M8000 only.
The PSUBP is shown in FIGURE B-34 and FIGURE B-35.
FIGURE B-34
PSU Backplane (PSUBP_A)
FIGURE B-35
PSU Backplane (PSUBP_B)
The PSUBP has the following characteristics.
B-38
■
Because the PSUBP is a built-in unit in the cabinet, only cold replacement is
available for the PSUBP.
■
The PSUBP connects the output of PSU to each unit through system backplane
(BP_A or BP_B).
■
PUSBP_A allows connection to up to five PSUs.
■
The PUSBP_B enables connection to up to four PSUs.
■
The PSUBP has a label bearing the printed component number and manufacture's
serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.24
FAN Backplane
The FAN backplane (FANBP) is a PCB that has a connector for connecting multiple
FAN units (FAN_A or FAN_B). Three types of FANBP are available: the FANBP_C
for the M8000, and FANBP_A and FANBP_B for the M9000.
FIGURE B-36, FIGURE B-37, and FIGURE B-38 show the FANBP.
FIGURE B-36
FAN Backplane (FANBP_A)
Appendix B
Components
B-39
FIGURE B-37
FAN Backplane (FANBP_B)
FIGURE B-38
FAN Backplane (FANBP_C)
FANBP has the following characteristics.
B-40
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
■
Because the FANBP is a built-in unit in the cabinet, only cold replacement is
available for the FANBP.
■
FANBP connects FAN unit to the cabinet.
■
Up to 10 FAN units can be mounted on FANBP_A, and up to six FAN units can be
mounted on FANBP_B and FANBP_C each.
■
The FANBP has circuits mounted for monitoring the fan rotation speed, switching
between high/low speed, and controlling the CHECK LED. It is controlled by the
XSCF.
■
The FNBP has a label bearing the printed component number and manufacture's
serial number.
Appendix B
Components
B-41
B.25
Media Backplane
The Media backplane (MEDBP) is a PCB that has connectors for mounting the CDRW/DVD-RW drive unit motherboard and the tape drive unit motherboard. On the
CD-RW/DVD-RW drive unit motherboard and tape drive unit motherboard, the
CD-RW/DVD-RW drive unit and the tape drive unit can be mounted, respectively.
One MEDBP unit can be mounted on the M8000 and M9000 (base cabinet), and two
units can be mounted on the M9000 (with expansion cabinet).
The MEDBP is shown in FIGURE B-39.
FIGURE B-39
Media Backplane (MEDBP)
The MEDBP has the following characteristics.
B-42
■
Only cold replacement is available for MEDBP.
■
The MEDBP has a label bearing the printed component number and
manufacture's serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
B.26
Switch Backplane
Switch backplane (SWBP) is a PCB that has connectors to connect Media BP to
multiple IOUs. The SWBP is a unit that switches between IOUs to set which IOU is
to control the CD-RW/DVD-RW drive unit and the tape drive unit (TAPEU). One
SWBP unit can be mounted on the M8000 and M9000 (base cabinet), and two units
can be mounted on the M9000 (with expansion cabinet).
FIGURE B-40 shows the SWBP.
FIGURE B-40
Switch Backplane (SWBP)
SWBP has the following characteristics.
■
Only cold replacement is available for the SWBP.
■
Twelve ports of expander are mounted to control eight IOUs, one CD-RW/DVDRW drive unit, and one tape drive unit.
■
XSCF controls the settings for switching.
■
It supplies the power (12 V, 5 V) to Media backplane (CD-RW/DVD-RW drive
unit and tape drive unit).
Appendix B
Components
B-43
■
B-44
The SWBP has a label bearing the printed component number and manufacture's
serial number.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
C
External Interface Specifications
This appendix describes the specifications of the connectors provided on the SPARC
Enterprise server unit:
■
■
■
■
C.1
“Serial Port” on page 1
“UPC Port” on page 2
“USB Port” on page 2
“Connection Diagram for Serial Cable” on page 3
Serial Port
TABLE C-1
Serial Port
Setting
12 3 4 5 6 7 8
Pin No>
Signal name
Input/output
Description
1
RTS
Output
Request to Send
2
DTR
Output
Data Terminal Ready
3
TXD
Output
Send Data
4
GND
—
Ground
5
GND
—
Ground
6
RXD
Input
Receive Data
7
DSR
Input
Data Set Ready
8
CTS
Input
Clear to Send
C-1
C.2
UPC Port
TABLE C-2
UPC Port
Setting
5 4 3 2 1
9 8 7 6
C.3
Signal name
Input/output
Description
1
ER
Output
Equipment power-on report signal
2
NC
—
Not connected
3
NC
—
Not connected
4
NC
—
Not connected
5
SG
—
Ground
6
*BPS
Input
UPS hardware error signal
7
*BTL
Input
Battery end of discharge warning
signal
8
NC
---
Not connected
9
*ACOFF
Input
Power failure detection signal
USB Port
TABLE C-3
USB Port
Setting
1 2 3 4
C-2
Pin No>
Pin No>
Signal name
Name
Abbreviation
1
VBUS
Output
Power supply
2
-DATA
Input/output
Data
3
+DATA
Input/output
Data
4
GND
—
Ground
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
C.4
Connection Diagram for Serial Cable
FIGURE C-1
Connection diagram for serial cable
Chapter C
External Interface Specifications
C-3
C-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
APPENDIX
D
UPS Controller
This appendix explains the UPS controller (UPC) that controls the uninterruptible
power supply (UPS) unit.
D.1
Overview
The UPS is a unit to ensure a stable supply of power to the system even in the event
of a power supply failure or a massive blackout.
When the UPC port on the server is connected to the UPS signal cable that has a
UPC interface, if the UPS detects a power failure, it notifies the server of the failure
so that the server can perform emergency shutdown processing. This stops the
server safely.
D.2
Signal Cable
Use the shielded pair cable with the following specifications.
■
Direct current resistance (roundtrip/pair): 400 W/km or less
■
Cable length: Up to 10 m (33 ft.)
D-1
D.3
Configuration of Signal Lines
This section provides the signal definitions.
FIGURE D-1 shows the configuration of signal lines when the UPS is connected.
FIGURE D-1
Connection between the Server and UPS
UPS
TABLE D-1
7
*BTL
6
*BPS/*UALM
9
*ACOFF
5
SG
UPS cable
Server
Signal Line Definitions
Signal name
Definitions
Pin number
Remarks
*BPS/*UALM
Indicates a UPS error.
6
Normal state: Off
Abnormal state: On
*BTL
Indicates that the battery level of the UPS has
decreased and warns that battery power will be
unavailable after a certain period of time has
elapsed.
7
Normal state: Off
Abnormal warning:
On
(Note 1)
D-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
TABLE D-1
Signal Line Definitions (Continued)
Signal name
Definitions
Pin number
Remarks
*ACOFF
Indicates that a power failure has occurred in the
commercial power supply connected to the UPS.
9
Normal state: Off
Abnormal state: On
(Note 2)
SG
Signal ground
5
ER
(Equipment Ready)
Indicates that the server is operating
1
Do not connect
anything to this pin.
On: Indicates that the contact is closed.
Off: Indicates that the contact is open.
Note 1: Use a UPS that can normally supply power from the battery at least 10 to 60 seconds after *BTL is turned on.
Note 2: Use a UPS that can normally supply power from the battery within 2 seconds of a momentary commercial power interruption
even if *ACOFF is not turned on.
Appendix D
UPS Controller
D-3
D.4
Power Supply Conditions
TABLE D-2 and TABLE D-3 show the power supply conditions of the UPC interface.
D.4.1
Input Circuit
TABLE D-2
Input Power Supply Conditions
Signal name
Input conditions
*BPS/*UALM
No voltage relay contact
*BTL
Contact rating DC 12 V, 10 mA or greater (0.5 A maximum)
*ACOFF
Use of a gold-plated contact or reed relay is recommended.
Remarks: Limit the signal line chattering period to 1 ms.
D.4.2
Output Circuit
TABLE D-3
Output Power Supply Conditions
Signal name
Output conditions
ER
Output voltage
Output current
D-4
VOH
DC 3.1 V (minimum)
VOL
DC 0 - 0.4 V (maximum)
IOH
-4 mA (maximum)
IOL
4 mA (maximum)
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
D.5
UPS Cable
The UPS cable specifications are as follows:
■
Connector shape
D-SUB9 pin male (receptacle side: female)
DEU-9PF-F0
■
Terminal array
FIGURE D-2 shows the pin signals of the UPC port and UPS cable.
Do not use any pin that is not defined (Pin 2, 3, 4, and 8 in FIGURE D-2). The pins
on the cable side are as follows:
FIGURE D-2
Correspondence between the UPC Ports and the UPS Cable Pins
UPC port side
Pin
number
UPS cable side
Signal name
(Note)
1
ER
2
---
3
---
4
---
5
SG
6
*BPS/*UALM
SG
7
*BTL
8
---
*BTL
*ACOFF
*BPS/*UALM
9
*ACOFF
Note: Do not connect to the ER signal pin.
Remarks: If a UPS cable is required, it must be prepared separately. For details,
contact your sales representative.
Appendix D
UPS Controller
D-5
D.6
Connections
This section explains the connection between the UPC port and UPS.
D-6
■
UPC#0 connects to UPS#0, and UPC#1 connects to UPS#1.
■
Single power feed uses only UPC#0.
■
Dual power feed uses UPC#0 and UPC#1.
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
D.7
UPC Port
This chapter describes the location of the UPC port and the UPS connections.
■
UPC#0 connects with UPS#0. UPC#1 connects with UPS#1.
■
The single power feed uses UPC#0 only.
■
The dual power feed option uses UPC#0 and UPC#1.
FIGURE D-3 show the location of the UPC port in the main unit.
FIGURE D-3
UPC Port of SPARC Enterprise M8000/M9000 Server
UPC#1
UPC#0
SPARC Enterprise M8000 Server
(Front View)
SPARC Enterprise M9000 Server
(Front View)
Appendix D
UPS Controller
D-7
D-8
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
APPENDIX
E
XSCF Unit Replacement When XCP
1040 or 1041 Is in the Server
The server must be updated to XCP 1050 or later.
Note – XCP 1050 or later cannot be downgraded to XCP 1040 or XCP 1041.
1. Turn off all domains.
2. Turn the Mode switch to Service.
3. Turn off all main line switches.
4. Replace XSCFU#0 with spare material, and reconnect cables to XSCFU#0.
5. Extract XSCFU#1
6. Turn on all main line switches.
The following log may be recorded, but you can ignore it:
XSCF> showlogs error -r -M
Date: Oct 01 18:02:06 JST 2007
Status: Alarm
FRU: /XSCFU_B#1
Msg: XSCF_B shortage
Code: 80008000-55000000-010a101f00000000
Occurred: Oct 01 18:02:06.604 JST 2007
7. Set the date and time on the XSCFU.
8. Turn off all main line switches.
9. Reinsert the XSCFU#1 that you extracted in Step 5.
10. Turn off all of the server’s main line switches for 30 seconds.
11. After 30 seconds, turn the main line switches back on.
E-1
12. Wait until XSCF firmware reaches the ready state.
This can be confirmed when the READY LEDs of XSCFU_B#0 and XSCFU_B#1
remain lit.
13. Log in to XSCFU#0 using a serial connection or LAN connection.
14. Confirm no abnormality occurred by using showlogs error -v and showstatus
commands.
XSCF> showlogs error -v
XSCF> showstatus
If you encounter any hardware abnormality of the XSCF, contact service engineer.
15. Confirm and update the imported XCP image again.
XSCF> flashupdate -c update -m xcp -s 1050
Specify the XCP version to be updated. In this example, it is 1050. XSCFU#1 will be
updated, and then XSCFU#0 updated, again.
When the firmware update for XSCFU#0 is complete, XSCFU#1 is active.
16. Log in to XSCFU#1 using a serial connection or LAN connection.
17. Confirm completion of the update by using the showlogs event command.
XSCF> showlogs event
Confirm no abnormality is found during the update.
18. Confirm that both the current and reserve banks of XSCFU#0 display the
updated XCP versions.
XSCF> version -c xcp
XSCF#1 (Active )
XCP0 (Reserve): 1050
XCP1 (Current): 1050
XSCF#0 (Standby)
XCP0 (Reserve): 1050
XCP1 (Current): 1050
If the Current and Reserve banks on XSCF#0 do not indicate XCP revision 1050,
contact service engineer.
E-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
19. Confirm switching over between XSCFs works properly.
XSCF> switchscf -t Standby
The XSCF unit switch between the Active and Standby states.
Continue? [y|n] :y
a. When the READY LED on XSCFU_B#1 remains lit, log in to XSCFU#0 using
a serial connection or LAN connection.
b. Confirm switching over between XSCFs using the following commands:
XSCF> showhardconf
Confirm XSCFU#1 is now the standby, and that XSCFU#0 has become the active.
XSCF> showlogs error
Confirm no new errors have been recorded since the check in Step 14.
XSCF> showlogs event
Confirm a message "XSCFU entered active state from standby state".
XSCF> showstatus
Confirm a message “No failures found in System Initialization”.
20. Power on all domains.
XSCF> poweron -a
21. Log in to XSCFU#0 and confirm all domains start up properly.
XSCF> showlogs power
22. Check that there are no new errors.
XSCF> showlogs error
■
In case an abnormality is encountered, take appropriate maintenance action and
contact service engineer.
■
If no abnormality is found, proceed to Step 23.
Appendix E
XSCF Unit Replacement When XCP 1040 or 1041 Is in the Server
E-3
23. Turn the key switch on the operator panel from Service to Lock.
E-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
CHAPTER
Abbreviations
A
E
ACS
AC Section
ACSTPH
ACS Three-Phase
B
BP
Backplane
BUI
Browser User Interface
C
EMI
Electromagnetic Interference
F
FAN
FAN Unit
FRU
Field-Replaceable Unit
H
HDD
Hard Disk Drive
CMB
CPU Memory Board
CMU
CPU/Memory Board Unit
I
CLI
Command-Line Interface
IOB
I/O Board
CLKU
Clock Control Unit
IOU
I/O Unit
CPUM
CPU Module
IOUA
IOU Onboard Device Card_A
M
D
DAT
Digital Audio Tape
MAC
Memory Access Controller
DDC
DC to DC Converter
MEDBP
Media Backplane
DPF
Dual Power Feed
DR
Dynamic Reconfiguration
O
OBP
OpenBoot PROM
OPNL
Operator Panel
Abbreviations-1
P
PCICS
PCI Cassette
PCI-ES
PCI-Express Short
PFC
Power Factor Correction
PHP
PCI Hot Plug
POST
Power-On Self-Test
PSU
Power Supply Unit
R
RCI
Remote Cabinet Interface
RDPF
Rack-mountable Dual Power Feed
S
SAS
Serial Attached SCSI
SATA
Serial ATA (Advanced Technology
Attachment)
SC
System Controller
SNSU
Sensor Unit
SWBP
Switch Backplane
T
TAPEU
Tape Drive Unit
U
UPS
Uninterruptible Power Supply
X
XBU
Crossbar Unit
XSCF
eXtended System Control Facility
XSCFU
eXtended System Control Facility Unit
Abbreviations-2
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
Index
A
AC section, B-25
AC section replacement, 18-1
ACS
cold replacement, 18-5
active replacement, 4-2, 4-3
CD-RW/DVD-RW, 14-5
CMU, 6-12
DDC, 19-3
DVDU, 14-5
FAN, 8-8
HDD, 12-3
IOU, 7-8
PCI, 13-17
PSU, 9-8
TAPEU, 15-5
XSCFU, 11-5
addition and deletion, 24-1
addition and deletion of CMU, IOU, Fan, PSU, and
PCI cards, 25-1
addition and deletion of M9000 expansion
cabinet, 24-16
addition and deletion of power cabinet, 24-13
addition and deletion of rack mountable dual power
feed, 24-2
air filter (I/O unit), cleaning, 3-5
air filter (server), cleaning, 3-2
Antistatic Precautions, 1-6
B
backplane, B-35
backplane replacement, 20-1
C
CD-RW/DVD-RW drive unit, B-19
CD-RW/DVD-RW drive unit replacement, 14-1
certification standards, 2-12
checking
firmware configuration, 2-26
hardware configuration, 2-24
software configuration, 2-25
cleaning
air filter (I/O unit), 3-5
air filter (server), 3-2
tape drive unit, 3-1
CLKU
cold replacement, 16-4
clock control unit, B-16
clock control unit replacement, 16-1
CMU
active replacement, 6-12
cold replacement, 6-23
hot replacement, 6-12
cold replacement, 4-2, 4-17
ACS, 18-5
CD-RW/DVD-RW, 14-11
CLKU, 16-4
CMU, 6-23
DDC, 19-7
DVDU, 14-11
FAN, 8-14
fan BP, 20-1
HDD, 12-5
Index-1
IOU, 7-17
MEDBP, 22-5
OPNL, 10-4
PCI, 13-20
PSU, 9-12
PSU BP, 20-1
servers BPs, 20-1
SNSU, 21-4
SWBP, 23-5
TAPEU, 15-9, 15-11
XBU, 17-4
XSCFU, 11-12
commands for checking hardware configuration and
status, 2-24
commands for checking the messaging output, 2-30
commands for checking the monitoring output, 2-29
component, B-1
components to access and doors to open, 5-2
Conditions of memory mount, 6-40
confirming the hardware, 4-7, 4-15, 4-20
corresponding components and doors that can be
accessed, 5-2
CPU and DIMM replacement, 6-29
CPU memory board unit, B-4
CPU module, B-5
crossbar unit, B-14
crossbar unit replacement, 17-1
E
error conditions, 2-26
error display by LED, 2-30
EZ label, 2-12
F
FAN
active replacement, 8-8
cold replacement, 8-14
hot replacement, 8-8
FAN backplane, B-39
fan BP
cold replacement, 20-1
FAN unit, B-30
FAN unit replacement, 8-1
fmadm config command, 2-35, 2-36
fmadm faulty command, 2-35
fmadm repair Command, 2-36
fmdump command, 2-35
fmstat command, 2-37
FRU removal preparation, 4-1
FRU status, 2-23
FRU, disconnecting from a domain, 4-3
FRU, incorporating into a domain, 4-6
G
Grounding Port Connection Locations, 1-11
D
DDC, B-37
active replacement, 19-3
cold replacement, 19-7
DDC replacement, 19-1
devices, 2-38
diagnostic method flow chart, 2-22
diagnostics methods, 2-21
DIMM replacement, 6-44
disconnecting
CMU/IOU, 4-3
FRU from a domain, 4-3
PCI card, 4-4
disconnecting and replacing a FRU, 4-5
disconnecting off and replacing a FRU, 4-12
doors, how to open and close, 5-1
downloading the error log information, 2-26
Index-2
H
hard disk drive, B-10
hard disk drive (HDD) replacement, 12-1
hardware configuration, 2-23
HDD
active replacement, 12-3
cold replacement, 12-5
hot replacement, 4-2, 4-12
CMU, 6-12
FAN, 8-8
IOU, 7-8
PCI, 13-17
PSU, 9-8
how to open and close doors, 5-1
how to remove a door, 5-3
how to remove a side panel, 5-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010
I
I/O box downlink card, B-13
I/O unit, B-8
I/O unit (IOU) replacement, 7-1
ID label, 2-11
incorporating
CMU/IOU, 4-6
FRU into a domain, 4-6
internal components access, 5-1
iostat command, 2-38
IOU, B-12
active replacement, 7-8
cold replacement, 7-17
hot replacement, 7-8
K
key, 2-1, 2-16
L
labels about handling, 2-15
LED, 2-18
LED display, 2-31
LED error display, 2-30
M
M9000 expansion cabinet, 24-1
MEDBP
cold replacement, 22-5
media backplane, B-42
media backplane replacement, 22-1
memory, B-7
messaging output, 2-29
Method of Removing Static Electricity, 1-6
Removing Static Electricity on a CMU and an
IOU, 1-8
mode switch, 2-20
monitoring output, 2-28
N
netstat command, 2-43
O
operator panel, 2-16, B-21
operator panel location, 2-16
operator panel replacement, 10-1
OPNL
cold replacement, 10-4
overview
ACSs, 18-1
BP, 20-1
CLKU, 16-1
CMU, 6-1
DDC, 19-1
DVDU, 14-1
FAN, 8-2
HDD, 12-1
IOU, 7-2
MEDBP, 22-1
operator panel, 10-1
PCI slot devices, 13-1
PSU, 9-1
RDPF, 24-2
SNSU, 21-1
SWBPs, 23-1
TAPEU, 15-1
XBUs, 17-1
XSCFU, 11-1
P
PCI
active replacement, 13-17
cold replacement, 13-20
hot replacement, 13-17
PCI cassette, B-10
PCI slot device replacement, 13-1
periodic maintenance, 3-1
ping command, 2-44
power cabinet, 24-1, B-32
power supply unit, B-24
power supply unit (PSU) replacement, 9-1
POWER switch, 2-20
powering the server off, 4-18
powering the server on, 4-19
predictive self healing message, 2-28
predictive self-healing tools, 2-27
product overview, 2-1
production faceplate label, 2-11
prstat command, 2-47
prtconf command, 2-41
prtdiag command, 2-39
ps command, 2-46
Index-3
PSU
active replacement, 9-8
cold replacement, 9-12
hot replacement, 9-8
PSU backplane, B-38
PSU BP
cold replacement, 20-1
TAPEU
active replacement, 15-5
cold replacement, 15-9, 15-11
traditional Solaris troubleshooting commands, 2-37
troubleshooting, 2-1
troubleshooting commands, 2-34
types of Replacement Procedures, 4-2
R
U
rack-mountable dual power feed, B-34
rack-mountable dual power feed option, 24-1
rating label, 2-11
replacement of CPU/memory board unit (CMU),
CPU, and DIMM, 6-1
replacing a CPU module, 6-30
UPS controller (UPC)
Cable Connector, D-5
Configuration of Signal Lines, D-2
Connection, D-6
overview, D-1
Power supply conditions, D-4
Signal Cable, D-1
UPC Port, D-1
S
sensor unit, B-23
sensor unit replacement, 21-1
server and system configuration, 2-23
servers BPs
cold replacement, 20-1
showlogs command, 2-34
SNSU
cold replacement, 21-4
software and XSCF firmware configurations, 2-24
Solaris troubleshooting commands, traditional, 2-37
standard label, 2-11
state display, 2-18
status display, 2-32
SWBP
cold replacement, 23-5
switch backplane, B-43
switch backplane replacement, 23-1
switches, 2-20
switching off the main line switch, 4-34
system configuration, A-1
system name plate label, 2-11
system views, 2-1
X
XBU
cold replacement, 17-4
XSCF unit, B-17
XSCF unit replacement, 11-1
XSCFU
active replacement, 11-5
cold replacement, 11-12
required update, E-1
T
tape drive unit, B-20
tape drive unit replacement, 15-1
Tape drive unit, cleaning, 3-1
Index-4
SPARC Enterprise M8000/M9000 Servers Service Manual • June 2010

advertisement

Was this manual useful for you? Yes No
Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Related manuals

Download PDF

advertisement