Silicon Graphics Total Performance 9300 User Manual

Download

SGI® Total Performance 9300 RAID User’s Guide

007-4644-001

CONTRIBUTORS

Written by Matt Hoy

Illustrated by Kelly Begley, Matt Hoy, and Jim Ostrom

Production by Karen Jacobson

Engineering contributions by Bill Andrews, Marty Castilla, Bill Lusche, Robert Novak,

© 2003, Silicon Graphics, Inc. All rights reserved; provided portions may be copyright in third parties, as indicated elsewhere herein. No permission is granted to copy, distribute, or create derivative works from the contents of this electronic documentation in any manner, in whole or in part, without the prior written permission of Silicon Graphics, Inc.

LIMITED RIGHTS LEGEND

Use, duplication, or disclosure by the Government is subject to restrictions as set forth in the Rights in Data clause at FAR 52.227-14 and/or in similar orsuccessor clauses inthe FAR, orin the DOD,DOE or NASAFAR Supplements. Unpublishedrights reserved under the Copyright Laws of the United States. Contractor/manufacturer is Silicon Graphics, Inc., 1600 Amphitheatre Pkwy., Mountain View, CA 94043-1351.

TRADEMARKS AND ATTRIBUTIONS

Silicon Graphics, SGI, the SGI logo, IRIX, and Origin are registered trademarks, and Altix, CXFS, FailSafe, Octane2, and Silicon Graphics Fuel are trademarks, of Silicon Graphics, Inc., in the United States and.or other countries worldwide.

Brocade and Silkworm are registered trademarks of Brocade Communications Systems, Inc. QLogic is a trademark of QLogic Corporation.

Cover design by Sarah Bolles, Sarah Bolles Design, and Dany Galgani, SGI Technical Publications

Henry Ortiz, and Sammy Wilborn

Record of Revision

Version Description

001 May 2003

Original publication

007-4644-001 iii

Contents

Record of Revision . . . . . . . . . . . . . . . . . . . . . . iii

Figures . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Tables . . . . . . . . . . . . . . . . . . . . . . . . . . xi

About This Guide. . . . . . . . . . . . . . . . . . . . . . .xiii

Important Information . . . . . . . . . . . . . . . . . . . . .xiii

Chapter Descriptions . . . . . . . . . . . . . . . . . . . . . . xiv

Related Publications . . . . . . . . . . . . . . . . . . . . . . xv

Conventions . . . . . . . . . . . . . . . . . . . . . . . xvii

Product Support . . . . . . . . . . . . . . . . . . . . . . xvii

Reader Comments . . . . . . . . . . . . . . . . . . . . . . xviii

1. Introduction to the SGI TP9300 . . . . . . . . . . . . . . . . . . . 1

System Features . . . . . . . . . . . . . . . . . . . . . . . 1

Performance Features . . . . . . . . . . . . . . . . . . . . 2

RAID Controller Features . . . . . . . . . . . . . . . . . . . 3

Availability Features . . . . . . . . . . . . . . . . . . . . . 4

Supported Platforms . . . . . . . . . . . . . . . . . . . . . 4

Adapter and Switch Compatibility . . . . . . . . . . . . . . . . . 5

System Components . . . . . . . . . . . . . . . . . . . . . . 5

Storage Enclosures . . . . . . . . . . . . . . . . . . . . . 6

TP9300 Rack . . . . . . . . . . . . . . . . . . . . . . . 8

2. Storage Enclosures . . . . . . . . . . . . . . . . . . . . . . 11

Overview . . . . . . . . . . . . . . . . . . . . . . . . . 12

Disk Drives . . . . . . . . . . . . . . . . . . . . . . . . . 16

RAID Controllers . . . . . . . . . . . . . . . . . . . . . . . 17

Environmental Status Modules (ESMs) . . . . . . . . . . . . . . . . . 19

007-4644-001 v

Contents

Fans . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Power Supplies . . . . . . . . . . . . . . . . . . . . . . . . 21

Tray ID Switch . . . . . . . . . . . . . . . . . . . . . . . . 23

Link Rate Switch . . . . . . . . . . . . . . . . . . . . . . . 24

SFP Transceivers . . . . . . . . . . . . . . . . . . . . . . . 25

3. Operating Storage Enclosures . . . . . . . . . . . . . . . . . . . 27

Turning the Power On . . . . . . . . . . . . . . . . . . . . . . 27

Turning the Power On after an Unexpected Shutdown . . . . . . . . . . . . 29

Troubleshooting an Overtemp Condition . . . . . . . . . . . . . . . 29

Turning the Power On after an Overtemp Shutdown . . . . . . . . . . . . 30

Turning the Power On after an Emergency Shutdown . . . . . . . . . . . 31

Turning the Power Off . . . . . . . . . . . . . . . . . . . . . . 32

Monitoring Status with Software . . . . . . . . . . . . . . . . . . . 33

Checking the Indicator Lights. . . . . . . . . . . . . . . . . . . . 34

Front Indicator Lights . . . . . . . . . . . . . . . . . . . . . 34

Rear Indicator Lights . . . . . . . . . . . . . . . . . . . . . 36

Moving the Enclosure . . . . . . . . . . . . . . . . . . . . . . 40

Removing the Enclosure . . . . . . . . . . . . . . . . . . . . 40

Reinstalling the Enclosure . . . . . . . . . . . . . . . . . . . 43

4. Replacing Enclosure Components . . . . . . . . . . . . . . . . . . 45

Replacing a Failed Disk Drive. . . . . . . . . . . . . . . . . . . . 45

Replacing a Failed Fan . . . . . . . . . . . . . . . . . . . . . . 48

Replacing a Failed Power Supply Module . . . . . . . . . . . . . . . . 50

Replacing a Failed SFP Transceiver . . . . . . . . . . . . . . . . . . 52

Replacing a Failed RAID Controller Module . . . . . . . . . . . . . . . 54

Replacing a Failed Environmental Status Module . . . . . . . . . . . . . . 56

Upgrading Drives . . . . . . . . . . . . . . . . . . . . . . . 59

Adding a Drive to an Empty Slot . . . . . . . . . . . . . . . . . 59

Adding Drives of Larger Capacity . . . . . . . . . . . . . . . . . 61

Adding Expansion Enclosures . . . . . . . . . . . . . . . . . . . 63

5. Cabling . . . . . . . . . . . . . . . . . . . . . . . . . . 65

Connecting the Drive Loop Cables . . . . . . . . . . . . . . . . . . 65

vi 007-4644-001

Contents

Connecting the System to the Hosts . . . . . . . . . . . . . . . . . . 69

Examples of Direct Host Connection Cabling . . . . . . . . . . . . . . 69

Examples of Fabric/Switch Host Connection Cabling . . . . . . . . . . . 72

Connecting Ethernet and Serial Cables for Array Management. . . . . . . . . 73

Powering On the System . . . . . . . . . . . . . . . . . . . . . 75

A. Specifications and Requirements . . . . . . . . . . . . . . . . . . 79

Rack Specifications . . . . . . . . . . . . . . . . . . . . . . 79

Dimensions . . . . . . . . . . . . . . . . . . . . . . . 81

Weight . . . . . . . . . . . . . . . . . . . . . . . . . 82

Area Requirements . . . . . . . . . . . . . . . . . . . . . 84

Wiring and Power . . . . . . . . . . . . . . . . . . . . . 85

Environmental . . . . . . . . . . . . . . . . . . . . . . 90

Enclosure Specifications . . . . . . . . . . . . . . . . . . . . . 91

Dimensions . . . . . . . . . . . . . . . . . . . . . . . 91

Weights . . . . . . . . . . . . . . . . . . . . . . . . 92

Wiring and Power . . . . . . . . . . . . . . . . . . . . . 93

Environmental . . . . . . . . . . . . . . . . . . . . . . 94

Air Flow . . . . . . . . . . . . . . . . . . . . . . . . 95

B. SGI Field Engineering Compliance Statements . . . . . . . . . . . . . . 97

Electromagnetic Emissions . . . . . . . . . . . . . . . . . . . . 97

Radio and Television Interference . . . . . . . . . . . . . . . . . . 97

Product Safety . . . . . . . . . . . . . . . . . . . . . . . . 98

Electrostatic Discharge . . . . . . . . . . . . . . . . . . . . . 98

Shielded Cables . . . . . . . . . . . . . . . . . . . . . . . 99

FCC Warning . . . . . . . . . . . . . . . . . . . . . . . . 99

VDE 0871/6.78 . . . . . . . . . . . . . . . . . . . . . . . .100

European Union Statement . . . . . . . . . . . . . . . . . . . .100

International Special Committee on Radio Interference (CISPR). . . . . . . . . .101

Canadian Department of Communications Statement . . . . . . . . . . . .101

Attention . . . . . . . . . . . . . . . . . . . . . . . . .101

Japanese Class A Compliance Statement . . . . . . . . . . . . . . . .101

007-4644-001 vii

Figures

Figure 1-1 TP9300 Tower and Rackmount Storage Enclosures (Front View). . 6 Figure 1-2 TP9300 Controller and Expansion Enclosures (Rear View) . . . 7

Figure 1-3 TP9300 Rack with Enclosures Installed . . . . . . . . . 9

Figure 1-4 Rack Features . . . . . . . . . . . . . . . . 10

Figure 2-1 Storage Enclosure Components (Front View) . . . . . . . 13

Figure 2-2 Storage Enclosure Components (Rear View) . . . . . . . 15

Figure 2-3 Disk Drives and Drive Bay Numbering . . . . . . . . . 16

Figure 2-4 2882 RAID Controller . . . . . . . . . . . . . . 18

Figure 2-5 Environmental Status Modules . . . . . . . . . . . 19

Figure 2-6 Fan Units . . . . . . . . . . . . . . . . . 20

Figure 2-7 Power Supplies . . . . . . . . . . . . . . . . 22

Figure 2-8 Tray ID Switch . . . . . . . . . . . . . . . . 23

Figure 2-9 Link Rate Switch . . . . . . . . . . . . . . . 24

Figure 2-10 SFP Transceiver. . . . . . . . . . . . . . . . 25

Figure 3-1 Turning the Power On and Off . . . . . . . . . . . 28

Figure 3-2 Storage Enclosure Front Indicator Lights . . . . . . . . 35

Figure 3-3 Storage Enclosure Rear Indicator Lights . . . . . . . . . 37

Figure 3-4 Removing the Side Brackets from the Enclosure . . . . . . 41

Figure 3-5 Removing the Empty Drive Enclosure . . . . . . . . . 42

Figure 4-1 Removing and Installing a Drive Module . . . . . . . . 46

Figure 4-2 Removing and Installing a Fan Module . . . . . . . . . 49

Figure 4-3 Fault Lights for Fan Module Replacement . . . . . . . . 49

Figure 4-4 Removing and Installing a Power Supply Module . . . . . . 51

Figure 4-5 Removing and Installing an SFP Transceiver . . . . . . . 53

Figure 4-6 Removing and Installing a RAID Controller Module . . . . . 55

Figure 4-7 Removing SFPs from the Environmental Status Module . . . . 57

Figure 4-8 Removing and Installing an Environmental Status Module . . . 58

Figure 4-9 Adding New Expansion Enclosures to an Existing Array . . . . 64

007-4644-001 ix

Figures

Figure 5-1 Drive Loop Cabling for Systems with Two Enclosures . . . . . 66

Figure 5-2 Drive Loop Cabling for Systems with Three or More Enclosures . . 67

Figure 5-3 Drive Loop Cabling for Systems with Eight Enclosures . . . . 68

Figure 5-4 Direct Host Connection with Failover . . . . . . . . . 70

Figure 5-5 Direct Host Connection without Failover . . . . . . . . 71

Figure 5-6 Connecting to Hosts via a Fibre Channel Switch . . . . . . 72

Figure 5-7 Connecting Ethernet Cables . . . . . . . . . . . . 74

Figure 5-8 Power Connections and Switches of the Enclosures . . . . . 75

Figure 5-9 PDU Cabling. . . . . . . . . . . . . . . . . 76

Figure A-1 Populated Rack . . . . . . . . . . . . . . . . 80

Figure A-2 Dimensions of the System Rack . . . . . . . . . . . 81

Figure A-3 Rack Weights . . . . . . . . . . . . . . . . 83

Figure A-4 System Area Requirements . . . . . . . . . . . . 84

Figure A-5 Power Connections to Controller and Expansion Enclosures . . . 87

Figure A-6 AC Power Connectors and Receptacles . . . . . . . . . 89

Figure A-7 Dimensions of the System Enclosure . . . . . . . . . . 91

Figure A-8 Air Flow in the System Enclosure . . . . . . . . . . . 95

x 007-4644-001

Tables

Table 3-1 Storage Enclosure Front Indicator Lights . . . . . . . . 36

Table 3-2 Storage Enclosure Rear Indicator Lights . . . . . . . . . 38

Table A-1 Rack, Crate, and Enclosure Weights . . . . . . . . . . 82

Table A-2 AC Power Requirements (Domestic and International) . . . . 86

Table A-3 Environmental Requirements . . . . . . . . . . . . 90

Table A-4 Enclosure Weights . . . . . . . . . . . . . . . 92

Table A-5 Enclosure Module Weights . . . . . . . . . . . . 92

Table A-6 Site Wiring Voltages for System Enclosures

(Single-phase Line-to-neutral) . . . . . . . . . . . 93

Table A-7 Altitude Requirements for System Enclosures . . . . . . . 94

Table A-8 Environmental Requirements for System Enclosures . . . . . 94

007-4644-001 xi

About This Guide

This guide provides an overview of the SGI Total Performance 9300 (SGI TP9300). It covers routine operation and replacement procedures and provides troubleshooting and reference information for all customer-replaceable components.

The SGI TP9300 is a highly scalable RAID storage system designed for continuous availability. Some of the features of the system include hot-swappable components, easy expansion, and redundant power and cooling systems.

This guide is intended for system operators and service technicians who have extensive knowledge of Fibre Channel network technology and computer system operation, maintenance, and repair.

Use this guide to learn about all of the following:

• The parts of your system: the rack, controller enclosures, and expansion enclosures

• Operating your system correctly

• Replacing failed components

Important Information

Danger: Never look into the end of a ﬁber-optic cable to conﬁrm that light is being

emitted (or for any other reason). Most ﬁber-optic laser wavelengths (1300 nmand 1550nm) are invisible to the eye and cause permanent eye damage. Shorterwavelength lasers (for example, 780 nm) are visible and can cause signiﬁcant eye damage. Use only an optical power meter to verify light output.

007-4644-001 xiii

About This Guide

Danger: Never look into the end of a ﬁber-optic cable on a powered device with any

type of magnifying device, such as a microscope, eye loupe, or magnifying glass. Such activity causes a permanent burn on the retina of the eye. Optical signals cannot be veriﬁed by looking into the ﬁber end.

Chapter Descriptions

This guide contains the following chapters:

• Chapter 1, “Introduction to the SGI TP9300,” describes the features and components

• Chapter 2, “Storage Enclosures,” gives a detailed overview of the system enclosure

• Chapter 3, “Operating Storage Enclosures,” describes the functions and basic

of the system.

and its components.

operation of the system enclosures, including power on and troubleshooting with the indicator lights.

• Chapter 4, “Replacing Enclosure Components,” gives detailed procedures for replacing failed components.

• Chapter 5, “Cabling,” describes cabling between controller and expansion enclosures and between controllers and hosts.

• Appendix A, “Speciﬁcations and Requirements,” describes component speciﬁcations and requirements for purposes of installation and maintenance.

• Appendix B, “SGI Field Engineering Compliance Statements,” describes the regulatory and compliance information for the system.

xiv 007-4644-001

Related Publications

This guide is part of a document set that fully supports the installation, operation, and service of the TP9300. See the following documents for more information about your system. If a document number ends in “X,” use the latest available version of that document.

• SGI TP9300 RAID Installation and Upgrade Guide (108-0401-00x)

• SGI TP9400 and TP9500 RAID IRIX Administration Guide(007-4306-00x)

• SGI TP9400 and SGI TP9500 Software Concepts Guide (007-4305-00x)

• SGI Storage Area Network Installation Instructions (108-0252-00x)

About This Guide

This guide gives complete instructions on how to unpack, install, and conﬁgure the SGI TP9300 and its components. It also contains upgrade information.

This guide gives complete instructions on how to install the TPSSM software for host and/or client operation.

This guide explains the terminology and features of the TPSSM storage management software.

This guide is included with the Fibre Channel switch and provides information on storage area network installation and topologies.

You can obtain SGI documentation, release notes, or man pages in the following ways:

• See the SGI Technical Publications Library at http://docs.sgi.com. Various formats

are available. This library contains the most recent and most comprehensive set of online books, release notes, man pages, and other information.

• If it is installed on your SGI system, you can use InfoSearch, an online tool that

provides a more limited set of online books, release notes, and man pages. With an IRIX system, select Help from the Toolchest, and then select InfoSearch. Or you can type infosearch on a command line.

• You can also view release notes by typing either grelnotes or relnotes on a

command line.

• You can also view man pages by typing man <title> on a command line.

007-4644-001 xv

About This Guide

SGI systems include a set of IRIX man pages, formatted in the standard UNIX “man page” style. These are found online on the internal system disk (or CD-ROM) and are displayed using the man command. For example, to display the man page for the Add_disk command, type the following on a command line:

man Add_disk

Important system conﬁguration ﬁles and commands are documented on man pages. References in the documentation to these pages include the name of the command and the section number in which the command is found. For example, “Add_disk(1)” refers to the Add_disk command and indicates that it is found in section 1 of the IRIX reference.

For additional information about displaying reference pages using the man command, see man(1).

In addition, the apropos command locatesman pages based on keywords. For example, to display a list of man pages that describe disks, type the following on a command line:

apropos disk

For information about setting up and using apropos, see apropos(1) and makewhatis(1M).

xvi 007-4644-001

Conventions

About This Guide

The following conventions are used throughout this document:

Convention Meaning

Command This ﬁxed-space font denotes literal items such as commands, ﬁles,

routines, path names,signals, messages, and programming language structures.

variable The italic typeface denotes variable entries and words or concepts

being deﬁned. Italic typeface also is used for book titles.

user input This ﬁxed-space font denotes literal items that the user enters in

interactive sessions. Output is shown in nonbold, ﬁxed-space font.

[ ] Brackets enclose optional portions of a command or directive line.

... Ellipses indicate that a preceding element can be repeated. man page(x) Man page section identiﬁers appear in parentheses after man page

names.

GUI element This font denotes the names of graphical user interface (GUI)

elements such as windows, screens, dialog boxes, menus, toolbars, icons, buttons, boxes, ﬁelds, and lists.

Product Support

SGI provides a comprehensive product support and maintenance program for its products:

• If you are in North America, contact the Technical Assistance Center at

+1 800 800 4SGI or contact your authorized service provider.

• If you are outside North America, contact the SGI subsidiary or authorized

distributor in your country.

007-4644-001 xvii

About This Guide

Reader Comments

If you have comments about the technical accuracy, content, or organization of this document, contact SGI. Be sure to include the title and document number of the manual with your comments. (Online, the document number is located in the front matter of the manual. In printed manuals, the document number is located at the bottom of each page.)

You can contact SGI in any of the following ways:

• Send e-mail to the following address: techpubs@sgi.com

• Use the Feedback option on the Technical Publications Library website: http://docs.sgi.com

• Contact your customer service representative and ask that an incident be ﬁled in the SGI incident tracking system.

• Send mail to the following address:

Technical Publications SGI 1600 Amphitheatre Pkwy., M/S 535 Mountain View, California 94043-1351

• Send a fax to the attention of “Technical Publications” at +1 650 932 0801.

SGI values your comments and will respond to them promptly.

xviii 007-4644-001

1. Introduction to the SGI TP9300

System Features

Chapter 1

The SGI Total Performance 9300 is a high-bandwidth RAID Fibre Channel storage system. This chapter gives a brief overview of the features of the TP9300 and its primary components in the following sections:

• “System Features” on page 1

• “System Components” on page 5

The features of the TP9300 are listed in the following sections:

• “Performance Features” on page 2

• “RAID Controller Features” on page 3

• “Availability Features” on page 4

• “Supported Platforms” on page 4

• “Adapter and Switch Compatibility” on page 5

007-4644-001 1

1: Introduction to the SGI TP9300

Performance Features

The SGI TP9300 RAID storage system has the following basic features:

• Outstanding performance, built on multi-channel end-to-end Fibre Channel

• Continuous availability, with constant monitoring and optional redundancy of all

• Dynamic scalability, making it easy to grow all subsystem resources without

• Superior connectivity, allowing simultaneous connections to multiple servers

• Vast storage capacity with support for large numbers of drives.

• Storage management facilities for installation, conﬁguration, expansion, and

• Redundant power supplies and hot-swappable components.

technology.

active components.

disruption.

directly or by way of storage area networks (SANs). Support for optical host connections.

monitoring.

• Easily upgradeable to meet a variety of performance requirements.

• Integrated RAID controllers.

• Battery backup for cache data.

Note: JBOD is not supported.

2 007-4644-001

RAID Controller Features

The RAID controllers in the TP9300 have the following features:

• 112-drive maximum conﬁguration

• 4 RAID levels (0, 1, 3, and 5)

• 2 Gbit/s front end (FE) and back end (BE) Fibre Channel arbitrated loop (FC-AL)

• Immediate LUN availability (ILA)

• Transparent disk drive rebuilds

• Variable stripe size per controller (16K, 32K, and 64K)

• Mirrored cache

• Drive roaming during power off

• Cache coherency

• Transparent failover and failback

• Automatic error recovery

System Features

• Write through, write back, or read ahead support

• Automatic detection of failed drives

• Automatic drive rebuilds, using “hot spare” drive

• Hot-swappable drives

• SAN mapping server to LUN mapping

• Automatic ﬁrmware ﬂashing: In a dual controller conﬁguration, the ﬁrmware of the

replacement controller is automatically ﬂashed to match the ﬁrmware of the surviving controller.

007-4644-001 3

1: Introduction to the SGI TP9300

Availability Features

The TP9300 has the following availability features:

• Dual power feeds with dual power supplies

• Redundant cooling

• Battery back-up (BBU) maintains cache in case of power failure

• IRIX path failover

• Dynamic hot-sparing

• Non-disruptive component replacement

• Enclosure services interface (ESI) for SCSI enclosure services (SES)

Supported Platforms

The TP9300 supports the following hardware and software platforms:

• Software: IRIX, CXFS, FailSafe

• Hardware: SGI Altix 3000, SGI Origin 200, Origin 300, Onyx 300, Origin 2000, Origin 3000 series, and Onyx 3000 series servers.

4 007-4644-001

Adapter and Switch Compatibility

The TP9300 supports the following host bus adapters (HBAs) and switches:

Note: The TP9300 does not support copper Fibre Channel HBAs.

• QLogic 2200 optical 33/66-MHz HBA

• QLogic 2310 optical 66-MHz HBA

• QLogic 2342 optical 66-MHz dual channel HBA

• Brocade family SAN switches

• SilkWorm 2400 8-port switch

• SilkWorm 2800 16-port switch

• SilkWorm 3200 2 Gbit/s 8-port switch

• SilkWorm 3800 2 Gbit/s 16-port switch

• IRIX release level 6.5.16 or later

System Components

This section contains pictures and brief descriptions of the components that make up the TP9300 in the following sections:

• “Storage Enclosures” on page 6

• “TP9300 Rack” on page 8

007-4644-001 5

1: Introduction to the SGI TP9300

Storage Enclosures

The TP9300 is comprised of storage enclosures. These enclosures provide all of the logic, power, and I/O functions of the system. Each enclosure can be operated as an independent storage system, or multiple enclosures can be cabled together to create much larger disk arrays.

There are two types of storage enclosures: controller enclosures and expansion enclosures. The only difference between the types of enclosures is that the controller enclosure contains two integrated RAID controllers. The expansion enclosures do not contain RAID controllers and must be cabled to controller enclosures. Figure 1-1 shows a front view of both a deskside tower and a rackmount controller enclosure.

Note: Controller enclosures and expansion enclosures look identical from the front.

Front bezel

Drives (14 total)

Front bezel

Figure 1-1 TP9300 Tower and Rackmount Storage Enclosures (Front View)

6 007-4644-001

Drives (14 total)

System Components

Figure 1-2 shows a rear view of both a controller enclosure and an expansion enclosure.

2882 RAID controllers

FanFan

Controller enclosure

Power supply

Enviornmental status modules

Fan

Expansion enclosure

Power supply Power supply

Power supply

Figure 1-2 TP9300 Controller and Expansion Enclosures (Rear View)

See Chapter 2, “Storage Enclosures” for more information about the controller and expansion enclosures and their components.

Fan

007-4644-001 7

1: Introduction to the SGI TP9300

TP9300 Rack

The SGI TP9300 rack has the following features:

• 72-in. high x 22-in. wide x 36-in. deep.

• 38 rack units (1 rack unit = 1.75 in.).

• Removable rear panel.

• Heavy-duty casters for ease of movement.

• Dual power distribution units (PDUs).

• Optional stabilizer foot to reduce tipping.

• Convenient access to power and data cables through openings in the top and

Figure 1-3 shows a TP9300 rack with controller and expansion enclosures installed in it.

bottom of the rack.

8 007-4644-001

Controller enclosure

Expansion enclosures

System Components

Figure 1-3 TP9300 Rack with Enclosures Installed

007-4644-001 9

1: Introduction to the SGI TP9300

Figure 1-4 shows some of the features of the rack.

Standard EIA rails

Cable access holes

Casters

Removable stabilizer foot

Figure 1-4 Rack Features

For detailed speciﬁcations of the rack, see “Rack Speciﬁcations” on page 79.

10 007-4644-001

Chapter 2

2. Storage Enclosures

The storage enclosure is the basic building block of the TP9300. Each storage enclosure provides redundant power and cooling to its individual components. The storage enclosures also house the I/O components of the storage system.

There are two types of storage enclosures: controller enclosures and expansion enclosures. As their name implies, controller enclosures contain the RAID controllers for the storage system. The expansion enclosures provide expansion disk drive space. The two types of storage enclosures share many of the same components. The only difference between controller enclosures and expansion enclosures is the presence or absence of the RAID controllers.

This chapter describes the components in both the controller and expansion enclosures in the following sections:

• “Overview” on page 12

• “Disk Drives” on page 16

• “RAID Controllers” on page 17

• “Environmental Status Modules (ESMs)” on page 19

• “Fans” on page 20

• “Power Supplies” on page 21

• “Tray ID Switch” on page 23

• “Link Rate Switch” on page 24

• “SFP Transceivers” on page 25

007-4644-001 11

2: Storage Enclosures

Overview

Important: Several references are made in this chapter to the storage management software (TPSSM). For complete information on the operation and use of this software, see SGI TP9400 and SGI TP9500 Software Concepts Guide (007-4305-00x).

Figure 2-1 shows a front view of the storage enclosure. The front of the enclosure contains the following components:

• Front bezel: The bezel is a removable decorative cover with holes for viewing the status LEDs and openings for the drive bays.

• Disk drives: Up to 14 sled-mounted drives can be installed in the bays on the front of the enclosure.

• LEDs: There are two LEDs for each of the 14 drive bays that indicate the status of the drive: the drive active LED and the drive fault LED. There are also two LEDs that indicate the status of the entire enclosure: the enclosure fault LED and the enclosure power LED.

• Alarm mute button: This button silences the audible alarm of the controller enclosure. The expansion enclosure does not have an alarm; therefore, the expansion enclosure does not use this button.

Note: Controller enclosures and expansion controllers look identical when viewed from the front.

12 007-4644-001

Drive status LEDs

Front bezel

Disk drive (14 total)

Enclosure power LED

Figure 2-1 Storage Enclosure Components (Front View)

Overview

Enclosure fault LED

Alarm mute

007-4644-001 13

2: Storage Enclosures

Figure 2-2 shows a rear view of both a controller enclosure and an expansion enclosure. The rear of the enclosures contain the following components:

• RAID controllers: Controller enclosures contain two 2882 RAID controllers.

• Environmental status modules (ESMs): Expansion enclosures have two ESM modules in place of the RAID controllers.

• Fans: Two fans provide redundant cooling for the enclosure.

• Link rate/ enclosure ID switches: These switches indicate the connection speed and set the ID number of the enclosure.

• Power supplies: Two power supplies provide redundant power for the enclosure.

14 007-4644-001

Controller enclosure

Overview

2882 RAID controllers

FanFan

Power supply

Link rate/ enclosure ID switches

Enviornmental Status Modules

Fan

Expansion enclosure

Power supply

Link rate/ enclosure ID switches

Figure 2-2 Storage Enclosure Components (Rear View)

Power supply

Fan

Power supply

007-4644-001 15

2: Storage Enclosures

Disk Drives

Each storage enclosure can contain up to 14 low-proﬁle Fibre Channel disk drives. Controller enclosures can be cabled to as many as seven expansion enclosures for a total capacity of 112 disks in one storage system. Each disk is mounted in a sled for ease of installation and removal. These drive sleds have a lever and latch mechanism that allows you to cam the drive in and out of the enclosure. Figure 2-3 shows the drive sled and drive bay numbering scheme for the TP9300.

Each drive has two status LEDs (see Figure 2-3) as follows:

• Drive active LED: This LED illuminates green when a drive is in the drive bay, and it blinks when the drive is active.

• Drive fault LED: This LED illuminates amber when there is a problem with the drive.

Locking lever

Drive active LED

Drive fault LED

Drives (14 total)

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Drive sled latch

Figure 2-3 Disk Drives and Drive Bay Numbering

16 007-4644-001

RAID Controllers

Each TP9300 controller enclosure contains two 2882 RAID controller modules. These controllers slide into the same opening that the ESMs occupy on the expansion enclosures. Each controller contains RAID logic, cache memory, a rechargeable battery for the cache, and additionalenvironmental monitoring circuitry. Each controller also has two labels on its rear panel: one shows the MAC address of the controller and one shows the battery service date (see Figure 2-4).

The RAID controllers provide the following features and capabilities:

• Support for RAID levels 0, 1, 3, and 5

• Dual-porting for failover

• Support for 2 hosts per controller (4 per controller module)

• 256 MB of cache memory with 7-day battery backup

• Array sizes up to 112 drives

• Data transfer rates up to 200 MB/s

The RAID controller has the following connectors (see Figure 2-4):

• 2 Fibre Channel host ports with small form-factor pluggable transceivers (SFPs)

• 1 Ethernet host port

• 1 Fibre Channel drive expansion port

• 1 RS-232 diagnostic port

The RAID controller has the following LED indicators (see Figure 2-4):

• Host 1 link and speed LEDs

• Host 2 link and speed LEDs

• Battery charging/charged LED

• Cache active LED

• Fault LED

• Expansion loop LED

• Expansion loop bypass LED

007-4644-001 17

2: Storage Enclosures

For more information about these LEDs, see “Checking the Indicator Lights” on page 34.

2882 controller

Latch

Lever

Pull ring

MAC address label

Battery label

Host 2 connector

Host 1 connector

Host 1 link

Host 1 speed

Host 2 speed

Host 1 link

Figure 2-4 2882 RAID Controller

Ethernet connector

RS-232 Serial port

Fault

Cache active

Battery charging/charged

Expansion loop connector

Expansion loop bypass

Expansion loop

18 007-4644-001

Environmental Status Modules (ESMs)

The expansion enclosures contain environmental status modules instead of 2882 RAID controllers (see Figure 2-5). Each environmental status module contains an environmental services monitor board and two slots for SFPs (small form-factor pluggables). The environmental services monitor board is the interface between the controller enclosure and the expansion enclosure, and it monitors expansion enclosure status. If there are internal problems in the expansion enclosure, the environmental services monitor board lights the appropriate fault LED. For more information about this LED, see “Checking the Indicator Lights” on page 34.

ESM ESM

Environmental Status Modules (ESMs)

Figure 2-5 Environmental Status Modules

007-4644-001 19

2: Storage Enclosures

Fans

The storage enclosures each have two fan units. These fan units are mounted in the rear of the enclosure (see Figure 2-6). They pull air through the front bezel to cool the internal components of the enclosure. The air is exhausted out the rear of the fan units. The fans provide redundant cooling of the enclosure. If one fan unit fails, the other fan unit runs at a higher speed to compensate for the loss. Each fan unit has a fault light that illuminates if there is a problem with the units.

Vent

Pull knob

Fault light

Figure 2-6 Fan Units

Fans

Latch

20 007-4644-001

Power Supplies

Each storage enclosure contains two power supplies. These power supplies provide power to the other components in the enclosure by converting incoming AC voltage to DC voltage. The power supplies are redundant: if one supply fails, the other power supply can keep the enclosure running until the failed unit can be replaced. Each power supply has the following components on its rear panel (see Figure 2-7):

• Locking lever

• AC power input connector

• Power switch

• Power light

• Fault light

During normal operation, the power light is illuminated. If there is a problem with the power supply, the fault light comes on.

For speciﬁcation information about the power supplies, see “Wiring and Power” on page 93.

007-4644-001 21

2: Storage Enclosures

Power supplies

Lever

AC power connector

Power switch

Power light

Fault light

Figure 2-7 Power Supplies

22 007-4644-001

Tray ID Switch

The trayID switch is located between the power supplies on the rear of the enclosure (see Figure 2-8). This switch assigns a unique identiﬁer to every enclosure in the system. The controller enclosure should always have its tray ID set to “00.” The ﬁrst expansion enclosure should be set to “01.” Each subsequent expansion enclosure should be assigned the next highest ID. If more than one enclosure is set to the same ID number, the conﬂict LED illuminates.

Note: Each enclosure in a storage system must have a unique ID. If you are installing multiple systems in one rack, there maybe enclosures with the same ID numbersbecause the systems are independent.

Enclosure ID switch (Tray number)

Conflict LED

Figure 2-8 Tray ID Switch

007-4644-001 23

2: Storage Enclosures

Link Rate Switch

The link rate switch is located on the same panel as the enclosure ID switch, between the power supplies on the rear of the enclosure (see Figure 2-9). The link rate switch controls the speed of the Fibre Channel connections to the enclosure. This switch should always be set to 2 Gbit/s. When there is a 2 Gbit/s link present, the 2-Gbit link rate LED illuminates green.

Link rate switch

2-Gbit link rate LED

Figure 2-9 Link Rate Switch

Note: The link rate switches on all enclosures in a system should be 2 Gbit/s.

24 007-4644-001

SFP Transceivers

An SFP (small form-factor pluggable) transceiver is a module that ﬁts into the interface connectors on the RAID controllers and the ESM modules (see Figure 2-10). Every Fibre Channel cable that connects to a RAID controller or ESM module requires an SFP transceiver.

SFP tranceiver Fiber-optic cable

Figure 2-10 SFP Transceiver

You can hot-swap a failed SFP transceiver, which means you can replace it while the controller enclosure is in operation. If you replace the SFP transceiver and continue to experience problems, the enclosure may have defective components or connections. Check the storage management software (TPSSM) for indications of other component failures.

007-4644-001 25

Chapter 3

3. Operating Storage Enclosures

This chapter describes the operation of the storage enclosures in the following sections:

• “Turning the Power On” on page 27

• “Turning the Power On after an Unexpected Shutdown” on page 29

• “Turning the Power Off” on page 32

• “Monitoring Status with Software” on page 33

• “Checking the Indicator Lights” on page 34

• “Moving the Enclosure” on page 40

Note: All of the procedures in this chapter apply to both the controller enclosures and the expansion enclosures.

Turning the Power On

If the storage enclosure was turned off due to an unexpected shutdown or a power failure, follow the instructions in “Turning the Power On after an Unexpected Shutdown” on page 29.

Danger: Severe electrical shock can occur. Never turn on the power to any

equipment when there is evidence of ﬁre, water, or structural damage. If there is evidence of damage, call the factory or appropriate service organization for assistance. Depending on the current service agreements, you may need to send the unit back to the factory for repair/replacement.

007-4644-001 27

3: Operating Storage Enclosures

To turn on the enclosure after a normal shutdown (as described in “Turning the Power Off” on page 32), turn on both power switches on the rear of the enclosure (see Figure 3-1) or the main circuit breaker, whichever is applicable. You must turn on both power supply switches to take advantage of the redundant power supplies.

Note: Always wait at least 30 seconds between the time you turn a power switch off and the time you turn it back on again.

Note: To speed drive spin-up, it is recommended that you start the expansion enclosures before or at the same time as the controller enclosure.

Power supply switches

Figure 3-1 Turning the Power On and Off

Note: The activity and fault indicators above the drive bays may ﬂash intermittently as the drives spin up. Wait until the enclosure has ﬁnished powering on before checking the indicator lights on the front of the enclosure. All indicators should be green. If they are not, use the storage management software (TPSSM) to diagnose the problem.

28 007-4644-001

Turning the Power On after an Unexpected Shutdown

The enclosure (and the entire storage array) may shut down unexpectedly under the following conditions:

• If the internal temperature of the enclosure exceeds the maximum operating

temperature (an overtemp condition). See the “Troubleshooting an Overtemp Condition” section for more information on overtemp conditions.

• If there is a general power failure or a loss of power to the storage array.

• If you are forced to shut down the storage array without performing the normal

shutdown procedures (in “Turning the Power Off” on page 32) due to an emergency situation.

Note that in each of these cases, some data may be lost.

Warning: To avoid damage to the hardware, take special care when restarting the

enclosure after an unexpected shutdown.

If the enclosure shuts down unexpectedly, but there is still power to the site, use TPSSM to determine if the enclosure has overheated.

• If an overtemp condition is indicated, follow the steps in “Turning the Power On

after an Overtemp Shutdown” on page 30 to restart the drive enclosure.

• If the enclosure has shut down due to a power failure or an emergency shutdown,

follow the steps in “Turning the Power On after an Emergency Shutdown” on page 31 to restart the enclosure.

Troubleshooting an Overtemp Condition

If both fan units fail or are unable to maintain an internal temperature below 70ºC (158

F), one or both of the power supplies in the enclosure will shut down. If both power

supplies shut down, the enclosure is inoperable.

TPSSM warns you if the temperature of the enclosure is rising (before it has risen sufﬁciently to shut down the power supplies). The ﬁrst warning comes when the enclosure temperature exceeds 40ºC (104ºF). The enclosure shuts down if the temperature rises above 70ºC (158ºF).

007-4644-001 29

3: Operating Storage Enclosures

In the expansion enclosures, the overtemp indicator on the environmental status module illuminates if the temperature reaches 40ºC (104ºF). If both power supplies shut down, the overtemp indicator cannot come on. See Figure 3-3 on page 37 for the location of the overtemp indicators on expansion enclosures.

The controller enclosure does not have an overtemp indicator. Check the storage management software for temperature information about the controller enclosures.

Turning the Power On after an Overtemp Shutdown

To restart the enclosure after an unexpected shutdown due to an overtemp condition, follow these steps:

1. Turn off both power switches (see Figure 3-1 on page 28).

2. Do whatever is necessary to cool the enclosure (replace the fans, use external fans to cool the room, and so on).

Danger: Severe electrical shock can occur. Never turn on the power to any

3. Check all components and cables for visible damage. Do not power on the enclosure if you ﬁnd evidence of damage.

Note: To speed drive spin-up, it is recommended that you start the expansion enclosures before or at the same time as the controller enclosure.

4. Once the internal temperature is below 40ºC (104ºF), turn on the power switches and wait for the enclosure to power on.

5. Use TPSSM and the drive fault indicators (see “Checking the Indicator Lights” on page 34) to check the overall status of the enclosure and its components. Repair any faults found.

30 007-4644-001

Turning the Power On after an Emergency Shutdown

To restart the enclosure after a power failure or emergency shutdown, follow these steps:

1. After the emergency situation is over or power is restored to the site, turn off all

power switches (see Figure 3-1 on page 28).

Danger: Severe electrical shock can occur. Never turn on the power to any

2. Check all components and cables for visible damage. Do not power on the enclosure

if you ﬁnd evidence of damage.

Note: To speed drive spin-up, it is recommended that you start the expansion enclosures before or at the same time as the controller enclosure.

Turning the Power On after an Unexpected Shutdown

3. Turn on the power to the enclosures (see Figure 3-1 on page 28).

4. Use TPSSM and the drive fault indicators (see “Checking the Indicator Lights” on

page 34) to check the overall status of the enclosure and its components. Repair any faults found.

007-4644-001 31

3: Operating Storage Enclosures

Turning the Power Off

The enclosure is designed to run continuously, 24 hours a day. However, you may need to turn the power off for maintenance, such as upgrading the drives or replacing certain modules.To turn the power off, follow these steps:

1. Use TPSSM to determine the status of your system components and any special instructions before proceeding. The operating system software may require you to perform other procedures before turning off the power.

2. Stop all I/O activity to the enclosure.

If applicable, use TPSSM to logically disconnect the enclosure from the host. Make sure that all the drive activity indicators on the front of the enclosure are not blinking (indicating I/O activity) and that the fast write cache indicator light on the applicable RAID controller is off (not blinking).

3. Make sure that all the enclosure fault indicators are off (see Figure 3-2 on page 35 and Figure 3-3 on page 37).

If a fault indicator is on, correct the problem before turning off the power. Use TPSSM to diagnose and ﬁx the problem.

Warning: To shut off all power to an enclosure, you must turn off both power

switches and disconnect both power cords. An enclosure has two power switches and two power cords. The enclosure continues to operate if both switches are not turned off and both cords are not disconnected.

4. Turn off both power switches on the rear of the enclosure or the main circuit breaker, whichever is applicable (see Figure 3-1 on page 28).

Important: Once the power is off, you must wait at least 30 seconds before you turn it back on again.

5. After you have performed the necessary maintenance procedure, power on the enclosure using the procedure in “Turning the Power On” on page 27.

32 007-4644-001

Monitoring Status with Software

Use storage management software (TPSSM) to monitor enclosure status. You should run the software constantly and check it frequently.

TPSSM provides the best method to diagnose and repair failures. Thissoftware helps you do the following:

• Determine the nature of the failure.

• Locate the failed component.

• Provide recovery procedures to repair the failure.

Although the enclosure has fault indicators, these lights do not necessarily indicate which component has failed or needs to be replaced, orwhich type of recovery procedure you must perform. In some cases (such as loss of redundancy in various components), the fault light does not even come on. Only TPSSM can detect the failure.

For example, the recovery procedure for an impending drive failure (a predictive failure analysis, or PFA, ﬂag on a drive) varies depending on the drive status (hot spare, unassigned, RAID level, current volume status, and so on). Depending on the circumstances, a PFA ﬂag on a drive can indicate a high risk of data loss (if the drive is in a RAID0 volume) or a minimal risk (if the drive is unassigned). Only TPSSM can identify the risk level and provide the necessary recovery procedures. Note also that in the case of PFA ﬂags, the global fault and drive fault indicators do not come on, so just checking the indicators will not notify you of the failure, even if the risk of data loss is high.

Monitoring Status with Software

In addition, recovering from a failure may require you to perform procedures other than replacing the component (such as backing up the volume or failing a drive before removing it). TPSSM provides these procedures.

007-4644-001 33

Caution: If the software recovery procedures are not followed, data loss can result.

Note: For more information on the storage management software (TPSSM), see the SGI

TP9400 and SGI TP9500 RAID Administration Guide (007-4306-00x), the SGI TP9400 and SGI TP9500 Software Concepts Guide (007-4305-00x), and the SGI Storage Area Network Installation Instructions (108-0252-00x).

3: Operating Storage Enclosures

Checking the Indicator Lights

The enclosure’s indicator lights display the status of the enclosure and its components. Green indicators mean a normal operating status; amber indicators mean a possible failure.

It is important that you check all the indicators on the front and rear of the enclosure when you turn on the power. Besides checking for faults, you can use the indicators on the front of the enclosure to determine if the drives are responding to I/O transmissions from the host.

Important: Except as described in the notes following Figure 3-2 and Figure 3-3, an amber light indicates a component failure. If you see an amber indicator, run TPSSM to diagnose and repair the problem.

To check the enclosure indicators and operating status, follow the guidelines in Table 3-1 on page 36 (front indicators) and Table 3-2 on page 38 (rear indicators). If any indicators show anything other than a “normal” status, run TPSSM to diagnose and repair the problem.

Front Indicator Lights

This section describes the front indicator lights for the controller and expansion enclosures. Figure 3-2 shows the front indicators for the controller and expansion enclosures.

34 007-4644-001

Checking the Indicator Lights

Drive active (green)

Front bezel

Rack front view

Drive fault (amber)

Drives (14 total)

Global power

Global fault

Figure 3-2 Storage Enclosure Front Indicator Lights

Note: The normal operating state of all indicators on the front panel is green. If an amber indicator is on, or a green indicator is off, use TPSSM to determine the nature of the fault and the recovery procedure.

Exception: If the drive fault indicators are blinking, it means that TPSSM is locating a component. It does not indicate a failure.

007-4644-001 35

3: Operating Storage Enclosures

Table 3-1 describes the front indicator lights on the controller and expansion enclosures.

Table 3-1 Storage Enclosure Front Indicator Lights

Indicator Light Color Normal Operation

Drive activity indicator Green On, steady

On, blinking

Problem Indicator Condition Indicated

Off No power to enclosure; no power to storage array;

drive not properly seated in drive enclosure; drive

not spun up.

Drive fault indicator Amber Off

On, blinking

On, steady Drive failure; drive failed by user.

Global power indicator Green On Off No power to drive enclosure; no power to storage

array; power supply failure; overtemp condition.

Global fault indicator

a. Always use the storage management software (TPSSM) to precisely identify a failure.

b. The drive activity indicator blinks if data is being processed on the drives; otherwise, the indicator is on, steady.

c. The drive fault indicator blinks when the storage management software (TPSSM) is locating a drive, volume, or storage array. Otherwise, it

is off.

d. Not all enclosure component failures will turn this light on. See “Monitoring Status with Software” on page 88 for more information.

Amber Off On Enclosure component failure.

Rear Indicator Lights

This section describes the rear indicator lights for the controller and expansion enclosures. Figure 3-3 shows the rear indicators on the expansion enclosure and the RAID controller. Controller enclosures have RAID controller LEDs, but not ESM LEDs.

36 007-4644-001

Checking the Indicator Lights

Enclosure and ESM LEDs

Fan fault

Host 1 connector

In bypass

Power

Host 2 connector

Power

Fault

Ethernet connector

Overtemp

Out bypass

Power supply

fault

2-Gbit link rate

RS-232 Serial port

In bypass

Power

Fault

Conflict

Power

Expansion loop connector

Overtemp

Out bypass

Power supply

fault

Fan fault

RAID controller module LEDs

Host 1 link

Host 1 speed

Figure 3-3 Storage Enclosure Rear Indicator Lights

007-4644-001 37

Host 2 speed

Host 2 link

Fault

Cache active

Battery charging/charged

Expansion loop bypass

Expansion loop

3: Operating Storage Enclosures

Table 3-2 describes the rear indicators on the expansion enclosure and RAID controller.

Note: The normal operating state of all indicators on the rear panel is green. If an amber indicator is on, or a green indicator is off, use TPSSM to determine the nature of the fault and the recovery procedure.

Table 3-2 Storage Enclosure Rear Indicator Lights

Component Indicator Light Color Normal Operation

RAID controller

Host 1 link Host 2 link

Green On Off No connection to host

module

Host 1 speed Host 2 speed

Green On: 2 Gbit/s data

rate from host

Off: 1 Gbit/s data rate from host

Battery charging/charged

Green On: battery charged

Flashing: battery charging

Cache active Green On Off Cache failure

Fault Amber Off On General failure

Expansion loop Green On Off

Expansion loop

Amber Off On

bypass

Environmental status module

Environmental card fault indicator

Amber Off On,

Problem Indicator Condition Indicated

None

Off Battery failure

Failure of environmental card canister.

steady

Environmental card

Green On, steady Off No power to drive enclosure; no power to

power indicator

In, out bypass

Amber Off

indicators

Overtemp indicator

Amber Off On,

38 007-4644-001

On, steady

storage array; overtemp condition; power supply failure.

steady

On,

SFP failure; cable failure; controller

enclosure minihub failure.

Overtemp condition; fan failure.

steady

Table 3-2 Storage Enclosure Rear Indicator Lights (continued)

Checking the Indicator Lights

Component Indicator Light Color Normal Operation

ID conﬂict indicator Amber Off On,

Problem Indicator Condition Indicated

Enclosure number is set incorrectly;

steady

enclosure number is in use by another

subsystem on the loop.

Fan canister Fan fault indicator Amber Off On,

Fan canister failure; overtemp condition.

steady

Power supply canister

Power supply fault indicator

Power supply power indicator

Amber Off On,

Power supply failure; overtemp condition.

steady

Green On, steady Off Power supply off; power supply

disconnected; power supply is seated incorrectly; no power to drive enclosure; no power to storage array.

a. Always use the storage management software (TPSSM) to precisely identify a failure. b. If nothing is connected to the corresponding connector, this fault light is on. If a cable is connected and it is functioning properly, this light is

off.

c. This indicates a problem only if an SFP is plugged into the corresponding connector (see footnote a). Also, more than one bypass indicator

will be on if the connection fails. Usually, if an SFP fails, its bypass indicator will come on, as will the bypass indicator of the module to which it is connected.

d. The overtemp indicator comes on if the internal drive enclosure temperature reaches 40ºC (104ºF). See “Troubleshooting an Overtemp

Condition” in this chapter for more information.

007-4644-001 39

3: Operating Storage Enclosures

Moving the Enclosure

Before moving the enclosure to a new location or before removing the enclosure from its rack, it is highly recommended that you ﬁrst remove all drive sleds from the enclosure. Doing so helps safeguard the equipment and ensures a smoother transition to the new location. Before removing the drive sleds, label each one so that you can reinstall them in the correct order.

Instructions for depopulating the enclosure is provided in Chapter 4, “Replacing Enclosure Components”.

Note: If you are removing an expansionenclosure so that you can connect itto a different controller enclosure, you must delete all volumes and hot spares from the drives before removing the unit. If you are moving the enclosure a signiﬁcant distance (such as to another building or city), it is recommended that you pack it in its original shipping container.

Removing the Enclosure

To remove the enclosure from the rack, follow these steps:

1. Unsnap the side brackets from the front of the enclosure (see Figure 3-4).

2. Unscrew the front bezel from the mounting pins.

3. Using a ﬂat-blade screwdriver, carefully remove the bezel from the front of the drive enclosure. You must remove the bezel before you can remove the enclosure from the rack.

40 007-4644-001

Moving the Enclosure

Side brackets

Figure 3-4 Removing the Side Brackets from the Enclosure

4. It is highly recommended that you remove all of the modules from the enclosure to

make it lighter and easier to remove. See Chapter 4, “Replacing Enclosure Components” for information about depopulating an enclosure.

007-4644-001 41

3: Operating Storage Enclosures

5. Remove the front and rear mounting screws (see Figure 3-5). There are two mounting screws at the rear, one for each support rail, and two mounting screws in the front, one in each ﬂange.

Rear mounting

screws

Drive enclosure

48279

Support rails

installed in rack

Remove

screw from

flange hole.

Do not remove this screw from the vertical rail.

Figure 3-5 Removing the Empty Drive Enclosure

6. Slide the enclosure out of the front of the rack.

7. Repeat steps 1 through 6 for any other enclosures to be moved.

42 007-4644-001

Reinstalling the Enclosure

To reinstall the enclosure into a rack, follow these steps:

1. Install the support rails in the rack. See the procedure in the SGI TP9300 RAID

Installation and Upgrade Guide (108-0401-00x).

2. From the front of the rack, slide the enclosure into the rack along the support rails

(see Figure 3-5).

3. Fasten the rear and front mounting screws (see Figure 3-5).

4. Slide the top of the front bezel over the indicator lights and snap into place (see

Figure 3-4).

5. Wearing antistatic protection, reinstall the components into the enclosure. See

Chapter 4, “Replacing Enclosure Components” for more information.

Moving the Enclosure

007-4644-001 43

Chapter 4

4. Replacing Enclosure Components

This chapter describes how to replace each component in the enclosure in the following sections:

• “Replacing a Failed Disk Drive” on page 45

• “Replacing a Failed Fan” on page 48

• “Replacing a Failed Power Supply Module” on page 50

• “Replacing a Failed SFP Transceiver” on page 52

• “Replacing a Failed RAID Controller Module” on page 54

• “Replacing a Failed Environmental Status Module” on page 56

• “Upgrading Drives” on page 59

• “Adding Expansion Enclosures” on page 63

Replacing a Failed Disk Drive

When the storage management software (TPSSM) instructs you to replace a failed drive module, read “Disk Drives” on page 16, and then follow the steps in this section.

Warning: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

007-4644-001 45

4: Replacing Enclosure Components

1. Check TPSSM for instructions on drive failure recovery procedures. Follow any instructions provided by the software before proceeding.

Caution: Removing the wrong drive can cause data loss. Ensure that you remove

only the failed drive. The drive fault indicator of the failed drive module illuminates amber. Removing a drive module while its green activity indicator is blinking can cause data loss and may cause the host controller to mark the drive as failed. If you remove an active drive accidentally, wait at least 30 seconds and then reinstall it. See TPSSM for further recovery instructions.

2. Locate the failed drive module (its fault light should be on).

3. Wearing antistatic protection, lift (open) the lever of the failed drive module.

4. Pull out the drive module (see Figure 4-1).

Drive lever

Drive

Figure 4-1 Removing and Installing a Drive Module

5. Wait at least 30 seconds for the drive to spin down. Then remove the failed drive module from the enclosure.

6. Unpack the new drive module. Save all packing materials in case you need to return the module.

46 007-4644-001

Replacing a Failed Disk Drive

7. Review all documentation shipped with the new drive module for updated

replacement procedures and other information. If necessary, modify the remaining steps to meet the system requirements. Kits often contain the most current servicing information. If the kit instructions conﬂict with those in this procedure, follow the kit instructions.

Caution: If the temperature of the shipping or storage environment differs

signiﬁcantly from the temperature of the environment in which the drive module will be installed, acclimate the drive module before you install it.

8. Wearing antistatic protection, slide the new drive module fully into the slot.

Caution: Partial insertion of a drive may cause the controller to mark other drives on the same bus as failed. Install a drive in one complete motion. Make sure that you insert it all the way into the slot and lock it into place.

9. Lower (close) the lever. Wait for the new drive to spin up.

10. Check the drive activity and fault indicators above the new drive module. The

activity indicator should be on (either blinking or steady) and the fault indicator should be off.

Note: The fault indicator may ﬂash intermittently while the drive spins up. The activity indicator blinks when data is being written to the new drive module.

• If the activity indicator is off, the drive module could be installed incorrectly. Remove the drive module, wait 30 seconds, and then reinstall it.

• If the fault indicator stays on, or if the activity indicator stays off, the new drive may be bad. Use TPSSM to determine the problem.

11. See the applicable software procedures (for TPSSM or other system software) for instructions on bringing the drive module back online.

Important: Depending on your storage array conﬁguration, the storage array may automatically reconstruct data to the new drive module. If the array uses hot spares, it may have to complete reconstruction on the hot spare before it copies the data to the replaced drive. This increases the time required to complete this procedure. Use TPSSM to determine the current status of the new drive and the progress of any reconstruction or copyback.

007-4644-001 47

4: Replacing Enclosure Components

Replacing a Failed Fan

When TPSSM instructs you to replace a failed fan module, read “Fans” on page 20, and then follow these steps:

Caution: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

1. Check TPSSM for instructions on fan module failure recovery procedures. Follow any instructions provided before proceeding.

2. Unpack the new fan module. Set the new fan module on a dry, level surface near the enclosure. Save all packing materials in case you need to return the module.

3. Review all documentation shipped with the new fan module for updated replacement procedures and other information. If necessary, modify the remaining steps to meet the system requirements. Kits often contain the most current servicing information. If the kit instructions conﬂict with those in this procedure, follow the kit instructions.

4. If TPSSM instructs you to do so, turn off the power to the enclosure. Otherwise, leave the power on.

Caution: To prevent the enclosure from overheating while in use, complete the fan

module replacement procedure within 15 minutes from the time you remove the failed fanmodule to the time you install the new fan module. Both fan modules must be installed in the enclosure to provide sufﬁcient air circulation within the enclosure.

5. Locate the failed fan module (its fault light should be on).

6. Wearing antistatic protection, slide the latch to unlock the module (see Figure 4-2). The latch is at the bottom of the left-side fan module, and at the top of the right-side fan module.

7. Pull the knob to remove the fan module.

8. Determine the correct orientation for the new fan module. If replacing the module on the left, orient the unit so the latch is at the bottom. If replacing the right-side module, orient the unit so the latch is at the top.

9. Slide the new module into the slot.

48 007-4644-001

Fan

Latch

Figure 4-2 Removing and Installing a Fan Module

10. Push ﬁrmly until the latch snaps into place.

11. If you turned the power off, turn it back on.

Replacing a Failed Fan

12. Check the fault indicators as follows (see Figure 4-3):

• If the fan fault indicator is on or the fans are not spinning, the module might be installed incorrectly. Remove the fan module and reinstall it.

• If any of the other fault indicators shown in Figure 4-3 are on, use TPSSM to determine the problem.

Global fault

Front Rear

indicator

Fan canister fault indicator

Figure 4-3 Fault Lights for Fan Module Replacement

007-4644-001 49

Overtemp indicators

Fan canister fault indicator

55008-f

4: Replacing Enclosure Components

Replacing a Failed Power Supply Module

When instructed by TPSSM to replace a failed power supply module, read “Power Supplies” on page 21, and then follow these steps:

Warning: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

1. Check TPSSM for instructions on power supply failure recovery procedures. Follow any instructions provided before you proceed.

2. Unpack the new power supply module. Set the new power supply module on a dry, level surface near the enclosure. Save all packing materials in case you need to return the module.

3. Review all documentation shipped with the new power supply module for updated replacement procedures and other information. If necessary, modify the remaining steps to meet the system requirements. Kits often contain the most current servicing information. If the kit instructions conﬂict with those in this procedure, follow the kit instructions.

Danger: Shock can occur. Make sure the power supply is turned off and

unplugged before you remove or install it.

4. Locate the failed power supply (its fault light should be on).

5. Turn off the power and unplug the power cord from the failed module.

6. Remove the power cord from the power cord clip.

7. Wearing antistatic protection, grasp the pull-ring on the locking lever and squeeze the tab against the ring to unlatch the lever (see Figure 4-4).

8. Pull open the lever and remove the failed module.

9. Check the lever to make sure the orientation is the same as the module it is replacing. If not, move the lever to the pivot post on the other side.

50 007-4644-001

Replacing a Failed Power Supply Module

Locking lever

Power supply

Figure 4-4 Removing and Installing a Power Supply Module

Danger: Shock can occur. Make sure the power supply is turned off and

unplugged before you remove or install it.

10. Make sure that the power switch on the new power supply module is turned off.

11. Wearing antistatic protection, slide the new power supply module into the slot.

12. Close the lever and lock it into place.

13. Secure the power cord in the power cord clip.

14. Plug in the power cord and turn on the power switch.

15. Check that the power indicator on the new canister is on and that the fault indicator

is off (see Figure 4-4).

• If the power indicator is off, the canister might be installed incorrectly. Remove it and reinstall it.

• If the fault indicator is on, or the power indicator stays off, use TPSSM to determine the problem.

007-4644-001 51

4: Replacing Enclosure Components

Replacing a Failed SFP Transceiver

To replace a failed SFP transceiver, read “SFP Transceivers” on page 25, and then follow these steps:

Caution: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

1. Check TPSSM for instructions on SFP failure recovery procedures. Follow any instructions provided before you proceed.

2. Unpack the new SFP transceiver. Set it on a dry, level surface near the enclosure. Save all packing materials in case you need to return the SFP transceiver.

3. Review all documentation shipped with the new SFP transceiver for updated replacement procedures and other information. If necessary, modify the remaining steps to meet the system requirements. Kits often contain the most current servicing information. If the kit instructions conﬂict with those in this procedure, follow the kit instructions.

4. Locate the failed SFP transceiver.

Important: When an SFP transceiver fails, its bypass indicator glows amber, and so does the bypass indicator of the module to which it is connected (including the fault light on the controller enclosure). Make sure you remove the correct SFP transceiver. Use the fault lights and TPSSM to locate the failed SFP transceiver.

5. Wearing antistatic protection, remove the Fibre Channel cables from the failed SFP transceiver (see Figure 4-5).

Caution: Bending or damaging Fibre Channel cables can result in degraded

performance or data loss. Fibre Channel cables are fragile. Do not pinch the cables with tie wraps, step on them, or bend at them sharp angles.

52 007-4644-001

Fiber-optic cable

SFP transceiver

Replacing a Failed SFP Transceiver

Interface connector

Figure 4-5 Removing and Installing an SFP Transceiver

ESM

Note: Figure 4-5 shows SFPs being replaced in an expansion enclosure. The procedure is the same for controller enclosures.

6. Remove the failed SFP transceiver from the environmental status module or RAID

controller.

7. Insert the new SFP transceiver into the environmental status module or RAID

controller.

8. Connect the Fibre Channel cables to the new SFP transceiver.

9. Check the bypass indicator next to the new SFP transceiver as follows (see

Figure 4-5):

• If the bypass indicator is on, the SFP transceiver might be installed incorrectly.

• If the bypass indicator stays on, or a fault indicator on the environmental status

007-4644-001 53

Remove the SFP transceiver and reinsert it.

module comes on, use TPSSM to determine the problem.

4: Replacing Enclosure Components

Replacing a Failed RAID Controller Module

When TPSSM instructs you to replace a failed RAID controller module, read “RAID Controllers” on page 17, and then follow these steps:

Caution: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

1. Check TPSSM for instructions on RAID controller module failure recovery procedures. Follow any instructions provided before you proceed.

2. Unpack the new controller module. Set the new module on a dry, level surface near the enclosure. Save all packing materials in case you need to return the controller.

3. Review all documentation shipped with the new RAID controller module for updated replacement procedures and other information. If necessary, modify the remaining steps to meet the system requirements. Kits often contain the most current servicing information. If the kit instructions conﬂict with those in this procedure, follow the kit instructions.

4. Locate the failed RAID controller module (its fault indicator should be on).

5. Disconnect the Fibre Channel cables from the SFP transceivers in the failed controller module and then remove the SFP transceivers (see Figure 4-5). Label each cable to ensure that all cables are properly reconnected to the new module.

Caution: Bending or damaging Fibre Channel cables can result in degraded

performance or data loss. Fibre Channel cables are fragile. Do not pinch the cables with tie wraps, step on them, or bend at them at sharp angles.

6. Wearing antistatic protection, push down on the latch centered above the RAID controller module. The levers will pop out of the locked position (see Figure 4-6).

7. Grasp the pull-rings and pull on the levers to remove the failed canister.

8. Wearing antistatic protection, slide the new RAID controller module all the way into the empty slot.

9. Close both levers until the latch locks into place.

10. Install the SFP transceivers and Fibre Channel interface cables in their original locations.

54 007-4644-001

Replacing a Failed RAID Controller Module

Controller

Lever

Figure 4-6 Removing and Installing a RAID Controller Module

11. Check the power and fault indicators on the new controller as follows (see

Figure 4-5):

• If the power indicator is off, the controller module may not have been inserted correctly. Remove the module and reinsert it.

• If the fault indicator is on, the power indicator stays off, or any other fault indicator is on, use TPSSM to determine the problem.

12. See TPSSM for instructions on bringing the RAID controller online.

007-4644-001 55

4: Replacing Enclosure Components

Replacing a Failed Environmental Status Module

When TPSSM instructs you to replace a failed environmental status module, read “Environmental Status Modules (ESMs)” on page 19, and then follow the steps in this section.

Caution: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

1. Check TPSSM for instructions on environmental status module failure recovery procedures. Follow any instructions provided before you proceed.

2. Unpack the new environmental status module. Set the new module on a dry, level surface near the enclosure. Save all packing materials in case you need to return the module.

3. Review all documentation shipped with the new environmental status module for updated replacement procedures and other information. If necessary, modify the remaining steps to meet the system requirements. Kits often contain the most current servicing information. If the kit instructions conﬂict with those in this procedure, follow the kit instructions.

4. Locate the failed environmental status module (its fault indicator should be on).

5. Disconnect the Fibre Channel cables from the SFP transceivers in the failed module, and then remove the SFP transceivers (see Figure 4-7). Label each cable to ensure that all cables are properly reconnected to the new module.

Caution: Bending or damaging Fibre Channel cables can result in degraded

56 007-4644-001

performance or data loss. Fibre Channel cables are fragile. Do not pinch the cables with tie wraps, step on them, or bend at them at sharp angles.

Fiber-optic cable

SFP transceiver

Replacing a Failed Environmental Status Module

Interface connector

Figure 4-7 Removing SFPs from the Environmental Status Module

ESM

6. Wearing antistatic protection, push down on the latch centered above the

environmental status module. The levers will pop out of the locked position (see Figure 4-8).

7. Grasp the pull-rings and pull on the levers to remove the failed module.

8. Wearing antistatic protection, slide the new environmental status module all the

way into the empty slot.

9. Close both levers until the latch locks into place.

10. Install the SFP transceivers and Fibre Channel interface cables in their original

locations.

11. Check the power and fault indicators on the new module (see Figure 4-7).

• If the power indicator is off, the module may not have been inserted correctly.

• If the fault indicator is on, the power indicator stays off, or any other fault

12. See TPSSM for instructions on bringing the environmental status module online.

007-4644-001 57

Remove the module and reinsert it.

indicator is on, use TPSSM to determine the problem.

4: Replacing Enclosure Components

Figure 4-8 Removing and Installing an Environmental Status Module

Latch

Lever

55020

58 007-4644-001

Upgrading Drives

You can upgrade drives in the following two ways:

• Add drives to empty slots in the enclosure (see “Adding a Drive to an Empty Slot”

on page 59).

• Replace existing drives with drives of larger capacity (see “Adding Drives of Larger

Capacity” on page 61).

Adding a Drive to an Empty Slot

Follow these instructions to install an additional drive into an empty slot in the enclosure. You can install additional drives while the enclosure is powered on and running.

Warning: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

Upgrading Drives

1. Read the pertinent information in the software manuals supplied with your system

regarding drive upgrades and installation.

2. Check TPSSM and repair all reported problems.

3. Locate the blank drive module that you will replace with a drive, as follows:

• Because a blank drive module contains no drive, its activity indicator (on the front bezel above the drive module) will not be on.

• Verify the drive position is blank using TPSSM.

Caution: Removing the wrong drive can cause data loss. Make sure you remove

007-4644-001 59

only a blank drive module. If you remove an active drive accidentally, wait at least 30 seconds and then reinstall it. Check TPSSM for further recovery instructions.

4. Wearing antistatic protection, lift (open) the lever of the blank drive module.

5. Pull out the blank drive module (see Figure 4-1 on page 46).

6. Unpack the new drive module. Save all packing materials in case you need to return the module.

4: Replacing Enclosure Components

7. Review all documentation shipped with the new drive module for updated replacement procedures and other information.

Caution: If the temperature of the shipping or storage environment differs

signiﬁcantly from the temperature of the environment in which the drive module will be installed, acclimate the drive module before you install it.

8. Slide the new drive module fully into the slot.

9. Lower (close) the lever.

10. Check to be sure that the drive activity indicators are on and that the fault indicators above the new drive module are off.

Note: The fault indicator may ﬂash intermittently while the drive spins up.

• If the activity indicator is off, the drive module may not be installed correctly.

• If the fault indicator stays on, or the activity indicator stays off, the new drive

Remove the drive module, wait 30 seconds, and then reinstall it.

may be bad. Use TPSSM to determine the problem.

11. Conﬁgure the new drive using TPSSM.

If you need to install more than one drive module, repeat the steps in this section.

60 007-4644-001

Adding Drives of Larger Capacity

This section provides guidelines for upgrading the disk drives in a storage array. Carefully read your software documentation and this entire section to determine if you should follow this procedure, a modiﬁed version of this procedure, or use a different procedure provided by your operating system. Instructions provided with your software should supersede the instructions in the section.

Warning: Electrostatic discharge can damage sensitive components. Use a grounded

wrist strap or other antistatic precautions before handling enclosure components.

The only method for upgrading disk drives is to replace them all at the same time. This method requires you to back up the enclosure and shut down the storage array before replacing the drives. After replacing all the drives, you must reconﬁgure the enclosure and restore the data from backup media. This is the safest way to exchange drives without losing data. However, this method may take a long time to complete because of the backup, reconﬁguration, and restoration processes. In addition, other users will not be able to use the enclosure until you ﬁnish the procedure. You must use this method on RAID 0 logical units (LUNs).

Upgrading Drives

Follow these instructions to replace all drives at the same time. All the data currently on the drives will be lost when you replace the drives. You must use this method if you are upgrading drives containing RAID 0 volumes.

1. Read the following:

• “Upgrading Drives” on page 59.

• Information in your software documentation.

• Documentation shipped with the new drives.

Read all precautionary notes, kit instructions, and other information. Kit instructions often contain the most current information regarding the drives and their installation, plus upgrade or servicing procedures. Compare the kit instructions with this procedure to determine if you need to modify this procedure.

2. Check TPSSM and repair all reported problems.

3. Perform a complete backup of the drives that you are replacing. You need the

backup to restore data on the drives later in this procedure.

007-4644-001 61

4: Replacing Enclosure Components

4. Shut down all I/O activity on the enclosure. Make sure the drive activity indicators on the front are not blinking.

5. Turn off both power switches on the rear of the enclosure.

6. Wearing antistatic protection, lift (open) the lever of a drive module and pull out the drive module (see Figure 4-1 on page 46).

7. Repeat step 6 for all of the drive modules in the enclosure.

8. Unpack a new drive. Save the packing material and documentation in case you need to return the drive.

9. Slide the new drive module fully into a slot.

10. Lower (close) the lever.

11. Repeat steps 8 through 10 until you have installed all the new drives.

12. Turn on both power switches on the enclosure.

13. Check to be sure that the drive activity indicators are on and that the fault indicators above the new drive modules are off.

Note: The fault indicator may ﬂash intermittently while the drive spins up.

• If the activity indicator is off, the drive module may not be installed correctly.

Remove the drive module, wait 30 seconds, and then reinstall it.

• If the fault indicator stays on, or the activity indicator stays off, the new drive

may be bad. Use TPSSM to determine the problem.

14. Use TPSSM to conﬁgure the new drives. See your software documentation for detailed instructions.

15. Restore the data from backup to all the drives.

62 007-4644-001

Adding Expansion Enclosures

You can add new expansion enclosures to an existing storage array without turning off the array or interrupting data ﬂow.

Caution: You can add only new expansion enclosures to the storage array. This means

that there must be no existing information on the drives in the expansion enclosure you want to install. If the enclosure you want to install currently contains volumes or conﬁgured hot spare drives, you must delete them before adding the enclosure to a new conﬁguration.

You need torefer to the installation manual accompanying your controller and expansion enclosures to complete this procedure.

To install one or more new expansion enclosures into an existing loop, follow these steps:

1. Follow the instructions in the SGI TP9300 RAID Installation and Upgrade Guide to set

up and mount the new expansion enclosures and to connect them to each other.

Adding Expansion Enclosures

2. Using Figure 4-9 as a guide, connect the cables from the controller enclosure to the

new expansion enclosure(s). Then, if applicable, connect the cables from the existing expansion enclosure(s) to the new expansion enclosure(s).

Warning: Electrostatic discharge can damage sensitive components. Use a

grounded wrist strap or other antistatic precautions before handling enclosure components.

3. After you have connected the cables to the new expansion enclosures, the

controllers will locate the new disk drives.

Important: Depending on how long it takes you to complete step 2, TPSSM may report a loss of redundancy error on the expansion enclosures in the loop. This error will not interfere with data ﬂow, and will disappear after you ﬁnish installing the expansion enclosure(s).

4. Use TPSSM to check the status of the new drives and correct any errors found.

5. Use TPSSM to conﬁgure the new drives.

007-4644-001 63

4: Replacing Enclosure Components

Controller A Drive channel 1

Drive loop A

Controller B Drive channel 2

Drive loop B

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

Controller enclosure ID 00

Expansion enclosure ID 01

Expansion enclosure ID 02

Expansion enclosure ID 03

Expansion enclosure ID 04

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

Expansion enclosure ID 05

Expansion enclosure ID 06

Expansion enclosure ID 07

Figure 4-9 Adding New Expansion Enclosures to an Existing Array

64 007-4644-001

Chapter 5

5. Cabling

This chapter describes the cabling between the controller enclosure, the front-end host, and the expansion enclosures. It also covers Ethernet, serial port, and power connections in the following sections:

• “Connecting the Drive Loop Cables” on page 65

• “Connecting the System to the Hosts” on page 69

• “Powering On the System” on page 75

Connecting the Drive Loop Cables

The system is designed to support redundant drive loops. A redundant drive loop consists of one or more expansion enclosures connected to the controller enclosure using two sets of data cables. If one data path fails, the controller uses the other data path to maintain access to the drive group.

Follow these instructions to cable a group of enclosures into a redundant drive loop. If the drive loop contains only one enclosure, skip this procedure and proceed to “Connecting the System to the Hosts” on page 69.

Warning: Electrostatic discharge can damage sensitive components. Use

appropriate antistatic precautions before handling any components.

To complete this procedure, you will need two optic interface cables.

007-4644-001 65

5: Cabling

1Gb

X10

Tray Number

Conflict

2Gb

1Gb

X10

Tray Number

Conflict

2Gb

1. Connect the drive loops in the ﬁrst two enclosures (see Figure 5-1). Starting with the controller enclosure, connect an interface cable from the drive loop connector on the left RAID controller module to the drive channel in connector on the left environmental status module in the expansion enclosure.

2. Connect an interface cable from the drive loop connector on the right RAID controller module to the drive channel out connector on the right environmental status module in the expansion enclosure. This creates a redundant drive loop between the two enclosures (see Figure 5-1).

ESM A Drive channel in

Controller A Drive channel 1

Drive loop A

2Gb

1Gb

Conflict

X10

Tray Number

Controller B Drive channel 2

Controller enclosure ID 00

Drive loop B

ESM B Drive loop out

Disk drive

2Gb

1Gb

Conflict

X10

Tray Number

enclosure ID 01

Figure 5-1 Drive Loop Cabling for Systems with Two Enclosures

3. Connect additional drive enclosures to drive loops A and B by connecting the drive

channel out

connectors on the new enclosure to the drive channel in connectors of

the expansion enclosure directly beneath it (see Figure 5-2). Then connect the drive

loop out

cable from the left side RAID controller to the left-side drive channel in

connector on the new expansion enclosure. The drive channel out connector on the ﬁrst expansion enclosure should remain connected to the right-side RAID controller drive loop connector.

66 007-4644-001

Connecting the Drive Loop Cables

1Gb

X10

Tray Number

Conflict

2Gb

1Gb

X10

Tray Number

Conflict

2Gb

1Gb

X10

Tray Number

Conflict

2Gb

ESM A Drive channel in

Controller A Drive channel 1

Drive loop A

Controller B Drive channel 2

2Gb

1Gb

Conflict

X10

Tray Number

Storage system enclosure ID 00

Drive loop B

ESM A Dive loop out

2Gb

1Gb

Conflict

X10

Tray Number

ESM B Drive loop out

Expansion enclosure ID 01

Drive loop B

ESM B Drive loop out

2Gb

1Gb

Conflict

X10

Tray Number

Expansion enclosure ID 02

ESM B drive channel in

Figure 5-2 Drive Loop Cabling for Systems with Three or More Enclosures

4. Repeat step 3 to add up to seven expansion enclosures to the drive loop (see

Figure 5-3).

007-4644-001 67

5: Cabling

Controller A Drive channel 1

Drive loop A

Controller B Drive channel 2

Drive loop B

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

Controller enclosure ID 00

Expansion enclosure ID 01

Expansion enclosure ID 02

Expansion enclosure ID 03

Expansion enclosure ID 04

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

2Gb

1Gb

Conflict

X10

Tray Number

Expansion enclosure ID 05

Expansion enclosure ID 06

Expansion enclosure ID 07

Figure 5-3 Drive Loop Cabling for Systems with Eight Enclosures

68 007-4644-001

Connecting the System to the Hosts

You can connect your system to a host computer (or multiple host computers) in many ways. The following ﬁgures show typical examples of connecting eitherby direct connect or fabric connect. All examples assume dual-controller high -availability (HA) conﬁgurations. The following topics are covered:

• “Examples of Direct Host Connection Cabling” on page 69

• “Examples of Fabric/Switch Host Connection Cabling” on page 72

• “Connecting Ethernet and Serial Cables for Array Management” on page 73

Warning: Electrostatic discharge can damage sensitive components. Use

appropriate antistatic precautions before handling any components.

Examples of Direct Host Connection Cabling

Connecting the System to the Hosts

The TP9300 can be connected directly to one or two hosts. These connections can be conﬁgured with or without failover. Failover is a method of transferring I/O requests from one controller to another when a component failure or other error occurs on the data path to the ﬁrst controller. Depending on the operating system of your host(s), you may be able to choose from several failover options.

Note: If the host’s operating system does not support failover, the host can only be connected to one controller. See Figure 5-5.

The following sections contain more information:

• “Direct Connection with Failover” on page 70

• “Direct Connection without Failover” on page 71

007-4644-001 69

5: Cabling

Direct Connection with Failover

Figure 5-4 shows the possible conﬁgurations for direct connection systems with failover.

Direct connection to single

host with dual HBAs

Host

Controller AController B Controller AController B

Controller enclosure

Controller A

Host connectors

(SFP transceivers)

Direct connection to dual

hosts with dual HBAs

Host 1 Host 2

Controller enclosure

Controller B

Host connectors

(SFP transceivers)

Figure 5-4 Direct Host Connection with Failover

70 007-4644-001

Direct Connection without Failover

Figure 5-5 shows the possible conﬁgurations for direct connection systems without failover.

Connecting the System to the Hosts

Direct connection to dual

hosts with single HBAs

Host 1 Host 2

Controller B

Controller enclosure

Controller A

Host connectors

(SFP transceivers)

Figure 5-5 Direct Host Connection without Failover

Controller B

Host connectors

(SFP transceivers)

Controller A

007-4644-001 71

5: Cabling

Examples of Fabric/Switch Host Connection Cabling

Figure 5-6 shows an example of connecting the TP9300 to hosts using a Fibre Channel switch. For more information, see the “SGI Storage Area Network Installation Instructions” (108-0252-00x) shipped with the Fibre Channel switch.

Figure 5-6 shows a simple switch conﬁguration.

Host system with two host bus adapters

Switch Switch

Note: Other configurations are possible

Controller enclosure

Controller A

host connectors

(SFP transceivers)

Controller B

host connectors

(SFP transceivers)

Figure 5-6 Connecting to Hosts via a Fibre Channel Switch

72 007-4644-001

Connecting Ethernet and Serial Cables for Array Management

There are two methods of managing the storage system: in-band management and out-of-band management. In-band management requires that only the Fibre Channel host interface is present and functional. Out-of-band management requires that Ethernet or serial connections are cabled to each controller.

Note: For more information on storage management, see the SGI TP9400 and SGI TP9500 RAID Administration Guide (007-4306-00x), the SGI TP9400 and SGI TP9500 Software Concepts Guide (007-4305-00x), and the SGI Storage Area Network Installation Instructions

(108-0252-00x).

To connect Ethernet or serial (RS-232) interface cables to the controller enclosure for direct management of the storage array, follow the steps in this section. Optionally, you can use the Ethernet or serial connectors for diagnostic procedures.

Warning: Electrostatic discharge can damage sensitive components. Use

appropriate antistatic precautions before handling any components.

Connecting the System to the Hosts

007-4644-001 73

5: Cabling

1. To cable the system for direct management via Ethernet, connect a pair of Ethernet interface cables from the storage management station to the Ethernet connectors for controller A (left) and controller B (right) on the rear of the controller enclosure (see Figure 5-7).

2. To cable the system for direct management via the RS-232 serial ports, connect a pair of serial cables from the storage management station to the serial connectors for controller A (left) and controller B (right) on the rear of the controller enclosure.

Storage Management

Station

Ethernet Connector

Ethernet Network

Ethernet Connector

Controller B

Controller Enclosure

Ethernet Connector

Controller A

Figure 5-7 Connecting Ethernet Cables

74 007-4644-001

Powering On the System

1Gb

X10

Tray Number

Conflict

2Gb

To connect the controller and the expansion enclosures to the AC power source and power on the system, follow these steps:

Warning: Electrostatic discharge can damage sensitive components. Use

appropriate antistatic precautions before handling any components.

1. Make sure the switches on the power distribution units (on the rear of the rack

inside the access panel) are in the OFF position.

2. Make sure both power switches on each of the controller and expansion enclosures

are off.

3. Plug the power cords into each power supply of the storage enclosures (see “AC

power in” in Figure 5-8). Plug the left-side power cords into the left PDU and plug the right-side power cords into the right PDU.

Powering On the System

2Gb

1Gb

Conflict

X10

Tray Number

AC power in

Power switch

AC power in

Power switch

Figure 5-8 Power Connections and Switches of the Enclosures

Note: See Appendix A, “Speciﬁcations and Requirements” for details on power requirements.

4. To keep the power cords from being accidently pulled from the power connections,

secure the cords in the strain relief power cord clips.

007-4644-001 75

5: Cabling

1Gb

X10

Tray Number

Conflict

2Gb

5. To ensure power redundancy, plug each power distribution unit into a separate AC power circuit (see Figure 5-9).

Controller enclosure power supplies

2Gb

1Gb

Conflict

X10

Tray Number

Expansion enclosure power supplies

Power distribution units (PDUs)

Figure 5-9 PDU Cabling

Breaker

switches

76 007-4644-001

Powering On the System

Note: To speed drive spin-up, it is recommended that you start the expansion enclosures before or at the same time as the controller enclosure. This procedure powers up all components in the system at the same time by turning on the power distribution units.

6. Turn on both power switches on the rear of each expansion enclosure. Then turn on

both power switches on the rear of the controller enclosure (see Figure 5-8).

7. When all controller and drive enclosure power switches are in the ON position, ﬂip

the breaker switches of both PDUs to the ON position (see Figure 5-9).

8. Check the indicator lights on the front and rear of each component. If any lights

indicate a fault, see “Checking the Indicator Lights” on page 34.

007-4644-001 77

Appendix A

A. Speciﬁcations and Requirements

This appendix describes the technical speciﬁcations and requirements of the TP9300 system rack and components in the following sections:

• “Rack Speciﬁcations” on page 79

• “Enclosure Speciﬁcations” on page 91

Rack Speciﬁcations

The system is housed in a 22-inch wide rack that contains two AC distribution boxes. Standard EIA rails provide mounting holes for installing 19-inch wide devices. There are roller casters and stability feet on the bottom of the rack for moving and leveling during installation and relocation.

Figure A-1 shows a rack that contains eight enclosures: one controller enclosure and seven expansion enclosures.

007-4644-001 79

A: Specifications and Requirements

Controller enclosure

Expansion enclosures

Figure A-1 Populated Rack

For more information on installation and cabling schemes, see the SGI TP9300 RAID Installation and Upgrade Guide (108-0401-00x).

80 007-4644-001

Dimensions

Rack Specifications

Figure A-2 shows the system rack and its dimensions.

183 cm (72 in.)

Figure A-2 Dimensions of the System Rack

007-4644-001 81

56 cm

(22 in.)

91 cm

(36 in.)

A: Specifications and Requirements

Weight

The total weight of the system depends on the quantity of enclosures installed. Table A-1 lists the overall weight of the rack, plus the maximum weight of an enclosure. You can use these weights to estimate the total weight of your system, based on the number of enclosures installed in the rack. Figure A-3 provides an example of calculating the weight of a rack.

Table A-1 Rack, Crate, and Enclosure Weights

Rack (Empty)

121 kg (270 lb)

Crate (Empty)

97 kg (215 lb)

System Enclosure (Maximum)

44.5 kg (98.1 lb)

82 007-4644-001

Silicon Graphics Total Performance 9300 User Manual

Specifications and Main Features

Frequently Asked Questions

User Manual