Sun Microsystems Sun StorEdge T3 Installation, Operation And Service Manual

Download

Page 1

Sun StorEdge™T3 and T3+ Array

Installation, Operation,

and Service Manual

Sun Microsystems, Inc. 901 San Antonio Road Palo Alto, CA 94303-4900 U.S.A. 650-960-1300

Part No. 816-0773-10 August 2001, Revision A

Send comments about this document to: docfeedback@sun.com

Page 2

This productor document is distributed underlicenses restrictingits use, copying, distribution, anddecompilation. No part of thisproduct or document maybe reproducedin any form by anymeans without prior written authorizationof Sun and its licensors,if any.Third-party software, includingfont technology,is copyrighted and licensed from Sun suppliers.

Parts ofthe product may be derivedfrom BerkeleyBSD systems, licensed from the University ofCalifornia. UNIX is a registered trademarkin the U.S.and other countries, exclusively licensedthrough X/Open Company,Ltd.

Sun, SunMicrosystems, the Sun logo, AnswerBook2,docs.sun.com, SunSolve, JumpStart, StorTools, SunStorEdge, and Solaris are trademarks, registered trademarks, or servicemarks of Sun Microsystems, Inc. in theU.S. and other countries. AllSPARCtrademarks are used under license and aretrademarks or registered trademarks of SPARC International,Inc. in the U.S. and other countries.Products bearing SPARC trademarks are basedupon an architecture developed by Sun Microsystems, Inc.

The OPENLOOK and Sun™ Graphical UserInterface was developed by SunMicrosystems, Inc. for its users and licensees.Sun acknowledges the pioneeringefforts of Xerox in researching and developing the conceptof visual or graphical userinterfaces for the computer industry. Sun holds anon-exclusive license from Xerox to the Xerox Graphical User Interface, which license alsocovers Sun’s licensees who implementOPEN LOOK GUIsand otherwise comply with Sun’swritten license agreements.

Federal Acquisitions:Commercial Software—GovernmentUsers Subject to Standard License Terms and Conditions.

DOCUMENTATION IS PROVIDED “AS IS” AND ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONS AND WARRANTIES, INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULARPURPOSE OR NON-INFRINGEMENT, ARE DISCLAIMED, EXCEPT TO THE EXTENT THAT SUCH DISCLAIMERS ARE HELD TO BE LEGALLY INVALID.

Ce produitou document est distribué avecdes licences qui en restreignent l’utilisation, la copie,la distribution, et la décompilation.Aucune partie dece produit ou document nepeut êtrereproduite sousaucune forme, par quelque moyenque ce soit, sans l’autorisation préalable et écrite deSun et de ses bailleursde licence, s’il y ena. Le logiciel détenu pardes tiers, et qui comprend latechnologie relative aux polices de caractères, estprotégé parun copyright et licencié pardes fournisseurs de Sun.

Des partiesde ce produit pourront être dérivées des systèmes BerkeleyBSD licenciés par l’Université deCalifornie. UNIX est une marque déposée auxEtats-Unis et dans d’autres payset licenciée exclusivement par X/Open Company, Ltd.

Sun, SunMicrosystems, le logo Sun, AnswerBook2,docs.sun.com, SunSolve, JumpStart, StorTools, SunStorEdge, et Solaris sont des marques de fabriqueou des marques déposées, oumarques deservice, de Sun Microsystems, Inc. aux Etats-Uniset dans d’autres pays. Toutes les marques SPARCsont utilisées sous licence etsont des marques de fabrique ou desmarques déposéesde SPARC International, Inc. auxEtatsUnis etdans d’autres pays. Les produits portantles marques SPARC sont basés sur une architecture développéepar Sun Microsystems, Inc.

L’interfaced’utilisation graphique OPEN LOOK etSun™ a été développée parSun Microsystems,Inc. pour ses utilisateurs etlicenciés. Sun reconnaît lesefforts depionniers de Xerox pour la recherche et le développementdu concept des interfaces d’utilisationvisuelle ou graphique pour l’industriede l’informatique. Sun détient unelicence non exclusive de Xeroxsur l’interface d’utilisation graphique Xerox, cette licence couvrant égalementles licenciés de Sun quimettent en place l’interface d’utilisationgraphique OPEN LOOK et quien outrese conforment aux licences écritesde Sun.

LA DOCUMENTATION EST FOURNIE “EN L’ETAT” ET TOUTES AUTRES CONDITIONS, DECLARATIONS ET GARANTIES EXPRESSES OU TACITES SONT FORMELLEMENT EXCLUES,DANS LA MESURE AUTORISEE PAR LA LOI APPLICABLE,Y COMPRISNOTAMMENT TOUTE GARANTIE IMPLICITE RELATIVE A LA QUALITE MARCHANDE, A L’APTITUDE A UNE UTILISATION PARTICULIERE OU A L’ABSENCE DE CONTREFAÇON.

Please

Recycle

Page 3

Preface xi

1. Sun StorEdge T3 and T3+ Array Overview 1-1

1.1 Product Description 1-2

1.2 Array Features 1-3

1.3 Components 1-5

1.3.1 Disk Drive 1-5

1.3.2 Controller Card 1-6

1.3.2.1 Sun StorEdge T3 Array Controller 1-6

1.3.2.2 Sun StorEdge T3+ Array Controller 1-7

1.3.3 Interconnect Card 1-7

1.3.4 Power and Cooling Unit 1-8

1.4 Architecture 1-8

1.5 Supported Configurations 1-9

2. Installation 2-1

2.1 Preparing for the Installation 2-2

2.2 Inspecting the Array 2-3

2.3 Editing the Host Files 2-4

2.4 Installing the Array 2-7

2.5 Connecting the Cables 2-8

Contents iii

Page 4

2.6 Powering On and Verifying the Hardware Configuration 2-14

2.7 Establishing a Network Connection 2-16

2.8 Verifying the Firmware Level and Configuration 2-19

2.9 Installing the Administration Tools 2-25

2.10 Defining and Mounting Volumes 2-25

2.10.1 Workgroup Configuration 2-27

2.10.2 Enterprise Configuration 2-29

2.11 Changing the Default Configuration 2-32

2.12 Connecting the Host System 2-33

2.12.1 Verifying the Data Host Connection 2-33

2.12.2 Establishing Logical Volumes on the Host 2-34

2.12.3 Creating Soft Partitions on the Data Host 2-35

2.12.4 Establishing Alternate Pathing on the Host 2-35

3. Operation 3-1

3.1 Powering Off and On 3-1

3.1.1 Powering Off 3-1

3.1.2 Powering On 3-2

3.2 Reconfiguring the Array Settings 3-3

3.2.1 Unit Volume Configurations 3-3

3.2.2 RAID Levels 3-4

3.2.2.1 RAID 0 3-5

3.2.2.2 RAID 1 3-5

3.2.2.3 RAID 5 3-5

3.2.3 Cache Modes 3-6

3.2.4 Disabling and Reconstructing a Drive 3-8

3.3 Monitoring the Array 3-9

3.3.1 Sun StorEdge Component Manager 3-9

3.3.2 SNMP Notification 3-9

3.3.3 syslog Error Reports 3-10

iv Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 5

4. Troubleshooting 4-1

4.1 Analyzing the Problem 4-1

4.1.1 Host-Generated Message 4-1

4.1.2 Sun StorEdge Component Manager 4-2

4.1.3 Telnet Session 4-2

4.1.4 Array LEDs 4-3

4.1.4.1 Disk Drive LEDs 4-3

4.1.4.2 Power and Cooling Unit LEDs 4-4

4.1.4.3 Interconnect Card LEDs 4-6

4.1.4.4 Controller Card LEDs 4-7

4.2 Channel Connection Failures 4-9

4.3 FRU Failures 4-10

5. Service 5-1

5.1 Preparing for Service 5-2

5.2 Removing and Replacing Components 5-2

5.2.1 Disk Drives 5-3

5.2.1.1 Disk Drive LEDs 5-3

5.2.1.2 Removing and Replacing a Disk Drive 5-4

5.2.2 Power and Cooling Units 5-7

5.2.2.1 Power and Cooling Unit LEDs 5-8

5.2.2.2 Removing and Replacing a Power and Cooling Unit 58

5.2.3 UPS Battery 5-10

5.2.3.1 Battery Maintenance 5-10

5.2.3.2 Removing and Replacing the UPS Battery 5-11

5.2.4 Interconnect Cards 5-11

5.2.4.1 Interconnect Card LEDs 5-11

5.2.4.2 Removing and Replacing an Interconnect Card 5-12

5.2.5 Controller Card 5-14

5.2.5.1 Controller Card LEDs 5-14

Contents v

Page 6

5.2.5.2 Removing and Replacing a Controller Card 5-14

5.2.6 Chassis 5-16

5.3 Upgrading Firmware 5-16

5.3.1 Upgrading Sun StorEdge T3+ Array Controller Firmware 5-17

5.3.1.1 To Upgrade the Level 2 Image 5-17

5.3.1.2 To Upgrade the Level 3 Image 5-19

5.3.2 Upgrading Sun StorEdge T3 Array Controller Firmware 5-20

5.3.2.1 To Upgrade Controller Firmware 5-21

5.3.2.2 To Upgrade the Controller EPROM Firmware: 5-22

5.3.3 Upgrading Interconnect Card Firmware 5-24

5.3.4 Upgrading Disk Drive Firmware 5-25

A. Specifications A-1

A.1 Power Specifications A-1

A.2 Environmental Specifications A-2

A.3 Mechanical Specifications A-3

A.4 Cable Specifications A-3

B. Illustrated Parts B-1

Glossary Glossary-1

vi Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 7

Figures

FIGURE 1-1 Sun StorEdge T3 Array 1-3

FIGURE 1-2 Disk Drives (Front View) 1-5

FIGURE 1-3 Controller Card (Rear View) 1-6

FIGURE 1-4 Sun StorEdge T3 Array Controller Card Front Panel 1-6

FIGURE 1-5 Sun StorEdge T3+ Array Controller Card Front Panel 1-7

FIGURE 1-6 Interconnect Cards (Rear View) 1-7

FIGURE 1-7 Power and Cooling Units (Rear View) 1-8

FIGURE 1-8 Workgroup Configuration 1-9

FIGURE 1-9 Enterprise Configuration 1-10

FIGURE 2-1 Removing the Front Panel 2-3

FIGURE 2-2 Serial Number and MAC Address on Pull-out Tab 2-4

FIGURE 2-3 Sun StorEdge T3 Array Cables and Adapters 2-9

FIGURE 2-4 Sun StorEdge T3+ Array Cables 2-9

FIGURE 2-5 Connecting the Fiber-Optic Cable to the FC-AL Connector on a

Sun StorEdge T3+ Array 2-10

FIGURE 2-6 Connecting the Fiber-Optic Cable and MIA to the FC-AL Connector on a

Sun StorEdge T3 Array 2-11

FIGURE 2-7 Connecting the Ethernet Cable 2-11

FIGURE 2-8 Connecting the Power Cords 2-12

FIGURE 2-9 Connecting the Interconnect Cables 2-13

vii

Page 8

FIGURE 2-10 Fully Cabled Partner Group (Enterprise Configuration) 2-14

FIGURE 2-11 Power and Cooling Unit LEDs and Power Button Location 2-15

FIGURE 2-12 Disk Drive Physical Numbering 2-26

FIGURE 3-1 Power Switch Locations 3-2

FIGURE 4-1 Disk Drive LEDs (Viewed Through Front Cover) 4-3

FIGURE 4-2 Power and Cooling Unit LEDs 4-4

FIGURE 4-3 Interconnect Card LEDs 4-6

FIGURE 4-4 Sun StorEdge T3 Array Controller Card LEDs 4-7

FIGURE 4-5 Sun StorEdge T3+ Array Controller Card LEDs 4-8

FIGURE 5-1 Removing the Front Panel 5-4

FIGURE 5-2 Disk Drive Numbering 5-4

FIGURE 5-3 Releasing the Latch Handle 5-5

FIGURE 5-4 Removing a Disk Drive 5-6

FIGURE 5-5 Power Cords Connected to the Power and Cooling Units 5-7

FIGURE 5-6 Removing the Power and Cooling Unit 5-9

FIGURE 5-7 Removing the Interconnect Card 5-13

FIGURE 5-8 Removing the Controller Card 5-15

FIGURE B-1 Disk Drive in Canister B-2

FIGURE B-2 Sun StorEdge T3 Array Controller Card B-2

FIGURE B-3 Sun StorEdge T3+ Array Controller Card B-3

FIGURE B-4 Interconnect Card B-3

FIGURE B-5 Power and Cooling Unit B-3

FIGURE B-6 Cables, Adapters, and Cords B-4

FIGURE B-7 Door Assembly (Front Panel) B-5

viii Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 9

Tables

TABLE 1-1 Sun StorEdge T3 and T3+ Array Key Features—73-Gbyte Drive 1-4

TABLE 2-1 Drive Status Messages 2-27

TABLE 3-1 Volume Configuration Examples 3-4

TABLE 3-2 RAID 0 Data Blocks 3-5

TABLE 3-3 RAID 1 (1+0) Data Blocks 3-5

TABLE 3-4 RAID 5 Data Blocks 3-6

TABLE 3-5 Cache Transition States 3-7

TABLE 3-6 syslog Error Reporting Levels 3-10

TABLE 4-1 Disk Drive LED Descriptions 4-4

TABLE 4-2 Power and Cooling Unit LED Descriptions 4-5

TABLE 4-3 Interconnect Card LED Descriptions 4-6

TABLE 4-4 FC-AL Channel-Active LED Descriptions (Sun StorEdge T3 array) 4-7

TABLE 4-5 Controller Status LED Descriptions (Sun StorEdge T3 array) 4-7

TABLE 4-6 Sun StorEdge T3+ Array Controller Card LED Descriptions 4-8

TABLE A-1 Power Specifications A-1

TABLE A-2 Environmental Specifications A-2

TABLE A-3 Physical Specifications A-3

TABLE A-4 Cable Specifications A-3

TABLE B-1 Illustrated Parts B-1

Page 10

x Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 11

Preface

This manual contains information on how to install, operate, and service the Sun StorEdge™ T3 and T3+ arrays. Most of the procedures in this manual can be performed by an experienced system administrator.

Before You Read This Book

Read the safety information in the Sun StorEdge T3 and T3+ Array Regulatory and Safety Compliance Manual that is included with the array before attempting the installation. Make sure you have prepared for the installation by reviewing the Sun

StorEdge T3 and T3+ Array Site Preparation Guide and the Sun StorEdge T3 and T3+ Array Configuration Guide. These guides are available on the Sun™ documentation

web site or from your Sun representative.

Work with your Sun representative to determine if you require any external hardware or software products for using this device. Being prepared with this knowledge and the appropriate tools will simplify the installation.

How This Book Is Organized

This manual is organized as follows:

Chapter 1 provides an overview of the Sun StorEdge T3 and T3+ arrays, describing features, components, architecture, and supported configurations.

Chapter 2 describes how to install the array in either a single-unit or partner-group configuration.

Page 12

Chapter 3 describes options for operating your array, such as reconfiguring the default settings and monitoring array activity.

Chapter 4 provides information on how to approach troubleshooting the array and references troubleshooting tools.

Chapter 5 describes how to service the major components of the array and how to upgrade array firmware.

Appendix A contains listings of array specifications.

Appendix B contains an illustrated parts list.

Glossary is a list of words and phrases and their definitions. Glossary terms are italicized in the text.

Using UNIX Commands

This document contains some information on basic UNIX®commands and procedures such as booting the devices. For further information, see one or more of the following:

■ AnswerBook2™ online documentation for the Solaris™ software environment

■ Other software documentation that you received with your system

xii Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 13

Typographic Conventions

Typeface Meaning Examples

AaBbCc123 The names of commands, files,

and directories; on-screen computer output

AaBbCc123

AaBbCc123 Book titles, new words or terms,

What you type, when contrasted with on-screen computer output

words to be emphasized

Command-line variable; replace with a real name or value

Edit your.login file. Use ls -a to list all files.

% You have mail.

% su Password:

Read Chapter 6 in the User’s Guide. These are called class options. Yo u must be superuser to do this.

To delete a file, type rm filename.

Shell Prompts

Shell Prompt

C shell machine_name%

C shell superuser machine_name#

Bourne shell and Korn shell $

Bourne shell and Korn shell superuser #

Sun StorEdge T3 and T3+ array :/:

Preface xiii

Page 14

Accessing Sun Documentation Online

You can find the Sun StorEdge T3 and T3+ array documentation and other select product documentation for Network Storage Solutions at:

http://www.sun.com/products-n-solutions/hardware/docs/ Network_Storage_Solutions

Sun Welcomes Your Comments

Sun is interested in improving its documentation and welcomes your comments and suggestions. You can email your comments to Sun at:

docfeedback@sun.com

Please include the part number (816-0773-10) of your document in the subject line of your email.

Preface xv

Page 16

xvi Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 17

CHAPTER

Sun StorEdge T3 and T3+ Array Overview

This chapter provides a high-level introduction to the Sun StorEdge T3 and T3+ arrays, and is organized as follows:

■ “Product Description” on page 1-2

■ “Array Features” on page 1-3

■ “Components” on page 1-5

■ “Architecture” on page 1-8

■ “Supported Configurations” on page 1-9

1-1

Page 18

1.1 Product Description

The Sun StorEdge T3 array is a high-performance, modular, scalable storage device that contains an internal RAID controller and nine disk drives with Fibre Channel connectivity to the data host. Extensive features include redundant components, notification of failed components, and the ability to replace components while the unit is online. The Sun StorEdge T3+ array provides the same features as the Sun StorEdge T3 array, and includes an updated controller card with direct fiber-optic connectivity and additional memory for data cache. The controller cards of both array models are described in more detail later in this chapter.

The array can be used either as a standalone storage unit or as a building block, interconnected with other arrays of the same type and configured in various ways to provide a storage solution optimized to the host application. The array can be placed on a table top or rackmounted in a server cabinet or expansion cabinet.

The array is sometimes called a controller unit, which refers to the internal RAID controller on the controller card. Arrays without the controller card are called expansion units. When connected to a controller unit, the expansion unit enables you to increase your storage capacity without the cost of an additional controller. An expansion unit must be connected to a controller unit to operate because it does not have its own controller.

In this document, the Sun StorEdge T3 array and Sun StorEdge T3+ array are referred to as the array, except when necessary to distinguish between models.

reliability, availability, and serviceability (RAS)

Note – The Sun StorEdge T3 and T3+ arrays are similar in appearance. In this

document, all illustrations labeled Sun StorEdge T3 array also apply to the Sun StorEdge T3+ array, except when necessary to distinguish specific model features. In these instances, the array model is specified.

1-2 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 19

FIGURE 1-1 Sun StorEdge T3 Array

1.2 Array Features

Sun StorEdge T3 and T3+ arrays provide high reliability while being simple to administer and maintain. The array features a design that supports high availability, high performance, configuration flexibility, and scalability of bandwidth, capacity, and input/output operations per second (IOPS). these key features, depicting an array with nine 73-Gbyte drives.

Note – Other drive capacities are available. Check with your Sun representative for

more information on storage capacities that are currently available.

TABLE 1-1 provides an example of

Chapter 1 Sun StorEdge T3 and T3+ Array Overview 1-3

Page 20

TABLE 1-1 Sun StorEdge T3 and T3+ Array Key Features—73-Gbyte Drive

Feature Sun StorEdge T3 Array Sun StorEdge T3+ Array

Capacity per unit 657 GB (9 x 73 GB drives) 657 GB (9 x 73 GB drives)

256 MB cache 1 GB cache

Performance per controller

95+ MB/sec for large RAID 5 reads from disk (64K >)

90+ MB/sec for large RAID 5 writes to disk 90+ MB/sec for large RAID 5 writes to disk

4,150 IOPS for 8 KB (100% cache hit reads) 7,200 IOPS for 8 KB (100% cache hit reads)

Connectivity per controller

Scalability (one to two units, one to two controllers)

FC-AL host interface FC-AL host interface (LC-SFF to SC)

10BASE-T Ethernet network port 10/100BASE-T Ethernet network port

1,314 GB maximum capacity 1,314 GB maximum capacity

256 to 512 MB cache 1 to 2 GB cache

1 to 2 host Fibre Channel interfaces 1 to 2 host Fibre Channel interfaces

90 to 190 MB/sec bandwidth 90 to 190 MB/sec bandwidth

4,100 to 8,150 IOPS (100% cache hit reads) 7,200 to 14,360 IOPS (100% cache hit reads)

Reliability/ redundancy

RAID 0/1 (1+0)/5 RAID 0/1 (1+0)/5

Redundant back-end data loops Redundant back-end data loops

Redundant interconnect cards Redundant interconnect cards

Passive centerplane connector board Passive centerplane connector board

Redundant power and cooling units and power cables

Redundant back-end FC-AL interconnect cables

Redundant power and cooling units and power cables

Redundant back-end FC-AL interconnect cables

Redundant controller configuration Redundant controller configuration

Administration Telnet access (CLI interface) Telnet access (CLI interface)

Component Manager (GUI interface) Component Manager (GUI interface)

SNMP Version 1 SNMP Version 1

1-4 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 21

1.3 Components

The array contains four basic components that can be easily replaced:

■ Disk drive

■ Controller card

■ Interconnect card

■ Power and cooling unit

All components plug into a centerplane; there is no internal cabling. For information on how to remove and replace these components, see Chapter 5.

Note – The array centerplane and external chassis are physically connected, and are

available as one field-replaceable unit (FRU). Only a qualified field-service representative should replace this FRU.

1.3.1 Disk Drive

Behind the panel at the front of the array are nine Fibre Channel disk drives, numbered drive 1 through drive 9 from left to right. Each drive is in an enclosed canister that is easily installed and removed from the array. Drive light-emitting- diodes (LEDs), which are visible through the front panel, indicate drive activity and status. See Chapter 4 for more information on LED locations and descriptions. See Chapter 5 for information on how to service the disk drives.

FIGURE 1-2 Disk Drives (Front View)

Chapter 1 Sun StorEdge T3 and T3+ Array Overview 1-5

Page 22

1.3.2 Controller Card

The Sun StorEdge T3 and T3+ array each have a specific controller card that contains slightly different features on the exterior and a different design on the internal controller board. The general features of each array controller card are described in this section. The location of the controller card for both models is shown in

FIGURE 1-5. Chapter 5 provides information on how to service the controller card.

Controller card

FIGURE 1-3 Controller Card (Rear View)

1.3.2.1 Sun StorEdge T3 Array Controller

The Sun StorEdge T3 array controller card contains the RAID controller hardware and firmware, as well as a host Fibre Channel interface, an Ethernet 10BASE-T host interface, an RJ-11 serial port for special service procedures, and 256 Mbytes of highspeed synchronous-dynamic-random-access-memory (SDRAM) data cache.

Fibre Channel

interface

FIGURE 1-4 Sun StorEdge T3 Array Controller Card Front Panel

1-6 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

10BASE-T

Ethernet interface

RJ-11 serial port

diagnostic connector

Page 23

1.3.2.2 Sun StorEdge T3+ Array Controller

The Sun StorEdge T3+ array controller card contains the RAID controller hardware and firmware, a new Fibre Channel LC-SFF (small form factor) connector that provides direct fiber-optic cable connection, two RJ-45 ports for a 10/100BASE-T Ethernet network connection and a serial connection for special service procedures, and 1 Gbyte of high-speed SDRAM data cache.

Fibre Channel

LC-SFF connector

Ethernet interfaceinterface

FIGURE 1-5 Sun StorEdge T3+ Array Controller Card Front Panel

1.3.3 Interconnect Card

The interconnect card contains the interface circuitry and two connectors for interconnecting multiple arrays. It provides loop-switch capability and contains an environmental monitor for the array. Each array contains two interconnect cards for redundancy. For information on how to service the interconnect card, see Chapter 5.

RJ-45

10/100BASE-T

RJ-45 serial port

diagnostic connector

FIGURE 1-6 Interconnect Cards (Rear View)

Chapter 1 Sun StorEdge T3 and T3+ Array Overview 1-7

Interconnect card 2

Interconnect card 1

Page 24

1.3.4 Power and Cooling Unit

The power and cooling unit contains a power supply, two cooling fans, an integrated uninterruptible power source (UPS) battery, and LED status indicators for AC power. Each array contains two power and cooling units for redundancy. See Chapter 5 for information on how to service the power and cooling unit and for information on UPS battery maintenance.

Power and cooling unit 2

Power and cooling unit 1

FIGURE 1-7 Power and Cooling Units (Rear View)

1.4 Architecture

The Sun StorEdge T3 and T3+ arrays are uniquely designed to be modular, scalable, reliable, serviceable, high-performance storage building blocks with flexible configurations. The design allows for multiple arrays to be combined in various ways to provide complete storage solutions that are optimized for all applications—from transaction processing to decision support and highperformance computing, from workgroup environments to data centers.

The arrays have redundant components for high reliability. Components are hot swappable and field replaceable for serviceability. Controller units have a cached hardware RAID controller for high performance. Expansion units can be added to scale capacity. Controllers can be added to expansion units to scale performance. Controller units can be paired in an enterprise (partner group) configuration,

1-8 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 25

providing controller and data path redundancy and mirrored caches for high availability. Enterprise configurations support host-based alternate pathing for enhanced availability.

Data and administrative paths are completely independent for reliability, security, serviceability, and ease of use. The network-based administrative path allows for centralized configuration and monitoring of large numbers of Sun StorEdge T3 and T3+ array configurations, providing storage to multiple application servers.

Each disk drive has a drive label, a small portion of which is reserved for the system area. Approximately 150 Mbytes is reserved for the system area, which contains the configuration data, boot firmware, and file system information. The system area information is mirrored across all nine drives for redundancy so that data can be recovered from the other functional drives.

Finally, the unique switched-loop architecture provides the flexibility to configure multiple units together for scalability and availability, while enabling loops to be reconfigured dynamically for diagnosis and recovery in case of loop-related failures.

1.5 Supported Configurations

Two array configurations are supported:

■ Workgroup. This standalone array is a high-performance, high-RAS configuration

with a single hardware RAID cached controller. The unit is fully populated with redundant hot-swap components and nine disk drives.

FC-AL connection

Ethernet connection

LAN

FIGURE 1-8 Workgroup Configuration

Chapter 1 Sun StorEdge T3 and T3+ Array Overview 1-9

Application host

Management host

Ethernet port

Page 26

■ Enterprise. Also called a partner group, this is a configuration of two controller

units paired using interconnect cables for back-end data and administrative connections. The enterprise configuration provides all the RAS of single controller units, plus redundant hardware RAID controllers with mirrored caches, and redundant host channels for continuous data availability for host applications.

Alternate master unit

Ethernet

Interconnect cables

Master unit

FC-AL connection

FIGURE 1-9 Enterprise Configuration

connection

Ethernet connection

LAN

Application host

Host-bus adapters

Management host

Ethernet port

Note – Sun StorEdge T3 array configurations, both workgroup and enterprise,

require a media-interface adapter (MIA) connected to the Fibre Channel port in order to connect a fiber-optic cable. The MIA and FC-AL connection are detailed in Chapter 2. Sun StorEdge T3+ array configurations support direct FC-AL connections.

In an enterprise configuration, there is a master controller unit and an alternate master controller unit. In all default enterprise configurations, the master controller unit is the array positioned at the bottom of an array stack in either a rackmounted or tabletop installation. The alternate master controller unit is positioned on top of the master controller unit. The positioning of the master and alternate master controller units is important for cabling the units together correctly, understanding IP address assignments, interpreting array command-line screen output, and determining controller failover and failback conditions.

In this manual, the terms master controller unit and master unit are used interchangeably, as are the terms alternate master controller unit and alternate master unit.

1-10 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 27

CHAPTER

Installation

This chapter describes how to install Sun StorEdge T3 and T3+ arrays in either a workgroup (single) or enterprise (partner group) configuration.

Note – This chapter is for installing new array units only. To reconfigure existing

single array units into an enterprise configuration, contact your SunService™ representative.

Caution – In an enterprise configuration, interconnected arrays must be of the same

model type; that is, you can only connect two Sun StorEdge T3 arrays together or two Sun StorEdge T3+ arrays together. A mixed array partner group is not compatible.

Caution – For security reasons, connect the array to a private network only.

The following sections describe specific steps of the installation. Follow these directions in the order provided to ensure a successful installation.

■ “Preparing for the Installation” on page 2-2

■ “Inspecting the Array” on page 2-3

■ “Editing the Host Files” on page 2-4

■ “Installing the Array” on page 2-7

■ “Connecting the Cables” on page 2-8

■ “Powering On and Verifying the Hardware Configuration” on page 2-14

■ “Establishing a Network Connection” on page 2-16

■ “Verifying the Firmware Level and Configuration” on page 2-19

■ “Installing the Administration Tools” on page 2-25

■ “Defining and Mounting Volumes” on page 2-25

■ “Changing the Default Configuration” on page 2-32

■ “Connecting the Host System” on page 2-33

2-1

Page 28

2.1 Preparing for the Installation

Before you install the array, make sure you have completed the following tasks:

■ Prepare the area according to the guidelines and procedures in the Sun StorEdge

T3 and T3+ Array Site Preparation Guide.

■ Review safety requirements for the array installation in the Sun StorEdge T3 and

T3+ Array Regulatory and Safety Compliance Manual.

■ Obtain all the equipment necessary to complete the installation, such as

additional cabling or adapters. Refer to the Sun StorEdge T3 and T3+ Array Configuration Guide or contact a Sun sales representative to help you determine

your configuration needs for additional hardware and software.

The array is designed to be easily installed. A Phillips screwdriver is required for installing the array into a rack; a flat blade screwdriver is helpful for removing and replacing components.

Note – The array configuration requires an Ethernet connection for each controller

unit. The Sun StorEdge T3+ array includes an Ethernet cable (category 5) with each array for the 10/100BASE-T network connection. The Sun StorEdge T3 array requires a category 3 Ethernet network cable for the 10BASE-T network connection. This cable is not included with the Sun StorEdge T3 array ship kit and must be purchased separately.

The array uses nominal input voltages of 100–120 VAC or 200–240 VAC. Sun products are designed to work with single-phase power systems that have a grounded neutral conductor. See Appendix A for additional electrical specifications.

Caution – To reduce the risk of electrical shock, do not connect Sun products into

another type of power source. Contact your facilities manager or a qualified electrician if you are unsure what type of power is supplied to your building.

2-2 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 29

2.2 Inspecting the Array

Caution – This procedure requires two people to lift and move the array. Use care

to avoid injury. An array can weigh up to 67 pounds (30 kg).

1. Unpack the array.

You should have the following items:

■ Sun StorEdge T3 array or Sun StorEdge T3+ array

■ One 5m fiber-optic cable (unique to the array model)

■ Two power cords

The cables are illustrated in

2. Inspect the array for evidence of damage.

If the array is damaged, keep all contents and packing materials for the shipping company’s agent to inspect.

3. Save the packing materials for future use.

4. Remove the front panel of the array by pressing in on the side latches and pulling the cover off (

FIGURE 2-1).

FIGURE 2-3.

Latch

FIGURE 2-1 Removing the Front Panel

Latch

Chapter 2 Installation 2-3

Page 30

5. Locate the pull-out tab at the left side of the array next to the first disk drive (

FIGURE 2-2).

This tab contains the array serial number and media access control (MAC) address. Record this information; you will need the MAC address for the installation.

Pull-out tab

FIGURE 2-2 Serial Number and MAC Address on Pull-out Tab

6. Make sure that all the disk drives, the power and cooling units, the interconnect cards, and the controller card (if present) are firmly seated in the unit.

7. Replace the front panel.

Caution – You must replace the front panel for the array to meet regulatory

emissions requirements.

2.3 Editing the Host Files

The JumpStart™ feature automatically downloads the assigned IP address to the array. To enable this feature, you must edit your host file before cabling the array. After you cable the array and power on, the IP address is automatically assigned. Before you begin, make sure you have the following:

■ MAC address. See FIGURE 2-2 for the MAC address location. See Step 1 of the

following procedure for the format of the MAC address.

■ IP address. For this information, contact the person who maintains your network.

■ Array name. This is the user-assigned name of the array you are installing.

2-4 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 31

Note – If you are installing an enterprise configuration (partner group), you need

the MAC address that will be assigned only to the master controller unit (also called “master unit”). In a partner group, this is the array that is located on the bottom. In Step 1, enter the MAC address for the master unit. When you install the array in the next section, make sure that the master unit is placed on the bottom of the configuration.

Do not use the MAC address of the array that you are placing on the top. In a partner group, the array that is placed on top is referred to as the alternate master controller unit or alternate master unit.

To set the network IP address for the array:

1. On a host connected to the same subnet as the array, edit the /etc/ethers file by

adding the MAC address and array name.

For example:

8:0:20:7d:93:7e array-name

In this example:

■ 8:0:20:7d:93:7e is the MAC address.

■ array-name is the name of the array you are installing.

2. Edit the /etc/hosts file with the IP address and array name.

For example:

192.129.122.111 array-name

In this example: 192.129.122.111 is the assigned IP address.

3. Edit the /etc/nsswitch.conf file to reference the local system files.

To ensure that the Solaris software environment uses the changes made to the

/etc/ethers and /etc/hosts files, edit the host and ethers entries in the /etc/nsswitch.conf file so that the files parameter appears before the [NOTFOUND=return] statements.

hosts: nis files [NOTFOUND=return] ethers: nis files [NOTFOUND=return]

Chapter 2 Installation 2-5

Page 32

4. Determine if the reverse address resolution protocol (RARP) daemon is running by typing:

# ps -eaf | grep rarpd

■ If the RARP daemon is running, you must kill the process and restart it for the

host to recognize the changes you just made. To kill the process, type:

# kill process ID

The process ID was provided in the ps -eaf | grep rarpd command output.

■ If the RARP daemon is not running, proceed to the next step.

5. Start the RARP daemon in the Solaris software environment by typing:

# /usr/sbin/in.rarpd -a &

The IP address will automatically download to the array after you install the array and power on.

Note – In some cases, the array could time out before it receives the RARP request

through an Ethernet switch. If this happens, the array cannot receive the assigned IP address. If the array should time out before receiving the RARP request, it could be due to an improper spanning-tree setting of the Ethernet switch. Refer to your switch vendor documentation for information on spanning-tree settings and how to change them. Changing this setting properly will enable the array to receive the RARP request before timing out.

2-6 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 33

2.4 Installing the Array

This section describes the procedures for installing either a workgroup or enterprise configuration; see Chapter 1 for an explanation of each configuration. The array can either be placed on a tabletop or rackmounted in a server cabinet or expansion cabinet. This section provides guidelines for tabletop placement of an array configuration. To rackmount the array in a cabinet, you must order a Sun StorEdge T3 array rackmount kit. Instructions for rackmounting arrays are included in the kit.

The array is designed to sit on a desk or table. Use the following guidelines to prepare a location for your system.

■ Choose a desk or table that can support up to 67 pounds (30 kg) for one fully

configured array or 135 pounds (60 kg) for two arrays.

■ Leave enough space in front and in back of the array to access components.

■ Provide a minimum space of 6 inches (15 cm) in front and in back of the array for

adequate air flow.

■ Keep power and interface cables clear of foot traffic. Route cables inside walls,

under the floor, through the ceiling, or in protective channels. Route interface cables (excluding fiber-optic cables) away from motors and other sources of magnetic or radio frequency interference.

■ Make sure that the fiber-optic and power cable lengths do not exceed cable length

limitations. See Appendix A for cable lengths.

■ Ensure that the operating environment for the array does not exceed the

specifications. See Appendix A for environmental specifications.

Caution – Use two people to lift the array to avoid injury. It can weigh up to 67

pounds (30 kg).

1. Place the array horizontally in the designated location.

Caution – Do not place the array in a vertical position. It is not designed to stand

vertically.

2. If you are installing a second array, stack it on top of the first array.

The second array is the alternate master unit. The first array, or the bottom array in the stack, is the master unit.

Chapter 2 Installation 2-7

Page 34

Caution – Do not stack more than two arrays together. Stacking more than two

arrays could damage them.

You are now ready to connect the cables.

2.5 Connecting the Cables

This section explains how to connect cables for both a workgroup (standalone) and for an enterprise (partner group) configuration.

Note – If you are connecting a single array, skip the steps that start with “Partner

group only” and proceed to the next step.

Before you begin, make sure you have all of the required cables and adapters:

■ Fiber-optic cable, one per unit

■ AC power cords, two per unit

■ Interconnect cables, one pair of cables per partner group

■ Ethernet interface cable, 1 per controller unit:

■ 10BASE-T (category 3) interface cable for the Sun StorEdge T3 array (must be

purchased separately)

■ 10/100BASE-T (category 5) interface cable for the Sun StorEdge T3+ array

(included with array packaging)

■ Media interface adapter (MIA), required for Sun StorEdge T3 array only

Sun StorEdge T3 array cables and adapters are shown in T3+ array cables are shown in

FIGURE 2-4

FIGURE 2-3. Sun StorEdge

Note – Your array packaging might not have included interconnect cables. If you

want to install an enterprise configuration and do not have interconnect cables, contact your Sun sales representative.

2-8 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 35

Fiber-optic cable

MIA

Interconnect cable

AC power cord

FIGURE 2-3 Sun StorEdge T3 Array Cables and Adapters

SC connector

Fiber-optic cable

Interconnect cable

Ethernet cable (category 5)

FIGURE 2-4 Sun StorEdge T3+ Array Cables

LC-SFF connector

AC power cord

Chapter 2 Installation 2-9

Page 36

1. Connect the fiber-optic cable to the host bus adapter (HBA).

■ For a Sun StorEdge T3+ array, make sure to attach the standard connector (SC) on

the fiber-optic cable to the HBA.

■ For a Sun StorEdge T3 array, you can connect either end of the cable to the HBA.

2. Connect the other end of the fiber-optic cable to the FC-AL connector on the controller card at the back of the master (bottom) unit.

■ For a Sun StorEdge T3+ array, connect the LC-SFF (small form factor) end of the

cable to the FC-AL interface (

■ For a Sun StorEdge T3 array, use a media interface adapter (MIA) to connect the

fiber-optic cable to the FC-AL interface on the controller card (

FIGURE 2-5).

FIGURE 2-6).

3. Partner group only: Repeat Step 1 and Step 2 to connect a second fiber-optic cable to the alternate master (top) unit and to a separate HBA.

FC-AL connector

LC-SFF connector

Fiber-optic cable

FIGURE 2-5 Connecting the Fiber-Optic Cable to the FC-AL Connector on a

Sun StorEdge T3+ Array

2-10 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 37

FC-AL connector

MIA

FIGURE 2-6 Connecting the Fiber-Optic Cable and MIA to the FC-AL Connector on a

Fiber-optic cable

Sun StorEdge T3 Array

4. Attach one end of the Ethernet cable to the Ethernet port on the controller card of the master unit (

■ For Sun StorEdge T3+ arrays, use a category 5 Ethernet cable.

■ For Sun StorEdge T3 arrays, use a category 3 Ethernet cable.

FIGURE 2-7).

Ethernet port

FIGURE 2-7 Connecting the Ethernet Cable

Ethernet cable

Chapter 2 Installation 2-11

Page 38

5. Attach the other end of the Ethernet cable to a network hub or router.

Note – You must use a shielded Ethernet interface cable to comply with regulatory

emissions requirements.

6. Partner group only: Repeat Step 4 and Step 5 to connect the alternate master unit to a hub or router using a second Ethernet cable.

Connect this cable to a hub or router on the same network as the master unit.

7. Attach the two power cords to the two power and cooling units on the back of the master unit (

FIGURE 2-8).

AC power cords

FIGURE 2-8 Connecting the Power Cords

8. Partner group only: Repeat Step 7 to connect the power cords to the alternate master unit.

Caution – Do not power on the units yet. If you are connecting a single array unit,

proceed to Section 2.6, “Powering On and Verifying the Hardware Configuration” on page 2-14 for instructions on powering on the unit. If you are connecting a partner group, proceed to Step 9.

2-12 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 39

9. Partner group only: Connect the interconnect cables to the interconnect cards as shown in

FIGURE 2-9.

Make sure you connect the cables to the correct interconnect card connectors exactly as shown in the figure. This cable connection determines the master and alternate master relationship. Tighten the retaining screws.

The remaining connectors are reserved for expansion units.

Alternate master controller unit

Master controller unit

FIGURE 2-9 Connecting the Interconnect Cables

Chapter 2 Installation 2-13

Page 40

A fully cabled partner group is shown in FIGURE 2-10.

Power connectors (cords not shown)

Alternate master controller unit

Interconnect cables

Application host

Master controller unit

FC-AL connections

Ethernet connections

FIGURE 2-10 Fully Cabled Partner Group (Enterprise Configuration)

2.6 Powering On and Verifying the Hardware Configuration

To power on and verify the hardware configuration:

1. Connect the power cords from each array into AC outlets.

Use alternate power sources to make sure that power and cooling unit features are redundant.

HBAs

Ethernet port

Management host

2. Verify that AC power is present on each power and cooling unit.

When AC power is present, the AC LED on each power and cooling unit is solid amber and the fans will turn at low speed.

2-14 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 41

3. Press the power button on each power and cooling unit on all arrays to power on the units.

When the arrays are powered on, the AC LED and power supply (PS) LED on the power and cooling units are green.

AC LED Power button PS LED

FIGURE 2-11 Power and Cooling Unit LEDs and Power Button Location

4. Check the LEDs at the front and back of the each array to ensure that all components are receiving power and are functional.

While the drives are spinning up, the LEDs will blink. The array boot time might take several minutes, after which all LEDs should be solid green, indicating the unit is receiving power and there is no drive activity.

For more information on the LEDs and how to interpret them, see Section 4.1.4, “Array LEDs” on page 4-3.

Note – The batteries in the power and cooling units recharge after you power on the

unit. While the batteries are recharging, write-behind cache is disabled.

Chapter 2 Installation 2-15

Page 42

2.7 Establishing a Network Connection

Use the Ethernet network connection to initialize the network addresses and to administer the array. The IP address was automatically downloaded using the JumpStart program when you powered on the array.

Note – Have the gateway IP address information before beginning this procedure.

Contact the person who maintains your network for this information.

Note – The telnet session will not connect until the controller LEDs are green.

1. On the host, use the telnet command with the array name (or IP address) to

connect to the array.

For example:

# telnet array-name Trying 129.150.47.101... Connected to 129.150.47.101. Escape character is ’^]’.

pSOSystem (129.150.47.101)

Note – The telnet session will verify whether your network connection is good. If

you cannot connect through the telnet session, see the troubleshooting information in the Sun StorEdge T3 and T3+ Array Administrator’s Guide or the Sun StorEdge Component Manager User’s Guide to analyze the problem.

2. Type root as the login and press Return at the password prompt.

The array displays the banner information, which appears similar to the following:

:/:<1>

2-16 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 43

3. Use the passwd command to set a password for the root account.

Press Return when prompted for the OLD password.

:/:<1>passwd OLD password: <Return> NEW password: New password NEW password (confirm): New password :/:<2>

Note – It is important to set a root password for security reasons.

4. Set the gateway using the set gateway command.

The gateway address enables you to access the array outside the subnet.

For example:

:/:<3>set gateway 129.150.47.1

5. Set the netmask using the set netmask command.

The netmask specifies the network mask used to implement IP subnetting.

:/:<4>set netmask 255.255.255.0

6. Set the host name using the set hostname command.

:/:<5>set hostname hostname

7. Use the tzset command to set the time zone and confirm the setting.

The time zone is off-set from Greenwich mean time (GMT), also known as universal time coordinated (UTC). The numerical value used with the tzset command is the difference in the number of hours between your time zone and the GMT, based on international standardized time zone designations.

For example, if you are located in the Pacific Standard Time (PST) time zone, the difference would be minus (-) eight hours (0800) from GMT as shown:

:/:<6>tzset -0800 :/:<7>tzset GMT-0800 :/:<8>

Chapter 2 Installation 2-17

Page 44

8. Set the date using the date command.

The date syntax is yyyymmddHHMM.SS.

For example:

:/:<9>date 200005171620.30 Wed May 17 16:20:30 2000 :/:<10>

9. Reset the array using the reset command.

Answer y when prompted for confirmation. A reset disconnects the telnet session with the array.

:/:<110>reset Reset the system, are you sure? [N]: Y

The array reboots. This can take up to several minutes.

10. On the host, exit the telnet session by pressing Control and the right bracket (])

symbol.

The telnet prompt is displayed. Type quit.

# <Control>] telnet>quit #

2-18 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 45

2.8 Verifying the Firmware Level and Configuration

The array contains various types of firmware that are particular to the array model.

The Sun StorEdge T3 array contains four types of firmware:

■ Controller firmware

■ Controller erasable programmable read-only memory (EPROM) firmware

■ Disk drive firmware

■ Interconnect card firmware

The Sun StorEdge T3+ array contains the following firmware:

■ Controller firmware:

■ Level 1

■ Level 2

■ Level 3

■ Disk drive firmware

■ Interconnect card firmware

Use this procedure to make sure you are running the latest firmware versions on your particular array model, and that the array configuration information indicates that the unit is ready for operation. You can check the firmware versions and array information in a telnet session with the array.

Note – Verification of level 1 firmware on a Sun StorEdge T3+ array is not required

at this time. Access to level 1 firmware operations is limited to qualified service providers.

1. On the host, use the telnet command with the array name (or IP address) to

connect to the array.

For example:

# telnet array-name Trying 129.150.47.101... Connected to 129.150.47.101. Escape character is ’^]’.

pSOSystem (129.150.47.101)

Chapter 2 Installation 2-19

Page 46

2. Log in to the array by typing root and your root password at the prompts.

The array prompt is displayed.

3. Type ver to display the header information:

■ An example of header information for a Sun StorEdge T3 array:

:/:<5>ver

In this example Sun StorEdge T3 array controller firmware is listed as

Release 1.17b.

■ An example of header information for a Sun StorEdge T3+ array:

:/:<5>ver

In this example, the Sun StorEdge T3+ array is shown running 2.00 level 3 controller firmware.

2-20 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 47

4. Type fru list to display the firmware versions for the disk drives and

interconnect card on both array models.

For example (enterprise configuration):

:/:<6>fru list ID TYPE VENDOR MODEL REVISION SERIAL

------ ----------------- ----------- ----------- ------------- -------u1ctr controller card 0034 501-5710-02( 0200 123456 u2ctr controller card 0034 501-5710-02( 0200 123455 u1d1 disk drive SEAGATE ST118202FSUN 9329 LK478728 u1d2 disk drive SEAGATE ST118202FSUN 9329 LK493799 u1d3 disk drive SEAGATE ST118202FSUN 9329 LK493800 u1d4 disk drive SEAGATE ST118202FSUN 9329 LK494457 u1d5 disk drive SEAGATE ST118202FSUN 9329 NK040486 u1d6 disk drive SEAGATE ST118202FSUN 9329 LK339351 u1d7 disk drive SEAGATE ST118202FSUN 9329 LK150715 u1d8 disk drive SEAGATE ST118202FSUN 9329 NK040157 u1d9 disk drive SEAGATE ST118202FSUN 9329 LKG79907 u2d1 disk drive SEAGATE ST118202FSUN 9329 LKG90019 u2d2 disk drive SEAGATE ST118202FSUN 9329 LKH18597 u2d3 disk drive SEAGATE ST118202FSUN 9329 LKH15606 u2d4 disk drive SEAGATE ST118202FSUN 9329 LKH16563 u2d5 disk drive SEAGATE ST118202FSUN 9329 LKG88883 u2d6 disk drive SEAGATE ST118202FSUN 9329 LKH61907 u2d7 disk drive SEAGATE ST118202FSUN 9329 LKG90719 u2d8 disk drive SEAGATE ST118202FSUN 9329 LKG95442 u2d9 disk drive SEAGATE ST118202FSUN EA29 LKG61774 u1l1 loop card SCI-SJ 375-0085-01- 5.03 Flash 007924 u1l2 loop card SCI-SJ 375-0085-01- 5.02 Flash 007382 u2l1 loop card SCI-SJ 375-0085-01- 5.02 Flash 003343 u2l2 loop card SCI-SJ 375-0085-01- 5.02 Flash 003344 u1pcu1 power/cooling unit TECTROL-CAN 300-1454-01( 0000 001455 u1pcu2 power/cooling unit TECTROL-CAN 300-1454-01( 0000 001408 u2pcu1 power/cooling unit TECTROL-CAN 300-1454-01( 0000 001686 u2pcu2 power/cooling unit TECTROL-CAN 300-1454-01( 0000 001445 u1mpn mid plane SCI-SJ 375-0084-01- 0000 000650 u2mpn mid plane SCI-SJ 375-0084-01- 0000 000649

In this example:

■ Disk drive firmware version is listed as Revision 9329

■ Interconnect card (loop card) firmware version is listed as Revision 5.03

Flash

■ Sun StorEdge T3 array controller EPROM firmware is listed as Revision 0200.

■ Sun StorEdge T3+ array level 2 controller firmware, the level 2 image is 0200.

Chapter 2 Installation 2-21

Page 48

5. Refer to the SunSolveSMweb site at http://sunsolve.sun.com for the latest firmware release information:

a. In the SunSolve Online column, click on Patches.

b. In the Patches column, click on PatchPro.

c. Click on Storage Products.

Refer to the patch README file for details on firmware and specific firmware patch requirements.

Note – If you do not have access to the SunSolve web site, contact your SunService

provider for patch information.

6. If you need to upgrade the firmware, do so now by following the instructions in Section 5.3, “Upgrading Firmware” on page 5-16.

If you do not need to upgrade firmware, proceed to Step 7.

7. Type port list to display how the paths are mapped from the host to the volumes.

An example of port list output for a single controller unit:

:/:<7>port list

port targetid addr_type status host wwn u1p1 1 hard online sun 50020f23000058de

Use the world wide name (wwn) from the port list output to map the volume to the Solaris environment using the format(1M) utility on the data host. In the following example, the output for the format command would show the same wwn, listed in the last line, as the wwn in the port list output.

# format Searching for disks...done

AVAILABLE DISK SELECTIONS:

0. c0t0d0 <SUN4.2G cyl 3880 alt 2 hd 16 sec 135> /pci@1f,4000/scsi@3/sd@0,0

1. c2t1d0 <SUN-T300-0113 cyl 34145 alt 2 hd 64 sec 128> /pci@6,2000/SUNW,ifp@1/ssd@w50020f23000002ba,0

2-22 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 49

8. Type sys stat to view the controller status.

■ An example of sys stat output for a single controller unit:

:/:<9>sys stat Unit State Role Partner

----- --------- ------ ------ 1 ONLINE Master

■ An example of sys stat output for a partner group:

:/:<10>sys stat Unit State Role Partner

----- --------- ------ ------ 1 ONLINE Master 2 2 ONLINE AlterM 1

9. Type vol mode to display the cache and mirror settings and to determine whether

the volume is mounted.

The default settings for a single controller unit are as follows:

:/:<11>vol mode

volume mounted cache mirror v0 yes writebehind off

Chapter 2 Installation 2-23

Page 50

10. Type fru stat to display the health status of each FRU.

All FRUs must be in optimal condition, as shown in this example for a workgroup configuration:

:/:<12>fru stat CTLR STATUS STATE ROLE PARTNER TEMP

------ ------- ---------- ---------- ------- ---u1ctr ready enabled master - 29.0

DISK STATUS STATE ROLE PORT1 PORT2 TEMP VOLUME

------ ------- ---------- ---------- --------- --------- ---- -----u1d1 ready enabled data disk ready ready 30 v0 u1d2 ready enabled data disk ready ready 30 v0 u1d3 ready enabled data disk ready ready 34 v0 u1d4 ready enabled data disk ready ready 34 v0 u1d5 ready enabled data disk ready ready 32 v0 u1d6 ready enabled data disk ready ready 36 v0 u1d7 ready enabled data disk ready ready 37 v0 u1d8 ready enabled data disk ready ready 41 v0 u1d9 ready enabled standby ready ready 34 v0

LOOP STATUS STATE MODE CABLE1 CABLE2 TEMP

------ ------- ---------- ------- --------- --------- ---u1l1 ready enabled master - - 29.5 u1l2 ready enabled slave - - 30.0

POWER STATUS STATE SOURCE OUTPUT BATTERY TEMP FAN1 FAN2

------ ------- --------- ------ ------ ------- ------ ------ -----u1pcu1 ready enabled line normal normal normal normal normal u1pcu2 ready enabled line normal normal normal normal normal

Proceed to Section 2.9, “Installing the Administration Tools” on page 2-25 to continue with the installation.

2-24 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 51

2.9 Installing the Administration Tools

The Sun StorEdge T3 and T3+ arrays can be used with management software, such as Sun StorEdge Component Manager, to aid in their operation. Sun StorEdge Component Manager provides a graphical user interface (GUI) to monitor and manage one or more arrays that are connected to a host.

Sun StorEdge Component Manager software enables you to administer some of the components of the array, and it constantly monitors system health. Alarm notification and remote reporting alert the administrator to abnormal activities or conditions that require attention.

The Sun StorEdge Component Manager Installation Guide provides instructions for installing the software and covers the necessary steps to verify the installation, launch the software, and uninstall the software. The Sun StorEdge Component Manager User’s Guide describes how to use the GUI to administer Sun StorEdge T3 and T3+ arrays.

2.10 Defining and Mounting Volumes

This section includes information on defining and mounting volumes for both a workgroup and enterprise configuration. For information on the commands used in this section, refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide.

The array has been set up with a default RAID 5 configuration. To view the configuration, use the vol list command while in a telnet session with the array. For example:

<1>vol list

:/:

volume capacity raid data standby v0 143.2 GB 5 u1d1-9 none

In this example:

■ v0 is the volume.

■ 143.2 GB is the total disk space of the volume.

■ 5 is the RAID level.

■ u1d1-9 is the number of the array (u1) and the numbers of the disk drives (d1

through d9).

■ none indicates that there is no standby (hot spare) disk drive.

Chapter 2 Installation 2-25

Page 52

The physical location of the disk drives is shown in FIGURE 2-12.

d1 d2 d3 d4 d5 d6 d7 d8 d9

FIGURE 2-12 Disk Drive Physical Numbering

■ If you want to change the default configuration, do not continue with this section.

Instead, refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide for information on how to remove the default volume and reconfigure the disk drives.

■ If you have a workgroup configuration and want to mount the default volume,

proceed to Section 2.10.1, “Workgroup Configuration” on page 2-27.

■ If you have an enterprise configuration and want to mount the default volumes,

skip to Section 2.10.2, “Enterprise Configuration” on page 2-29.

Note – If you are creating new volumes or changing the volume configuration, you

must first manually rewrite the label of the previous volume using the

autoconfigure option of the format(1M) utility. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide for more information on this procedure.

Note – The default configuration does not include a hot spare. If you want to add a

hot spare to the configuration, you must remove the existing volume and re-create the configuration. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide for information on adding a hot spare.

2-26 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 53

2.10.1 Workgroup Configuration

To display and mount the default volume of a single controller unit:

1. On the array, type vol stat to check the status of the volume.

:/:<2> vol stat v0 u1d1 u1d2 u1d3 u1d4 u1d5 u1d6 u1d7 u1d8 u1d9 unmounted 0 0 0 0 0 0 0 0 0

Where:

■ v0 is the name of the volume.

■ unmounted is the status of the volume.

■ u1d1–u1d9 are the disk drives in the volume.

■ 0 is the status of the disk drives in the volume.

All the disk drives should show a status of 0. Other possible status messages are listed in the following table.

TABLE 2-1 Drive Status Messages

Value Description

0 Drive mounted

2 Drive present

3 Drive is spun up

4 Drive is disabled

5 Drive has been replaced

7 Invalid system area on drive

9 Drive not present

D Drive is disabled and is being reconstructed

S Drive substituted

■ If the vol stat output shows the volume as mounted, skip to Step 4.

■ If the vol stat output shows the volume as unmounted, proceed to Step 2.

2. Use the vol mount command to mount the volume.

:/:<3>vol mount volume-name

Chapter 2 Installation 2-27

Page 54

3. Use the vol stat command to verify that the volume is mounted.

:/:<5> vol stat v0 u1d1 u1d2 u1d3 u1d4 u1d5 u1d6 u1d7 u1d8 u1d9 mounted 0 0 0 0 0 0 0 0 0

Once the volume is mounted, it becomes available to the attached host system.

Note – If you are running on the Solaris 7, 11/99 operating environment or a later

release of the Solaris operating environment, skip Step 4 and proceed to Step 5. Later versions of the Solaris operating environment automatically recognize added storage devices without additional command operations.

4. Use the luxadm(1M) command to recognize the new volume.

The luxadm(1M) command probes for new devices. Refer to the luxadm(1M) man page for more information on this command.

a. Make sure there is a /dev/es directory on the host system. If not, type:

# mkdir /dev/es

The /dev/es directory is necessary for running the luxadm command.

b. On the host system, type luxadm insert:

# luxadm insert

Note – If the luxadm utility is not available, you must do a reconfiguration reboot

(boot -r) to ensure that the host recognizes the new volumes. It is preferable, however, to use the luxadm command for this procedure instead of the boot -r command.

2-28 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 55

5. Verify that the new volume is available to the host system by using the

format(1M) command on the data host.

7# format Searching for disks...done

AVAILABLE DISK SELECTIONS:

0. c0t0d0 <SUN4.2G cyl 3880 alt 2 hd 16 sec 135> /pci@1f,4000/scsi@3/sd@0,0

1. c2t1d0 <SUN-T300-0113 cyl 34145 alt 2 hd 64 sec 128> /pci@6,2000/SUNW,ifp@1/ssd@w50020f23000002ba,0

Proceed to Section 2.12, “Connecting the Host System” on page 2-33 to continue the installation.

2.10.2 Enterprise Configuration

To define and mount the default volumes of an enterprise configuration, you must first make sure the volume on the master unit is mounted, and then add a volume on the alternate master as follows:

1. On the array, type vol stat to check the status of the volume.

Only the volume on the master controller unit is displayed.

:/:<1>vol stat v0 u1d1 u1d2 u1d3 u1d4 u1d5 u1d6 u1d7 u1d8 u1d9 unmounted 0 0 0 0 0 0 0 0 0

Where:

■ v0 is the name of the volume.

■ unmounted is the status of the volume.

■ u1d1–u1d9 are the disk drives in the volume.

■ 0 is the status of the disk drives in the volume.

All the disk drives should show a status of 0. Possible status messages are listed in

TABLE 2-1.

■ If the vol stat output shows the volume as mounted, skip to Step 4.

■ If the vol stat output shows the volume as unmounted, proceed to Step 2.

Chapter 2 Installation 2-29

Page 56

2. Use the vol mount command to mount the volume.

:/:<2>vol mount v0

3. Type vol stat to verify that the volume on the master controller unit is

mounted.

:/:<3> vol stat v0 u1d1 u1d2 u1d3 u1d4 u1d5 u1d6 u1d7 u1d8 u1d9 mounted 0 0 0 0 0 0 0 0 0

4. Use the vol add command to create the default volume on the alternate master as

follows:

a. Define the volume name (vol add v1).

b. Define the drives where the volume will reside (data u2d1-9).

c. Define the RAID level (raid 5).

:/:<4>vol add v1 data u2d1-9 raid 5

Where:

■ v1 is the volume name.

■ u2d1-9 indicates the location of the volume: unit 2, disk drives 1 through 9.

■ raid 5 is RAID level 5.

5. Type vol stat to check the status of the volumes.

The status of all drives must be 0. For example:

:/:<5>vol stat

v0 u1d1 u1d2 u1d3 u1d4 u1d5 u1d6 u1d7 u1d8 u1d9 mounted 0 0 0 0 0 0 0 0 0 v1 u2d1 u2d2 u2d3 u2d4 u2d5 u2d6 u2d7 u2d8 u2d9 unmounted 0 0 0 0 0 0 0 0 0

2-30 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 57

6. Use the vol init command to initialize the volume on the alternate master.

:/:<6>vol init v1 data

Depending on system activity at the time of initialization, it can take up to an hour to initialize a volume. Only one volume can be initialized at a time.

7. Use the vol mount command to mount the volume on the alternate master.

:/:<7>vol mount v1

8. Type vol list to confirm that you created the volume correctly.

:/:<8>vol list

volume capacity raid data standby v0 143.2 GB 5 u1d1-9 none v1 143.2 GB 5 u2d1-9 none

Note – If you are running on the Solaris 7, 11/99 operating environment or a later

release of the Solaris operating environment, skip Step 9 and proceed to Step 10. Later versions of the Solaris operating environment automatically recognize added storage devices without additional command operations.

9. Use the luxadm(1M) command to recognize the new volume.

The luxadm(1M) command probes for new devices. Refer to the luxadm(1M) man page for more information on this command.

a. Make sure there is a /dev/es directory on the host system. If not, type:

# mkdir /dev/es

The /dev/es directory is necessary for running the luxadm command.

b. On the host system, type luxadm insert:

# luxadm insert

Chapter 2 Installation 2-31

Page 58

Note – If the luxadm utility is not available, you must do a reconfiguration reboot

(boot -r) to ensure that the host recognizes the new volumes. It is preferable, however, to use the luxadm command for this procedure instead of the boot -r command.

10. Verify that the new volume is available to the host system by using the

format(1M) command on the data host.

# format Searching for disks...done

AVAILABLE DISK SELECTIONS:

0. c0t0d0 <SUN4.2G cyl 3880 alt 2 hd 16 sec 135> /pci@1f,4000/scsi@3/sd@0,0

1. c2t1d0 <SUN-T300-0113 cyl 34145 alt 2 hd 64 sec 128> /pci@6,2000/SUNW,ifp@1/ssd@w50020f23000002ba,0

2. c1t1d0 <SUN-T300-0113 cyl 34145 alt 2 hd 64 sec 128>

/pci@6,0/SUNW,socal@1,0/sf@0,0/ssd@w50020f2300000121,0

2.11 Changing the Default Configuration

The default configuration for the Sun StorEdge T3 and T3+ array is a RAID 5 volume.

■ If you want to change the default configuration, do so now before continuing the

installation process. Configuration options for the array are discussed in Chapter 3. The command-line procedures for reconfiguring the array are in the Sun StorEdge T3 and T3+ Array Administrator’s Guide. Once you have reconfigured the array, return to this chapter and proceed to Section 2.12, “Connecting the Host System” on page 2-33.

■ If you are not changing the default configuration, proceed to Section 2.12,

“Connecting the Host System” on page 2-33.

2-32 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 59

2.12 Connecting the Host System

This section contains procedures that are performed on the host system for the array. Have the documentation for the host available to assist you in completing these tasks.

■ 2.12.1 "Verifying the Data Host Connection”

■ 2.12.2 "Establishing Logical Volumes on the Host”

■ 2.12.3 "Creating Soft Partitions on the Data Host”

■ 2.12.4 "Establishing Alternate Pathing on the Host”

2.12.1 Verifying the Data Host Connection

Once the physical connection between the host and the array has been established, you can use a utility, such as the format(1M) command in the Solaris environment, to verify the existence of a volume on the array. For example:

# format Searching for disks...done

c1t1d0: configured with capacity of 133.38GB

AVAILABLE DISK SELECTIONS:

0. c0t2d0 <drive type unknown> /sbus@1f,0/SUNW,fas@e,8800000/sd@2,0

1. c0t3d0 <SUN2.1G cyl 2733 alt 2 hd 19 sec 80> /sbus@1f,0/SUNW,fas@e,8800000/sd@3,0

2. c1t1d0 <SUN-T300-0113 cyl 34145 alt 2 hd 64 sec 128> /sbus@1f,0/SUNW,socal@1,0/sf@0,0/ssd@w50020f2300000121,0

Specify disk (enter its number):

In this example, device number 2 is the array, as identified by the SUN-T300-0113 label.

Chapter 2 Installation 2-33

Page 60

Note – If you are creating new volumes or changing the volume configuration, you

must first manually rewrite the label of the previous volume using the

autoconfigure option of the format(1M) utility. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide for more information on this procedure.

2.12.2 Establishing Logical Volumes on the Host

Using the format(1M) utility within the Solaris environment, you can view the disk space available on the array from the host. At this point in the installation process, you can use this disk space as you would any physical disk. This includes performing operations such as installing a file system, using the device as a raw device, or partitioning the device. Refer to the data host documentation for more information on establishing logical volumes.

Note – Two commands are available to determine the drive volume capacity of the

array. Within the array, use the vol command to display the raw volume capacity. Within the Solaris environment, use the format(1M) utility (among others) to display the capacity of a newly created volume. The volume capacities differ between the internal array vol command and the Solaris environment format(1M) command; the Solaris format(1M) command reports a smaller volume capacity. This is due to the Solaris environment reporting volume sizes using 1,024 bytes per Kbyte capacities.

For example, using the array vol command, a 4-drive, 18-Gbyte RAID 0 volume would report a 71.6-Gbyte capacity within the array. This is calculated as follows:

139857920 blks * 512 bytes/blk * 1 Gbyte/1000

The format(1M) command in the Solaris environment reports a 66.69 Gbyte capacity for the same volume using 1,024 bytes per Kbyte, calculated as follows:

139857920 blks * 512 bytes/blk * 1 Gbyte/1024

There is no loss of actual disk capacity.

2-34 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

= 71.6 Gbytes

= 66.69 Gbytes

Page 61

2.12.3 Creating Soft Partitions on the Data Host

The array’s native volume management can support a maximum of two volumes or logical unit numbers (LUNs) per array unit. This can result in very large volumes (128-Gbytes in a configuration of single 7+1 RAID 5 LUN plus hot spare, with 18-Gbyte drives). Some applications cannot use such large volumes effectively. There are two solutions that can be used separately or in combination.

■ First, use the partitioning utility available on the data host’s operating system. In

the Solaris environment, this is the format utility, which can create up to seven distinct partitions per volume. Note that in the case of the configuration described above, if each partition is equal in size, this will result in 18-Gbyte partitions, which still may be too large to be used efficiently by legacy applications.

■ Second, you can use third-party software on the host to create as many partitions

as desired from a given volume. In the Solaris environment, VERITAS Volume Manager can be used for this purpose.

For information on using the format utility, refer to the format(1M) man page. For more information on third-party software such as VERITAS Volume Manager, refer to the documentation for that product.

Note – This completes the installation process for an array workgroup installation.

If you are installing an enterprise configuration, you must continue to the next section.

2.12.4 Establishing Alternate Pathing on the Host

This task applies only when two controller units are configured as a partner group in an enterprise configuration.

In a partner group, controller caches are mirrored so that if one controller fails, no data is lost. Any data that is in the cache of the failed controller, but is not yet written to disk, is preserved in the cache of the partner controller and is written to disk by means of one of the back-end Fibre Channel loops connecting the two units.

The back-end Fibre Channel loops are used to maintain host access to the disks of the unit with the failed controller (or any failure in the host data path to that controller). This requires the use of a host-based alternate pathing mechanism such as the Dynamic Multi-Pathing (DMP) feature of VERITAS Volume Manager, Sun Enterprise Alternate Pathing software, or Sun StorEdge Traffic Manager software. When a failure occurs in the host data path, the Alternate Pathing mechanism reroutes data to the other array controller in the partner group. The controller then redirects the data across one of the back-end loops to the disks of the other controller, thus maintaining data availability to the host application.

Chapter 2 Installation 2-35

Page 62

For more information and for instructions on installing and configuring Alternate Pathing software, refer to the documentation for the Alternate Pathing tool selected.

For the Alternate Pathing software to function properly, you must enable multipathing support on the array for enterprise configurations. To do so, use the following procedure:

1. If you are not already in a telnet session with the array, start one.

If you are in a telnet session, skip to Step 2.

a. On the host, use the telnet command with the array name (or IP address) to

connect to the array.

# telnet array-name Trying 129.150.47.101... Connected to 129.150.47.101. Escape character is ’^]’.

pSOSystem (129.150.47.101)

b. Log in to the array by typing root and your password at the prompts.

2. At the array prompt, enable Alternate Pathing software as follows:

■ For hosts running Sun StorEdge Traffic Manager Alternate Pathing software, type:

:/:<1>sys mp_support mpxio

■ For hosts running other Alternate Pathing software, type:

:/:<1>sys mp_support rw

3. Use the sys list command to verify that the mirror setting is set to auto.

:/:<2>sys list blocksize : 64k cache : auto mirror : auto mp_support : rw <or mpxio> naca : off rd_ahead : on recon_rate : med sys memsize : 128 MBytes cache memsize : 1024 MBytes

2-36 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 63

Note – For data hosts running Solaris operating environments earlier than Solaris 7,

11/99 or later, run the luxadm insert command for the host to recognize the new paths.

4. Type port list to display the world wide name (wwn).

An example of port list output for a partner group:

:/:<7>port list

port targetid addr_type status host wwn u1p1 1 hard online sun 50020f23000003c3 u2p1 2 hard online sun 50020f2300000341

5. Use the wwn from the port list output to correlate the volume to the Solaris environment by using the format(1M) utility on the data host.

In the following example, the output for the format command would show the same wwn, listed in the second to last line (u1) and last line (u2), as the wwn in the port list output.

format

Searching for disks...done

AVAILABLE DISK SELECTIONS:

0. c0t0d0 <SUN4.2G cyl 3880 alt 2 hd 16 sec 135> /pci@1f,4000/scsi@3/sd@0,0

1. c1t0d0 <SUN-T300-0113 cyl 34145 alt 2 hd 64 sec 128> /pci@6,2000/SUNW,ifp@1/ssd@w50020f23000003c3,0

2. c2t1d0 <SUN-T300-0113 cyl 34145 alt 2 hd 64 sec 128> /pci@6,2000/SUNW,ifp@1/ssd@w50020f2300000341,0

6. On the host, exit the telnet session by pressing Control and the right bracket (])

symbol (Solaris operating environment only) or using the equivalent command for your operating system.

This command takes you to the telnet prompt; type quit.

Note – After multipathing support is enabled and the host is rebooted, from the

host side you will see two paths available to the same volume: a primary path and an alternate path. This means that if you are using the format(1M) utility in the Solaris environment, you will see two paths for each volume.

Chapter 2 Installation 2-37

Page 64

This completes the installation process for an enterprise configuration. For information on available array settings and monitoring options, see Chapter 3.

2-38 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 65

CHAPTER

Operation

This chapter describes the basic operations that can be performed on Sun StorEdge T3 and T3+ arrays:

■ “Powering Off and On” on page 3-1

■ “Reconfiguring the Array Settings” on page 3-3

■ “Monitoring the Array” on page 3-9

3.1 Powering Off and On

You do not need to power off the array except under the following circumstances:

■ Hardware reconfiguration

■ Relocation

■ Centerplane and chassis replacement

■ Emergency, such as a threat of flooding

3.1.1 Powering Off

Caution – This procedure will power off the array and all arrays connected to it.

Note – If the array is operating in the Solaris environment and there are volumes

mounted, unmount the volumes on the data host before powering off the array.

1. Ensure that the data path between the host and the array has been quiesced.

There must not be any I/O activity.

3-1

Page 66

2. In a telnet session with the array, type:

:/:<1>shutdown Shutdown the system, are you sure? [N]: y

3. Press the power button once on each power and cooling unit to turn the switch off (

FIGURE 3-1).

Power switches

FIGURE 3-1 Power Switch Locations

3.1.2 Powering On

When powering on a configuration, always power on the arrays before booting the host server.

● Power on the master unit by pressing the power button on each power and

cooling unit.

At this point, all power and cooling unit LEDs on the array and any units connected to it will turn green, indicating that power has been restored.

Note – If the green power and cooling unit LEDs on connected units do not light,

press the power switches on those units.

3-2 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 67

3.2 Reconfiguring the Array Settings

The array is set with the following default configuration:

■ RAID 5 volume; no hot spare

■ Cache mode set to auto

■ Autodisable and autoreconstruction of disk drives

These default settings and other possible configurations are described in the following sections. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide for instructions on how to change the array default settings.

Some of the terminology used in this chapter is defined as follows:

■ Volume, also known as a logical unit number (LUN), is the basic structure you

create across the drive group to retrieve and restore data. A volume is a set of physical drives that has been configured into RAID level 0, 1, or 5.

■ Hot spare is a drive containing no data that acts as a standby in case another drive

fails in a RAID 1 or 5 volume. The hot spare drive adds another level of redundancy: if a drive fails, the hot spare takes over for the failed drive until you replace it. When you replace the failed drive, data from the hot spare is copied back to the replaced drive and the hot spare returns to a standby status.

■ Parity is additional information stored along with the data that enables the

controller to rebuild data after a drive failure. Only RAID 5 volumes have parity. RAID 1 uses mirroring, not parity, which enables the controller to reconstruct data after a drive failure.

3.2.1 Unit Volume Configurations

The drives in an array unit can be grouped into one or two volumes (LUNs). The minimum number of drives to compose a volume is two; the maximum is nine. A hot spare drive can be configured in the unit as an option. If you are configuring a hot spare in the volume, you can use only drive 9 in the array as a hot spare. You can have only one hot spare per array, which can be shared by two volumes.

If you choose to include a hot spare disk drive in your configuration, add it when you initially create the volume. If you want to add a hot spare at a later date, you must remove the existing volume and re-create the configuration.

You can also configure the two volumes into smaller partitions or combine multiple volumes into a single volume by using specialized third-party software such as VERITAS Volume Manager.

Chapter 3 Operation 3-3

Page 68

Note – If you are creating new volumes or changing the volume configuration, you

must first manually rewrite the label of the previous volume by using the

autoconfigure option of the format(1M) utility. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide for more information on this procedure.

Examples of valid volume configurations are shown in

TABLE 3-1 Volume Configuration Examples

Volume 1 Volume 2 Hot spare

9 disk RAID 5 None

8 disk RAID 5 None X

9 disk RAID 1 None

8 disk RAID 1 None X

2 disk RAID 1 7 disk RAID 5

2 disk RAID 1 6 disk RAID 5 X

2 disk RAID 1 7 disk RAID 1

2 disk RAID 1 6 disk RAID 1 X

4 disk RAID 0 5 disk RAID 1

2 disk RAID 0 6 disk RAID 5 X

3.2.2 RAID Levels

The RAID level determines how the controller reads and writes data and parity on the drives. The Sun StorEdge T3 and T3+ array can be configured with RAID level 0, RAID level 1 (1+0), or RAID level 5. The default configuration is a RAID level 5.

TABLE 3-1.

The following sections describe the data layout for each of the three RAID levels.

3-4 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 69

3.2.2.1 RAID 0

Data blocks in a RAID 0 volume are striped across all the drives in the volume in order. There is no parity data, so RAID 0 uses the full capacity of the drives. There is no redundancy; however, if a single drive fails, all data on the volume is lost.

TABLE 3-2 shows the data layout in a RAID 0 volume.

TABLE 3-2 RAID 0 Data Blocks

Stripe Drive1 Drive2 Drive3 Drive4 Drive5 Drive6 Drive7 Drive8 Drive9

0012345678

19ABCDEF1011

2 12131415161718191A

1. Block numbers are written in hexadecimal notation.

3.2.2.2 RAID 1

Each data block in a RAID 1 (1+0) volume is mirrored on two drives, and the blocks are striped across all the drives in the volume. If one of the mirrored pair fails, the data from the other drive is used. Because the data is mirrored in a RAID 1 configuration, the volume has only half the capacity of the assigned drives. For example, if you create a four-drive RAID 1 volume with 18-Gbyte drives, the resulting data capacity is 2 Χ 18 = 36 Gbytes. RAID 1 (1+0) volume with 9 drives.

TABLE 3-3 RAID 1 (1+0) Data Blocks

Stripe Drive1 Drive2 Drive3 Drive4 Drive5 Drive6 Drive7 Drive8 Drive9

0012345678

1123456780

29ABCDEF1011

1. Block numbers are written in hexadecimal notation.

TABLE 3-3 shows the data blocks in a

3.2.2.3 RAID 5

In RAID level 5, data is striped across the drives in the volumes in segments, with parity information striped across the drives as well. Because of this parity, if a single drive fails, data can be recovered from the remaining drives. Two drive failures cause all data to be lost. A RAID 5 volume has the data capacity of all the drives in the LUN less one. For example, a 5-drive RAID 5 volume with 18-Gbyte drives has a

Chapter 3 Operation 3-5

Page 70

data capacity of 4Χ18 = 72 Gbytes. TABLE 3-4 shows the data blocks in a RAID 5 volume. The location of the parity drive in the stripe rotates from right to left in order to eliminate a performance bottleneck.

TABLE 3-4 RAID 5 Data Blocks

Stripe Drive1 Drive2 Drive3 Drive4 Drive5 Drive6 Drive7 Drive8 Drive9

001234567P

19ABCDEFP8

2 121314151617P 1011

3 1B 1C 1D 1E 1F P 18 19 20

4 24252627P 20212223

5 2D 2E 2F P 28 29 2A 2B 2C

6 36 37 P 30 31 32 33 34 35

7 3F P 38 39 3A 3B 3C 3D 3E

8 P 40 41 42 43 44 45 46 47

9 48494A4B4C4D4E4FP

1. Block numbers are written in hexadecimal notation.

3.2.3 Cache Modes

Data cache is used for the following purposes:

■ Buffering. Data that is being transferred between the host and the drives.

■ Read caching. Data for future retrieval, to reduce disk I/O as much as possible.

■ Write caching. Data that is used to build up stripes of data, eliminating the read-

modify-write overhead. Write caching improves performance for applications that are writing to disk.

Four cache modes are supported: auto, write-behind, write-through, and none. The default mode for the Sun StorEdge T3 and T3+ array is auto.

■ In the auto cache mode, if the array has full redundancy available, the caching

mode is write-behind. If any component is nonredundant, the cache mode is set to write-through. Read caching is always performed. This mode provides the best performance while retaining full redundancy protection.

■ The write-behind cache mode forces write-behind caching to always be used. Data

loss may occur in this mode if the units are not a fully redundant enterprise configuration and an array system failure occurs.

3-6 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 71

Caution – To avoid data loss in the write-behind cache mode, make sure you use

host-based mirroring.

■ The write-through cache mode forces write-through caching to always be used.

Read caching is also performed in this mode.

■ The none cache mode forces no write caching and no read caching.

Caution – If one of the redundant components fail, the cache mode is set to

write-through. In the write-through mode, mirroring is not used and data is not being cached. If you were to view the cache mode at this time using the vol mode command, the setting for cache is displayed as writethrough and the setting for mirror is displayed as on. This means that mirroring is enabled, but it is not being used.

The following table highlights the system states that can cause write-behind cache to be disabled. In cases where write-behind cache has been disabled, the array is designed to protect the data in cache. If any hardware component fails that might compromise the safety of this data, the array will disable write-behind cache, causing outstanding cached data to be flushed to disk. Write-behind caching is re-enabled when the failed condition has been corrected.

TABLE 3-5 Cache Transition States

Failure or Problem Write-Behind Cache State

Disk hard failure (RAID 5) Disable

Disk hard failure (RAID 1) Disable

Disk hard failure (RAID 0) Disable

Disk reconstruct Disable

Disk copy-back operation Disable

Interconnect Card 1 Disable

Interconnect Card 2 Disable

Interconnect Card 1+2 Disable (system down)

Controller Disable (controller failover)

Battery Disable

Battery warranty expires Disable

Battery refresh Disable (hold time low)

28-day battery refresh Disable (during 12-minute battery discharge)

Chapter 3 Operation 3-7

Page 72

TABLE 3-5

Failure or Problem Write-Behind Cache State

Power and cooling unit 1 Disable

Power and cooling unit 2 Disable

Power and cooling unit 1+2 Disable (system down)

Battery DC not OK Disable

Host cable Enable (LUN failover)

Media Interface Adapter Enable (LUN failover)

Host GBIC Enable (LUN failover)

Missing FRU Disable

1. Battery discharge may be slightly longer or shorter than 12 minutes.

Cache Transition States (Continued)

3.2.4 Disabling and Reconstructing a Drive

If a drive fails, it is automatically disabled by the system. This prevents any further access to the drive. The data on the drive is available by reconstructing it using RAID redundancy data in the volume. When a drive is disabled, the volume is operating without any further redundancy protection, so the failed drive needs to be replaced as soon as possible.

If the volume has a hot spare configured and that drive is available, the data on the disabled drive is reconstructed on the hot-spare drive. When this operation is complete, the volume is operating with full redundancy protection, so another drive in the volume may fail without loss of data.

After a drive has been replaced, the original data is automatically reconstructed on the new drive. If no hot spare was used, the data is regenerated using the RAID redundancy data in the volume. If the failed drive data has been reconstructed onto a hot spare, once the reconstruction has completed, a copy-back operation begins where the hot spare data is copied to the newly replaced drive.

3-8 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 73

3.3 Monitoring the Array

You can monitor the array remotely by using any of the following tools:

■ Sun StorEdge Component Manager software

■ Simple Network Management Protocol (SNMP) notification

■ syslog error reports

3.3.1 Sun StorEdge Component Manager

The Sun StorEdge Component Manager software provides a graphical user interface (GUI) to monitor and manage one or more arrays that are connected to a host. The software enables you to administer some of the components of the array and also constantly monitors system health. Alarm notification and remote reporting alert the administrator to abnormal activities or conditions that require attention. Sun StorEdge Component Manager can run on any host connected to the same network as the array.

For more specific information regarding Sun StorEdge Component Manager and its functions, refer to the Sun StorEdge Component Manager User’s Guide. For information on installing the software, refer to the Sun StorEdge Component Manager Installation Guide.

3.3.2 SNMP Notification

SNMP notification directs the software to send remote notification of array events to the administrator ’s host by using SNMP traps. An SNMP trap is a UDP datagram sent to a set of hosts that are identified by their IP addresses. The contents of the trap message include header information that identifies the type of trap and the data associated with the trap message.

When an event occurs in the array unit that requires attention, a trap is asynchronously sent to the host to report the event. The host(s) to which a trap is sent is configured by the user. The host software that receives the trap is an SNMP management package.

Sun StorEdge T3 and T3+ arrays use the /etc/syslog.conf file to trigger the SNMP trap. This file contains four levels of error warning to the administrator. For instructions on how to set up SNMP notification for the array, refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide.

Chapter 3 Operation 3-9

Page 74

An SNMP MIB file is provided with the array for use with SNMP-based applications. The MIB file is located in the following directory on the array:

/web/snmp/t300.mib

To use this file, use ftp to transfer this MIB file from the array to your SNMP management host and refer to your specific SNMP management documentation for correct MIB file installation procedures.

Note – SNMP notification can be used in combination with the Sun StorEdge

Component Manager software.

3.3.3 syslog Error Reports

A syslog daemon exists in the hardware RAID controller that records system messages and provides for remote monitoring. There are four possible levels of messages.

TABLE 3-6 syslog Error Reporting Levels

Message Level Indication Example

Error Critical system event

requiring immediate user intervention of attention.

Warning A system event that will

eventually require user intervention.

Notice A system event that may

be a side effect of other events or possibly a normal condition.

Information A system event that has

no effect on the health of the system.

An over-temperature condition or a removed FRU.

A FRU being disabled and a recovery procedure being executed.

The power switch turned off.

The healthy state of a FRU.

You can configure the level of error reporting to receive only certain types of messages, such as error and warning messages only. For more information on configuring remote system logging, refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide.

3-10 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 75

CHAPTER

Troubleshooting

This chapter describes how to approach troubleshooting the array. Use the information in this chapter along with the Sun StorEdge T3 and T3+ Array Administrator’s Guide or Sun StorEdge Component Manager User’s Guide to determine the service required. This chapter is organized as follows:

■ “Analyzing the Problem” on page 4-1

■ “Channel Connection Failures” on page 4-9

■ “FRU Failures” on page 4-10

4.1 Analyzing the Problem

To begin the problem analysis, check one or a combination of the following four resources, as described in the following sections.

■ “Host-Generated Message” on page 4-1

■ “Sun StorEdge Component Manager” on page 4-2

■ “Telnet Session” on page 4-2

■ “Array LEDs” on page 4-3

4.1.1 Host-Generated Message

The data host sees the array as one or two logical volumes. A host-generated message might indicate that the host cannot communicate with the array through the FC-AL channel, or that an excessive number of channel errors are occurring. If the host loses access to the array through the channel connection, then any host messages regarding the array will refer only to the logical volume(s). To see details of problems internal to the array, use the tool(s) that you installed on your management console.

4-1

Page 76

In an enterprise configuration where Alternate Pathing failover has been established, the failure of a channel path or array controller causes the host to redirect I/O from the failed channel to the second FC-AL connection.

4.1.2 Sun StorEdge Component Manager

Sun StorEdge Component Manager communicates with the array unit by means of an Ethernet connection. The software must be installed, configured, and manually started on the management host; it does not have to be the same host that is directly connected by way of the Fibre Channel to the array.

The software recognizes the unit through the Ethernet IP address and monitors it through a polling routine. The Sun StorEdge Component Manager GUI displays information on the unit, including the suspected problem area. The software can be configured to route error messages to a log file, generate an email, or page an on-call person. The log file contains information that you can use to determine the fault and what caused the message to be sent.

4.1.3 Telnet Session

The telnet session is a direct network link to the array unit by means of the command-line interface (CLI). You can run individual commands to query and repair the unit from this interface. The telnet session requires access to the unit’s IP address. You can run the CLI from any host that is part of the same network.

You can quickly determine the status of the array from the CLI. You can also examine the syslog file of the array file system through the CLI. The syslog file contains a record of events that have occurred in the unit.

For more information on how to set up the syslog file and interpret it, and for information on how to use the CLI commands, see the Sun StorEdge T3 and T3+ Array Administrator’s Guide.

4-2 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 77

4.1.4 Array LEDs

Each FRU component within the array (except for the chassis) has a set of LEDs that indicate health and status, as shown in the following sections:

■ “Disk Drive LEDs” on page 4-3

■ “Power and Cooling Unit LEDs” on page 4-4

■ “Interconnect Card LEDs” on page 4-6

■ “Controller Card LEDs” on page 4-7

Even though the LED indicates a FRU failure, you should always verify the FRU status using either the CLI or Sun StorEdge Component Manager before replacing the component. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide or the Sun StorEdge Component Manager User’s Guide for instructions.

Note – See Chapter 5 for instructions on how to service a failed FRU.

4.1.4.1 Disk Drive LEDs

The top of each disk drive has LEDs that indicate drive activity and status. These LEDs can be seen with the front cover on the unit. LED states and describes each state.

TABLE 4-1 lists the possible drive

Disk drive LEDs

FIGURE 4-1 Disk Drive LEDs (Viewed Through Front Cover)

Chapter 4 Troubleshooting 4-3

Page 78

TABLE 4-1 Disk Drive LED Descriptions

Drive Activity (green)

Drive Status (amber) Description

Off Off Drive not installed (not recognized)

Slow blink Off Drive is spinning up or down

Solid Off Drive is OK and is idle

Flashing Off Drive is OK and is active

n/a Solid A drive reconstruct or firmware download is in

progress

n/a Slow blink Drive failure; OK to replace drive

Note—Verify a drive failure using the CLI or Sun StorEdge Component Manager.

Note – See Chapter 5 for instructions on how to service a failed disk drive.

4.1.4.2 Power and Cooling Unit LEDs

Each power and cooling unit has an AC LED and a power supply (PS) LED.

TABLE 4-2 lists the possible conditions of these LEDs and describes each state.

AC LED PS LED

FIGURE 4-2 Power and Cooling Unit LEDs

4-4 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 79

TABLE 4-2 Power and Cooling Unit LED Descriptions

AC LED (green or amber)

PS LED (green or amber) Description

Off Off Power is off

No AC input

Amber Off Power is off

Power switch turned off AC power is available

Green Off Occurs when

array is shut down:

• Power and cooling unit is disabled

• AC power is available

Green Green Normal operating state:

• Power and cooling unit is receiving AC power

• Power switch is turned on

• AC power is available

Amber Amber Switch is off;

array will power off after power

and cooling unit is disabled

Green Amber Indicates one or more of the following:

• Over-temperature condition; PCU disabled

• DC power not available; PCU disabled

• Both fans fault; PCU disabled

• Battery on refresh cycle

Green Blinking green Battery on refresh cycle

Green Blinking amber Indicates one or more of the following:

• PCU disabled

• One fan fault

• Battery hold-time low; PCU remains enabled

• Battery out of warranty; PCU remains enabled

• Battery life-span failure; PCU remains enabled

Note—Verify a power and cooling unit failure using the CLI or Sun StorEdge Component Manager.

Note – See Chapter 5 for instructions on how to service a failed power and cooling

unit.

Chapter 4 Troubleshooting 4-5

Page 80

4.1.4.3 Interconnect Card LEDs

Each interconnect card has a status LED for each interconnect cable. TABLE 4-3 lists the possible interconnect card status LED states with descriptions of each state.

Interconnect card 1

LED

FIGURE 4-3 Interconnect Card LEDs

TABLE 4-3 Interconnect Card LED Descriptions

Interconnect Card Status LED (green or amber) Description

Off Interconnect card not installed (not recognized)

Green-solid Interconnect card OK

Cable OK (if present)

Green-slow blink Interconnect card OK

Cable bad; OK to replace cable

Amber-solid Interconnect card firmware download in progress

Amber-slow blink Interconnect card failure; OK to replace interconnect card

Note—Verify an interconnect card failure using the CLI or Sun StorEdge Component Manager.

Interconnect card 2LED

Note – See Chapter 5 for instructions on how to service a failed interconnect card.

4-6 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 81

4.1.4.4 Controller Card LEDs

The controller card LEDs for a Sun StorEdge T3 array differ from that of a Sun StorEdge T3+ array. The controller LEDs of both array models are described in the following sections.

Sun StorEdge T3 Array Controller LEDs

The Sun StorEdge T3 array controller card has a channel-active LED for the host interface port, a controller status (online) LED, and a reset switch. possible states of the channel-active LED and describes each state. possible states and descriptions for the controller status LED.

TABLE 4-4 lists the TABLE 4-5 lists the

FC-AL channel-active LED

Controller

Reset switch

status LED

FIGURE 4-4 Sun StorEdge T3 Array Controller Card LEDs

TABLE 4-4 FC-AL Channel-Active LED Descriptions (Sun StorEdge T3 array)

Channel Active LED (green) Description

Off Port disabled

Amber Port enabled and not receiving signal from host

Green Port enabled and idle

Blinking green Port enabled and active

TABLE 4-5 Controller Status LED Descriptions (Sun StorEdge T3 array)

Controller Status LED (green or amber) Description

Off Controller not installed (not recognized)

Green Controller OK

Chapter 4 Troubleshooting 4-7

Page 82

TABLE 4-5 Controller Status LED Descriptions (Sun StorEdge T3 array) (Continued)

Controller Status LED (green or amber) Description

Amber Controller boot, shutdown, or firmware download in progress

Blinking amber Controller failure; OK to replace controller

Note—Verify a controller card failure using the CLI or Sun StorEdge Component Manager.

Note – See Chapter 5 for instructions on how to service a failed controller card.

Sun StorEdge T3+ Array Controller Card LEDs

The Sun StorEdge T3+ array controller card has two channel-active LEDs, one for the FC-AL interface port and one for the Ethernet port, and a controller status (online) LED.

TABLE 4-6 lists the possible states of the controller card LEDs and describes each

state.

FC-AL active LED

Ethernet active LED

FIGURE 4-5 Sun StorEdge T3+ Array Controller Card LEDs

TABLE 4-6

LED Action Description

FC-AL Channel Active LED (green)

Sun StorEdge T3+ Array Controller Card LED Descriptions

Off Port disabled

Green Port enabled and idle

10/100BASE-T active LED

Controller online status LED

Blinking green Port enabled and active

Ethernet Active LED (green)

Off Link invalid

Green Link valid and idle

Blinking green Link valid and active

4-8 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 83

TABLE 4-6

LED Action Description

10/100 BASE-T Active LED (green)

Sun StorEdge T3+ Array Controller Card LED Descriptions (Continued)

Off Port disabled (10 Mbps rate)

Green Port enabled and idle (100 Mbps

rate)

Controller Status LED (green or amber)

Blinking green

Off Controller not installed (not

Port enabled and active

recognized)

Green Controller OK

Amber Controller boot, shutdown, or

firmware download in progress

Blinking amber Controller failure; OK to replace

controller

Note—Verify a controller card failure using the CLI or Sun StorEdge Component Manager.

4.2 Channel Connection Failures

Host-to-array channel failures occur when the connection between the array and the host is either severed or intermittent. The components that make up this data channel connection are:

■ Host bus adapter (HBA), which resides on the host

■ GBIC adapter, used to connect the FC-AL cable to an SBus HBA

■ Fibre Channel cable that connects the array to the host

■ Channel interface port in the array

Errors in the host data channel are outside the scope of the array. To determine failures in the data path, you must use host-based application diagnostics, such as the StorTools™ product for the Solaris operating environment.

Refer to the documentation of the selected diagnostics tool for information on identifying data channel failures.

Chapter 4 Troubleshooting 4-9

Page 84

4.3 FRU Failures

FRU failures are reported by means of an alarm or message notification from the tool used to monitor the array. The failed FRU will most likely display a blinking amber LED. Before replacing a FRU based on the LED indicator, verify its failure by using the command-line interface (CLI) or the health-check mechanism in Sun StorEdge Component Manager.

To check for a FRU failure using the CLI, refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide. To use Sun StorEdge Component Manager to diagnose a failed FRU, refer to the Sun StorEdge Component Manager User’s Guide.

4-10 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 85

CHAPTER

Service

This chapter contains information about removing and replacing the fieldreplaceable units (FRUs) in an array and upgrading software and firmware.

Caution – Any FRU that is removed must be replaced within 30 minutes or the

array and all attached arrays will automatically shut down and power off.

This chapter is divided into the following sections:

■ “Preparing for Service” on page 5-2

■ “Removing and Replacing Components” on page 5-2

■ Disk Drives

■ Power and Cooling Units

■ UPS Battery

■ Interconnect Cards

■ Controller Card

■ Chassis

■ “Upgrading Firmware” on page 5-16

■ Upgrading Sun StorEdge T3+ Array Controller Firmware

■ Upgrading Sun StorEdge T3 Array Controller Firmware

■ Upgrading Interconnect Card Firmware

■ Upgrading Disk Drive Firmware

5-1

Page 86

5.1 Preparing for Service

Follow these procedures to help prevent damaging FRUs:

■ Wear a grounding wrist strip.

■ Before handling a FRU, discharge any static electric charge by touching a ground

surface.

■ Do not remove a FRU from its antistatic protective bag until it is ready for

installation.

■ When removing a FRU from the array, immediately place it in an antistatic bag

and packaging.

■ Handle a FRU only by its edges and avoid touching the circuitry.

■ Do not slide a FRU over any surface.

■ Avoid having plastic, vinyl, and foam in the work area.

■ Limit body movement (which builds up static electricity) during FRU installation.

Caution – Although the array will remain operational for up to 30 minutes with a

FRU removed, replace a failed FRU as soon as you have removed it from the array.

5.2 Removing and Replacing Components

This section contains instructions for servicing the following array FRUs:

■ “Disk Drives” on page 5-3

■ “Power and Cooling Units” on page 5-7

■ “UPS Battery” on page 5-10

■ “Interconnect Cards” on page 5-11

■ “Controller Card” on page 5-14

■ “Chassis” on page 5-16

5-2 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 87

5.2.1 Disk Drives

Caution – Replace only one disk drive in an array at a time to ensure that no data

is lost. Ensure that the disk drive is fully reenabled before replacing another disk drive in the same array.

The default configuration for the array is to automatically spin up and reenable a replaced disk drive, then automatically reconstruct the data from the parity or hotspare disk drives. Disk drive spinup takes about 30 seconds, and reconstruction of the data on the disk drive can take one or more hours depending on system activity.

Note – In some cases, if disk drives are removed and replaced on a powered-down

or incompletely booted array, the internal system monitoring facilities may not correctly detect the drive replacement. Therefore, replace disk drives on poweredup, fully-booted arrays only.

5.2.1.1 Disk Drive LEDs

The top of each disk drive has LEDs that indicate drive activity and status. For information about the disk drive LEDs and how to interpret them to diagnose a drive failure, see Section 4.1.4.1, “Disk Drive LEDs” on page 4-3.

Note – Even if the LED indicates a drive failure, you should always verify the FRU

status using either the CLI or Sun StorEdge Component Manager before replacing the drive. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide or the Sun StorEdge Component Manager User’s Guide for instructions.

Chapter 5 Service 5-3

Page 88

5.2.1.2 Removing and Replacing a Disk Drive

1. Observe static electricity precautions.

2. Remove the front panel by pressing in on the side latches and pulling the front panel forward.

Latch

FIGURE 5-1 Removing the Front Panel

3. Locate the disk drive that needs to be replaced.

Disk drives are numbered from 1 to 9 starting on the left side of the array.

Disk 1

FIGURE 5-2 Disk Drive Numbering

5-4 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Disk 9

Page 89

4. Use a coin or small screwdriver to press in and release the drive latch handle.

FIGURE 5-3 Releasing the Latch Handle

5. Use the latch handle to slowly pull the disk drive out an inch.

Wait 30 seconds before pulling the drive out completely, giving the disk drive time to spin down.

6. Remove the disk drive from the array (

FIGURE 5-4).

Caution – Any disk drive that is removed must be replaced within 30 minutes or

the array and all attached arrays will automatically shut down and power off.

Chapter 5 Service 5-5

Page 90

FIGURE 5-4 Removing a Disk Drive

7. Push in the latch handle on the removed disk drive to protect it from damage.

8. Release the latch handle on the disk drive to be installed.

9. Insert the new disk drive gently on the middle of the rails and push it in until it is seated with the centerplane connector.

10. Use a coin or small screwdriver to press in and lock the latch handle.

11. Replace the front panel.

Caution – You must replace the front panel for the array to meet regulatory

emissions requirements.

12. Using Sun StorEdge Component Manager or the CLI, verify that there are no disk drive errors.

Refer to the Sun StorEdge Component Manager User’s Guide or the Sun StorEdge T3 and T3+ Array Administrator’s Guide for more information.

5-6 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 91

5.2.2 Power and Cooling Units

The power and cooling unit has two active power sources: standby and primary power. Standby power, which is used to power the microcontroller on the interconnect card, is activated when AC power is present. Primary power, which is used to power all remaining circuits and disk drives, is activated when AC or battery power is present and the power switch is on.

Each power and cooling unit has a power switch in the rear upper center of the unit. Turning off the power on a power and cooling unit affects only that power and cooling unit. Therefore, to power off all primary power to the unit, both power switches on both power and cooling units must be turned off. After the switches are turned off, system primary power does not actually turn off until the controller has performed an orderly shutdown, including writing any data cache to disk. This process can take up to two minutes.

Separate power cords are used for the connector on each power and cooling unit to provide redundant cabling. The power cords need to be connected to alternate AC power sources for full redundancy.

Power switches

FIGURE 5-5 Power Cords Connected to the Power and Cooling Units

Caution – Do not handle the power and cooling unit when the power cord is

connected. Line voltages are present within the power and cooling unit when the power cord is connected even if the power switch is off.

At the rear of the power and cooling unit is a recessed PC card connector. Do not touch this connector or allow any metal object to touch it. The power and cooling unit contains the UPS battery backup. There are no serviceable parts inside this unit.

Chapter 5 Service 5-7

Page 92

5.2.2.1 Power and Cooling Unit LEDs

Each power and cooling unit has an AC LED and a power supply (PS) LED. For information about the power and cooling unit LEDs and how to interpret them to diagnose a failure, see Section 5.2.2.1, “Power and Cooling Unit LEDs” on page 5-8.

Note – Even if the LED indicates a power and cooling unit failure, you should

always verify the FRU status using either the CLI or Sun StorEdge Component Manager before replacing the unit. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide or Sun StorEdge Component Manager User’s Guide for more information.

5.2.2.2 Removing and Replacing a Power and Cooling Unit

Caution – To ensure correct airflow for system cooling, both power and cooling

units must be in the installed position for normal operation. A failed power and cooling unit should be removed only when a replacement power and cooling unit is available to be inserted.

Caution – Replace only one power and cooling unit at a time to prevent system

interruption.

To replace a power and cooling unit:

1. Power off the power and cooling unit by pressing the power switch (

Make sure that the AC LED is amber and the PS LED is off.

2. Disconnect the power cord from the AC outlet.

3. Disconnect the power cord from the power and cooling unit connector by squeezing both sides of the connector and pulling it straight out (

4. Unlock the power and cooling unit by using a coin or small screwdriver to push in and release the two latch handles (

5. Pull the power and cooling unit out of the array.

Put one index finger through each of the latch handles. With your thumbs on the top of the chassis for support, pry the power and cooling unit out of its connectors with an upward rotation. Once it is out approximately .5 inches (2 cm), it will be free to slide out of the frame on its rails.

5-8 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

FIGURE 5-6).

FIGURE 5-5).

Page 93

Caution – Any power and cooling unit that is removed must be replaced within 30

minutes or the array and all attached arrays will automatically shut down and power off.

Latch handle

FIGURE 5-6 Removing the Power and Cooling Unit

6. Insert the new power and cooling unit.

7. Lock the new power and cooling unit by pushing in both latch handles.

8. Insert the power cord into the power and cooling unit connector.

9. Connect the power cord to the AC outlet.

Verify that the AC LED on the power and cooling unit is amber, indicating that AC power is present.

10. Push the power and cooling unit power switch on.

Verify that both LEDs on the power and cooling unit are green, indicating that the unit is receiving power.

Chapter 5 Service 5-9

Page 94

5.2.3 UPS Battery

The UPS battery is located within the power and cooling unit. The battery provides backup in case of a complete AC power failure and will sustain power to the array long enough to flush cache data to the drives.

During a power failure, if the battery is flushing cache data to the drives, battery power becomes depleted. Once AC power is available, the battery recharges. While the battery is recharging, write-behind cache mode is disabled and write-through cache mode is enabled until the battery is fully recharged. The battery recharge could take up to 12 hours, depending on the length of the power outage and the amount of cache data that was flushed to the drives.

5.2.3.1 Battery Maintenance

The array has been factory configured to initiate a battery refresh cycle that occurs automatically once every 28 days. This battery refresh cycle involves a discharge of the battery followed by a recharge. The refresh cycle is sequential; the second power and cooling unit does not start a refresh until the first power and cooling unit successfully completes the refresh cycle, ensuring that only one battery in an array is refreshed at a time. This refresh cycle verifies the health of the battery. During the refresh, if a problem is detected with the battery, future refresh operations are suspended until the problem is fixed. If this happens, write-behind caching is turned off automatically as a safety precaution. A refresh cycle does not begin unless both power and cooling units are in optimal condition.

A battery refresh operation in progress is indicated in the syslog file. During the battery refresh operation, the battery health is determined by the discharge time of the battery. This time is written into the syslog file and is displayed as battery hold time. If the discharge time of the battery is greater than six minutes (360 seconds), the battery is considered healthy. If the battery is not healthy, the system will disable the battery, disable write cache, and indicate that the power and cooling unit needs to be replaced. The following is an example of the battery hold time as displayed in the syslog file:

May 21 00:19:22 driver24 BATD[1]: N: u1pcu1: hold time was 1159 seconds.

You can use the refresh -s command to view an active refresh operation. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide for more information on this command.

5-10 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 95

The battery service life is two years. When the battery approaches the end of its life span, warning messages are sent to the syslog file. The first message is sent 45 days before the end of the life span, followed by a warning message every 5 days thereafter. Replace the power and cooling unit that contains the battery as soon you receive these warning messages.

You can view information on the battery, including the life span, using the CLI or Sun StorEdge Component Manager. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide or the Sun StorEdge Component Manager User’s Guide for instructions.

5.2.3.2 Removing and Replacing the UPS Battery

The UPS battery itself is not a serviceable FRU. To replace the battery, you must replace the entire power and cooling unit. The warning message will indicate which power and cooling unit needs to be replaced. See Section 5.2.2.2, “Removing and Replacing a Power and Cooling Unit” on page 5-8 for instructions.

5.2.4 Interconnect Cards

The interconnect cards contain the interface circuitry and two connectors on each card for interconnecting multiple Sun StorEdge T3 array or Sun StorEdge T3+ arrays. The interconnect card also contains the environmental monitor for the array. If an interconnect card or cable fails, the array continues to function using the remaining interconnect card and cable.

The interconnect cards in an array are assigned a master and slave status, similar to the controller card. If you unplug or remove the master interconnect card from the unit and then put it back, the new status is reported as “slave.” The corresponding interconnect card that was the slave then becomes the master. You can view the status of the interconnect cards using the fru stat command on the array. There is no performance impact to the array if the interconnect cards change status, and it is not necessary to alter the state back to the original master and slave configuration.

For more information on the fru stat command and other array commands, refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide.

5.2.4.1 Interconnect Card LEDs

Each interconnect card has an interconnect status LED for each interconnect cable. For the location of and information about the interconnect card status LEDs and how to interpret them to diagnose a failure, see Section 4.1.4.3, “Interconnect Card LEDs” on page 4-6.

Chapter 5 Service 5-11

Page 96

Note – Even if the LED indicates an interconnect card failure, you should always

verify the FRU status using either the CLI or Sun StorEdge Component Manager before replacing the unit. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide or Sun StorEdge Component Manager User’s Guide for more information.

5.2.4.2 Removing and Replacing an Interconnect Card

Caution – Use the interconnect cables to cable only Sun StorEdge T3 or T3+ arrays

together using the interconnect card connectors. Do not use these cables for any other FC-AL connection.

Caution – The interconnect card FRUs are extremely sensitive to static electricity.

Use proper grounding wrist straps and antistatic procedures when handling these FRUs. See Section 5.1, “Preparing for Service” on page 5-2 for a list of static electricity precautions.

Caution – Replace one interconnect card at a time. Follow the procedure as

described to ensure that there is no interruption in system operation or loss of data.

To prevent interruption of system operation during interconnect card replacement, make sure that:

■ In a workgroup configuration, remove only the failed interconnect card. Leave the

second interconnect card intact in the array.

Caution – Removing both interconnect cards at the same time in a single unit will

cause the unit to shut down immediately. Remove only one card at a time.

■ In an enterprise configuration, remove the interconnect cable only from the failed

interconnect card. Leave the interconnect cable attached to the working interconnect card.

To replace an interconnect card:

1. Ensure that the interconnect card to be replaced is showing failure status.

2. Observe static electricity precautions, as described in Section 5.1, “Preparing for Service” on page 5-2.

If you have a single controller unit configuration, proceed to Step 4.

5-12 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 97

3. Remove the interconnect cable from the failed interconnect card only.

Mark the connector with either 1 or 2.

4. Unlock the failed interconnect card by pushing in on the latch handle.

Latch handle

FIGURE 5-7 Removing the Interconnect Card

5. Pull the interconnect card out using the latch handle.

Caution – The interconnect card that is removed must be replaced within 30

minutes or the array and all attached arrays will automatically shut down and power off.

6. Insert the new interconnect card, making sure that the card sits on the frame.

7. Lock the new interconnect card in place by pushing in on the latch handle.

8. Reconnect the interconnect cable to the interconnect card.

9. Verify that the LEDs on the interconnect card show that the interconnect card has initialized properly.

You can verify the status of the interconnect card using either the CLI or Sun StorEdge Component Manager. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide or the Sun StorEdge Component Manager User’s Guide for instructions.

Chapter 5 Service 5-13

Page 98

5.2.5 Controller Card

The controller card is redundant only when the array is configured in an enterprise configuration; that is, when the array is connected to another array that contains a controller card.

In an enterprise configuration, if there is a controller card failure in a master unit, the controller would failover to the alternate master, which would then assume master controller status. After the failed controller card is replaced in the master unit, that unit assumes the role of alternate master. The new master unit (previously the alternate master) retains all the necessary information, such as the unit IP address, to continue functioning without system interruption. The enterprise configuration can be returned back to its original, prefailover configuration of master and alternate master by performing a reset command.

In a workgroup configuration, system activity is interrupted until the controller card is replaced and is back online. When a controller card is replaced, the array automatically downloads the previous settings to the new controller card. The array is fully functional once the new controller card is installed and the controller is online.

5.2.5.1 Controller Card LEDs

The controller card LEDs for a Sun StorEdge T3 array differ from that of a Sun StorEdge T3+ array. For the locations of and information about the LEDs on both array models and how to interpret the LEDs to diagnose a failure, see Section 4.1.4.4, “Controller Card LEDs” on page 4-7.

Note – Even if the LED indicates a controller card failure, you should always verify

the FRU status using either the CLI or Sun StorEdge Component Manager before replacing the unit. Refer to the Sun StorEdge T3 and T3+ Array Administrator’s Guide or Sun StorEdge Component Manager User’s Guide for more information.

5.2.5.2 Removing and Replacing a Controller Card

A controller card can be replaced without system interruption only if the array is part of an enterprise configuration.

To replace the controller card:

1. Observe static electricity precautions.

2. Ensure that the controller card is showing failure status.

5-14 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Page 99

3. Remove the Ethernet cable from the Ethernet connector.

4. Remove the fiber-optic cable from the FC-AL connector.

If you are removing a Sun StorEdge T3 array controller card, you must also remove the media interface adapter from the FC-AL connector.

5. Unlock the controller card by pushing in on the latch handles (

FIGURE 5-8).

6. Pull the controller card out using the latch handles.

Caution – The controller card that is removed must be replaced within 30 minutes

or the array and all attached arrays will automatically shut down and power off.

Latch handle

FIGURE 5-8 Removing the Controller Card

7. Insert the new controller card.

8. Lock the new controller card by pushing in the latch handles.

9. Insert the fiber-optic cable into the FC-AL connector.

10. Insert the Ethernet cable into the Ethernet connector.

11. Check the controller status LED to determine when the controller is operational.

While the controller boots, the controller status LED is solid amber. When the controller is operational, the LED is green.

Chapter 5 Service 5-15

Page 100

Note – After replacing the controller card, you might need to upgrade the controller

firmware and EPROM firmware to make sure you are running the same firmware versions that you had on the previous controller card. To upgrade the controller firmware, see “Upgrading Firmware” on page 5-16.

5.2.6 Chassis

The array chassis rarely needs to be replaced. However, the chassis part number is available to replace the centerplane and chassis if necessary. (These must be replaced together, because they are factory aligned.)

Problems with the centerplane would most likely occur due to an electrical short or a bent or broken pin connector. These problems would first appear as a failure of another FRU component, such as an interconnect failure or drive failure. If replacing the FRU that appears to be failed does not correct the problem, then examine the centerplane connector that the FRU connects to for bent or broken pins. If nothing is obvious, install another spare FRU to verify that it is not a failed FRU component causing the problem. If you have eliminated all possibility of a FRU component failure and the problem still remains, it is likely to be a centerplane failure.

If there is a centerplane failure, contact your SunService representative. Only trained, qualified service providers should replace an array chassis and centerplane.

5.3 Upgrading Firmware

The firmware upgrade procedures that follow must be done through the Ethernet connection. The latest firmware versions are located on the SunSolve web site:

http://sunsolve.sun.com

To download the firmware patches from the SunSolve web site:

1. In the SunSolve Online column, click on Patches.

2. In the Patches column, click on PatchPro.

3. Click on Storage Products.

Refer to the patch README file for details on firmware and specific patch install requirements.

5-16 Sun StorEdge T3 and T3+ Array Installation, Operation, and Service Manual • August 2001

Sun Microsystems Sun StorEdge T3 Installation, Operation And Service Manual

Specifications and Main Features

Frequently Asked Questions

User Manual

Contents

Figures

Tables

Preface

Before You Read This Book

How This Book Is Organized

Using UNIX Commands

Typographic Conventions

Shell Prompts

Related Documentation

Accessing Sun Documentation Online

Sun Welcomes Your Comments

Sun StorEdge T3 and T3+ Array Overview

1.1 Product Description

1.2 Array Features

1.3 Components

1.3.1 Disk Drive

1.3.2 Controller Card

1.3.2.1 Sun StorEdge T3 Array Controller

1.3.2.2 Sun StorEdge T3+ Array Controller

1.3.3 Interconnect Card

1.3.4 Power and Cooling Unit

1.4 Architecture

1.5 Supported Configurations

Installation

2.1 Preparing for the Installation

2.2 Inspecting the Array

2.3 Editing the Host Files

2.4 Installing the Array

2.5 Connecting the Cables

2.6 Powering On and Verifying the Hardware Configuration

2.7 Establishing a Network Connection

2.8 Verifying the Firmware Level and Configuration

2.9 Installing the Administration Tools

2.10 Defining and Mounting Volumes

2.10.1 Workgroup Configuration

2.10.2 Enterprise Configuration

2.11 Changing the Default Configuration

2.12 Connecting the Host System

2.12.1 Verifying the Data Host Connection

2.12.2 Establishing Logical Volumes on the Host

2.12.3 Creating Soft Partitions on the Data Host

2.12.4 Establishing Alternate Pathing on the Host

Operation

3.1 Powering Off and On

3.1.1 Powering Off

3.1.2 Powering On

3.2 Reconfiguring the Array Settings

3.2.1 Unit Volume Configurations

3.2.2 RAID Levels

3.2.2.1 RAID 0

3.2.2.2 RAID 1

3.2.2.3 RAID 5

3.2.3 Cache Modes

3.2.4 Disabling and Reconstructing a Drive

3.3 Monitoring the Array

3.3.1 Sun StorEdge Component Manager

3.3.2 SNMP Notification

3.3.3 syslog Error Reports

Troubleshooting

4.1 Analyzing the Problem

4.1.1 Host-Generated Message

4.1.2 Sun StorEdge Component Manager

4.1.3 Telnet Session

4.1.4 Array LEDs

4.1.4.1 Disk Drive LEDs

4.1.4.2 Power and Cooling Unit LEDs

4.1.4.3 Interconnect Card LEDs

4.1.4.4 Controller Card LEDs

4.2 Channel Connection Failures

4.3 FRU Failures

Service

5.1 Preparing for Service

5.2 Removing and Replacing Components

5.2.1 Disk Drives

5.2.1.1 Disk Drive LEDs