nStor 4100FS User Manual

4100FS.book Page 1 Tuesday, July 8, 2003 4:35 PM

USER'S GUIDE

NexStor® 4100FS

STORAGE SYSTEMS

4100FS.book Page 2 Tuesday, July 8, 2003 4:35 PM

4100FS.book Page 3 Tuesday, July 8, 2003 4:35 PM

USER'S GUIDE

NexStor® 4100FS

STORAGE SYSTEMS

4100FS.book Page 4 Tuesday, July 8, 2003 4:35 PM

nStor Corporation, Inc.

Restricted Rights and Liability

No part of this manual may be reproduced, stored in a retrieval system, or transmitted, in any form or by any
means, electronic, mechanical, recording, or otherwise, in whole or part, without prior written permission from
nStor Corporation, Inc.

nStor Corporation, Inc. shall not be liable for any damages or for the loss of any information resulting from the
performance or use of the information contained herein. Your rights to the software are governed by the license
agreement included with any accompanying software. nStor Corporation reserves the right to periodically revise
this manual without notice. Product features and specifications described are subject to change without notice.

For Regulatory information, see Appendix C.

Copyright

nStor

Corporation

, Inc.
6190 Corte Del Cedro Road
Carlsbad, CA 92009

Copyright © 2003 nStor Corporation, Inc. All Rights Reserved.

nStor, AdminiStor, NexStor, the NexStor logo, and the nStor logo are trademarks or registered trademarks of nStor
Corporation, Inc.

Other company and product names herein may be trademarks or registered trademarks of their respective
companies.

Agency Notes

WARNING: Drives and controller/adapter cards described in this manual should only be installed in UL-listed and CSA

certified computers that give specific instructions on the installation and removal of accessory cards (refer to your
computer installation manual for proper instructions).

ATTENTION: Les lecteurs et cartes contrôleurs décrits ici ne doivent être montés que sur des ordinateurs homologués (UL et

CSA) et livrés avec des manuels contenant les instructions d’installation et de retrait des accessoires. Reportez­vous au manuel d’installation de votre ordinateur.

SERVICE NOTE: Remove the power cables prior to servicing this equipment.

!

4100FS.book Page i Tuesday, July 8, 2003 4:35 PM

Contents

About this Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Welcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

1 Getting Started . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

At a Glance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Front Bezel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

AC Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Cooling Fan Module . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

SAF-TE Disk I/O Card . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Host SCSI I/O Card . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Fibre Channel Host I/O Card . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

RAID Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

RS-232 Service Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Control and Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Status Indicator LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Drive LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Audible Alarm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2 Theory of Controller Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Operating Modes Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

Stand-Alone Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Active-Active Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

Active-Passive Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

Understanding Mirrored Operations . . . . . . . . . . . . . . . . . . . . . . 29

A Word about Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Minimizing Downtime for Maximum Data Availability . . . . . . . . . 30

How Available are Clusters? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Application of Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

i

4100FS.book Page ii Tuesday, July 8, 2003 4:35 PM

Table of Contents

3 Setup and Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Storage System Detailed Installation . . . . . . . . . . . . . . . . . . . . . . 34

Installing the Storage System Enclosure into the Rack Cabinet . . 34

Installing the Storage System into the Tower Stand . . . . . . . . . . 37

Completing the Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

Enclosure Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

Topology Host Cabling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

Upgrades . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69

Powering On the Storage System . . . . . . . . . . . . . . . . . . . . . . . . 70

Powering Off the Storage System . . . . . . . . . . . . . . . . . . . . . . . . 70

4 SAN Solution Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Active-Active:Dual Port Mode for SAN . . . . . . . . . . . . . . . . . . . . 71

Active-Active:Single Port Mode for SAN . . . . . . . . . . . . . . . . . . . 76

SAN Configuration Not Supported . . . . . . . . . . . . . . . . . . . . . . . 80

True LUN Sharing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

5 Accessing the RAID Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Accessing the Disk Array Administrator Software/VT-100 . . . . . . 82

Menu System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

Updating RAID Controller Firmware . . . . . . . . . . . . . . . . . . . . . 86

6 Monitoring Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Enclosure Component Monitoring . . . . . . . . . . . . . . . . . . . . . . . 89

Status Indicator LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90

Drive LEDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

Drive Carrier LitePipes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Audible Alarm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

One-Touch Annunciation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Enclosure SAF-TE Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . 98

Enclosure Fan Speed Control . . . . . . . . . . . . . . . . . . . . . . . . . . . 100

SAFTE Commands Debug . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Uploading SAF-TE Controller Card Firmware . . . . . . . . . . . . . . . 103

7 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

General Enclosure Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Common SCSI Bus Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . 107

Terminal Emulator and COM Port Problems . . . . . . . . . . . . . . . . 109

Host SCSI Channel Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . 110

ii

4100FS.book Page iii Tuesday, July 8, 2003 4:35 PM

Device SCSI Channel Problems . . . . . . . . . . . . . . . . . . . . . . . . . 110

Problems During Bootup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

Controller Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

Common Problems and Interpreting the LEDs . . . . . . . . . . . . . . 112

Warning and Error Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Warnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Disk Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Disk Channel Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

8 Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Removing the Front Bezel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

Replacing the Cooling Fans . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121

Replacing a Power Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

Replacing a Disk Drive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

Replacing the SAF-TE Disk I/O and Host SCSI I/O Card . . . . . . . 127

Replacing the Host I/O Card . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

Replacing a RAID Controller . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Replacing a “Killed” Controller When in Active-Active/

Active-Passive Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Replacing the Enclosure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134

Table of Contents

A Technical Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137

B Port and Transceiver Information . . . . . . . . . . . . . . . . . . . . . . . . . . 141

VHD/CI SCSI Connectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

SAF-TE Service Port . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

RAID Controller Service Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . 144

Null-Modem Cable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

Optical SFP Transceiver . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146

C Regulatory Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

Compliance Information Statement . . . . . . . . . . . . . . . . . . . . . . . 147

FCC Class A Radio Frequency Interference Statement . . . . . . . . . 148

Class A Taiwanese Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Class A Japanese Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

CE Notice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Power Cord Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

iii

4100FS.book Page iv Tuesday, July 8, 2003 4:35 PM

Table of Contents

iv

4100FS.book Page v Tuesday, July 8, 2003 4:35 PM

Preface

About this Manual

Welcome

Congratulations on the purchase of your new NexStor 4100FS Series RAID

Storage System from nStor Corporation. The NexStor 4100FS Series is a very

high-performance fully fault-tolerant Fibre Channel-to-Ultra160 SCSI storage

system. It’s unique 2U design is optimized to fit in the compact space of today’s

data center rack enclosures and as a deskside tower system.

At its core is the

pluggable

1-inch high Ultra320 or Ultra160 disk drives

Ultra160 SCSI

storage enclosure which supports up to twelve hot

all in a 2U (3.47-inch) form

factor enclosure. Full component redundancy is provided through hot pluggable

cooling fan module and independent power supplies. RAID functionality is

provided through hot pluggable embedded

Fibre Channel-to-Ultra160 SCSI

RAID

Controllers. The NexStor 4100FS is available in two model configurations, a single

controller configuration designed for Stand-Alone topologies, or a dual controllers

configuration for Active-Active or Active-Passive topologies.

Product Identification

Storage Enclosure Number of Controllers Model of Controller

NexStor 4000S

NexStor 4110FS

NexStor 4120FS

0 JBOD

1 NexStor-Chaparral JFS224

2 NexStor-Chaparral JFS224

v

4100FS.book Page vi Tuesday, July 8, 2003 4:35 PM

About this Manual

This user’s guide is your complete documentation to set up the storage system

hardware, add components, cable the storage system components, replace parts,

and diagnose/repair your system.

For information on the software control, management and configuration, refer to

one of the following: VT-100 Software Guide or the AdminiStor Software Guide.

Features

The NexStor 4100FS Series Storage Systems are designed for mission critical
applications requiring the highest performance with uncompromised data
reliability, such as mid-range and enterprise server storage. They maintain
exceptionally high throughput and are ideally suited for high bandwidth data
intensive applications, such as electronic commerce, digital video, CAD, seismic
research, digital pre-press, and 3-D imaging.

The following are major features of the NexStor

■

2 Gb Fibre Channel-to-Ultra160 RAID storage system enclosure.

■

Hot pluggable Fibre Channel-to-Ultra160 RAID Controllers.

Supports up to 48 drives in RAID configurations.

■

■

Designed to fit standard 19-inch rack enclosures and a deskside tower.

■

Sequential data transfers from disk arrays at over 160 MB/sec.

Greater than 18,000 IOPs capability from a single controller.

■

■

Active-Active failover with dual hot-pluggable RAID controllers.

■

Operating system independent – no special software or drivers required.

Four 2 Gb Fibre Channel host ports.

■

■

Up to four Ultra160 MB per second SCSI disk channels.

■

Support for up to 48 disk drives organized in up to 24 arrays and a total of
64 logical unit numbers (LUNs).

■

All SCSI channels are backward compatible Ultra2 LVD and
Ultra/Fast/Asynchronous single-ended SCSI modes.

■

All SCSI channels support SPI-3 Cyclic Redundancy Check (CRC) and
Domain Validation.

4100FS

Series Storage Systems:

■

Support for cache memory options from 64 MB up to 512 MB using
standard PC-133 compatible SDRAM. DIMMs must be qualified by nStor.

■

Support for RAID levels 0, 1, 3, 4, 5, 10, and 50.

vi

Features

4100FS.book Page vii Tuesday, July 8, 2003 4:35 PM

■

Online capacity expansion allowing reconfiguration without interruptions.

Drive Roaming for no loss of RAID and LUN configurations.

■

■

Advanced disk utilities, array verification and recovery, and spare pooling.

■

Firmware can be upgraded either in-band over the SCSI bus or out-of-band
via the RS-232 service port.

■

Continuous runtime diagnostics for warnings and automatic shutdown for
out-of-spec temperature and voltages, battery failures, and internal errors.

About this Manual

Features

vii

4100FS.book Page viii Tuesday, July 8, 2003 4:35 PM

About this Manual

viii

Features

4100FS.book Page 1 Tuesday, July 8, 2003 4:35 PM

Chapter 1

Getting Started

This chapter provides a description of the enclosure components and its onboard

monitoring systems.

The Components section identifies and gives a complete description of each

major component. The Monitoring section describes the enclosure’s LEDs, and

the manner in which the normal and abnormal conditions are presented.

M
R
A
L
A
T
E
S
E

R

R

E

S

E

T

A

L

A

R

M

Rack-Mount Model

Tower Model

NexStor 4100FS Series Storage System

1

4100FS.book Page 2 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

At a Glance

The following illustrations show the featured components of the NexStor 4100FS

Series Storage System. Familiarize yourself with its components prior to installing

and using the storage system.

Drive Status LEDs

(left column of LEDs)

Drive Activity LEDs

(right column of LEDs)

Power On LED

Channel Status LED

Power Supply Status LED

Fan Status LED

R

E
S

E

T

A

L

A

R

M

350-watt hot-pluggable

independent power supplies

Dual in-line 80-CFM hot

swappable cooling fans

SAFTE Disk I/O Card

Host SCSI I/O Card

Host I/O Card

SAF-TE Service & Controller

Service Ports

Controller

Cover Plate

Alarm Reset Button

C

h

a

n

n

e

l 1

C

C

h

a

h

nn

a

n

e

n

l 3

e

l 2

C

h

a

C

T
x

n

H

n

3

e

P

l 0

­1

Rx

Tx

P

-

C

2

T

Rx

R

L

Tx

1

C

H

P

-

0

1

Rx

Tx

S
A

F

­T

P

E

­2

Rx

C

T
R

L

2

Controller 2

Controller 1

TERM PO

TO SUPPLY

ADD JUMPE

WER

J

P

1

J

P

3

J

TION

P

3

TION

R

TERM PO

TO SUPPLY

ADD JUMPE

WER

J

P

1

TERMINA

J

P

TO DISABLE

2

R

ADD JUMPE

J

P

4

R

TERMINA

J

P

TO

2

DISABLE

ADD JUMPE

J

P

4

R

Optional Dual
FC-to-Ultra160 SCSI
RAID Controllers

2

At a Glance

Component Views

4100FS.book Page 3 Tuesday, July 8, 2003 4:35 PM

Components

This section provides a descriptive overview of each of the major components
that comprise the NexStor 4100FS Storage System enclosure.

Front Bezel

The front bezel houses the Status LEDs, Drive LEDs, and alarm reset button. When
removed, the user has access to the disk drives. The front bezel can be installed or
removed without interruption to system activities.

Embedded within the front bezel is the electronic package that provides the
communication with the SAF-TE processors (SEP). The SAF-TE processors control
the signals to the front panel through a smart interconnect. Power is applied to the
front bezel through a smart interconnect edge connector, where a control circuit
monitors the bezel for a proper connection. When the bezel is properly installed
and power is applied to the enclosure, the bezel is immediately energized.

Chapter 1 - Getting Started

Refer to “Control and Monitoring” on page 15 for details on the monitoring
functions.

Drive LEDs

R

e

s

e

t A

la

r

Status LEDs

Alarm Reset Button

m

Removable Front Bezel

To remove the bezel and gain access to the disk drives, use a Phillips screwdriver
to release both bezel fasteners, then grasp and remove the bezel. The fasteners
rotate one-quarter turn clockwise to lock and counter-clockwise to unlock.

Components

3

4100FS.book Page 4 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

AC Power

The power system consists of two 350-watt hot-pluggable power supplies, each
with independent AC power cords and cooling fans providing the enclosure with
“N+1” redundant power. Each power supply has auto-switching circuitry for use
with either 100V or 240V AC systems.

Power On LED

Fault LED

Power Supply

Power is applied to the enclosure by pressing each of the two power supply
On/Off switches to their “On” position. A Power On LED located on each power
supply will be illuminated indicating that AC power has been applied. The front
bezels’ Power On LED will also be illuminated indicating that power has been
applied.

Each power supply also incorporates an amber general Fault LED. If the power
supply is installed and power is not applied to the power supply or the power
supply cooling fan fails, the Fault LED will illuminate, along with an audible alarm.

The front bezels’ Power Supply Status LED will illuminate green when both
power supplies are on and operating normally. If only one power supply is
operational, the Power Supply Status LED will be illuminated amber.

Each power supply has an AC Power Cord Module. The module has a power
cord bale incorporated into the design to secure the power cord once it has been
properly inserted. The bale prevents inadvertent disconnects.

4

AC Power

4100FS.book Page 5 Tuesday, July 8, 2003 4:35 PM

Cooling Fan Module

The cooling system consists of two high-performance (80-CFM) cooling fans

mounted in a single fan module which slides into a bay at the rear of the

enclosure. The design of the fan module provides for an easy-to-install user-

replaceable component in a live environment without interruption of service.

If any one fan should fail, cooling redundancy and efficiency are degraded. The

cooling fans and enclosure temperature are constantly monitored by the SES

processor for fault conditions. In the event of a fault condition the front panel Fan

Status LED will change from a green state to a solid amber state in the case of a fan

failure, or to a blinking amber green state in the case of an over-temperature

condition. In both cases an audible alarm sounds. The SES processor will also

provide notification data to monitoring software, such as StorView.

Chapter 1 - Getting Started

WARNING:

Do not operate the enclosure for extended periods of time (greater

than 5 minutes) with the cooling fan module removed.

Fan Speed Override Control
Jumpers JP1 (Fan 0)
and JP2 (Fan 1)

Cooling Fan Module

The

NexStor 4100FS Storage System

monitors temperature in three different areas,

the drive bay, the RAID Controllers, and the power supplies. There are several

steps the storage system performs to prevent component damage due to over

temperature conditions.

Cooling Fan Module

5

4100FS.book Page 6 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

If the drive bay area reaches a temperature of 50°C (122°F) an audible alarm will

sound, the front panel Fan Status LED will toggle amber green, and the monitoring

software will post a warning message. These notifications give the user a warning

that some condition is causing the enclosure temperature to exceed the preset

value, and an action is required by the user to determine the cause and take

corrective measures. It may be due to a blockage of air flow or a low fan speed.

If any RAID Controller reaches a temperature of 65°C (149°F) an audible alarm will

sound, the front panel Fan Status LED will alternate amber and green, and the

monitoring software will post a warning message. If the temperature on any RAID

Controller continues to rise and reaches 71°C (159°F) the controller will flush its

cache and shutdown. If it is the only RAID Controller (Simplex mode) or the only

remaining controller (surviving controller from a failed over operation) then the

controller will also spin down the disk drives at this temperature.

If any power supply reaches 85°C (185°F) the power supply will shut down.

SAF-TE Disk I/O card

The

has a firmware-based VT-100 interface which provides

an option to manage fan speed. This option provides a whisper mode fan

operation for noise sensitive environments. When enabled (default), and based on

a set of conditions, the software will manage the cooling fans RPM speed to

maintain the enclosure temperature while minimizing noise levels. Refer to

“Enclosure Fan Speed Control” on page 100 for more details on using this option.

A manual override of the fan speed control is available for special circumstance

environments. Referring to the illustration on the preceding page, two jumpers are

provided on the fan module printed circuit board to override the software control

of the fan speeds. This hardware setting routes full power voltage to the fans for

maximum operational speed, which is greater than the maximum speed set by the

automatic software control. This configuration is normally used when fan speed

noises are not an issue, and the ambient operating temperature is at or above 30°C

(86°F), thus ensuring that maximum available cooling is being provided.

The jumpers JP1 and JP2 by default are offset, which enables the automatic fan

speed control. The jumper JP1 controls Fan 0 and JP2 controls Fan 1. Placing the

jumper on both pins for each jumper will override the automatic setting and

configure the fans to maximum power.

6

Cooling Fan Module

4100FS.book Page 7 Tuesday, July 8, 2003 4:35 PM

SAF-TE Disk I/O Card

Chapter 1 - Getting Started

WARNING:

The SAF-TE Disk I/O card is NOT HOT SWAPPABLE. You must

POWER DOWN the enclosure prior to removing or installing this card.

The SAF-TE Disk I/O card provides the built-in environmental and system status

monitoring, as well as, host connectivity to the disk drives, It also houses the

switches for setting SCSI IDs, VT-100 communication protocols, and drive spin up

options.

This card has two SAF-TE processors (SEPs) that continuously monitors the

enclosure for temperature status, fan status, power supply status, and SCSI

channel status. The SEPs are responsible for reporting environmentals and system

status to the front bezel LEDs/audible alarms and external monitoring software.

The monitoring system is fully compliant with SAF-TE specification protocol

version 1.0.

n1

n2

atio

Channel 1

Termination Jumper

Channel 1

C

hannel 2

atio

ur

ur

onfig

onfig

C

C

RAID

C

N

C

F

N

G

F

1

G

S

2

R

P

A

A

ID

S

R

B

CHANNEL -1

J

P

8

TERMINATION

TO

DISABLE

ADD JUMPER

J

P

7

TERMINATION

TO

DISABLE

ADD JUMPER

CHANNEL -2

P

E

A

A

1

U

R

D

D

E

LY

S

2

R

S

M

E

L

T

T

S

R

T

T

R

T

Spare

S

RDA

A

0

0

1

1234 5678

Channel 2

Termination Jumper

1

D

Spare 2

BAUD RateSelect

S1BDDLR

rive Start

e

mot

lay D

e

e

R

M

UP (Off -1)

DOWN (On -0)

rt

ive Sta

Dr

At power up, the SAF-TE processors read the SCSI switch settings and configures

the system for the appropriate addresses. It then executes instructions from

firmware performing a self-test diagnostics. The firmware is flash upgradeable

using the SAF-TE RS-232 Service port located below the I/O card slots at the rear

of the enclosure. The firmware also contains the necessary functions for

enclosure management via a VT-100 interface.

NOTE:

Specific switch settings are discussed later in the Installation chapter.

SAF-TE Disk I/O Card

SAF-TE Disk I/O Card

7

4100FS.book Page 8 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

The SAF-TE Disk I/O card has two 68-pin VHD/CI SCSI connectors which

provides the daisy-chain drive connectivity to the expansion JBOD enclosures.

The connectors are labeled “Channel 1” and “Channel 2.”

NOTE:

Termination is automatic and provided internally on the drive mid-plane
circuit board. It is configured using two sets of jumpers on the card.

Changes are not required for the RAID enclosure’s SAF-TE Disk I/O card jumpers.

These jumpers enable and disable the automatic termination. Add the jumper

(installed on both pins) to disable the automatic termination. JP8 for Channel 1

and JP7 for Channel 2. (The default position is the jumper is offset or

unjumpered, that is automatic termination is enabled.)

In daisy-chain configurations, you will be required to make changes to the

jumper settings in the enclosure’s SAF-TE Disk I/O card in that appears at the end

of the daisy-chain. In each supported configuration, a NexStor 4000S (JBOD)

enclosure will be the enclosure at the end of the daisy chain. In some

configurations multiple JBOD enclosures can appear at the end of the chain.

Those enclosures will need to have their jumper settings, JP7 and JP8, set to the

jumpered position. During installation setup, instructions are provided to

correctly configure the daisy-chain enclosure settings.

8

SAF-TE Disk I/O Card

4100FS.book Page 9 Tuesday, July 8, 2003 4:35 PM

Host SCSI I/O Card

Chapter 1 - Getting Started

WARNING:

The Host

SCSI I/O

card is NOT HOT SWAPPABLE. You must POWER

DOWN the enclosure prior to removing or installing this card.

The Host SCSI I/O card is installed to provide the Channel 0 and Channel 3 SCSI

drive ports. These two drive channels are used for drive channel expansion

providing the connection to the NexStor 4000S JBOD enclosure(s).

NOTE:

The SAF-TE Disk I/O card provides Channels 1 and 2 SCSI drive ports.

TERMINATION

JP4

TO DISABLE

ADD JUMPER

JP2

Termination Jumpers

for Disk Channels 0 and 3

Channel 3

Channel 0

JP3

C

hannel -0

JP1

TERM POWER

TO SUPPLY

ADD JUMPER

C

hannel -3

Host SCSI I/O Card

The Host SCSI I/O cards are configured with automatic termination and do not

require any external terminators. Jumpers on the card are provided to configure

Term Power and Termination, however their settings do not require any changes

for the RAID enclosure. The default settings are: JP1 and JP2 unjumpered or

offset, and JP3 and JP4 installed on both pins.

Host SCSI I/O Card

9

4100FS.book Page 10 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

Fibre Channel Host I/O Card

WARNING:

The

Fibre Channel Host I/O

card is NOT HOT SWAPPABLE. You must

POWER DOWN the enclosure prior to removing or installing this card.

The Fibre Channel Host I/O card provides the four host ports configured with

internal active hubs and Small Form Factor Pluggable (SFP) module cages for

connectivity from the host computer (server) to each RAID controller. The cards

support both optical and copper SFP transceivers providing NCITS T11 Fibre

Channel compliant device connectivity at speeds of 1.0625 Gb per second or

2.125 Gb per second.

OFF = 2 GIG

O

N = 1 G

JUM

PER

IG

JP

5

T

P1

P2

P4

P3

x

P1

R

x

T
x

P

2

R

x

T

x

P4

R

x

T
x

P3

Jumper sets FC Loop Speed.
(On configures 1 Gb mode and
offset sets the 2 Gb mode.)

The card has a jumper installed to allow the user to set the Fibre Channel Loop

speed. With the jumper, (JP5), installed it c0onfigures the loop speed to 1 Gb and

with the jumper off set it configures the loop speed to 2 Gb mode.

In the Installation chapter you will find individual cabling diagrams for each

supported topology that assist with the host to enclosure cabling. To ensure

proper failover and failback operations and LUN presentation, cable your system

based on the diagram for your selected topology.

10

Fibre Channel Host I/O Card

Fibre Channel Host I/O Card

4100FS.book Page 11 Tuesday, July 8, 2003 4:35 PM

RAID Controllers

The enclosure houses one or two RAID Controllers. It supports configurations for
single controller Stand-Alone topologies and dual controller Active-Active/
Active-Passive topologies.

Chapter 1 - Getting Started

The RAID Controller is a Fibre Channel

-to-

Ultra160

SCS

I RAID controller. It has
four 2 Gb Fibre Channel host ports and four Ultra160 SCSI disk channel ports.
This controller is occasionally referred to throughout this manual and in the
software as a JFS224 model.

The

controller

operates in one of three different modes: Stand-Alone, Active-Active.
or Active-Passive. In the Stand-Alone mode, the controller operates autonomously.
In the Active-Active/Active-Passive mode, the two controllers operate as a pair. If
one controller fails in the Active-Active/Active-Passive mode, the other can take
over the failed controller’s work.

The Disk Array Administrator software is embedded in the controller firmware,
and requires no specific software drivers for the host operating system. These
utilities are accessed through the Java-based AdminiStor storage management
software or a VT-100 terminal. (See either the AdminiStor Software Guide or the
VT-100 Software Guide.)

Ejector Handles

RAID Controller

RAID Controllers

11

4100FS.book Page 12 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

Optical SFP Transceiver

The Fibre Channel Host I/O card uses a hot-swappable Small Form-Factor

Pluggable (SFP) transceiver. The optical transceiver provides operations up to 2.5

Gb per second. The transceiver includes a lost signal detect circuit which provides

TTL logic high output when an unusable input signal is detected.

The transceivers are available in a few different models which are determined by

the type of ejector mechanism.

The type A model uses a pivoting mechanism. Use the edge of your finger tip
or the finger nail to engage the ejector and pull towards you.

The type B model uses a similar mechanism to the type A except it rotates
about a pivot point. Use the edge of your finger tip or the finger nail to
engage the ejector and pull towards you.

The type C model uses a pull tab to eject the transceiver. Grasp and pull the
tab to release the transceiver.

Ejector Release Lever

Ejector Release Tab

Type A

Ejector Release Lever

Type C

SFP Optical Transceiver Models

Type B

The SFP transceiver is Class 1 Laser safety compliant and conforms to Class 1 eye

safety standards.

CAUTION:

Do not look into the laser light beam for any extended period of time.

Dust covers are provided to protect the

recommend using the dust covers provided when a connector is not in place.

12

Optical SFP Transceiver

optical

transceivers’ optics. We highly

4100FS.book Page 13 Tuesday, July 8, 2003 4:35 PM

Channel 1

Channel 3

CH 3

T

x

P

-1

R

Chapter 1 - Getting Started

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

J

P

R

L

P

J

Y

U

1

M

P

E

J

P

3

R

M

I

N

J

A

T

P

I

O

3

N

B

L

E

R

M

IN

A

T

IO

N

B

L

E

Channel 2

Channel 0

x

T

x

P

-2

R

x

T
x

CH 0

P

-1

R

x

T
x

P

-2

R

x

C

o

n

tro

lle

r 2

C

o

n

tro

lle

r 1

R

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

J

P

R

L

P

J

Y

U

1

T

E

D

I

S

A

M

P

E

R

T

E

D

I

S

A

M

P

E

R

M

J

P

P

E

T

2

R

O

A

D

D

J

J

U

P

4

J

P

T

2

O

A

D

D

J

J

U

P

4

Install the Dust Covers
when the optical transceiver
port is not in use.

Installing and Removing SFP Dust Covers

Optical SFP Transceiver

13

4100FS.book Page 14 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

RS-232 Service Ports

Located below the I/O card slots is the SAF-TE service port and two RAID

Controller service ports. Refer to the illustration under “At a Glance” for the port

locations on the rear panel.

Controller 1: Service Port

SAF-TE Service Port

CTRL 1

SAF-TE

CTRL 2

Controller 2: Service Port

SAF-TE Service and RAID Controller Service Ports

The SAF-TE Service port provides an RS-232 serial interface to the SAF-TE Disk

I/O card allowing firmware uploads, and maintenance/service monitoring of the

SEPs. This port is labeled “SAF-TE.”

The two RAID Controller service ports, which are labeled “CTRL 1” and “CTRL 2,”

provide an interface to each RAID Controller. They provide access to the

firmware-based Disk Array Administration software and provide options to

update the firmware. The ports are standard DB9 connectors which provides a

connection using a null-modem serial cable for VT-100 terminal and host serial

port communication via emulation software.

14

RS-232 Service Ports

4100FS.book Page 15 Tuesday, July 8, 2003 4:35 PM

Control and Monitoring

An integral part of the NexStor 4100FS Series enclosures is its control and monitor
capabilities.

The

SAF-TE processors
conditions such as enclosure temperature, cooling fans, power supplies, and
bus status
audible alarm notifications. This monitored information is also communicated to
external monitoring software.

Refer to “VT-100 Interface Enclosure Monitoring” on page 110 for complete
details.

The RAID Controllers provide monitoring data for its environmental condition
and logical arrays. They communicate that data to the front bezel LEDs,
third-party configuration and monitoring software, and the VT-100 Disk Array
Administration firmware-based software interface for management and
monitoring. (Refer to the software user’s guide for configuration, management,
and monitoring of the controllers and logical arrays.)

Chapter 1 - Getting Started

provide monitoring data for the enclosure environmental

SCSI

. This data is reported to the monitoring system to provide LED and

Drive Status LEDs

(left column of LEDs)

Drive Activity LEDs

(right column of LEDs)

Power On LED

Channel Status LED

Power Supply Status LED

Fan Status LED

Alarm Reset Button

Reset Alarm

Front Bezel LEDs and Reset Button Identification

The NexStor 4100FS Series incorporates a “One-Touch Annunciation” which
provides an LED readout of the current switch configuration, host interface type,
bus configuration, communication BAUD setting, drive spin-up settings, and
controller status. Refer to “One-Touch Annunciation” on page 85 for more details.

Control and Monitoring

15

4100FS.book Page 16 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

Status Indicator LEDs

The Status Indicator LEDs comprise the Power-On LED, Channel Status LED,
Power Supply Status LED, and Fan Status LED. These series of LEDs are grouped
on the right side of the front bezel directly above the Alarm Reset button. The
following is an description of each of these LEDs.

Power-On LED

The Power-On LED signifies that the enclosure is powered on and will be
illuminated green when power has been applied.

Channel Status LED

The Channel Status LED will indicate the status of the logical arrays.

Power Supply Status

The Power Supply Status LED indicates the condition of the power supplies. The
LED will illuminate steady green when both power supplies are functioning
normally and will change to amber if one power supply should fail or is turned off.

Fan Status

The Fan Status LED indicates the condition of the cooling fans. The LED will
illuminate green when both fans are functioning normally and will change to
amber if any of the fans fail.

Drive LEDs

The Drive LEDs are located on the left side of the front bezel in between the
ventilation ribs and comprise the Drive Status LEDs and Drive Activity LEDs. The
Drive LEDs are grouped in pairs and are in the general location of the disk drive
slot. There are 12 Drive Status and Drive Activity LEDs, one group o pair for each
disk drive.

Refer to “Drive LEDs” on page 91 for detailed information.

16

Status Indicator LEDs

4100FS.book Page 17 Tuesday, July 8, 2003 4:35 PM

Audible Alarm

An audible alarm will sound when any of the enclosure’s component condition
changes to an abnormal state. To silence the alarm, press the Alarm Reset button
located on the front bezel. The corresponding alarms’ LED will remain
illuminated until the condition returns to a normal state.

Chapter 1 - Getting Started

Audible Alarm

17

4100FS.book Page 18 Tuesday, July 8, 2003 4:35 PM

Chapter 1 - Getting Started

18

Audible Alarm

4100FS.book Page 19 Tuesday, July 8, 2003 4:35 PM

Chapter 2

Theory of Controller Operation

This chapter provides a functional overview and understanding of the supported

topologies and operating modes for the NexStor 4100FS Series Storage System.

With this information, you will be able to make the best choice based on the

supported topologies, to set up your storage solution.

There are three operating modes available for the NexStor 4100FS Series: Stand­Alone, Active-Active, and Active-Passive. The NexStor 4110FS model with its single
RAID controller supports the Stand-Alone operating mode, and the NexStor
4120FS with its dual RAID controllers supports the active-active or active-passive
operating modes. Either mode can function in single or dual port configurations.

Operating Modes Overview

The operating modes let you configure the RAID controllers to support a variety

of host environments.

■ Stand-Alone Dual Port – Use this mode when you have a NexStor 4110FS

and require single-HBA host access.

■ Active-Active Single Port – Use this mode when you have a NexStor 4120FS

and require both controllers to be active and processing host I/O

operations. This operating mode uses one of the FC host ports of each

RAID controller and supports single or dual host system connections.

19

4100FS.book Page 20 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

■ Active-Active Dual Port – Use this mode when you have a NexStor 4120FS

and require both controllers to be active and processing host I/O
operations. This operating mode uses one both FC host ports of each RAID
Controller and supports single or dual host system connections.

■ Active-Passive Dual Port – Use this mode when you have a NexStor 4120FS

and require only one RAID controller to be active. The other controller is
in a passive mode as a “hot” standby. This operating mode does not
provide the performance compared to either of the Active-Active modes.

The naming conventions are broken down into two parts: the first part refers to
the number of controllers and their state, and the second part refers to the state
of the controller’s host ports (not to be confused with the physical I/O ports).

Stand-Alone designates a single controller system, and Active-Active or
Active-Passive designates a dual controller system.

The Single Port or Dual Port component of the name convention refers to the
number of enabled Host ports on the active controller. Note that there are two
host ports on each controller (internally).

Operating Mode Number of

Stand-Alone Dual Port

(NexStor 4110FS)

Active-Active Single Port

(NexStor 4120FS)

Active-Active Dual Port

(NexStor 4120FS)

Active-Passive Dual Port

(NexStor 4120FS)

Controllers

12 4

21 4

22 4

2 2 (active controllers only) 4

Number of Active Host Ports per

Controller

Number of Disk

Channels

Another consideration when selecting one of these topologies is whether you will
enable or disable the internal hubs. Each NexStor 4100FS Series Storage System
enclosure has built-in internal hubs. These internal hubs, when enabled,
essentially replace the need for external hub hardware. They provide for up to
four (4) host connections. You can enable or disable this feature using the
AdminiStor software or the VT-100 interface.

20

Operating Modes Overview

4100FS.book Page 21 Tuesday, July 8, 2003 4:35 PM

Stand-Alone Mode

In Stand-Alone mode, the RAID Controller operates autonomously. The controller
has two Fibre Channel (FC) Host ports, which are labeled in the diagrams as “FC1”
and “FC2.” This operating mode uses both FC host ports.

Advantages:

■ Low cost entry-level RAID storage solution that is field upgradeable.

■ Allows the use of dual host ports to maximize bandwidth and host IOPs.

(For example: this configuration is ideal for sequential access patterns,
where disks can supply more than 160 MB/sec of data to a host.)

■ Host system has multiple paths to each array (with host driver).

■ All arrays appear on both host ports at the same LUN positions in Dual Port

mode. (If both ports are connected to the same host, the host will see a
given array twice unless it does special processing to detect duplicate paths
to array. A host device driver may provide this functionality.)

■ Hosts can detect redundant paths to an array by matching array serial

numbers from SCSI inquiry data.

Chapter 2 - Theory of Controller Operation

Disadvantages:

■ The controller is a single point of failure.

■ Third-party software is required for upstream path fail-over.

The following two diagrams depicts the logical function of this operating mode.

Drive I/O Connectors

Controller FC Host Ports

HOST I/O

P1

P2

P4

P3

T= Location of internal termination

Host I/O Connectors

RAID Controller

FC1

Active

Active

FC2

Controller 1 (A)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

T

Controller Disk Ports

Drives 7 - 12

Drives 1 - 6

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Stand-Alone Dual Port with Hubs Disabled

Stand-Alone Mode

21

4100FS.book Page 22 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

Controller FC Host Ports

HOST I/O

P1

HUB

P2

P4

P3

RAID Controller

FC1

Active

Active

FC2

HUB

Controller 1 (A)

T= Location of internal termination

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drive I/O Connectors

T

T

T

T

Drives 7 - 12

Drives 1 - 6

SEP

SEP

SAF-TE Disk I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Controller Disk Ports

Host I/O Connectors

Host SCSI I/O

Stand-Alone Dual Port with Hubs Enabled

Active-Active Mode

In the Active-Active mode, the two RAID Controllers cooperate to provide
redundancy. If one controller fails, the remaining controller will take over the
failed controller’s functionality. Each controller has two Fibre Channel Host ports,
they are labeled in the diagrams as FC1 and FC2.

In the Active-Active:Single Port mode, one FC Host port is normally active, and
the other is normally passive. In a failed over condition, the passive port of the
surviving controller becomes active and assumes the identity of the failed
controller. This failover process occurs without user intervention.

In the Active-Active:Dual Port mode, both FC Host ports are active and each

controller normally presents the LUNs it owns on both of its FC Host ports. This

provides the host with dual-ported access. However, when a controller fails, the

surviving controller presents its LUNs on one of its FC Host ports and the LUNs of

the failed controller on its other FC Host port.

22

Active-Active Mode

4100FS.book Page 23 Tuesday, July 8, 2003 4:35 PM

Presentation of LUNs in Active-Active:Single Port Mode

Chapter 2 - Theory of Controller Operation

Controller Status Controller 1

FC 1 Host Port

Both controllers online Controller 1

LUNs

Controller 1 Fails Inactive Inactive Controller 2

Controller 2 Fails Controller 1

LUNs

Controller 1

FC 2 Host Port

Passive Controller 2

Controller 2
LUNs

Controller 2

FC 1 Host Port

LUNs

LUNs

Inactive Inactive

Controller 2

FC 2 Host Port

Passive

Controller 1
LUNs

Presentation of LUNs in Active-Active:Dual Port Mode

Controller Status Controller 1

FC 1 Host Port

Both controllers online Controller 1

LUNs

Controller 1 Fails Inactive Inactive Controller 2

Controller 2 Fails Controller 1

LUNs

Controller 1

FC 2 Host Port

Controller 1
LUNs

Controller 2
LUNs

Controller 2

FC 1 Host Port

Controller 2
LUNs

LUNs

Inactive Inactive

Controller 2

FC 2 Host Port

Controller 2
LUNs

Controller 1
LUNs

In either mode, arrays can only be accessed by the controller that currently owns
them. One controller will have no visibility to the others’ arrays. If one controller
fails, the surviving controller will take ownership of all arrays. Pool spares and
unassigned disks are visible to both controllers.

The two controllers communicate via the back-end disk buses. Although many
different message types are sent between the two controllers, the majority of the
data is write cache data. In the default write-back caching mode, any data that the
host writes to one controller is copied to the other controller before a good status
is returned to the host. By mirroring the data, if a controller fails, the surviving
controller will have a copy of all data that has not been written to the disk.

NOTE: The controllers communicate with each other using SCSI initiator IDs 6

and 7 on each of the disk channels. You should not change these IDs.

Each controller will send heartbeat messages to its partner controller. If a
controller does not receive a heartbeat message within a set time period, it will
kill the other controller assuming that it has malfunctioned. After killing the other
controller, it will take ownership of the arrays and will activate its passive host
port to assume the defunct controller’s identity.

Active-Active Mode

23

4100FS.book Page 24 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

The following diagrams illustrate the logical function of the controllers when both
are active and then again when one controller has failed. First we present the
controllers in Single Port mode, then in Dual Port mode.

Controller FC Host Ports

HOST I/O

P1

P2

P4

P3

Host I/O
Connectors

HUB

HUB

FC1

FC2

Controller 1 (A)

FC2

FC1

Controller 2 (B)

RAID Controller

Active

Passive

RAID Controller

Active

Passive

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

T

T

T

T= Location of internal termination

T

Controller Disk Ports

Drive I/O Connectors

Active-Active Single Port Mode with Hubs Enabled and Both Controllers Active

Controller FC Host Ports

HOST I/O

P1

P2

P4

P3

Host I/O
Connectors

HUB

HUB

RAID Controller

FC1

Active

Active

FC2

Controller 1 (A)

ACTIVE (FAILED OVER)

RAID Controller

Inactive

FC2

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Drives 7 - 12

Drives 1 - 6

T

SEP

SEP

SAF-TE Disk I/O

Drive I/O Connectors

T

T

T

T

Host SCSI I/O

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

CH 1 CH 2

CH 0 CH 3

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Active-Active Single Port Mode with Hubs Enabled and Controller 2 Failed

24

Active-Active Mode

FC1

Controller 2 (B)

Inactive

FAILED

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T= Location of internal termination

T

Controller Disk Ports

4100FS.book Page 25 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

Controller FC Host Ports

FC1

FC2

Controller 1 (A)

FC2

FC1

Controller 2 (B)

RAID Controller

Active

Active

ACTIVE

RAID Controller

Active

Active

ACTIVE

Disk

CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

T

T

T

T= Location of internal termination

T

Controller Disk Ports

HOST I/O

P1

HUB

P2

P4

P3

Host I/O
Connectors

HUB

Active-Active Dual Port Mode with Hubs Enabled and Both Controllers Active

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Controller FC Host Ports

HOST I/O

P1

P2

P4

P3

RAID Controller

Drives 7 - 12

Drives 1 - 6

HUB

HUB

FC1

FC2

Controller 1 (A)

Active
(CTRL 1)

Active
(CTRL 2)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

ACTIVE (FAILED OVER)

RAID Controller

Host I/O
Connectors

FC2

FC1

Controller 2 (B)

Inactive

Inactive

FAILED

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

T= Location of internal termination

T

Controller Disk Ports

Active-Active Dual Port with Hubs Enabled and Controller 2 Failed

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Active-Active Mode

25

4100FS.book Page 26 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

In the failed-over condition, the active controller will present its native identity on
one FC Host port, and the failed-over controller’s identity on the other port. This
is slightly different when using a dual port configuration, where each array is
presented on both FC Host ports.

After the controller has failed-over, the user can replace the failed controller with
a new controller, and the arrays will fail-back to the new controller without
intervention. If you want to retry the failed controller, it’s necessary to release the
kill line that the survivor asserts. This is accomplished using AdminiStor software
or VT-100 Disk Array Administrator software (Other Controller Menu – Unkill
Other option).

If you replace a failed controller with a new controller, the new controller will
attempt to retrieve its configuration (see “Understanding Mirrored Operations” on
page 29) from the surviving controller. This means that the replacement
controller will boot up with the same configuration as the controller it replaced.

If the surviving controller for some reason does not contain a valid mirrored
configuration, you must manually configure the replacement controller. The correct
host target ID must be set, and the Active-Active operating mode must be enabled.

The Active-Active controller system makes some assumptions about the host
operating system’s SCSI processing. The host must retry incomplete commands
after a specific time period. This retry is necessary because a host may issue a
command to a controller, the controller fails over, and the new controller does
not have a copy of the command.

Advantages:

■ LUN isolation with the single port-single host system.

■ Dedicated LUNs with a single port-dual host system.

■ Access to all LUNs with a dual port-dual host quad cabling configuration.

■ Provides transparent fail-over and fail-back.

■ No single point of failure with the Active-Active-Dual Port configuration,

must have third party clustering and file share/file lock software installed.

Disadvantages:

■ Host HBA or the Fibre loop are single points of failure.

■ In single host configuration, the host system is also a single point of failure.

■ No upstream fail-over with a single host that has dual HBAs or dual host

systems.

■ Protected LUN sharing between two host systems is not supported without

the use of third party file share/file lock software.

26

Active-Active Mode

4100FS.book Page 27 Tuesday, July 8, 2003 4:35 PM

Active-Passive Mode

In the Active-Passive mode, there is one active and one passive RAID Controller.
The primary (active) controller operates with both of its FC Host ports enabled and
presents all LUNs on both ports to maximize bandwidth. The secondary (passive)
controller keeps both its FC Host ports passive, and does not service any LUNs. Both
controllers monitor each other’s heartbeats and the passive controller receives write
data and parity logs from the active controller.

If the active controller fails, normal fail-over occurs to the passive controller, which
becomes active by enabling both of its FC Host ports
arrays

.

Controller 1 is the default active controller. Controller 2 is always passive unless
Controller 1 fails. The default active controller is not selectable.

Chapter 2 - Theory of Controller Operation

and takes ownership of all

Because the active controller does not have to mirror the passive controllers data,
more write-back cache memory can be made available to the active controller. This
means that half the cache memory can be dedicated to writes and half to reads. This
is the same allocation used in Stand-Alone mode.

Each controller sends heartbeat messages via the SCSI buses to its partner
controller. If a controller does not receive a heartbeat message within a set time
period, it will kill the other controller assuming that it has malfunctioned. After
killing the other controller, it will take ownership of the arrays and will activate
its passive host port to assume the defunct controller’s identity.

NOTE: The controllers communicate with each other using SCSI initiator IDs 6

and 7 on each of the disk channels. You should not change these IDs.

On the following page, you will find diagrams which illustrates the Active-Passive
mode logical functions when the controllers are operating in a normal condition and
then in a failed condition.

Advantages:

■ Transparent failover and failback.

■ Single point of failure is the host system. (There are no single points of

failure when dual host systems are configured with clustering software.)

■ Upstream fail-over.

Disadvantages:

■ Lower performance than the Active-Active mode.

■ Upstream fail-over requires third party software.

Active-Passive Mode

27

4100FS.book Page 28 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

■ Careful administration is necessary, since all the arrays are reported twice

to the host operating system which requires the use of third party file
share/file lock software to prevent data overwrites.

HOST I/O

P1

P2

P4

P3

P1

P2

P4

P3

Controller FC Host Ports

HUB

HUB

FC1

FC2

Controller 1 (A)

RAID Controller

Active

Active

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

ACTIVE

RAID Controller

Host I/O
Connectors

FC2

FC1

Controller 2 (B)

Passive

Passive

PASSIVE

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

T= Location of internal termination

T

Controller Disk Ports

Active-Passive Dual Port with Hubs Enabled and Controller 1 Active

Controller FC Host Ports

HOST I/O

Host I/O
Connectors

HUB

HUB

RAID Controller

FC1

Inactive

Inactive

FC2

Controller 1 (A)

FAILED

RAID Controller

Active

FC2

FC1

Active

Controller 2 (B)

ACTIVE (FAILED OVER)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

T

T

T

T= Location of internal termination

T

Controller Disk Ports

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Active-Passive Dual Port with Hubs Enabled and Controller 1 Failed

28

Active-Passive Mode

4100FS.book Page 29 Tuesday, July 8, 2003 4:35 PM

Understanding Mirrored Operations

The mirroring feature causes configuration changes to be mirrored to the

opposite controller in an Active-Active configuration. The mirrored configuration

is stored in the other controller’s firmware (Flash). If a controller fails and is

replaced, then on boot up, it attempts to retrieve this configuration data from the

opposite controller. The intent is that the user will not need to reconfigure a new

controller when it is replaced.

NOTE: Dual RAID controller operations require that both controllers must have

If a replacement controller uses its mirrored configuration, you will see it print a

message during boot (saying it’s using the mirrored configuration), and then it

will reboot (rebooting is necessary for the mirrored configuration to take effect).

Chapter 2 - Theory of Controller Operation

the same version firmware installed.

Here’s how the mirrored configuration works in different scenarios:

■ Stand-Alone mode – The feature is not active.

■ Active-Active mode where both controllers are operating normally – Each

controller uses its own local configuration.

■ Active-Active mode where one controller boots and the other controller

does not boot due to some problem – The controller that boots uses its

local configuration.

■ Active-Active mode where one controller is up when a new replacement

controller boots – the new controller attempts to get its mirrored

configuration from the other controller. If it is not available, it will try to

use its local configuration. If that is not available, then it will use the

default settings.

NOTE: When Clear Configuration is selected in the Flash Utility menu, both the

local Flash configuration and the other controller’s mirrored configuration

image are cleared. Also, there is an option (J) in the Flash Utility menu

that can be used to tell a booting controller to temporarily ignore the

mirrored configuration. This can be used, if for whatever reason, the

mirrored configuration should not be used by a booting controller. This is

primarily a safeguard, because a retrieved mirrored configuration is

checked extensively before it is used.

Understanding Mirrored Operations

29

4100FS.book Page 30 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

A Word about Clustering

Minimizing Downtime for Maximum Data Availability

So-called open systems, such as Windows NT servers, just don’t provide the level

of availability that IS managers are familiar with on mainframes. A partial solution

to this problem is server clustering.

Clusters consist of two or more loosely coupled systems with a shared-disk

subsystem and software that handles failover in the case of a node (host) failure.

In most cases, hardware/software failover is performed automatically and is

transparent to users, although users will experience performance degradation as

processing is shifted to another cluster node. In some cases this failover can

occur in a matter of seconds.

High availability of data and applications is by far the most compelling reason to

go with clustering technology. For example, the accepted rule is that stand-alone

UNIX systems can provide 99.5% uptime. Adding a RAID subsystem can increase

the uptime to 99.9%. The goal of clustering is 99.99% availability.

Beyond clustering, fault-tolerant systems can provide 99.9999% uptime. At the

high end, continuous-processing systems offer virtually 100% uptime.

Although the increase from 99.5% to 99.99% availability may seem insignificantly

small, it adds up in terms of minutes per year of downtime. For example,

assuming a 7x24 operation, 99.5% uptime translates into 2,628 minutes — or

more than 43 hours of downtime per year. In contrast, 99.99% uptime translates

into less than one hour (52 minutes) of downtime per year.

Availability figures relate primarily to unplanned downtime. But the advantages

of clusters in terms of planned or scheduled downtime are even more significant.

If you figure two to sixteen hours per month for a server in a large shop.

Planned downtime requires shutting down stand-alone systems entirely. Result:

100% loss of processing for the duration of the downtime. But, with cluster, you

can shut down one node and off-load the processing to other nodes in the cluster

with no interruption of processing.

30

A Word about Clustering

4100FS.book Page 31 Tuesday, July 8, 2003 4:35 PM

High availability is not the only benefit of clustering. In some cases, users may

see advantages in the areas of performance, scalability, and manageability. In

reality, you can expect a 1.6x (80% efficiency) to 1.8x (90% efficiency)

performance increase as you go from one node to two nodes. Going from one

node to a four node cluster generally yields a 2.5x or 3x performance boost.

However, the cluster performance is application dependent. For example, READ

operations may yield a 1.8x performance increase going from one to two nodes,

but in a WRITE intensive application, you may only see a 1.4-1.6x improvement.

Although clusters seem to be relatively simple, they involve complex technology

that can be implemented in a variety of ways. The number of nodes supported

and type of interconnection used, and a number of other features differentiate

cluster implementations. One area of implementation is the manner in which

distributed lock manager is implemented. Some perform this at the user level and

others in the kernel, with the latter enhancing performance.

Chapter 2 - Theory of Controller Operation

In addition to the differing features you should consider the following:

Does the cluster:

• have the ability to hot load new nodes without bringing down the
whole cluster?

• provide automatic or manual failover?

• load balance?

• use a journalized file system?

• provide a fast cluster failover?

• allow for the nodes to be geographically located?

Minimizing Downtime for Maximum Data Availability

31

4100FS.book Page 32 Tuesday, July 8, 2003 4:35 PM

Chapter 2 - Theory of Controller Operation

How Available are Clusters?

The following tables outlines the maximum availability per downtime in the

different architectures:

Architecture Maximum Availability Downtime per Failure Downtime per Year

(in minutes)

Continuous
Processing

Fault-Tolerant 99.9999% Cycles 0.5 - 5

Clusters 99.9 - 99.999% Seconds to

High Availability 99.9% Minutes 500 - 10,000 (disk

Stand Alone
System

100.00% None 0

5 - 500

minutes

mirroring)

99.5% Hours 2,600 - 10, 000
(without disk
mirroring)

Application of Availability

The NexStor 4100FS Series Storage System implementation of availability within

its supported topologies are as follows:

Architecture Corresponding Topology

Continuous Processing Not Available

Fault-Tolerant Active-Passive:Dual Port

Cluster Active-Active:Dual Port

High Availability Active-Active:Single Port

Stand Alone System Stand-Alone:Dual Port

32

How Available are Clusters?

4100FS.book Page 33 Tuesday, July 8, 2003 4:35 PM

Chapter 3

Setup and Installation

Overview

This chapter describes the procedures to install and setup the NexStor 4100FS

Series Storage System. Each section will step you through the hardware

installation, cabling and topology configurations, and upgrades.

It is important to thoroughly review this information and perform the steps of

procedures in each applicable section, in the order in which they are presented.

This will ensure a smooth and trouble-free installation.

The installation is divided into three sections. The first section describes installing

the enclosure(s) into the rack cabinet.

The second

SAF-TE Disk I/O card switch settings and the enclosure-to-enclosure cabling. The

SAF-TE card switches settings determine the SCSI IDs, RAID Addressing, and disk

spinup options. These configurations are categorized by the number of disk

drives. For example, a NexStor 4110FS/4120FS with two daisy-chained NexStor

4000S JBOD enclosures will be described as the “RAID - 36 Drive Configuration.”

The third section, “Topology Host Cabling,” describes the cabling of the primary

section is

“Enclosure Configuration” which describes setting the

RAID storage enclosure to the host system(s).

operating modes: Stand-Alone Dual Port, Active-Active Single Port, Active-Active

Dual Port, or Active-Passive Dual Port.

The topologies are based on the

33

4100FS.book Page 34 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Storage System Detailed Installation

This first section describes preparing and installing the NexStor 4100FS Series

Storage System enclosure(s) into the rack cabinet or the enclosure into its tower

stand (“Installing the Storage System into the Tower Stand” on page 37).

After installing the hardware components, go to the Enclosure Configuration

section and set the SAF-TE Disk I/O card switches and enclosure cabling for your

drive configuration. Then locate the operating mode you wish to employ in the

Topology Host Cabling section and perform the steps to complete the host

cabling.

Installing the Storage System Enclosure into the Rack Cabinet

1 Install the storage enclosure(s) into the rack cabinet.

Select an appropriate location within your rack cabinet. You should consider

the location of the enclosure(s) in relationship to each other to ensure that

the cables will easily reach between enclosures.

CAUTION: The power supplies should be removed prior to installing the

enclosure into the rack cabinet. The enclosure chassis could be

damaged during installation due to the added weight of the power

supplies.

NOTE: Ensure that the selected location in the rack cabinet has adequate air

flow for the front to side and side to side areas.

2 Remove each enclosure from its shipping carton and inspect for obvious

damage. Place the enclosure on a flat surface to work from.

3 Remove the front bezel from the accessory box and store it in a location

where it will not be damaged. It will be installed later in the installation

procedures.

4 Remove the power supplies.

From the rear of the enclosure, remove each power supply by grasping its

handle and pressing in on the release latch with your thumb as you pull each

power supply from the enclosure. Repeat for the other power supply.

34

Storage System Detailed Installation

4100FS.book Page 35 Tuesday, July 8, 2003 4:35 PM

5 Locate the mounting hardware in the accessory kit (mounting rails, screws,

and nuts – on some rack installations you will use cage nuts and on some

racks they will be standard nuts).

Chapter 3 - Setup and Installation

NOTE:

It will be helpful to have an assistant available during the installation.

6 Lift and secure the enclosure into the rack cabinet.

a Position the enclosure in the cabinet at the desired location.

b Secure the left and right front chassis ears to the rack cabinet’s front

vertical members using the supplied screws and nuts. Ensure that they

are even horizontally.

Front Rack Vertical

Member

Mounting

Screw

Mounting

Screw

Nut

Nut

Chassis Ear

Attaching the Chassis Ears

c Install the rear mounting rails using the supplied screws and nuts.

From the rear of the rack cabinet, slide one of the mounting rails into the

slot provided on the left side of the enclosure.

Installing the Storage System Enclosure into the Rack Cabinet

35

4100FS.book Page 36 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Push the rail in the slot until it fits the depth of the rack cabinet drawing
the enclosure level and tight. It should mate with the rear rack cabinet
vertical member.

NOTE:

Be sure that the enclosure is level. Do this by verifying that the
same height mounting location slots on both the front and rear
rack cabinet vertical member being used.

d Secure the left side rail to the vertical member using the screws and nuts.

e Repeat the substeps 6(c) and 6(d) for the right side rail.

Rail Slot

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

JP1

P

R

L

J

Y

U

M

P

E

R

JP3

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

JP1

P

R

L

J

Y

U

M

T

JP2

E

P

R

E

M

T

R

O

I

N

D

JP3

A

A

T

IS

D

I

O

A

D

N

B

J

JP4

L

U

E

M

P

E

R

T

JP2

E

R

M

T

O

I

N

D

A

A

T

I

S

D

IO

A

D

N

B

J

JP4

L

U

E

M

P

E

R

C

o

n

tro

lle

r

2

C

o

n

tro

lle

r 1

Rear Rack Vertical Member

C

h

a

n

n

e

l

1

C

C

h

a

h

n

a

n

n

e

n

l

e

0

l

2

C

h

a

T

n

x

n

e

l

P

3

-1

R
x

T

x

P

-2
C
T

R

R

x

L

T

1

x

P

-1

R
x

S

A

T

F

x

T
E

P

-2

R

x

C
T

R
L

2

36

Nut

Nut

Mounting
Screws

7 Re-install the power supplies. Do this by aligning the power supply with its

open bay and slide the power supply in.

The power supply will fit flush and the latch will reset as the power supply
reaches its fully seated position.

8 Continue now with “Completing the Installation” on page 40.

Installing the Storage System Enclosure into the Rack Cabinet

Rail Slot

Attaching the Rails

4100FS.book Page 37 Tuesday, July 8, 2003 4:35 PM

Installing the Storage System into the Tower Stand

1 Remove the enclosure from its shipping carton and inspect for obvious

damage. Place it on a flat surface to work from.

2 Remove the front bezel from the accessory box and store it in a location

where it will not be damaged. It will be installed later in the installation

procedures.

3 Remove the power supplies.

From the rear of the enclosure, remove each power supply by grasping its

handle and pressing in on the release latch with your thumb as you pull each

power supply from the enclosure. Repeat for the other power supply.

4 Remove the cooling fan module.

Chapter 3 - Setup and Installation

Place your finger in the fan module handle and press with your thumb to

release the latch while pulling the module from the enclosure.

5 Remove the two rear mounting rails. Grasp and pull each rail from the

chassis.

NOTE: The mounting rails were installed in the slots at the factory for the

purpose of shipping.

6 Remove the tower stand case from its shipping carton and inspect for

obvious damage.

7 Locate the accessory kit in the tower shipping carton. It should contain eight

10-32 pan head screws and conversion instructions. (Note, the instructions

enclosed in the box are applicable to upgrading an existing installation.)

8 Rotate the enclosure chassis so that the power supply bays are on the top.

9 Carefully slide the enclosure chassis into the tower stand until it fits flush as

indicated in the illustration (A) on the following page.

10 Secure the top and bottom chassis ears to the tower stand using two each

10-32 pan head screws as indicated in the illustration (B) on the following

page.

11 Re-install the rear mounting rails into the slots at the rear of the chassis as

indicated in the illustration (C) on the following page.

Installing the Storage System into the Tower Stand

37

4100FS.book Page 38 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

12 Using the remaining two sets (4) of 10-32 pan head screws, secure the top

and bottom slide rails as indicated in the illustration (C) below.

Mounting
Screw

A

C

Mounting
Screws

Mounting
Screws

B

38

Installing the Storage System into the Tower Stand

Inserting and Securing the Chassis

4100FS.book Page 39 Tuesday, July 8, 2003 4:35 PM

13 Re-install the cooling fan module. Slide it into its open bay and ensure the

release latch resets indicating the fan module has completely seated.

14 Re-install the power supplies. Slide each power supply into its open bay and

ensure the release latch resets indicating the power supply has completely
seated.

15 Continue now with “Completing the Installation” on page 40.

Chapter 3 - Setup and Installation

Installing the Storage System into the Tower Stand

39

4100FS.book Page 40 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Completing the Installation

1 Install the disk drives.

a Remove each drive from its shipping container and remove the anti-static

protective packaging. Inspect each drive for obvious damage.

b From the front of the storage enclosure, install each disk drive into its

drive slot.

Align the carrier rails with the rail grooves in the drive bay. The drive
carrier tension clips ensure that the disk drive fits very tight, so it requires
some force to push the drive into its bay. Ensure that the drive seats
completely. Repeat this step to populate all the drive slots.

c Re-install the front bezel. Ensure that the bezel mounts to the two stud

post and the bezel lip fits under the chassis top.

Secure the front bezel. Using a Phillips screwdriver, rotate the fasteners
clockwise one-quarter turn to secure the bezel locks.

2 Install the power cords and secure them using the power cord bales.

CAUTION: Ensure that the power supply On/Off switches are in their OFF

position.

40

Completing the Installation

Installing Disk Drives

4100FS.book Page 41 Tuesday, July 8, 2003 4:35 PM

a Ensure that the orientation is such that when the power cord is inserted,

Chapter 3 - Setup and Installation

the bale will be the top of the cord and will fit over and onto the cord.

Bale fits over and onto
the power cord.

Attaching the Power Cord Bales

b Connect the other end of the power cords into a three-hole grounded

outlets or UPS power system.

c Repeat steps 2(a) and 2(b) for the other power cord.

3 Repeat the above steps for each additional storage system enclosure you will

be installing.

This completes the physical hardware installation.

Before You Continue...

The next section, Enclosure Configuration, includes steps and diagrams for
setting the SAF-TE Disk I/O card switches and attaching the required SCSI data
cables for each drive configuration. Locate the applicable configuration and refer
to the steps and diagrams to set the card switches and cable your system.

The Topology Host Cabling section includes steps and diagrams for cabling the
primary RAID storage enclosure to the host computer(s). Review the procedure
steps, then locate the applicable topology and cable your system based on the
diagrams.

The last section of this chapter provides the steps to properly power on or power
off your storage system.

Completing the Installation

41

4100FS.book Page 42 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Enclosure Configuration

In this section, configurations are determined by the number of enclosures and

available number of drives. Configurations with drive numbers greater then the

primary RAID enclosure’s 12 drives, will require attaching a NexStor 4000S JBOD

enclosure(s). Follow the steps of procedures for your configuration to set the

switches, jumpers, and connect the SCSI data cables.

RAID - 12 Drive Configuration

The NexStor 4110FS/4120FS can be setup to provide up to a 12 disk drive storage

solution using the primary RAID enclosure only.

1 Remove the SAF-TE Disk I/O card installed in the enclosure. Loosen the two

thumb screws and pull the card from its slot location.

rt

a

Channel 1

Termination Jumper

Channel 1

Channel 2

n1

n2

atio

atio

1

ur

ur

re

re

onfig

onfig

C

C

RAID

Spa

BAUD RateSelect

CNFG 1

CNFG 2

SPARE 1

RAID

SPARE 2

CHANNEL -1

JP

8

T
E
R

M

T

I

O
N
A

D

T

A
I
S

I

D

O

A

D

N

B

L

J

U

E

M
P
E

R

JP

7

T
E
R

M

T

I

O

N

A

D

T

A

I

S

I

D

O

A

D

N

B

L

J
U

E

M
P

E

R

CHANNEL -2

BAUD SEL

DLY STRT

RMT STRT

Channel 2

Termination Jumper

Spa

S

RDA

A

S1BDDLR

0

0

1

1234 5678

2

lay D

e

R

D

M

t

ive S

r

D

rive Start

e

ot

m

e

UP (Off -1)

DOWN (On -0)

SAF-TE Disk I/O Card Switch and Jumper Locations

2 Locate the switches on the card and set them as described in the following

switch setting diagrams.

The switch settings will assign specific SCSI IDs to the drive slots as
indicated, and reserve IDs 6 and 7 for the RAID Controller(s), and IDs 8 or 15
for the SAF-TE processors.

There are two possible switch settings for this configuration. You may choose
either setting to fit your needs.

42

Enclosure Configuration

4100FS.book Page 43 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

SAF-TE ID = 8/8

RAID Controllers = 6 & 7

Switch Settings

S

R

A

A

S1BDDLR

0

D

0

1

1234 5678

M

UP (1)

DOWN (0)

Switch Setting - Option 1 (NexStor 4110FS/4120FS)

SAF-TE ID = 15/15

RAID Controllers = 6 & 7

Switch Settings

S

RDA

A

S1BDDLR

0

0

1

1234 5678

M

UP (1)

DOWN (0)

Dual-Bus Mode

Slot 1

ID 1

Slot 2

ID 3

Slot 3 Slot 6 Slot 9 Slot 12

ID 5

Slot 4

ID 2

Slot 5

ID 4

ID 0

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 7

Slot 8

ID 1

ID 3

ID 5

Slot 10

Slot 11

Dual-Bus Mode

Slot 1

ID 9

Slot 2

ID 11

Slot 3 Slot 6 Slot 9 Slot 12

ID 13

Slot 4

ID 10

Slot 5

ID 12

ID 14

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 7

Slot 8

ID 9

ID 11

ID 13

Slot 10

Slot 11

ID 2

ID 4

ID 0

ID 10

ID 12

ID 14

Switch Setting - Option 2 (NexStor 4110FS/4120FS)

3 Re-install the SAF-TE Disk I/O card. Slide the card into the slot and ensure

that it seats completely. Secure the card by tighten the two thumb screws.

4 Connect a Fibre Channel data cable(s) from the host system(s) HBA(s) to the

primary RAID enclosure Host I/O card as indicated in “Topology Host
Cabling” on page 57.

This completes the RAID - 12 Drive Configuration setup, refer to “Topology Host

Cabling” on page 57 and “Powering On the Storage System” on page 70. Also,

refer to the AdminiStor software guide or the VT-100 software guide for

instructions on setting up and configuring the disk arrays.

RAID - 12 Drive Configuration

43

4100FS.book Page 44 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

RAID - 24 Drive Configuration

The NexStor 4110FS/4120FS can be setup to provide up to a 24 disk drive storage
solution. In this configuration you will attach a NexStor 4000S JBOD enclosure to
the primary RAID enclosure which will extend the drive channels (12 drives per
channel).

1 Remove the SAF-TE Disk I/O card installed in the primary RAID enclosure.

Loosen the two thumb screws and pull the card from its slot location.

Channel 1

Termination Jumper

C

hannel 1

Channel 2

rt

a

2

rive Start

D

m

lay

e

e

R

D

M

t

ive S

r

D

e

ot

UP (Off -1)

DOWN (On -0)

n1

n2

atio

atio

1

ur

ur

re

re

onfig

onfig

C

C

RAID

Spa

BAUD RateSelect

C

N

C

F

N

G

F

1

G

S

R

2

P

A

A

ID

S

R

B

CHANNEL -1

JP8

T
E
R

M

T

I

O

N

A

D

T

A
I
S

I

D

O

A

D

N

B

L

J

U

E

M

P
E
R

J

P

7

T

E
R
M

T

I

O
N
A

D

T

A

I

S

I

D

O

A

D

N

B
L

J

U

E

M
P

E

R

CHANNEL -2

P

E

A

A

U

1

R

D

D

E

L

S

2

R

Y

S

M

E

L

T

T

S

R

T

T

R

T

Channel 2

Termination Jumper

Spa

S

RDA

A

S1BDDLR

0

0

1

1234 5678

SAF-TE Disk I/O Card Switch and Jumper Locations

2 Locate the switches on the card and set them as described in the following

switch setting diagram.

The switch settings will assign specific SCSI IDs to the drive slots as
indicated, and reserve IDs 6 and 7 for the RAID Controller(s), and IDs 8 or 15
for the SAF-TE processors.

SAF-TE ID = 8/8

RAID Controllers = 6 & 7

Switch Settings

S

R

A

A

S1BDDLR

0

D

0

1

1234 5678

M

UP (1)

DOWN (0)

Slot 1

ID 1

Slot 2

ID 3

Slot 3 Slot 6 Slot 9 Slot 12

ID 5

Slot 4

Slot 5

Switch Setting - Primary RAID Enclosure (NexStor 4110FS/4120FS)

Dual-Bus Mode

Slot 7

Slot 8

ID 1

ID 3

ID 5

ID 2

ID 4

ID 0

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 10

Slot 11

ID 2

ID 4

ID 0

44

RAID - 24 Drive Configuration

4100FS.book Page 45 Tuesday, July 8, 2003 4:35 PM

3 Re-install the SAF-TE Disk I/O card into the primary RAID enclosure. Slide

the card into the slot and ensure that it seats completely. Secure the card by
tighten the two thumb screws.

4 Remove the SAF-TE Disk I/O card installed in the daisy-chain enclosure

(NexStor 4000S). Loosen the two thumb screws and pull the card from its slot
location.

The daisy-chain enclosure will be configured to dual-bus mode. This
provides the drive channel expansion from the primary enclosures’ Disk I/O
channels. Set the switches as described in the following diagram.

Chapter 3 - Setup and Installation

SAF-TE ID = 15/15

RAID Controllers = 6 & 7

Switch Settings

S

R

A

A

S1BDDLR

0

D

0

1

1234 5678

M

UP (1)

DOWN (0)

Slot 1

ID 9

Slot 2

ID 11

Slot 3 Slot 6 Slot 9 Slot 12

ID 13

Dual-Bus Mode

Slot 4

ID 10

Slot 5

ID 12

ID 14

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 7

Slot 8

ID 9

ID 11

ID 13

Slot 10

Slot 11

ID 10

ID 12

ID 14

Switch Setting - Dual-Bus Daisy-Chain Enclosure (NexStor 4000S)

5 (Daisy-chain Enclosure - NexStor 4000S) Change the jumper settings on the

SAF-TE Disk I/O card JP7 and JP8 as described.

Locate and add the two jumpers (JP7 and JP8), refer to the illustration on the
previous page. Add a jumper (installed on both pins) to JP8 for Channel 1
and JP7 for Channel 2.

This will ensure that the automatic termination feature functions properly.

6 Re-install the SAF-TE Disk I/O card into the daisy-chain enclosure. Slide the

card into the slot and ensure that it seats completely. Secure the card by
tighten the two thumb screws.

7 Connect the SCSI data cables from the NexStor 4110FS/4120FS SAF-TE Disk

I/O card to the NexStor 4000S enclosures’ SAF-TE Disk I/O card as indicated

in the following cabling diagram.

RAID - 24 Drive Configuration

45

4100FS.book Page 46 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

SAFTE Disk I/O

Chl 1

Power
Supply

Power
Supply

Cooling

Fans

SCSI Disk I/O

CTRL 1 CTRL 2SAFTE

Chl 0 Chl 3

Host I/O

P4P3P2P1

Chl 2

RAID Controller

NexStor 4110FS

SCSI Data Cable SCSI Data Cable

SAFTE Disk I/O

Chl 1

Power
Supply

Power
Supply

Cooling

Fans

CTRL 1 CTRL 2SAF-TE

Chl 2

NexStor 4000S

Enclosure Cabling Diagram (NexStor 4110FS used as example)

8 Connect a Fibre Channel data cable(s) from the host system(s) HBA(s) to the

primary RAID enclosure Host I/O card as indicated in “Topology Host
Cabling” on page 57.

This completes the RAID - 24 Drive Configuration setup, refer to “Topology Host

Cabling” on page 57 and “Powering On the Storage System” on page 70. Also,

refer to the AdminiStor software guide or the VT-100 software guide for

instructions on setting up and configuring the disk arrays.

46

RAID - 24 Drive Configuration

4100FS.book Page 47 Tuesday, July 8, 2003 4:35 PM

RAID - 36 Drive Configuration

The NexStor 4110FS/4120FS can be setup to provide up to a 36 disk drive storage
solution. In this configuration you will attach two NexStor 4000S JBOD
enclosures to the primary RAID enclosure which will extend the drive channels
(12 drives per channel).

1 Remove the SAF-TE Disk I/O card installed in the primary RAID enclosure.

Loosen the two thumb screws and pull the card from its slot location.

Channel 1

Termination Jumper

Chapter 3 - Setup and Installation

rt

a

2

D

lay

e

R

D

M

t

ive S

r

D

rive Start

e

ot

m

e

UP (Off -1)

DOWN (On -0)

n1

n2

atio

atio

1

ur

ur

re

onfig

C

N

C

F

N

G

F

1

G

S

2

R

P

A

A

ID

S

R

B

CHANNEL -1

JP8

TERMINATION

TO

DISABLE

ADD JUMPER

JP

7

TERMINATION

TO

DISABLE

C

h

a

nn

e

l 1

C

h

a

n

n

e

l 2

ADD JUMPER

CHANNEL -2

P

E

A

A

1

U

R

D

D

E

L

S

2

R

Y

S

M

E

L

T

T

S

R

T

T

R

T

Channel 2

Termination Jumper

C

0

1234 5678

re

onfig

C

RAID

Spa

Spa

BAUD Rate Select

S

RDA

A

S1BDDLR

0

1

SAF-TE Disk I/O Card Switch and Jumper Locations

2 Locate the switches on the card and set them as described in the following

switch setting diagram.

The switch settings will assign specific SCSI IDs to the drive slots as
indicated, and reserve IDs 6 and 7 for the RAID Controller(s), and IDs 8 or 15
for the SAF-TE processors.

SAF-TE ID = 8/8

RAID Controllers = 6 & 7

Switch Settings

S

RDA

A

S1BDDLR

0

0

1

1234 5678

M

UP (1)

DOWN (0)

Slot 1

ID 1

Slot 2

ID 3

Slot 3 Slot 6 Slot 9 Slot 12

ID 5

Switch Setting - Primary RAID Enclosure (NexStor 4110FS/4120FS)

Dual-Bus Mode

Slot 4

Slot 5

ID 4

ID 0

Drive IDs of the Drive Slots (Drive Channel Side)

ID 2

Slot 7

Slot 8

ID 1

ID 3

ID 5

Slot 10

Slot 11

ID 2

ID 4

ID 0

RAID - 36 Drive Configuration

47

4100FS.book Page 48 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

3 Re-install the SAF-TE Disk I/O card in the primary RAID enclosure. Slide the

card into the slot and ensure that it seats completely. Secure the card by

tighten the two thumb screws.

4 Remove the SAF-TE Disk I/O card installed in the first daisy-chain enclosure

(NexStor 4000S). Loosen the two thumb screws and pull the card from its slot

location.

The first daisy-chain enclosure will be configured to dual-bus mode. It

provides the drive channel expansion from the primary enclosures’ Disk I/O

channels. Set the switches as described in the following diagram.

SAF-TE ID = 15/15

RAID Controllers = 6 & 7

Switch Settings

S

R

A

A

S1BDDLR

0

D

0

1

1234 5678

M

UP (1)

DOWN (0)

Slot 1

ID 9

Slot 2

ID 11

Slot 3 Slot 6 Slot 9 Slot 12

ID 13

Dual-Bus Mode

Slot 4

ID 10

Slot 5

ID 12

ID 14

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 7

Slot 8

ID 9

ID 11

ID 13

Slot 10

Slot 11

ID 10

ID 12

ID 14

Switch Setting - Dual-Bus Daisy-Chain Enclosure (NexStor 4000S)

5 (First Daisy-chain Enclosure - NexStor 4000S) Change the jumper settings on

the SAF-TE Disk I/O card JP7 and JP8 as described.

Locate and add the two jumpers (JP7 and JP8), refer to the illustration on the
previous page. Add a jumper (installed on both pins) to JP8 for Channel 1
and JP7 for Channel 2.

This will ensure that the automatic termination feature functions properly.

6 Re-install the SAF-TE Disk I/O card in the first daisy-chain enclosure. Slide

the card into the slot and ensure that it seats completely. Secure the card by

tighten the two thumb screws.

7 Remove the SAF-TE Disk I/O card installed in the second daisy-chain

enclosure (NexStor 4000S). Loosen the two thumb screws and pull the card

from its slot location.

48

RAID - 36 Drive Configuration

4100FS.book Page 49 Tuesday, July 8, 2003 4:35 PM

The second daisy-chain enclosure will be configured to single-bus mode. It

provides the drive channel expansion from the primary enclosures’ I/O card

Channel 0. Set the switches as described in the following diagram.

Chapter 3 - Setup and Installation

SAF-TE ID = 15

RAID Controllers = 6 & 7

Switch Settings

S

RDA

A

S1BDDLR

0

0

1

1234 5678

M

UP (1)

DOWN (0)

Slot 1

ID 1

Slot 2

ID 3

Slot 3 Slot 6 Slot 9 Slot 12

ID 5

Single-Bus Mode

Slot 4

ID 2

Slot 5

ID 4

ID 0

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 7

Slot 8

Switch Setting - Single-Bus Daisy-Chain Enclosure (NexStor 4000S)

ID 9

ID 11

ID 13

Slot 10

Slot 11

ID 10

ID 12

ID 14

8 (Second Daisy-chain Enclosure - NexStor 4000S) Change the jumper settings

on the SAF-TE Disk I/O card JP7 and JP8 as described.

Locate and add the two jumpers (JP7 and JP8), refer to the illustration on the
previous page. Add a jumper (installed on both pins) to JP8 for Channel 1
and JP7 for Channel 2.

This will ensure that the automatic termination feature functions properly.

9 Re-install the SAF-TE Disk I/O card into the second daisy-chain enclosure.

Slide the card into the slot and ensure that it seats completely. Secure the

card by tighten the two thumb screws.

10 (Second Daisy-chain Enclosure - NexStor 4000S) Install the Single-Bus

Module in the Controller 1 slot.

a Remove the Controller Cover plate. Loosen the four thumb screws and

pull the cover plate from the enclosure.

b Slide the Single-Bus Module in the Controller 1 (lower) slot, and secure it

by pressing the latches into place.

RAID - 36 Drive Configuration

49

4100FS.book Page 50 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

JP1

P

R

L

J

Y

U

M

P

E

JP3

I

N

JP3

A

T

I

O

N

E

I

N

A

T

I

O

N

E

C

H
A
N

N
E
L

-1

J
P

5

T

E
R
M

T

O

P

O

S

W

A

U

D

P

E

D

P

R

L

J

Y

U
M
P

E

J

R

P
8

T
E
R

M

T

I

O
N
A

D
T

A
I
S

I

D

O

A

D

N

B

L

J
U

E

M
P

E

R

JP7

T
E
R
M

T

JP6

I

O

N

A

D

T

A
I
S

I

D

O

A

D

N

B
L

J
U

E

M
P

E

R

T

E
R
M

T

O
P
O

S

W

A

U

D

P

E

D

P

R

L

J

Y

U

M

C

P
E

H

R

A

N
N
E

L

-

2

R

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

JP1

P

R

L

J

Y

U

M

T

JP2

E

R

M

B

L

R

M

B

L

P

E

T

R

O

D

A

I

S

D

A

D

J

JP4

U

M

P

E

R

T

JP2

E

T

O

D

A

I

S

D

A

D

J

JP4

U

M

P

E

R

C
N

C

F

N

G

F

1

G

S

2

R

P

A

A

ID

S

R

B

P

E

A

A

1

U

R

D

D

E

L

S

R

Y

2

S

M

E

L

T

T

S

R

T

T

R

T

Single-Bus
Module

Cover Plate

Single-Bus Module Installation

c Re-install the Controller Cover plate. Tighten the four thumb screws, do

not overtighten.

11 Connect the SCSI data cables from the NexStor 4110FS/4120FS to the two

NexStor 4000S enclosures as indicated in the following enclosure cabling

diagram.

NOTE: When connecting the Single-Bus NexStor 4000S enclosure to the

primary RAID enclosure, you will connect to the Channel 0

connector on the Host SCSI I/O card installed in the NexStor

4110FS/4120FS enclosure.

50

RAID - 36 Drive Configuration

4100FS.book Page 51 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Power

Supply

Power

Supply

Power

Supply

SCSI Data Cable

Power
Supply

Power
Supply

Power
Supply

SAFTE Disk I/O

Cooling

Fans

CTLR 1 CTLR 2SAF-TE

NexStor 4110FS

SAFTE Disk I/O

Cooling

Fans

Ctlr 1 Ctlr 2SAF-TE

NexStor 4000S (Dual-Bus)

SAFTE Disk I/O

Cooling

Fans

Ctlr 1 Ctlr 2SAF-TE

Chl 1

I/O

Chl 0 Chl 3

Host I/O

P4 P3P2P1

Chl 1

Chl 1

SCSI Data Cable

Chl 2

RAID Controller

Chl 2

SCSI Data Cable

Chl 2

Single Bus

Module

NexStor 4000S (Single-Bus)

Enclosure Cabling Diagram (NexStor 4110FS used as example)

12 Connect a Fibre Channel data cable(s) from the host system(s) HBA(s) to the

primary RAID enclosure Host I/O card as indicated in “Topology Host
Cabling” on page 57.

This completes the RAID - 36 Drive Configuration setup, refer to “Topology Host

Cabling” on page 57 and “Powering On the Storage System” on page 70. Also,

refer to the AdminiStor software guide or the VT-100 software guide for

instructions on setting up and configuring the disk arrays.

RAID - 36 Drive Configuration

51

4100FS.book Page 52 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

RAID - 48 Drive Configuration

The NexStor 4110FS/4120FS can be setup to provide up to a 48 disk drive storage
solution. In this configuration you will attach three NexStor 4000S JBOD
enclosures to the primary RAID enclosure which will extend the drive channels
(12 drives per channel).

1 Remove the SAF-TE Disk I/O card installed in the primary RAID enclosure.

Loosen the two thumb screws and pull the card from its slot location.

Channel 1

Termination Jumper

C

h

a

nn

e

l 1

C

rt

a

2

rive Start

D

lay

e

e

R

D

M

t

ive S

r

D

e

ot

m

UP (Off -1)

DOWN (On -0)

n1

n2

atio

atio

1

ur

ur

re

onfig

C

N

C

F

N

G

F

1

G

S

2

R

P

A

A

ID

S

R

B

CHANNEL -1

JP8

TERMINATION

TO

DISABLE

ADD JUMPER

JP

7

TERMINATION

TO

DISABLE

ADD JUMPER

CHANNEL -2

h

a

n

n

el 2

P

E

A

A

1

U

R

D

D

E

L

S

2

R

Y

S

M

E

L

T

T

S

R

T

T

R

T

Channel 2

Termination Jumper

C

0

1234 5678

re

onfig

C

RAID

Spa

BAUD Rate Select

Spa

S

RDA

A

S1BDDLR

0

1

SAF-TE Disk I/O Card Switch and Jumper Locations

2 Locate the switches on the card and set them as described in the following

switch setting diagram.

The switch settings will assign specific SCSI IDs to the drive slots as
indicated, and reserve IDs 6 and 7 for the RAID Controller(s), and IDs 8 or 15
for the SAF-TE processors.

SAF-TE ID = 8/8

RAID Controllers = 6 & 7

Switch Settings

S

R

A

A

S1BDDLR

0

D

0

1

M

UP (1)

1234 5678

DOWN (0)

Switch Setting - Primary RAID Enclosure (NexStor 4110FS/4120FS)

Slot 1

ID 1

Slot 2

ID 3

Slot 3 Slot 6 Slot 9 Slot 12

ID 5

Slot 4

Slot 5

Dual-Bus Mode

Slot 7

Slot 8

ID 1

ID 3

ID 5

ID 2

ID 4

ID 0

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 10

Slot 11

ID 2

ID 4

ID 0

52

RAID - 48 Drive Configuration

4100FS.book Page 53 Tuesday, July 8, 2003 4:35 PM

3 Re-install the SAF-TE Disk I/O card in the primary RAID enclosure. Slide the

card into the slot and ensure that it seats completely. Secure the card by

tighten the two thumb screws.

4 Remove the SAF-TE Disk I/O card installed in the first daisy-chain enclosure

(NexStor 4000S). Loosen the two thumb screws and pull the card from its slot

location.

The first daisy-chain enclosure will be configured to dual-bus mode. This

provides the drive channel expansion from the primary enclosures’ Disk I/O

channels. Set the switches as described in the following diagram.

Chapter 3 - Setup and Installation

SAF-TE ID = 15/15

RAID Controllers = 6 & 7

Switch Settings

S

R

A

S1BDDLR

0

DA0

1

1234 5678

M

UP (1)

DOWN (0)

Slot 1

ID 9

Slot 2

ID 11

Slot 3 Slot 6 Slot 9 Slot 12

ID 13

Dual-Bus Mode

Slot 4

ID 10

Slot 5

ID 12

ID 14

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 7

Slot 8

ID 9

ID 11

ID 13

Slot 10

Slot 11

ID 10

ID 12

ID 14

Switch Setting - Dual-Bus Daisy-Chain Enclosure (NexStor 4000S)

5 (First Daisy-chain Enclosure - NexStor 4000S) Change the jumper settings on

the SAF-TE Disk I/O card JP7 and JP8 as described.

Locate and add the two jumpers (JP7 and JP8), refer to the illustration on the
previous page. Add a jumper (installed on both pins) to JP8 for Channel 1
and JP7 for Channel 2.

This will ensure that the automatic termination feature functions properly.

6 Re-install the SAF-TE Disk I/O card in the first daisy-chain enclosure. Slide

the card into the slot and ensure that it seats completely. Secure the card by

tighten the two thumb screws.

7 Remove the SAF-TE Disk I/O card installed in the second and third

daisy-chain enclosures (NexStor 4000S). Loosen the two thumb screws and

pull the card from its slot location.

RAID - 48 Drive Configuration

53

4100FS.book Page 54 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

The second and third daisy-chain enclosures will be configured to single-bus

mode through their switch settings. Therefore, the switch setting is the same

for both of these daisy-chained enclosures.

SAF-TE ID = 15

RAID Controllers = 6 & 7

Switch Settings

RDA

A

0

1

1234 5678

S

S1BDDLR

0

M

UP (1)

DOWN (0)

Single-Bus Mode

Slot 1

ID 1

Slot 2

ID 3

Slot 3 Slot 6 Slot 9 Slot 12

ID 5

Slot 4

ID 2

Slot 5

ID 4

ID 0

Drive IDs of the Drive Slots (Drive Channel Side)

Slot 7

Slot 8

ID 9

ID 11

ID 13

Slot 10

Slot 11

ID 10

ID 12

ID 14

Switch Setting - Single-Bus Daisy-Chain Enclosure (NexStor 4000S)

8 (Second & Third Daisy-chain Enclosure - NexStor 4000S) Change the jumper

settings on the SAF-TE Disk I/O card JP7 and JP8 as described.

Locate and add the two jumpers (JP7 and JP8), refer to the illustration on the
previous page. Add a jumper (installed on both pins) to JP8 for Channel 1
and JP7 for Channel 2.

This will ensure that the automatic termination feature functions properly.

9 Re-install the SAF-TE Disk I/O cards in each enclosure. Slide the card into the

slot and ensure that it seats completely. Secure the card by tighten the two

thumb screws.

10 (Second & Third Daisy-chain Enclosure) Install the Single-Bus Module in the

Controller 1 “lower” slot in both of these enclosures.

a Remove the Controller Cover plate. Loosen the four thumb screws and

pull the cover plate from the enclosure.

b Slide the Single-Bus Module in the Controller 1 (lower) slot, and secure it

by pressing the latches into place.

54

RAID - 48 Drive Configuration

4100FS.book Page 55 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

JP1

P

R

L

J

Y

U

M

P

E

R

JP3

M

I

N

JP3

A

T

I

O

N

L

E

M

I

N

A

T

I

O

N

L

E

C
H
A

N
N
E

L

-
1

J
P

5

T

E
R
M

T

O
P
O

S

A

W

U

D

P

E

D

P

R

L

J

Y

U

M

P
E

J

R

P

8

T

E
R
M

T

I

O

N

A

D

T

A

I

S

I

D

O

A

D

N

B
L

J

U

E

M
P
E

R

J

P

7

T
E
R

M

T

J

I

O

N

P

A

D

T

A

I

6

S

I

D

O

A

D

N

B

L

J
U

E

M

P
E
R

T
E
R

M

T

O

P

O

S

A

W

U

D

P

E

D

P

R

L

J

Y

U
M

C

P

E

H

R

A

N

N
E
L

­2

T

E

R

M

T

O

P

S

O

A

U

W

D

P

E

D

JP1

P

R

L

J

Y

T

E

R

I

S

A

B

E

R

T

E

R

I

S

A

B

E

R

U

M

JP2

P

E

T

R

O

D

A

D

D

J

JP4

U

M

P

JP2

T

O

D

A

D

D

J

JP4

U

M

P

C

N

C

F

N

G

F

1

G

S

2

R

P

A

A

ID

S

R

B

P

E

A

A

1

U

R

D

D

E

L

S

2

R

Y

M

E

S

L

T

T

S

R

T

T

R

T

Single-Bus
Module

Cover Plate

Single-Bus Module Installation

c Re-install the Controller Cover plate. Tighten the four thumb screws, do

not overtighten.

11 Connect the SCSI data cables from the NexStor 4110FS/4120FS to the three

NexStor 4000S enclosures as indicated in the following cabling diagram.

NOTE: When connecting the Single-Bus NexStor 4000S enclosure to the

primary RAID enclosure you will connect it to the Channel 0 and
Channel 1 connectors on the Host SCSI I/O card installed in the
NexStor 4110FS/4120FS enclosure.

RAID - 48 Drive Configuration

55

4100FS.book Page 56 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

SCSI Data Cable

Power
Supply

Power
Supply

Power
Supply

SCSI Data Cable

Power
Supply

Power
Supply

Power
Supply

Cooling

Fans

CTLR 1 CTLR 2SAF-TE

NexStor 4110FS

Cooling

Fans

CTRL 1 CTLR 2SAF-TE

NexStor 4000S (Dual-Bus)

Cooling

Fans

CTLR 1 CTLR 2SAF-TE

NexStor 4000S (Single-Bus)

SAFTE Disk I/O

Chl 1

I/O

Chl 0 Chl 3

Host I/O

P4 P3P2P1

SAFTE Disk I/O

Chl 1

SAFTE Disk I/O

Chl 1

SCSI Data Cable

Chl 2

RAID Controller

Chl 2

SCSI Data Cable

Chl 2

Single Bus

Module

56

12 Connect a Fibre Channel data cable(s) from the host system(s) HBA(s) to the

primary RAID enclosure Host I/O card as indicated in “Topology Host
Cabling” on page 57.

This completes the RAID - 48 Drive Configuration setup, refer to “Topology Host

Cabling” on page 57 and “Powering On the Storage System” on page 70. Also,

refer to the AdminiStor software guide or the VT-100 software guide for

instructions on setting up and configuring the disk arrays.

RAID - 48 Drive Configuration

Power
Supply

Power
Supply

Cooling

Fans

CTLR 1 CTLR 2SAF-TE

NexStor 4000S (Single-Bus)

Enclosure Cabling Diagram

SAFTE Disk I/O

Chl 1

Chl 2

Single Bus

Module

4100FS.book Page 57 Tuesday, July 8, 2003 4:35 PM

Topology Host Cabling

This section provides instructions and diagrams for the physical cabling of the
primary RAID enclosure and your host system(s).
by the number of RAID Controllers installed and the specific operating mode.
They are further categorized by the condition of the internal hubs.

Basic Connection Instructions

1 Install your host bus adapter(s) into the host system(s). Refer to your HBA

user’s guide and any applicable nStor Attach Kit for specific details.

2 For multiple LUN support, ensure that your host operating system is properly

configured to support this feature. Refer to your host operating system user’s
guide for information on how to perform this procedure.

3 Connect the Fibre Channel data cables. Refer to the topology described for

your installation on the following pages, and cable your system based on the
diagrams shown.

Chapter 3 - Setup and Installation

The topologies are determined

a Connect the required Fibre Channel data cable(s) to the host system’s

Fibre Channel host bus adapter port connector(s).

b Connect the other end of those Fibre Channel data cable(s) to the storage

enclosure Host Channel ports on the Host I/O card(s) as indicated.

4 Continue now with “Powering On the Storage System” on page 70.

5 After powering up all the system(s), you may need to set the operating

mode.

By default the RAID Controller’s operating mode are set to “Active-Active
Single Port.” Access the AdminiStor software or VT-100 on-board software
and make the necessary changes to the operating mode. The selections will
be:

• Stand-Alone Dual Port

• Active-Active Single Port

• Active-Active Dual Port

• Active-Passive Dual Port

6 Using AdminiStor or the VT-100 interface, configure the disk arrays as

desired.

Topology Host Cabling

57

4100FS.book Page 58 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Stand-Alone Dual Port:Host Cabling

This topology is chosen when the desired application requires a low cost entry
level, fault-tolerant disk storage solution. This solution provides a single controller
configuration that supports dual ported access with the internal hubs either enabled
or disabled. With the internal hub disabled, it is ideal in a host direct attached
application or SAN environments when connected to an external switch/hub.

Setup for a Single Host- One HBA (Hubs Disabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Host System #1

HBA 1

Controller FC Host Ports

HOST I/O

P1

P2

P4

P3

T= Location of internal termination

Host I/O Connectors

RAID Controller

FC1

Active

Active

FC2

Controller 1 (A)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

T

Controller Disk Ports

Logical Diagram - Single Host with a Single HBA (Hubs Disabled)

FC HBA 1

Host Computer #1

Drives 7 - 12

Drives 1 - 6

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Host Cabling Diagram - Single Host with a Single HBA (Hubs Disabled)

58

Stand-Alone Dual Port:Host Cabling

Power

Supply

Power

Supply

Cooling

Fans

Fibre Channel Data Cable

SAFTE Disk I/O

Chl 1 Chl 2

SCSI Disk I/O

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

NexStor 4110FS

Connect to P3

P4 P3P2P1

RAID Controller

4100FS.book Page 59 Tuesday, July 8, 2003 4:35 PM

Setup for a Single Host - Two HBAs (Hubs Disabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Chapter 3 - Setup and Installation

Host System #1

HBA 2

Host System #1

HBA 1

Controller FC Host Ports

HOST I/O

P1

P2

P4

P3

RAID Controller

Drives 7 - 12

Drives 1 - 6

Controller Disk Ports

Host I/O Connectors

FC1

Active

Active

FC2

Controller 1 (A)

T= Location of internal termination

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

T

Logical Diagram - Single Host with Two HBAs (Hubs Disabled)

FC HBA 2 FC HBA 1

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Host Computer #1

Fibre Channel Data Cable
Connect to P3

RAID Controller

Power
Supply

Fibre Channel Data Cable

Connect to P4

Power
Supply

Cooling

Fans

SAFTE Disk I/O

Chl 1 Chl 2

SCSI Disk I/O

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

P4 P3P2P1

NexStor 4110FS

Host Cabling Diagram - Single Host with Two HBAs (Hubs Disabled)

Stand-Alone Dual Port:Host Cabling

59

4100FS.book Page 60 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Setup for a Single Host - Two HBAs (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Host System #1

HBA 2

Host System #1

HBA 1

Controller FC Host Ports

HOST I/O

P1

P2

P4

P3

Host I/O Connectors

RAID Controller

Drives 7 - 12

FC1

Active

Disk
CH1

T

HUB

Active

FC2

HUB

Controller 1 (A)

T= Location of internal termination

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

Controller Disk Ports

Drives 1 - 6

Logical Diagram - Single Host with Two HBAs (Hubs Enabled)

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

60

Stand-Alone Dual Port:Host Cabling

FC HBA 2 FC HBA 1

Host Computer #1

Fibre Channel Data Cable
Connect to P3

RAID Controller

Power

Supply

Fibre Channel Data Cable

Connect to P4

Power

Supply

Cooling

Fans

SAFTE Disk I/O

Chl 1 Chl 2

SCSI Disk I/O

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

P4 P3P2P1

NexStor 4110FS

Host Cabling Diagram - Single Host with Two HBAs (Hubs Enabled)

4100FS.book Page 61 Tuesday, July 8, 2003 4:35 PM

Setup for a Multiple Hosts - Single HBAs (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Chapter 3 - Setup and Installation

Host System #3

HBA 1

Host System #4

HBA 1

Host System #2

HBA 1

Host System #1

HBA 1

Controller FC Host Ports

P1

P2

HOST I/O

HUB

HUB

FC1

FC2

Active

Active

RAID Controller

Disk

CH1

Disk

CH2

Disk

CH0

T

T

T

P4

Disk

P3

Controller 1 (A)

CH3

T

T= Location of internal termination

Host I/O Connectors

Controller Disk Ports

Logical Diagram - Multiple Host with Single HBAs (Hubs Enabled)

FC HBA 1

Host Computer #4

Host Computer #1

Drives 7 - 12

Drives 1 - 6

Drive I/O Connectors

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

FC HBA 1

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Fibre Channel Data Cable

Fibre Channel Data Cable

Host Computer #3

Fibre Channel Data Cable

Power

Power

Supply

Supply

Fibre Channel Data Cable

Connect to P2

FC HBA 1

Connect to P1

Cooling

Fans

SAFTE Disk I/O

Chl 1 Chl 2

SCSI Disk I/O

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

Host Computer #2

P4 P3P2P1

RAID Controller

NexStor 4110FS

Host Cabling Diagram - Multiple Host with Single HBAs (Hubs Enabled)

Stand-Alone Dual Port:Host Cabling

Connect to P4

FC HBA 1

Connect to P3

61

4100FS.book Page 62 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Active-Active Single Port:Host Cabling

This topology is chosen when the desired application requires a high-performance
robust full system level fault-tolerant disk storage solution and transparent
controller failover/failback.
with a single port HBA and is ideal when the host driver software will not
support LUNs that appear twice.

Setup for Single Host - Single HBA (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

This dual controller configuration supports a host

Host System #1 HBA #1

P1

P2

P4

P3

Jumper Cable

HOST I/O

Host I/O
Connectors

HUB

HUB

FC1

FC2

Controller 1 (A)

FC2

FC1

Controller 2 (B)

RAID Controller

Active

Passive

RAID Controller

Active

Passive

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

T

T

T

T= Location of internal termination

T

Logical Diagram - Single Host with Single HBA (Hubs Enabled)

FC HBA 1

Host Computer

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Host Cabling Diagram - Single Host with Single HBA (Hubs Enabled)

62

Active-Active Single Port:Host Cabling

Power
Supply

Power

Supply

Cooling

Fans

Fibre Channel Data Cable
Connect to P2

SAFTE Disk I/O

Chl 1

I/O

Chl 0 Chl 3

Host I/O

P4 P3P2P1

Ctlr 1 Ctlr 2SAF-TE

NexStor 4120FS

Chl 2

RAID Controller

RAID Controller

Fibre Channel Jumper Cable

4100FS.book Page 63 Tuesday, July 8, 2003 4:35 PM

Setup for Single Host - Dual HBAs (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Chapter 3 - Setup and Installation

Host System #1 HBA #2

Host System #1 HBA #1

HOST I/O

P1

Host I/O
Connectors

HUB

HUB

P2

P4

P3

RAID Controller

FC1

Active

Passive

FC2

Controller 1 (A)

RAID Controller

Active

FC2

FC1

Passive

Controller 2 (B)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

SEP

SEP

T

CH 1 CH 2

T

SAF-TE Disk I/O

T

CH 0 CH 3

T

T

T

T

T= Location of internal termination

T

Host SCSI I/O

Logical Diagram - Single Host with Dual HBAs (Hubs Enabled)

FC HBA 1 FC HBA 2

Host Computer #1

Fibre Channel Data Cable

Connect to P1

Fibre Channel Data Cable

Connect to P2

SAFTE Disk I/O

Power

Supply

Power

Supply

Cooling

Fans

Chl 1 Chl 2

I/O

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

RAID Controller

P3P2P1

P4

RAID Controller

NexStor 4120FS

Host Cabling Diagram - Single Host with Dual HBAs (Hubs Enabled)

Active-Active Single Port:Host Cabling

63

4100FS.book Page 64 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Setup for Dual Host - Single HBAs (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Host System #1 HBA #1

Host System #2 HBA #1

HOST I/O

P1

Host I/O
Connectors

HUB

HUB

P2

P4

P3

RAID Controller

FC1

Active

Passive

FC2

Controller 1 (A)

RAID Controller

Active

FC2

Passive

FC1

Controller 2 (B)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

T

T

T

T= Location of internal termination

T

Logical Diagram - Dual Host with Single HBAs (Hubs Enabled)

FC HBA 1

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

64

Active-Active Single Port:Host Cabling

Host Computer #2

FC HBA 1

Host Computer #1

Fibre Channel Data Cable

Power
Supply

Power
Supply

Connect to P1

Cooling

Fans

SAFTE Disk I/O

Chl 1 Chl 2

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

I/O

P4

Fibre Channel Data Cable

Connect to P2

RAID Controller

P3P2P1

RAID Controller

NexStor 4120FS

Host Cabling Diagram - Dual Host with Single HBAs (Hubs Enabled)

4100FS.book Page 65 Tuesday, July 8, 2003 4:35 PM

Setup for Dual Host - Dual HBAs (Quad Cabling - Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Chapter 3 - Setup and Installation

Host System #1 HBA #2

Host System #2 HBA #1

Host System #2 HBA #2

Host System #1 HBA #1

HOST I/O

P1

Host I/O
Connectors

HUB

HUB

P2

P4

P3

RAID Controller

FC1

Active

Passive

FC2

Controller 1 (A)

RAID Controller

Active

FC2

FC1

Passive

Controller 2 (B)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

T

T

T

T= Location of internal termination

T

Logical Diagram - Dual Host with Dual HBAs (Quad Cabling - Hubs Enabled)

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Host Computer #2

Power

Power

Supply

Supply

FC HBA 1FC HBA 2

Fibre Channel Data Cable

Connect to P2

Cooling

Fans

Fibre Channel Data Cable

Connect to P4

SAFTE Disk I/O

Chl 1 Chl 2

I/O

Chl 0 Chl 3

Host I/O

P4 P3P2P1

Ctlr 1 Ctlr 2SAF-TE

Host Computer #1

RAID Controller

RAID Controller

FC HBA 1

FC HBA 2

Fibre Channel Data Cable

Connect to P3

Fibre Channel Data Cable

Connect to P1

NexStor 4120FS

Host Cabling Diagram - Dual Host with Dual HBAs (Quad Cabling - Hubs Enabled)

Active-Active Single Port:Host Cabling

65

4100FS.book Page 66 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Active-Active Dual Port:Host Cabling

This topology is chosen when the desired application requires a high-performance
robust full system level fault-tolerant disk storage solution.
configuration supports multiple hosts, and failover and failback operations. All
LUNs are available to all hosts.

Setup for Dual Host - Dual HBAs (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

This dual controller

Host System #1 HBA #2

Host System #2 HBA #1

Host System #2 HBA #2

Host System #1 HBA #1

HOST I/O

P1

P2

P4

P3

Host I/O
Connectors

Logical Diagram - Dual Host with Dual HBAs (Hubs Enabled)

Host Computer #2

HUB

HUB

RAID Controller

FC1

Active

Active

FC2

Controller 1 (A)

RAID Controller

Active

FC2

FC1

Active

Controller 2 (B)

FC HBA 1FC HBA 2

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

T= Location of internal termination

T

Drives 7 - 12

Drives 1 - 6

Host Computer #1

FC HBA 1

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

FC HBA 2

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Power
Supply

Host Cabling Diagram - Dual Host with Dual HBAs (Hubs Enabled)

66

Active-Active Dual Port:Host Cabling

Fibre Channel Data Cable

Power
Supply

Connect to P2

Cooling

Fans

Fibre Channel Data Cable

Connect to P4

SAFTE Disk I/O

Chl 1 Chl 2

I/O

Chl 0 Chl 3

Host I/O

P4 P3P2P1

Ctlr 1 Ctlr 2SAF-TE

NexStor 4120FS

RAID Controller

RAID Controller

Fibre Channel Data Cable

Connect to P3

Fibre Channel Data Cable

Connect to P1

4100FS.book Page 67 Tuesday, July 8, 2003 4:35 PM

Active-Passive Dual Port:Host Cabling

This topology is chosen when the desired application requires a high-availability
high access fault-tolerant disk storage solution. This dual controller configuration
supports multiple hosts and failover and failback operations. All LUNs are available
to all hosts. The dual host configuration is ideal for shared storage clustering.

Setup for Single Host - Dual HBAs (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Chapter 3 - Setup and Installation

Host System #1 HBA #1

Host System #1 HBA #2

HOST I/O

P1

Host I/O
Connectors

HUB

HUB

P2

P4

P3

RAID Controller

FC1

Active

FC2

Active

Controller 1 (A)

RAID Controller

Passive

FC2

FC1

Passive

Controller 2 (B)

Passive

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Drives 7 - 12

Drives 1 - 6

T

T

T

T= Location of internal termination

T

Logical Diagram - Single Host with Dual HBAs (Hubs Enabled)

FC HBA 1FC HBA 2

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

Host Computer #1

Fibre Channel Data Cable

Connect to P1

Fibre Channel Data Cable

Connect to P2

SAFTE Disk I/O

Power
Supply

Power
Supply

Cooling

Fans

Chl 1 Chl 2

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

I/O

P4

RAID Controller

P3P2P1

RAID Controller

NexStor 4120FS

Host Cabling Diagram - Single Host with Dual HBAs (Hubs Enabled)

Active-Passive Dual Port:Host Cabling

67

4100FS.book Page 68 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Setup for Dual Host - Single HBAs (Hubs Enabled)

Connect the Fibre Channel data cables as described in the following diagrams.

Host System #1 HBA #1

Host System #2 HBA #1

HOST I/O

P1

Host I/O
Connectors

HUB

HUB

P2

P4

P3

RAID Controller

FC1

Active

FC2

Active

Controller 1 (A)

RAID Controller

Passive

FC2

FC1

Passive

Controller 2 (B)

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

Disk
CH1

Disk
CH2

Disk
CH0

Disk
CH3

T

T

T

T= Location of internal termination

T

Passive

Logical Diagram - Dual Host with Single HBAs (Hubs Enabled)

FC HBA 1

Drives 7 - 12

Drives 1 - 6

SEP

SEP

SAF-TE Disk I/O

Host SCSI I/O

T

CH 1 CH 2

T

T

CH 0 CH 3

T

68

Active-Passive Dual Port:Host Cabling

Host Computer #2

FC HBA 1

Host Computer #1

Fibre Channel Data Cable

Power
Supply

Power

Supply

Connect to P1

Cooling

Fans

SAFTE Disk I/O

Chl 1 Chl 2

Chl 0 Chl 3

Host I/O

Ctlr 1 Ctlr 2SAF-TE

I/O

P4

Fibre Channel Data Cable

Connect to P2

RAID Controller

P3P2P1

RAID Controller

NexStor 4120FS

Host Cabling Diagram - Dual Host with Single HBAs (Hubs Enabled)

4100FS.book Page 69 Tuesday, July 8, 2003 4:35 PM

Upgrades

At some point you may have a need to upgrade your storage system to increase

the storage capacities and/or the RAID capabilities. The NexStor 4100FS Series

Storage System provides you with this ability to meet your expanding data

storage requirements.

The RAID controllers support a feature know as “drive roaming,” where the

controllers can keep track of which drive belong to which logical arrays. In the

event the drives’ enclosure is changed or they are moved to another enclosure

attached to the same set of controllers, the arrays will not be lost or damaged.

The upgrade process is relatively easy, you simply refer to the new Enclosure

Configuration Cabling section to find the new configuration that supports the

number of enclosures/drives you will be adding. Set the enclosure SAF-TE Disk I/O

card(s) switch settings for the new configuration and cabling the storage system

enclosures as indicated. Then, if applicable, refer to the Topology Host Cabling

section to locate the new topology and re-cable your new configuration as

indicated.

Chapter 3 - Setup and Installation

NOTE: Changing operating mode topologies may have an effect on which

configuration you may use. Refer to the Enclosure Configuration Cabling

section and the Topology Host Cabling section to ensure that you have

selected a configurations that is supported. Refer to “Operating Modes

Overview” on page 19 for more information.

NOTE: You will not be required to move disk drives around in the enclosure(s),

the Drive Roaming feature will locate those drives when the entire

system is powered back up.

Upgrades

69

4100FS.book Page 70 Tuesday, July 8, 2003 4:35 PM

Chapter 3 - Setup and Installation

Powering On the Storage System

After you have the system setup and installed, you are ready to power on the

storage system enclosure(s).

NOTE: Ensure that none of the data cables or power cables are obstructing the

air flow exiting the cooling fan module.

1 Locate the power supply On/Off switches at the rear of the NexStor 4100FS

Series Storage System enclosure(s).

2 Press each power supply switch to its “On” position. (Repeat this step for

each of the attached storage system enclosure(s).)

All enclosures will perform a power on self test during their initial start.

3 Next power on the host computer(s) which is connected to the storage

system.

Powering Off the Storage System

NOTE: If you are planning to completely shut down the entire system, power

down the host system first.

1 If applicable, ensure that the controller(s) have been gracefully shut down.

Refer to “Shutdown Both” in the VT-100 or AdminiStor software guides.

2 Press each enclosure’s power supply On/Off switches to their “Off” position.

3 Repeat the above step for each of the attached storage system enclosure(s).

70

Powering On the Storage System

4100FS.book Page 71 Tuesday, July 8, 2003 4:35 PM

Chapter 4

SAN Solution Strategies

This chapter will describe connecting the NexStor 4120FS Series Storage Systems

providing an Active-Active/Active-Passive solution to a Storage Area Network

(SAN).

Active-Active:Dual Port Mode for SAN

At the heart of this strategy is the Active-Active Dual Port operating mode. In this

mode, four Fibre Channel (FC) ports are simultaneously active when both

controllers are online. In the event of a controller failure, two FC ports remain

available to service all host I/O. The table below illustrates how LUNs are

presented by each controller in this mode.

Controller 1

Host Port 0 (FC1)

Both Controllers Online Controller 1

LUNs

Controller 1 Failed Inactive Inactive Controller 2

Controller 2 Failed Controller 1

LUNs

Note that when a controller fails, the surviving controller continues to present its

LUNs to its Host Port 0, and changes to present the failed controller’s LUNs to its

Host Port 1. The Fibre Channel Node and Port World Wide Names corresponding

to the failed controller’s Host Port 0 (FC1) are also associated with its LUNs on

the surviving controller’s Host Port 1 (FC2).

Controller 1

Host Port 1 (FC2)

Controller 1

LUNs

Controller 2

LUNs

Controller 2

Host Port 0 (FC1)

Controller 2

LUNs

LUNs

Inactive Inactive

Host Port 1 (FC2)

Controller 2

Controller 2

LUNs

Controller 1

LUNs

71

4100FS.book Page 72 Tuesday, July 8, 2003 4:35 PM

Chapter 4 - SAN Solution Strategies

Hubs Enabled

In the following illustrations, the controllers are configured in an Active-Active Dual
Port mode with the internal hubs enabled.

In the example below, two hosts are connected via two independent loops using
the enclosure’s built-in hub bypass circuits. These circuits provide two internal
hubs, eliminating the need for the use of external Fibre Channel hardware, other
than the cables.

Disk I/O

Connectors

NexStor 4120FS

JFS RAID Controller

CH 1 CH 2

T

SEP

Drives 7 - 12

Disk

CH1

LUN 1

Active

FC1

T

Disk

LUN 1

FC2

HUB

SEP

SAF-TE Disk I/O Card

Drives 1 - 6

Disk

CH2

Active

HUB

CH0

CH 0 CH 3

T

Disk

CH3

Controller 1 (A)

T

I/O Card

T

Disk

CH1

Active

LUN 2

JFS RAID Controller

FC2

Disk

LUN 2

FC1

T

CH2

Active

T

Disk

CH0

T= Location of internal termination

T

Disk

CH3

Controller 2 (B)

Host 1

72

Active-Active:Dual Port Mode for SAN

See's LUN 1 & LUN 2

See's LUN 1 & LUN 2

FC HBA

FC HBA

HOST I/O

P2

P1

P4

See's LUN 1 & LUN 2

Host 2

P3

FC HBA

Host I/O

See's LUN 1 & LUN 2

FC HBA

Connectors

4100FS.book Page 73 Tuesday, July 8, 2003 4:35 PM

This example illustrates the conditions when Controller 2 fails. Note that both
hosts can still see Controller 1’s LUNs (LUN 1) and Controller 2’s LUNS (LUN 2).
However, each host HBA can only see one LUN, instead of both.

Chapter 4 - SAN Solution Strategies

Disk I/O

Connectors

NexStor 4120FS

JFS RAID Controller

CH 1 CH 2

T

SEP

Drives 7 - 12

Disk

CH1

LUN 1

Active

FC1

T

Disk

LUN 2

FC2

SEP

CH2

Active

CH 0 CH 3

T

SAF-TE Disk I/O Card

Drives 1 - 6

Disk

Disk

CH0

T

CH3

Controller 1

I/O Card

T

Disk

CH1

Active

LUN 2

JFS RAID Controller

FC2

Disk

LUN 2

FC1

T= Location of internal termination

T

T

Disk

CH2

CH0

Active

Disk

CH3

T

FAILED

Controller 2

See's LUN 1 & LUN 2

Host 1

FC HBA

HOST I/O

P1

See's LUN 1 & LUN 2

FC HBA

HUB

HUB

P2

Host 2

P3

P4

See's LUN 1 & LUN 2

FC HBA

Host I/O

Connectors

See's LUN 1 & LUN 2

FC HBA

Active-Active:Dual Port Mode for SAN

73

4100FS.book Page 74 Tuesday, July 8, 2003 4:35 PM

Chapter 4 - SAN Solution Strategies

Hubs Disabled

In the following illustrations, the two controllers are configured in an Active-Active
Dual Port mode with the internal hubs disabled.

In the example below, two hosts are connected via two independent switches.
Using external switches supports higher bandwidth, since the host ports are not
required to share bandwidth on a loop.

Disk I/O

Connectors

NexStor 4120FS

CH 1 CH 2

T

SEP

Drives 7 - 12

Disk

CH1

LUN 1

Active

JFS RAID Controller

FC1

T

Disk

LUN 1

FC2

CH2

Active

SEP

SAF-TE Disk I/O

Drives 1 - 6

Disk

T

CH0

CH 0 CH 3

T

Disk

CH3

Controller 1 (A)

Disk I/O

T

Disk

CH1

Active

LUN 2

JFS RAID Controller

FC2

T= Location of internal termination

T

T

Disk

Disk

CH2

CH0

Active

LUN 2

FC1

T

Disk

CH3

Controller 2 (B)

See's LUN 1 &

LUN 2

Host 1

74

Active-Active:Dual Port Mode for SAN

Switch 1

FC HBA

HOST I/O

P1

See's LUN 1 &

LUN 2

FC HBA

P2

P3

P4

Host I/O

Connectors

Switch 2

See's LUN 1 &

LUN 2

LUN 2

FC HBA

Host 2

See's LUN 1 &

FC HBA

4100FS.book Page 75 Tuesday, July 8, 2003 4:35 PM

In this example, two hosts are connected via two independent switches. Note
that both hosts can still see Controller 1’s LUNs (LUN 1) and Controller 2’s LUNs
(LUN 2). However, each host HBA can only see one LUN, instead of both.

Chapter 4 - SAN Solution Strategies

Disk I/O

Connectors

NexStor 4120FS

JFS RAID Controller

CH 1 CH 2

T

SEP

Drives 7 - 12

Disk

CH1

LUN 1

Active

FC1

T

LUN 2

Disk

FC2

SEP

CH2

Active

SAF-TE Disk I/O

Drives 1 - 6

Disk

CH0

CH 0 CH 3

T

Disk

CH3

Controller 1 (A)

T

Disk I/O

T

Disk

CH1

In-Active

LUN 2

JFS RAID Controller

FC2

Disk

LUN 2

FC1

T= Location of internal termination

T

T

Disk

CH2

CH0

In-Active

T

Disk

CH3

Controller 2 (B)

FAILED

Host 1

See's LUN 1

FC HBA

Switch 1

HOST I/O

See's LUN 2

FC HBA

P2

P1

P3

P4

Host I/O

Connectors

Switch 2

See's LUN 2

FC HBA

Active-Active:Dual Port Mode for SAN

75

Host 2

See's LUN 1

FC HBA

4100FS.book Page 76 Tuesday, July 8, 2003 4:35 PM

Chapter 4 - SAN Solution Strategies

Active-Active:Single Port Mode for SAN

There are several scenarios where it may be desirable to configure the controllers
in Active-Active Single Port mode. In general, these are situations where the hosts
have a single Fibre Channel port, and so the SAN is not truly fault-tolerant.

Hubs Disabled (Single Port Hosts and One Switch)

The following illustration depicts two single ported hosts connected to the two
controllers via a single switch.

Disk I/O

Connectors

NexStor 4120FS

JFS RAID Controller

CH 1 CH 2

T

SEP

Drives 7 - 12

Disk

CH1

LUN 1

Active

FC1

T

Disk

LUN 1

FC2

SEP

Drives 1 - 6

CH2

Passive

CH 0 CH 3

T

SAF-TE Disk I/O

Disk

Disk

CH0

CH3

T

Controller 1 (A)

JFS RAID Controller

I/O Card

T

Disk

LUN 2

FC2

CH1

Passive

Disk

LUN 2

FC1

CH2

Active

T

T

Disk

T= Location of internal termination

T

Disk

CH0

CH3

Controller 2 (B)

76

Active-Active:Single Port Mode for SAN

Host 1

Host Link 1

FC HBA

HOST I/O

P2

P1

P3

P4

Host I/O

Connectors

Switch

Host Link 2

Host 2

FC HBA

4100FS.book Page 77 Tuesday, July 8, 2003 4:35 PM

In this case, there would be an advantage to configuring the controllers in

Active-Active Dual Port mode, because of increased connectivity. However, the

host driver software would need to handle the fact that each LUN would appear

twice in the SAN fabric, and that one image of the LUNs would disappear if one

controller failed.

Depending on the host software being used, it may be more desirable to run in

Active-Active Single Port mode, since a controller failover would be transparent

to the hosts (as far as the number of LUN images and the port WWNs attached to

the LUNs). The advantages of using a switch is that the two hosts each have a

direct path to a controller port and don’t have to compete for access on the same

loop.

Chapter 4 - SAN Solution Strategies

Active-Active:Single Port Mode for SAN

77

4100FS.book Page 78 Tuesday, July 8, 2003 4:35 PM

Chapter 4 - SAN Solution Strategies

Hubs Disabled (Single External Hub)

The following illustration depicts the hosts in a single loop configuration using an
external hub. When a controller failover occurs, this configuration operates much
the same as the previous one which uses a switch. The advantage is that a hub is
less expensive. In Active-Active Dual Port mode, you must manage the multiple
images of the same LUN.

Disk I/O

Connectors

NexStor 4120FS

JFS RAID Controller

CH 1 CH 2

T

SEP

Drives 7 - 12

Disk

CH1

LUN 1

Active

FC1

T

Disk

LUN 1

FC2

Passive

SEP

Drives 1 - 6

CH2

CH 0 CH 3

T

SAF-TE Disk I/O

Disk

Disk

CH0

CH3

T

Controller 1 (A)

JFS RAID Controller

I/O Card

T

Disk

LUN 2

FC2

CH1

Passive

Disk

LUN 2

FC1

T= Location of internal termination

T

T

Disk

CH2

CH0

Active

T

Disk

CH3

Controller 2 (B)

78

Active-Active:Single Port Mode for SAN

Host 1

Host Link 1

FC HBA

HOST I/O

P1

P2

Hub

Host 2

P3

P4

Host I/O

Connectors

Host Link 2

FC HBA

4100FS.book Page 79 Tuesday, July 8, 2003 4:35 PM

Hubs Enabled

The illustration below depicts hosts in a single loop configuration with the
internal hubs enabled. The internal hubs eliminate the need for an external hub.
It allows one or two hosts to be connected to the Active-Active storage enclosure
without the cost of external hardware other than cables.

Chapter 4 - SAN Solution Strategies

Disk I/O

Connectors

NexStor 4120FS

JFS RAID Controller

CH 1 CH 2

T

SEP

Drives 7 - 12

Disk

CH1

LUN 1

Active

FC1

T

Disk

LUN 1

FC2

CH2

Passive

SEP

SAF-TE Disk I/O

Drives 1 - 6

Disk

CH 0 CH 3

T

CH0

T

Disk

CH3

Controller 1 (A)

Disk I/O

T

Disk

CH1

LUN 2

Active

JFS RAID Controller

FC2

Disk

LUN 2

FC1

T= Location of internal termination

T

T

Disk

CH2

CH0

Passive

T

Disk

CH3

Controller 2 (B)

Host Link 1

Host 1

HOST I/O

FC HBA

HUB

HUB

P2

P1

Cable

Jumper

P4

Host 2

P3

Host I/O

Connectors

Host Link 2

FC HBA

Active-Active:Single Port Mode for SAN

79

4100FS.book Page 80 Tuesday, July 8, 2003 4:35 PM

Chapter 4 - SAN Solution Strategies

SAN Configuration Not Supported

In the Active-Active:Dual Port configuration with two independent hubs or the
internal hubs, two single-ported hosts is not supported. The issue with this
configuration is that if a controller fails, each host loses access to one of the LUNs.

For example, if Controller 1 fails, Host 1 can no longer access LUN 2, because
Port 2 on Controller 2 has changed from presenting its own LUNs to presenting
Controller 1’s LUNs. A need for this SAN configuration can arise if multiple
single-ported hosts need access to LUNs from both controllers, and two hubs are
desired to improve performance by decreasing loop access contention.

Note that this SAN is only partially fault-tolerant because the hubs and controllers
have redundant hardware, but the hosts do not. The problem with this
configuration can be overcome by using a single switch at the cost of replacing
the two hubs with a switch. If loop access contention is not an issue, then the
single hub configuration works as well.

True LUN Sharing

When considering the presentation of LUNs for Active-Active controllers, a good
question is: “Why not just present all LUN on all ports?” The answer is that LUN
sharing greatly increases complexity and sacrifices performance. In order to do
LUN sharing between two controllers, the following must happen:

1 Some common data sharing method must be developed so that access to

resources shared by both controllers can be administered. Because of this
arbitration for resources, the performance of both controllers suffers as
compared to two controllers that must only control their own resources.

2 Read cache on one controller must be invalidated if the data is modified by a

write to the other controller.

3 Writes to disk must be controlled to ensure that simultaneous updates to the

same region do not occur by both controllers.

4 RAID utility functionality (parity verification and expansion, for example)

must be controlled to ensure that collisions between utilities and data
accesses on the two controllers don’t occur.

The Active-Active strategy chosen allows a host to access two separate Fibre Channel
ports (one from each controller) which are connected together via a switch or hub,
in order to support access to LUNs owned by different controllers. This is simpler
and provides higher performance than so-called “true LUN sharing” schemes.

80

SAN Configuration Not Supported

4100FS.book Page 81 Tuesday, July 8, 2003 4:35 PM

Chapter 5

Accessing the RAID Controllers

There are two types of software user interfaces provided to access, configure and

manage the RAID Controllers; they are AdminiStor Storage Management software

and the controller’s onboard Disk Array Administrator software.

AdminiStor is a HTML/Java application that provides a GUI interface through a

standard web browser for configuration and management of the disk arrays.

The onboard firmware-based Disk Array Administrator software provides a

VT-100 terminal interface which is accessed by directly attaching to either one or

both controllers service ports. The ports are located on the rear of the enclosure.

For detailed software information. refer to the AdminiStor Software Guide or the

VT-100 Software Guide.

With either software product, you can:

■ View component status.

■ Create and manage disk arrays.

■ Monitor system status.

■ Manage drive spares.

■ Configure the controller.

■ Manage disk drives and enclosure components.

At the end of this chapter are procedures for updating the RAID Controller’s

onboard firmware.

81

4100FS.book Page 82 Tuesday, July 8, 2003 4:35 PM

Chapter 5 - Accessing the RAID Controllers

Accessing the Disk Array Administrator Software/VT-100

You gain access to the RAID Controller onboard Disk Array Administrator software

through a VT-100 terminal connected to one of the RS-232 controller service ports

located at the rear of the enclosure. You must use a null-modem serial cable to

connect the terminal to either the Controller 1 (CTRL 1) or Controller 2 (CTRL 2)

service ports. It is recommended to connect to the specific controller to which you

wish to access. For example; in an active-active configuration where Controller 1 is

the primary controller, access the arrays and utilities through the CTRL1 port.

Configure your host system or terminal RS-232 port to use the following settings:

Setting Value

Terminal Emulation VT-100 or ANSI (for color support)

Font Terminal

Translations None

Columns 80

Set the communications parameters for the terminal program as follows:

Setting Value

Baud Rate 115,200

Data Bits 8

Stop Bits 1

Parity None

Flow Control None

Connector COM1 (typically)

To access the controllers using a VT-100 terminal:

1 From the computer or terminal connected to one of the controller ports, start

your terminal or terminal emulation software.

Be sure that your terminal emulation software is set to use the correct COM
port on your computer. See Terminal Emulator and COM Port Problems in
chapter 7 for more details on how it can auto-detect the baud rate.

82

Accessing the Disk Array Administrator Software/VT-100

4100FS.book Page 83 Tuesday, July 8, 2003 4:35 PM

The initial Boot and POST screens are displayed.

Chapter 5 - Accessing the RAID Controllers

Boot and POST Screen

2 Following the Boot and POST screens the System Menu is displayed.

System Menu Screen

3 You can now perform all of the functions described in the following

chapters. All steps start from the System Menu.

If an event has occurred, you will see a message about the problem. This
message will also be stored in the event log.

Accessing the Disk Array Administrator Software/VT-100

83

4100FS.book Page 84 Tuesday, July 8, 2003 4:35 PM

Chapter 5 - Accessing the RAID Controllers

Menu System

Below and on the next page are charts of the menu system for the onboard Disk

Array Administrator software. These are provided to assist you with quickly

locating a specific software function within the menu system.

System Menu

Array Menu
Add an Array
Delete an Array
Pool Spare Menu
Display Drives
All Partitions Menu
Configuration Menu
Utilities Menu
Event Log Menu
Other Controller Menu
Shutdown/Restart

Array Menu

Array Status
Drive Status
Abort Initialization
Verify Function
Expand Function
Add Spare
Delete Spare
Change Array name
Trust Array
Switch Array Owner
Partition Menu
Add a Partition
Delete a Partition

Add an Array

Enter Array Name
Single Partition
Enter LUN
Select RAID Type
Number of Drives
Select Drives
Chunk Size
Number of Spares

Pool Spare Menu

Display Pool Spare
Add Pool Spare
Delete Pool Spare

Partition Menu

Partition Status
Partition Statistics
Expand Partition
Change LUN
Change Partition Name
Delete This Partition

Verify Function

Start Verify
View Verify Status
Abort Verify

Expand Function

Start Expand
View Expand Status

Partition Statistics

View Statistics
Reset Statistics

84

Menu System

Disk Array Administrator Software Menu 1 of 2

4100FS.book Page 85 Tuesday, July 8, 2003 4:35 PM

Menu System (continued)

System Menu

Array Menu
Add an Array
Delete an Array
Pool Spare Menu
Display Drives
All Partitions Menu
Configuration Menu
Utilities Menu
Event Log Menu
Other Controller Menu
Shutdown/Restart

Configuration Menu

Set Date/Time
Host Configuration
Channel Configuration
SEP Configuration
Disk Configuration
Option Configuration
Backoff Percent
Utility Priority
Alarm Mute
New Sample Rate
Restore Defaults

Utilities Menu

Rescan
Hot Swap Pause
Hardware Information
LUN Information
Drive Utilities Menu
Overall Statistics

Other Controller Menu

Other Information
Kill Other
Unkill Other
Shutdown Other
Shutdown Both

Chapter 5 - Accessing the RAID Controllers

Set Date/Time

Set Time
Set Date

Host Configuration

Channel
Enable/Disable
Link Speed
Target ID
Controller LUN
Topology (Loop or Point-to-Point)

Channel Configuration

Channel
Bus Speed
Disable Domain Validation

SEP LUNs

SEP Settings
SEP LUN

Disk Configuration

Write-Back Cache
SMART

Option Configuration

Operating Mode
Cache Lock
Battery
Enable Trust Array
Dynamic Spare Configuration
Enclosure Features

SEP Settings

Poll Rate
Temperature
Slot Flags
Global Flags

Enclosure Services

Config Internal Hubs
Connect I. Hubs on FO

Drive Utilities Menu

Blinik Drive LED
Clear Metadata
Down Drive
Test Unit Ready
Display Drive Cache

Overall Statistics

View Statistics
View R/W Histogram
Reset All Statistics

Disk Array Administrator Software Menu 2of 2

Menu System

85

4100FS.book Page 86 Tuesday, July 8, 2003 4:35 PM

Chapter 5 - Accessing the RAID Controllers

Updating RAID Controller Firmware

The following information describes the procedures to upload new firmware to
the RAID Controllers. The firmware is uploaded when the controller’s are offline
and during the boot process.

1 Connect one end of the null-modem RS-232 cable to the RAID Controller 1

Service port located on the rear panel of the enclosure.

Controller 1: Service Port

CTRL 1

SAF-TE

CTRL 2

SAF-TE Service Port

Controller 2: Service Port

Connecting the Serial Cable

The cable is a female-to-female DB-9 null-modem serial cable.

2 Connect the other end of the cable to either a host system’s serial

communication port or a VT-100 type terminal.

3 On a host system, run a terminal emulation program or start the terminal.

4 Verify the communication parameters are as follows:

• 115,200 Baud

This is the speed at which the controller is communicating.

• 8 Data bits

• 1 Stop bit

• None (parity)

• Flow Control Off

5 Boot the RAID Controller Enclosure.

6 When the boot information appears on the screen press the <Space Bar>.

86

Updating RAID Controller Firmware

4100FS.book Page 87 Tuesday, July 8, 2003 4:35 PM

Chapter 5 - Accessing the RAID Controllers

FLASH Loader Screen

7 Select option “3” xmodem by pressing the <3> key.

FLASH Loader Waiting for Transfer Selection Screen

8 Using the mouse, click on the pull-down menu Transfers and select “Send.”

Updating RAID Controller Firmware

87

4100FS.book Page 88 Tuesday, July 8, 2003 4:35 PM

Chapter 5 - Accessing the RAID Controllers

9 Click the browse button and locate the new Firmware file and click “Send.”

The firmware file will have a “.fla” extension.

NOTE: Ensure that the protocol “Xmodem” is selected.

Send File Screen

From the Xmodem send screen you can monitor the progress of the upload.
You can safely stop the transfer without affecting your existing firmware any
time during the transfer until it has been completed. The upload does not
overwrite the firmware during the upload process, it writes the new code
into unused EEPROM space until completed, then copies the new firmware
code to the EEPROM active region.

If you elect to stop an upload progress, ensure that the stop (abort)
command was completed by typing <Control-X> at the cursor.

10 After the upload is successful, the Flash Loader will reboot the controllers.

During the reboot it will bring the second controller offline, if applicable, and
update its firmware.

88

Updating RAID Controller Firmware