ATTO Technology Diamond Array S Installation And Operating Manual

Diamond Storage Array

S-Class

Installation, Operations, Maintenance

M anual

© 2005 ATTO Technology Inc. All rights reserved. All brand or product names are trademarks of their respective holders. No part of this manual may be reproduced in any form or by any means without the express written permission of ATTO Technology Inc.

2/05

6.4 PRMA-0339-000

Contents

Preface

1.0 Diamond Storage Array Product Overview ..................................1

Diamond Storage Array features Fibre Channel model SCSI model

2.0 Diamond Storage Array Technical Overview ...............................3

ADXT Powers ATA to New Levels

3.0 Installation Instructions .................................................................5

Three major steps are required to set up the Diamond Storage Array Step 1: physical setup Step 2a: set up the Ethernet connection Step 2b: connect to Ethernet If the Diamond Storage Array is attached to a DHCP server If the Diamond Storage Array is not attached to a DHCP server and you wish to change the defaults Step 3: set up RAID configuration

3.1 Components .........................................................................7

Floor model Rack mount Floor model Rack mount General instructions

3.2.1 Connecting a Fibre Channel Array .......................11

Autoconfiguration Manual configuration

3.2.2 Connecting a SCSI Array .......................................13

To connect a SCSI array

4.0 Determining Drive and Sled Designations ...................................15

Numbering conventions Physical numbering Logical numbering Examples Unique serial number for each LUN

5.0 Accessing the Array .......................................................................17

Command Line Interface ATTO ExpressNAV In-band SCSI over Fibre Channel RS-232 port Ethernet port SNMP

5.1 In-band CLI Over the Fibre Channel ...................................19

I/O details

5.2 Serial Port Access ................................................................21

5.3 Ethernet Access: Telnet and SNMP Protocols ..................23

To connect to the Ethernet port To use Telnet To use SNMP

5.4 ATTO ExpressNAV: Browser-based Interface ...................25

Browser compatibility To optimize ExpressNAV in Internet Explorer Opening an ExpressNAV session Navigating ExpressNAV

5.4.1 ExpressNAV Pages ................................................27

Status Ethernet SNMP Serial Port Fibre Channel Storage Management RAID Clear Data Logical Units Rebuild Configuration Advanced To use the advanced page CLI commands Restart Help

5.5 Command Line Interface Use and Guidance .....................29

5.5.1 Summary of CLI Commands .................................31

5.5.2 General Use Commands ........................................33

FirmwareRestart Help RestoreConfiguration SaveConfiguration SystemSN VerboseMode

5.5.3 Ethernet Commands ..............................................35

EthernetSpeed FTPPassword IPAddress IPDHCP IPGateway IPSubnetMask SNMPTrapAddress SNMPTraps

SNMPUpdates Te ln e t Pa s sw or d TelnetTimeout TelnetUsername

5.5.4 Fibre Channel Configuration Commands ............37

FcConnMode FcDataRate FcFairArb FcFrameLength FcFullDuplex FcHard FcHardAddress FcNodeName FcPortInfo FcPortList FcPortName FcWWName

5.5.5 Serial Port Configuration Commands ..................39

SerialPortBaudRate SerialPortEcho SerialPortHandshake SerialPortStopBits

5.5.6 Diagnostic Commands ..........................................41

AudibleAlarm DiamondModel DiamondName DriveCopyStatus DriveInfo FcNodeName FcPortList FcPortName Help IdentifyDiamond Info LUNInfo SerialNumber SledFaultLED SMARTData Temperature VirtualDriveInfo

5.5.7 Maintenance Commands .......................................43

FcScsiBusyStatus FirmwareRestart MaxEnclTempAlrm MinEnclTempAlrm Temperature Zmodem

5.5.8 Drive Configuration Commands ...........................45

ATA Di sk S t at e AutoRebuild ClearDiskReservedAreaData DriveCopy DriveCopyHalt DriveCopyResume DriveCopyStatus DriveInfo DriveSledPower DriveWipe IdeTransferRate LUNInfo LUNState QuickRAID0 QuickRAID1 QuickRAID5 QuickRAID10 RAID5ClearData RAID5ClearDataStatus RAIDInterleave RAIDHaltRebuild RAIDManualRebuild RAIDRebuildState RAIDRebuildStatus RAIDResumeRebuild RebuildPriority ResolveLUNConflicts RestoreModePages SledFaultLED VirtualDriveInfo

6.0 Configuring Drives ..........................................................................49

JBOD (Just a Bunch of Disks) RAID Level 0 RAID Level 1 RAID Level 10 RAID Level 5 Interleave Hot Spare sleds Enhancing performance

6.1 JBOD .....................................................................................51

To set up the JBOD configuration

6.2 RAID Level 0 .........................................................................53

Sled-based versus disk-based To set up RAID Level 0 groups To remove RAID Level 0 groups

6.3 RAID Level 1 .........................................................................57

To set up RAID Level 1 groups To remove RAID groups Hot Spare sleds To set up RAID Level 1 with Hot Spare sleds

6.4 RAID Level 5 .........................................................................59

Configuring a fully-populated array To set up one RAID Level 5 group with one Hot Spare sled To set up two RAID Level 5 groups with two Hot Spare sleds Configuring a partially-populated array To set up one RAID Level 5 group with one Hot Spare sled Removing RAID groups

6.5 RAID Level 10 .......................................................................63

To set up RAID Level 10 groups To remove RAID groups Hot Spare sleds To set up RAID Level 10 with Hot Spare sleds

6.6 Rebuilding RAID Configurations ........................................65

To reset LUN status To synchronize mirrored drives automatically Rebuild priority To synchronize mirrored drives manually

6.7 RAID Interleave Parameters ................................................67

To change the RAID Interleave parameter

7.0 Hardware Maintenance ...................................................................69

To remove a management card To remove a disk drive sled (Exhibit 7-2) To remove a Host Interface Card To remove the power supply (Exhibit 7-4), To remove a blower assembly (Exhibit 7-4),

7.1 Hot Swap Operating Instructions .......................................71

Disk drives Power supplies Blower assemblies To replace a blower assembly

7.2 Optional Hot Spare Sled ......................................................73

To set up RAID Level 1 with Hot Spare sleds To set up RAID Level 10 with Hot Spare sleds To set up one RAID Level 5 group with one Hot Spare sled To set up two RAID Level 5 groups with Hot Spare sleds

8.0 Copying Drives ................................................................................75

Procedure

9.0 Updating Firmware .........................................................................77

Updating firmware using the RS-232 serial port To update firmware Updating firmware using the optional Ethernet card

10.0 System Monitoring and Reporting ..............................................79

RS-232 monitoring port and CLI Ethernet monitoring port and CLI Power On Self Test (POST) Ready LED Audible alarm Thermal monitoring Power supply monitoring System fault LED and error codes Disk drive activity and disk fault LEDs

10.1 Troubleshooting .................................................................82

Windows 2000 special instructions Error messages Specific situations and suggestions If a drive fails to respond If a power supply fails To determine if the problem exists with the HIC If you can’t access the CLI via Ethernet If LUNs do not display on the host

10.2 Resetting Defaults ..............................................................85

Default To return to Default settings Factory Default To reset to factory defaults

Appendix A

Appendix B

Appendix C

Appendix D

Appendix E

ATA Disk Technology .....................................................i

Information Commands Results ....................................ii

Product Safety .................................................................vi

Safety compliances EMC specifications Radio and television interference

Specifications ..................................................................vii

Environmental and physical Rack mount dimensions Floor mount dimensions

Warranty ...........................................................................viii

Manufacturer limited warranty Contact ATTO Technology, Inc.

Preface

This guide will take the technology-savvy user through the installation and maintenance of the Diamond Storage Array.

The Diamond Storage Array was designed to meet your need for large amounts of easily accessible storage using proprietary Aggregated Data Transfer Technology (ADXT multiple, high performance ATA disk drives with the sustained data transfer rates required by sophisticated computer users.

Your comments help us improve and update our products. Contact us

) to merge the power of

Disclaimer

ATTO Technology, Inc.

155 CrossPoint Parkway Amherst, New York 14068 (716) 691-1999 • voice (716) 691-9353 • fax http://www.attotech.com/diamond

ATTO Technology can also be reached via e-mail at the following addresses

Sales Support Technical Support

: sls@attotech.com

: techsupp@attotech.com

Although reasonable efforts have been made to assure the accuracy of the information contained herein, this publication could include technical inaccuracies or typographical errors. Manufacturer expressly disclaims liability for any error in this information and for damages, whether direct, indirect, special, exemplary, consequential or otherwise, that may result from such error including but not limited to loss of profits resulting from the use or misuse of the manual or information contained therein (even if Manufacturer has been advised of the possibility of such damages). Any questions or comments regarding this document or its contents should be addressed to Manufacturer.

Manufacturer provides this publication as is, without warranty of any kind, either express or implied, including, but not limited to, the implied warranties for merchantability or fitness for a particular purpose.

Information in this publication is subject to change without notice and does not represent a commitment on the part of Manufacturer. Changes may be made periodically to the information herein; these changes will be incorporated in new editions of the publication. Manufacturer reserves the right to make improvements and/or changes at any time in product(s) and/or program(s) described in this publication.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

1.0 Diamond Storage Array Product Overview

The Diamond Storage Array offers up to 24 ATA disk drives in a rack mount or floor model configuration.

The Diamond Storage Array is ideally suited for data intensive applications such as web hosting, email servers, on-line transaction processing, digital video editing, medical imaging and digital audio editing. Virtually any high performance computing system with a growing need for storage capacity can use the power of the Diamond Storage Array.

With the cost effective approach of using ATA disk drives, you can add more storage capacity as your needs grow without the budget-busting costs of other disk storage technologies. You can also improve the performance and capacity of the Diamond Storage Array cabinet as technology progresses by simply replacing disk drive sleds and Host Interface Cards.

The Diamond Storage Array is operating system independent and supports all popular computer

hardware platforms and network environments with a 2-gigabit Fibre Channel interface or an Ultra160 SCSI interface.

The Diamond Storage Array is a fully populated, turnkey solution with drives pre-installed. It is fully supported by highly trained customer service and engineering staffs.

The Diamond Storage Array uses Aggregated Data Transfer Technology (ADXT

) to merge the performance of multiple ATA drives to achieve sustained, full bandwidth data transfer rates. ADXT provides end users with the power and sophisticated data control needed to take moderately priced ATA disk drives, combine them in a disk storage array, and power them to the performance levels of SCSI or Fibre Channel disk arrays.

Diamond Storage Array features

• Up to 7.2 Terabytes initial configuration

(expandable with future drive technology)

• 24 ATA disk drive capacity

• Aggregated Data Transfer Technology

(ADXTJ) for high performance/scalability

• Ultra ATA 100 megahertz (Mhz)

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

• JBOD, RAID Level 0, RAID Level 1, RAID Level 10 and RAID Level 5 configurable

• Staggered drive spin-up to reduce peak power demand

• Tagged command queuing to process up to 255 simultaneous data requests

• RS-232 management for local management control; Ethernet option available for setup connection only

• ATTO ExpressNAV for management and information

• Two power supplies capable of 85-264 V (rated 100-240V AC) operation (340 watts each)

• UL, TUV and CE marked and compliant

• Internal thermal and power management

• Redundant hot swappable power supplies with integrated thermal and power management

• Hot spare sleds: replace degraded sleds with spares on the fly using software

• Floor model cabinet or 19” 3U rack mount

Exhibit 1.0-1 Back of rack mount model, Diamond Storage Array. Left: Fibre Channel. Right: SCSI.

browser-based interface

Fibre Channel model

• 2 gigabit Fibre Channel Port (single or dual channel)

• SFP-based Fibre Channel interface supports long wave and short wave optical cables

• Built-in hub for daisy-chaining

• Up to 9,500 I/Os per second per port

• Up to 180 MB/sec. sustained Fibre Channel transfer rates per interface

SCSI model

• Ultra 160 SCSI bus.

• Dual stacked VHDCI connectors for daisychaining and termination

• SCSI Target ID selection switch

• Support for single-ended and LVD SCSI

• No onboard termination

Product overview

2.0 Diamond Storage Array Technical Overview

The Diamond Storage Array uses Aggregated Data Transfer Technology (ADXT) to achieve the high data transfer performance you need. ADXT achieve sustained, full bandwidth Fibre Channel data transfer rates.

merges the performance of multiple ATA drives together to

Unlike other storage arrays which use expensive SCSI or Fibre Channel disk drives to achieve performance, the Diamond Storage Array uses lower cost ATA disk drives combined with an intelligent midplane to create a storage array with exceptional price and performance characteristics.

The intelligent midplane contains hardware and software which provide the proprietary ADXT

Exhibit 2.0-1 Data pathways and architecture for Fibre Channel operation

switched data management and data movement technology, to create a higher performance storage solution.

The Diamond Storage Array is made up of dual SCSI or Fibre Channel Host Interface Cards, the intelligent midplane, a system management card, and 12 independent disk drive sleds containing up to 24 ATA disk drives.

ADXT Powers ATA to New Levels

The original notion of RAID was to build high capacity, reliable storage subsystems using large numbers of inexpensive disk drives. Thus its original definition:

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Redundant Array of

Inexpensive Drives

became

Redundant Array of Independent Drives

. Over time that definition

and the inherent cost advantage in a RAID system was lost. The Diamond Series provides a high capacity, high performance and highly reliable

disk array that uses the merged power of ATA disk drives.

enclosure management services, serverless backup, data replication and LUN mapping.

Intelligent Midplane

The heart of the Diamond Series disk storage array is the intelligent midplane with ADXT

to sum or

aggregate

the data rates of individual ATA disk drives to create high data transfer rates. This technology enables features such as serverless backup, advanced error protection, metadata storage techniques, virtualization software, thermal management and advanced enclosure services.

The midplane contains a combination of custom Application Specific Integrated Circuits (ASICs), processors and proprietary embedded software. It is divided into three main processing sections which handle the data being read or written to the ATA disk drives from the Fibre Channel or SCSI host interfaces. The Virtual Device Manager (VDM), Data Routing Fabric (DRF) and ATA Interface Engines (AIE) organize data streams for storage or retrieval.

Virtual Device Manager

Data is accessed through virtual devices using an implementation of the standard SCSI protocol controlled by the Virtual Device Manager. The VDM adds capabilities such as RAID, data management services,

Data Routing Fabric

Incoming or outgoing data is routed between the ATA Interface Engines (AIE) and the Fibre Channel or SCSI interface by the custom Data Routing Fabric ASIC, a high speed, low latency transfer fabric with more than 2 GB/sec. of bandwidth supported by up to 512 Megabytes of memory.

ATA Interface Engine (AIE)

The interface to each pair of drives is through a custom ATA Interface Engine ASIC. The AIE implements the typically software-intensive ATA interface completely in silicon. Each AIE contains a dedicated ATA protocol processor to completely automate command and protocol processing. Automated command overlapping and queuing maximizes the performance of multi-threaded I/O. The AIE transfers data using double-clocking technology at 66 megabytes per second. The Diamond Storage Array contains 12 AIEs for a parallel transfer rate of 792 megabytes per second. In addition to guaranteeing data transfer integrity with automatic CRC checking, the command and status transfers are validated using a patent pending technology unique to the AIE.

Disk 2

Disk 1

AIE

Disk 2

Disk 1

AIE2AIE3AIE4AIE5AIE6AIE7AIE8AIE9AIE10AIE11AIE

DATA ROUTING FABRIC

text

VIRTUAL DEVICE MANAGER

FIBRECHANNEL

INTERFACE 1

FIBRECHANNEL

INTERFACE 2

Disk 2

Disk 1

3.0 Installation Instructions

If you are familiar with the Diamond Storage Array, Fibre Channel, SCSI and RAID configurations, you may set up and configure the Diamond Storage Array using these instructions. Find details, illustrations and other guidance for more involved operations and special cases in the rest of this manual.

CAUTIONCAUTION

Before configuring the Diamond Storage Array, ensure that any computer data to be stored on the array is properly backed up and verified. The operation and storage of computer data on any disk storage array should be accompanied by a regular program of computer data backups and verification for the data stored on such an array.

The Manufacturer is not responsible for the loss of any data stored on the Diamond Storage Array under any circumstances and any special, incidental, or consequential damages that may result thereof.

Three major steps are required to set up the Diamond Storage Array

1 Physically set up Diamond Storage Array

2 Connect to Ethernet

3 Configure the drives

Step 1: physical setup

1 Make sure the Diamond Storage Array is

mounted properly and has adequate air flow around it.

2 Insert the appropriate connector into the

interface card in the back of the Diamond Storage Array.

3 Connect the Fibre Channel or SCSI cable from

your host computer system to the connector. To use ExpressNAV browser-based management interface and configure your Diamond Storage Array, you must connect to the Ethernet port.

4 To set up the Ethernet connection: connect a

cross-over cable (for a direct connection to a PC) or regular network cable from a network device to the RJ45 Ethernet port on the Ethernet management card on the front of the Diamond Storage Array.

Step 2a: set up the Ethernet connection

The Diamond Storage Array supports service operations over the RS-232 serial port using standard terminal emulation software available with most systems.

1 Connect a DB-9 null modem serial cable

between the port marked RS-232 on the front of the Diamond Storage Array and the computer’s serial port. The cable must be no longer than three meters.

2 Make sure the power switches on the power

supplies on the rear of the unit are in the Standby position.

3 Plug in the power cords to the back of the unit,

then into an appropriate power source (100-240 VAC).

4 Reboot your host computer system.

5 Press the Stand-by power switch for each

power supply on the Diamond Storage Array to the ON position.

6 Upon successful power up and diagnostics, the

unit displays the POST (power up self test) information.

The Diamond is now in Command Line Interface mode. You may modify the setup of the Diamond Storage Array using the CLI (refer to

via ASCII-based Commands

on page 27), but the

CLI: Interface

easiest method to configure the array is by using ATTO ExpressNAV software, a graphical user management interface accessed through a standard Internet browser. Refer to

ExpressNAV: Browser-based Interface

ATTO

on page 23.

Step 2b: connect to Ethernet

If the Diamond Storage Array is attached to a DHCP server

1 At the Ready prompt after POST (refer to Step

6 above), type set IPDHCP enabled

2Type SaveConfiguration Restart

3At the Ready prompt after POST (see above),

type get IPAddress

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

4 Enter this address into your browser.

5 The ATTO ExpressNAV screen appears. Log in

using the Telnet defaults:

Username: Telnet

Password: Diamond

6 Follow the screens to find information about the

array or to configure the array from the factorydefault settings. The Diamond Storage Array may be set up in a JBOD, RAID Level 0, 5 or 10 configuration with or without Hot Spare sleds

If the Diamond Storage Array is not attached to a DHCP server and you wish to change the defaults

1 At the Ready prompt after POST (see above),

type set IPAddress [desired IP address].

2Type set IPSubnetMask [desired IP Subnet

Mask]

3Type set IPGateway [desired IP Gateway]

4Type SaveConfiguration Restart to save the

configuration and restart the Diamond Storage Array

5 After the powerup and POST complete, type

the IP address from step 1 above into your browser.

6 The ATTO ExpressNAV screen appears. After

logging in (refer to Step 2b: connect to Ethernet, Step 5 on page 6), follow the screens to find information about the array or to configure the array from the factory-default settings. The Diamond Storage Array may be

set up in a JBOD, RAID Level 0, 1, 5 or 10 configuration with or without Hot Spare sleds.

Step 3: set up RAID configuration

The simplest way to set up configurations is to use the ExpressNAV interface. Refer to

ExpressNAV: Browser-based Interface

ATTO

on page 23 for more information on the interface. After logging in (refer to

page 6

Step 2b: connect to Ethernet

, Step 5 on

), follow the screens to find information about the array or to configure the array from the factory-default settings

You may the CLI, the

QuickRAID

commands to

set up RAID and hot spare sleds, if required.

• Each Hot Spare sled configuration requires a certain number of Hot Spare sleds. These sleds, once designated as Hot Spares, are not available for other use.

• RAID Level 10: 10 drive sleds and two Hot Spare sleds (sleds 11 and 12).

Note

All arrays using RAID level 10 and Hot Spare sled options must be fully populated.

• RAID Level 5: 10 drive sleds and two Hot Spare sleds (sleds 11 and 12)

• RAID Level 5: five drive sleds (sled positions 1 through 5) and one Hot Spare sled (sled

12).

Installation

3.1 Components

The Diamond Storage Array has been designed to be easy to use, maintain and upgrade. It features a durable steel outer case and modular components in either a floor or a rack mount model.

Immediately upon receipt, check the shipping carton for damage from mishandling. Contact us at once via the means that is easiest for you (refer to

Contact ATTO Technology, Inc.

on page viii) if the carton has been mishandled or displays any signs of damage.

The front of the Diamond Storage Array provides access to the management card and disk drive sleds. The rear of the unit holds the Host Interface Cards, power supplies and blower assemblies.

CAUTIONCAUTION

All modular components must be replaced by qualified personnel only.

Floor model

The management system card is at the top front of the case. At its center is a DB-9 serial RS-232 port, a connection for setup, monitoring and upgrade of the unit from any computer system with an RS-232 interface. The optional 10/100 BaseT Ethernet management services card enables Telnet-based monitoring and management. It also provides the ability to update the firmware in the array via FTP.

LEDs to the port’s right indicate fault, unit ready, Host Interface Cards A and B installation status, and the power status for each power supply.

sled contains two hard drives. Up to 24 hard drives may be installed on the 12 sleds. Empty bays should be covered by blank faceplates or empty sleds. Access is provided by loosening two screws and gently pulling on the sled handle.

On the rear of the unit are blowers which support hard drive, cabinet and power supply cooling. The blowers are held in by removable screws. Correct operation is displayed by a LED at the top of each panel.

The power supplies for the array, also in the rear of the unit, are accessible by loosening two screws and pulling on the power supply module handle. The power standby on/off switch is at the top of each module. A yellow LED indicates and a green LED indicates

. The power cord

caution

socket is at the bottom of each power supply.

Between the power supplies and blower assemblies are two slots that hold the Host Interface Cards. The HIC is the connection point into the array and is available in three options: 1Gigabit Fibre Channel, 2-Gigabit Fibre Channel, or Ultra 160 SCSI. Host Interface Cards have and

on-line

fault

and

activity

LED indicators,

fault

depending on the model.

SCSI

The SCSI card faceplate has a rotary binary-coded hex switch to set the SCSI ID of the array. The SCSI card also has an in channel, to connect via cable to the unit’s communication source, and an out channel, available for daisychaining arrays together or to complete termination using an external LVD terminator.

Below the management system card are individual disk drive sleds which also have LEDs for each drive’s status. Each

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Rack mount

The system management card is at the left front of the case. At its center is a DB-9 serial RS-232 port which allows a connection for setup, monitoring and upgrade of the unit from any computer system with an RS-232 interface. The optional 10/100

BaseT Ethernet management services card enables Telnet-based

System Management Card

monitoring and management. It also provides the ability to update the firmware in the array via FTP. LEDs farthest to the left indicate fault, unit ready, Host Interface Cards A and B installation status, and the

Disk drive sleds (12)

power status for each power supply.

Power supplies

To the right of the system management card are individual disk drive sleds which also have LED indicators for each drive’s status. Each sled contains two hard drives. Up to 24 hard drives may be installed on the 12 sleds. Empty bays should be covered by blank faceplates or filled with empty sleds. Access is provided by loosening two screws and pulling on the sled handle.

In the rear of the unit are the blower assemblies which support hard drive, cabinet and power supply cooling. Correct operation is displayed by a LED at the top of each panel. The blowers are held in place by removable screws.

The power supplies for the array are accessible by loosening two screws and pulling on the power supply module handle. The power standby on/off switch is at the top of each module. A yellow LED indicates

caution

and a green LED indicates on. The power cord socket is at the bottom of each power supply.

Intelligent midplane (internal)

Host interface cards

Blower assemblies

Between the power supplies and blower assemblies are two slots that hold the Host Interface Cards. The HIC is the connection point into the array and is available in a 2-Gigabit Fibre Channel or Ultra 160 SCSI. Host Interface Cards

fault

and

have

on-line

fault

and

activity

LED

indicators, depending on the model.

SCSI

The SCSI card faceplate has a rotary binary-coded hex switch which allows you to set the SCSI ID of the array. The SCSI card also has an in channel, to connect by a cable to the unit’s communication source, and an out channel, available for daisy-chaining arrays together or to complete termination using an external LVD terminator.

Components

3.2 Physical Set Up

The Diamond Storage Array is shipped completely assembled with two 120 VAC power cords for use in the United States and Canada.

Immediately upon receipt, check the shipping carton for damage from mishandling. Contact us at once by the means easiest for you (refer to

Contact ATTO Technology, Inc.

on page viii) if the carton has been mishandled or displays any signs of damage.

Floor model

The Diamond Storage Array is heavy (about 92 pounds fully loaded) and requires two people to lift and carry it safely. Place the array on a level surface and make sure there is adequate space in the front and back of the unit for proper cooling and airflow. Continue with the general instructions.

Rack mount

The array fully loaded is heavy (about 86 pounds). The unit should be handled with care and requires two people to lift, carry and/or install it safely.

The Diamond Array can be mounted via several different methods in a 19” rack with 3U (5.25”) of vertical space required. Air flow should not be restricted in any way.

Installation in a rack may create a differential between the room ambient temperature and the internal ambient temperature in the rack. While the maximum internal operating temperature of the array is 47°C, you should not run the system at the maximum temperature for extended periods. Therefore, ensure that the room ambient temperature is kept below 30°C for best operation.

Spaced rail pairs in some rackmount cabinets.

You can mount the array using two sets of rail pairs spaced to accommodate the overall length of the unit (approximately 23 inches). Mount using the rack mount brackets on the front and rear of the unit fastened to the rail pairs using 10/32 pan head screws with lock and flat washers.

Rack mount cabinets with stationary shelf or tray system.

The shelf or tray must be able to support at least 125 pounds. The shelf or tray must be installed and secured to the rack before installing the array. Secure the front of the array to the rack with 10/32 screws, locks and flat washers.

Sliding shelf or tray type systems

should never

be used under any circumstances.

Two point open rack system.

The rack must be strong enough to support the Diamond Array properly. Mounting brackets should be moved to the centermost mounting holes and secured using 10/32 screws.

CAUTIONCAUTION

Do not mount multiple arrays on a two-rail rack or mount the array above the midpoint of a two-rail rack system. Do not mount the array on any kind of rail-type system. The array is too heavy and does not have the proper hole pattern for rails.

Insure the array has adequate air flow and continue with the general instructions.

Each side of the rack mount array chassis has three pairs of mounting holes. One is located near the front of the rack, one near the unit’s center of gravity, and one near the rear of the rack. The holes will accommodate 10/32 screws but the screws can protrude no farther than .375 inches into the rack.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

General instructions

1 Insert the proper connector into the Host

Interface Card in the back of the array. (Refer to

Connecting a Fibre Channel Array

Fibre Channel and page 13 for SCSI).

2 Connect the cable (Fibre Channel or SCSI)

from your host system to the Host Interface Card connector on the back of the array. The cable you use depends upon your application, the environment and distance.

3 Make sure the power switches on the power

supplies on the rear of the unit are in the standby position. Plug in the power cords to the back of the unit, then into an appropriate power source (100-240 VAC). The power source must be connected to a protective earth ground and comply with local electrical codes. Improper grounding may result in an electrical shock or damage to the unit.

Exhibit 3.2-1 Back side of a rack mount Fibre Channel Diamond Storage Array.

Connecting a SCSI Array

on page 11 for

4 Press the stand-by power switch for each

power supply to the ON position. When the green power LED on the back of the unit is lit, the power supply is fully operational and delivering power to the system. The power LED on the front of the array will light once the firmware begins to execute.

When the power is turned on, the LEDs on the front of the array will flash twice. Drives will spin up in groups of three, about every one to two seconds. The individual LEDs will blink. After all available drives have spun up, the individual drive LEDs will stay lit. When all available drives are operational, the ready LED on the top front panel of the management card will remain lit.

5 Reboot your computer

6 Determine the best configuration for your needs

(i.e. JBOD, RAID, etc.) and refer to the rest of this manual for more detailed explanations and configuration information.

Installation

3.2.1 Connecting a Fibre Channel Array

The Diamond Storage Array supports up to two Fibre Channel Host Interface Cards (HIC). Physical connections and CLI commands contribute to the Fibre Channel topology.

The cable you use depends upon your application, the environment and the distance required for your

2 Gb HIC uses

2 SFPs to connect up

to 2 FC cables

storage area network.

To comply with FCC and CE EMI for the 2-gigabit

Host Interface Card, use fiber optic cables.

Exhibit 3.2-2 Fibre Channel cable options

Cable length Cable type Cable size

<10 meters Unequalized copper

>10 <30 meters Equalized copper

Up to 175 meters multi mode fiber optic 62.5 micron

Up to 500 meters multi mode fiber optic 50 micron

Up to 10 kilometers single mode fiber optic 9 micron

The Diamond Storage Array may have two Fibre Channel Host Interface Cards (HIC). In 2 gigabit Fibre Channel arrays, each HIC is connected by a Fibre Channel cable via a SFP (small form factor pluggable) module into a point-to-point or loop Fibre Channel topology.

Install the SFP according to the manufacturer’s instructions.

Each HIC has two ports and an on-board hub. Each port has an SFP module to connect to Fibre Channel. Each HIC is independent of the other, so that one may be connected into a point-to-point topology and the other into a loop. However, if one port in a HIC is connected into a point-topoint topology, the other port cannot be used.

One of the advantages of using loop topology for Fibre Channel connections is that it allows arrays to be daisy-chained together.

Autoconfiguration

The array automatically determines which HICs are installed and if they are in loop or point-topoint topologies.

If you wish to see how the unit has been set up, go to the CLI commands and type

Info or go to the

Status page of the ExpressNAV interface. The

return displays the data rate and connection mode for each HIC (FC 0 and FC 1).

Manual configuration

You may manually configure the array using CLI commands (refer to

Guidance

Channel

(refer to

on page 29.) or access the

page of the ExpressNAV interface

ATTO ExpressNAV: Browser-based Interface

Command Line Interface Use and

Fibre

on page 25).

•

FCConnMode

topology for both HICs on an array. Options are loop only (loop), point-to-point only (ptp), loop preferred (loop-ptp) or point-to-point preferred (ptp-loop).

Loop connects to either an FC-AL arbitrated loop or a fabric loop port (FL_Port) on a switch.

Point-to-point (ptp) connects to a direct fabric connection, such as an F port on a switch.

Loop-ptp allows the array to determine what kind of connection to use, but tries to connect in loop mode first, then point-to-point mode.

Ptp-loop allows the card to determine what kind of connection to use, but tries to connect in point-topoint mode first, then loop mode.

• FcDataRate specifies the Fibre Channel data

rate at which both HICs on a Diamond operate. Choices are 1 gigabit, 2 gigabit and autodetection.

specifies the Fibre Channel

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Exhibit 3.2-3 Possible 2 gigabit Fibre Channel physical connections depending on which Fibre Channel connection mode has been selected.

point-to-

point mode

loop

mode

Diamond Storage Array A Diamond Storage Array B

connection

possible

loop

mode

daisy -chain

loop

mode

loop

mode

loop

mode

loop

mode

Fibre Channel connections

3.2.2 Connecting a SCSI Array

The SCSI Diamond Storage Array uses a VHDCI connector and SCSI cables to connect to a host. It will automatically detect the type of Host Interface Card it is using without any intervention.

To connect a SCSI array

1 Insert a SCSI VHDCI connector into the Host

Interface Card in the back of the array.

2 Connect the cable from your host system to one

of the VHDCI connectors on the Host Interface Card connector on the back of the array.

3 If the SCSI array is the last device on the bus,

you must attach a VHDCI terminator to one connector of the SCSI Host Interface Card or

Exhibit 3.2-3 SCSI cable options.

Bus speed, MB/sec.

max.

SCSI-1 5 8 6 - 8

Fast SCSI 10 8 6 - 8

Fast Wide SCSI 20 16 6 - 16

Wide Ultra SCSI 40 16 3 - 4

Wide Ultra SCSI 40 16 1.5 - 8

Wide Ultra 2 SCSI 80 16 - 12 16

Ultra 3 or Ultra160 SCSI 160 16 - 12 16

Bus width,

bits

connect a cable between the second connector and the next device on the SCSI bus.

4 The SCSI Host Interface Card has a rotary

binary-coded hex switch which allows you to set the SCSI ID of the HIC. Be sure the selected ID is different from all other SCSI devices on the bus.

Note

If slower devices are connected on the same SCSI bus as the Ultra 160 array, the bus will communicate at the rate of the slowest device.

Max. bus lengths,

meters

Single-ended LVD

Max. device

support

Exhibit 3.2-4 SCSI interface cards: left without terminators attached; right with a terminator attached.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

SCSI connections

4.0 Determining Drive and Sled Designations

The Diamond Storage Array has been designed with 12 sleds, each holding two drives. The easiest way to configure a array is to use all the drives on all the sleds. The firmware uses a numbering system to determine which drives and sleds it is affecting.

We recommend that all sled slots are filled contiguously, starting with the first slot next to the management card.

When configuring a Diamond Array with fewer than 12 drive sleds, you must consider several factors: RAID level, number of physical drives/sleds and the end configuration you are trying to achieve. Review the information about each configuration to determine how each configuration would be affected by using fewer sleds.

Numbering conventions

The Diamond Storage Array with firmware version 3.1 and newer uses a unique numbering convention to orient its drives and sleds to the controlling firmware. Older versions do not use this convention. (Refer to

Updating Firmware

on page 77 for information about updating the array firmware.)

Physical

refers to the physical drives in the array, the hardware that actually exists in a physical sense.

Logical (or virtual)

refers to what the host operating system recognizes as an entity. Two physical drives may be seen as one logical drive by the operating system.

Logical disks do not always map one-to-one with physical disks.

In RAID configurations, for example, several physical disk drives (or portions of several physical drives) are grouped into a

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

Disk 2

LUN 13

LUN 14

LUN 15

Disk 2

LUN 16

Rack Mount Drive Sleds

LUN 17

Disk 2

LUN 16

LUN 17

Disk 2

LUN 18

Disk 2

LUN 19

LUN 20

LUN 19

logical disk or Logical Unit (LUN). Each LUN is

broken into logical blocks of 512 bytes each,

numbered 0 through n (the Logical Block Number or LBN). A 100 GB LUN has approximately 200,000,000 logical blocks.

Physical numbering

The Diamond RAID Storage Array contains

• Up to 24 physical disk drives

• Two drives mounted on 12 physical drive sleds

• Sleds are numbered 1-12, starting at the top (floor units) or the left (rackmount units).

• Each sled is connected to its own internal ATA bus with two disk drives, numbered 1 and 2.

• Two green LEDs, labeled Drive 1 and Drive 2, indicate activity for the two drives. They remain solidly lighted when there is no activity.

Logical numbering

Logical numbering depends on the RAID configuration of the storage array. S Class firmware includes RAID Level 0, RAID Level 1, RAID Level 10 and RAID Level 5 capability. A variety of configurations may be selected using the QuickRAID0 command. The default configuration is QuickRAID0 0 or JBOD (Just a Bunch of Disks).

Disk 2

LUN 20

LUN 21

LUN 22

LUN 23

Disk 2

LUN 24

Mana

Drive 1

LUN 1

Drive 1

LUN 2

Drive 1

LUN 3

Drive 1

LUN 4

Drive 1

LUN 5

Drive 1

LUN 6

Drive 1 LUN 7

Drive 1

LUN 8

Drive 1

LUN 9

Drive 1 LUN 10

Drive 1 LUN 6

ement car d

Drive 2

LUN 16

Drive 2 LUN 17

LUN 18

Drive 2

LUN 19

Drive 2 LUN 20

Drive 2 LUN 21

LUN 22

Drive 2 LUN 13

Drive 2

LUN 14

Drive 2 LUN 15

Drive 2

LUN 20

LUN 19 Drive 1 LUN 12

Drive 2

LUN 17

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

Floor Model Drive Sleds

Management Card

Disk 1 LUN 1

Disk 1 LUN 2

Disk 1 LUN 3

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Disk 1 LUN 4

Disk 1 LUN 5

Disk 1 LUN 6

Disk 1 LUN 7

Disk 1 LUN 8

Disk 1 LUN 9

Disk 1

LUN 10

Disk 1

LUN 11

Disk 1

LUN 12

Drive 1 LUN 11

Drive 1

LUN 12

Drive 2 LUN 23

Drive 2 LUN 24

Examples

JBOD

mode uses 24 LUNs. Each LUN is equivalent to one physical drive. The array will operate with several sleds missing, but the empty sled(s) will be treated as offline and will not be configurable.

RAID Level 0 (

QuickRAID0 [n]

)

With a fully populated array, RAID 0 may be configured as 1, 2, 3, 4, 6, or 12 LUNs. As

QuickRAID0 1

, all 24 physical drives are configured as a single stripe group or LUN. You may also configure two LUNs of 12 drives each, three LUNs of eight drives each, four LUNs of six drives each, six LUNs of four drives each and 12 LUNS of two drives each. (See Exhibit 6.2-3 in

RAID Level 0

on page 53). The command assumes there are 24 drives available to configure the specified number of LUNS.

configured as six mirrored pairs, then striped into two LUNs.

RAID Level 5 (

QuickRAID5

)

with a fully populated Array may be configured as 1, 2, 3 or 4 LUNs. As

QuickRAID5 1

, all 24 physical drives are configured as a single RAID 5 LUN. You may also configure two LUNs of 12 drives each, three LUNs of eight drives each, or four LUNs of six drives each. RAID Level 5 parity will reduce the usable capacity of each LUN by the equivalent of one drive sled.

Unique serial number for each LUN

Each LUN in a system has a unique 24-character serial number which is updated when a system configuration changes. It is stored in a Device Association Table on each drive.

RAID Level 1 (

array will be configured into six groups, LUNs 16, with each LUN containing two physical sleds; each sled in the LUN will be a mirror image of the other sled in the LUN.

RAID Level 10 (

command first creates six mirrored groups, then stripes them into groups of one, two or three RAID 10 groups. Each group is a LUN. Issuing

QuickRAID10 2

Character

Placement

0-19 Any ASCII

20 1 for Raid1

21 A – Z LUN Configuration Iteration Character: starts at A and increments every time a member drive is

22 A – Z Generation Character: starts at A and increments every time a new configuration is stamped on the

23 0 Reserved for future use

QuickRAID1

QuickRAID10

)

has no options: the

)

The

QuickRAID10

, the 12 physical sleds are

Valid values

char

A for Raid10

0 for Raid0 5 for Raid5

X for JBOD

Example of unique serial number for a JBOD configuration:

Drive Serial Number

RAID Configuration Character

replaced in a RAID Group. Reverts back to A any time the Generation Character is updated

system. When it reaches Z, rolls over to

When a drive or sled is replaced in

QuickRAID

configurations that allow for hot swap, a new serial number will be computed and will in turn be stamped onto all member drives of the RAID group. The CLI command ExpressNAV interface page

LUNInfo

or the

Logical Units will

display the serial number for each drive.

Format of the 24 characters is defined in the chart below:

1231231231231231231XAC0

Numbering conventions

5.0 Accessing the Array

Communicate with the Diamond Storage Array through the Fibre Channel link, the RS-232 port or the Ethernet port using Command Line Interface commands.

You may configure and tune the Diamond Storage Array for different environments and applications, update the firmware, monitor internal power and temperature status, report on hardware diagnostics and log failures.

Three avenues are available

• In-band SCSI over Fibre Channel and over SCSI

• RS-232 port

• Telnet and SNMP over Ethernet

The rest of this manual describes how to access the array and use the Command Line Interface or the ExpressNAV interface.

Command Line Interface

The Command Line Interface provides access to the array through ASCII command lines.

An initial display, after powering up the unit or restarting the firmware, will contain the information in Exhibit 5.0-1. Once the initial display is complete, with the word

Ready

you are

in the Command Line Interface mode.

• In-band CLI over SCSI, where ASCII CLI

commands, may be issued via CLI commands except

• ID/value, where the application program uses a

SCSI CDB (command descriptor block) to select the buffer ID of the configuration parameters to be affected, and the new value of the parameter. Most configuration options are available.

Zmodem

Write Buffer.

are supported.

All

RS-232 port

The array provides remote service operations over the RS-232 serial port using standard terminal emulation software available with most systems.

You will need to set the following serial parameters in your terminal program

• Bits per second: 115200

• Data Bits: 8

• Parity: None

• Stop Bits: 1

• Flow Control: None.

• Terminal type: ASCII

• Echo: on

Help

Type available.

ATTO ExpressNAV

ATTO ExpressNAV is an integrated configuration tool accessible through an Ethernet connection. Platform independent, ExpressNAV contains all the current capabilities of the CLI in a userfriendly GUI console. A menu on each page provides access to information and configuration operations.

Refer to on page 25 for more information.

to display a list of all commands

ATTO ExpressNAV: Browser-based Interface

Ethernet port

The 10/100 BaseT Ethernet port provides Telnetor SNMP-based monitoring and management.

The default IP address is 10.0.0.1; the default subnet mask is 255.255.255.0. These parameters can be changed using the RS-232 port or Telnet. The array should first be configured for the network using the RS-232 port to establish the correct IP address.

SNMP, or Simple Network Management Protocol, is an application layer protocol that allows network devices to exchange management information. Through a combination of standard

In-band SCSI over Fibre Channel

In-band SCSI commands (

Buffer

) may be issued to the array Fibre Channel

Write Buffer

and

Read

port to manage configuration via two mechanisms

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

and custom MIBs (Management Information Base), the array provides status and error indications to an SNMP server, allowing the array to be managed with other devices in a complex

system through a common interface. You use CLI commands or the

SNMP

page of the ExpressNAV interface to configure up to six unique Trap addresses. A trap is a way for the array to tell the SNMP server that something has happened.

SNMP

SNMP, or Simple Network Management Protocol, is an application layer protocol that allows network devices to exchange management

Exhibit 5.0-1 An example of the POST information on screen after boot

Firmware version 5.40 release date Mar 30 2004, 10:43:06 Build 021G

Power-On Self-Test Completion Status: GOOD 64 Megabytes of RAM Installed.

Interface Port 0 is not installed. Interface Port 1 is 1.0624 Gb/s Fibre Channel.

information. Through a combination of standard and custom MIBs, the array provides status and error indications to an SNMP server, allowing the array to be managed with other devices in a complex system through a common interface.

You use CLI commands or the ExpressNAV interface to configure up to six unique Trap addresses. A trap is a way for the array to tell the SNMP server that something has happened.

SNMP

page of the

Interface 0 World Wide Name = 20 00 00 10 86 10 02 DC Interface 1 World Wide Name = 20 00 00 10 86 10 02 DC

Diamond Array Serial Number = "MIDP100197" Diamond Array Name = " " System Serial Number = "" Active Configuration = ATTO DiamondClass = (S)83 Internal Temperature = 26 C [5 - 47] ErrorLog Contents: NO ERRORS For help, type HELP.

Ready.

Interfaces

5.1 In-band CLI Over the Fibre Channel

In-band Command Line Interface (CLI) configures and manages the Diamond Storage Array using SCSI-based CLI commands over a Fibre Channel port connection.

In-band CLI allows a programmer to configure the Diamond Storage Array while it is moving data. Using a programmer’s interface, CLI commands as described previously in this manual may be implemented.

In-band CLI is implemented as part of LUN 0. It uses a different LUN than the array, and reports as a Storage Enclosure Services (SES) device (device type 0x0D).

LUN 0 is visible on all Fibre ports but is actually a single unit. The default value for LUN 0 is 0x00.

LUN 0 must be reserved for each Write Buffer/Read Buffer pair, using the SCSI Reserve command to insure integrity of the in-band CLI session.

1 An initiator (host) sends a SCSI Reserve

command to LUN 0.

• If LUN 0 is not reserved by another initiator, LUN 0 is now reserved and ready to begin a new CLI session.

• If the array configuration is reserved by a different CLI session (i.e. serial or Telnet), the in-band session will not be allowed to modify the array configuration. If you try, the results buffer of LUN 0 will return

Process X has the configuration reserved. ID of this session = Y Ready.

2 The initiator issues a SCSI Write Buffer

command to LUN 0. A must be accompanied by an ASCII buffer representing the CLI command string such as

set DiamondName Omega1

3 LUN 0 will execute the command line and

create feedback in the form of ASCII characters into a buffer. This buffer is 8 KB and circular. Retrieve the results by issuing a Read Buffer command before issuing another Write Buffer command.

Write Buffer

command

4 A subsequent Write Buffer command will

execute the new command line and overwrite the previous results in the buffer with new results.

5 LUN 0 can be released by issuing a SCSI

Release command to the LUN after each Write/Read Buffer Buffer pairs.

Initiator (Host) Diamond Array

Reserve LUN 0 return: “ok”

Write Buffer LUN 0 bid ‘AA’ “get Temperature”

Read Buffer LUN 0 bid ‘AA’

Release LUN 0 return: “ok”

pair, or multiple

executes the CLI command, stores output in buffer

return: ”Temperature=28C\r\n\Ready .\r\n\0”

Write/Read

I/O details

The buffer sent to the Services LUN during the data out phase of a Write Buffer command must be

• ASCII data

• maximum 80 bytes length

• terminated with either a carriage return character (0x0D), line feed character (0x0A) or NULL character (0x00)

• Characters following the first carriage return character, line feed character or NULL character are ignored.

The buffer retrieved from the Services LUN during the data-in phase of a Read Buffer command will be

• ASCII data

• 8 KBytes (8192 bytes) in length

• terminated with a NULL character (0x00)

• Characters following the NULL character are meaningless.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

CHECK_CONDITION,

INVALID_PARAMETER_IN_CDB

will be returned

to an initiator that specifies an incorrect Buffer

Exhibit 5.1-1 The SCSI command process: reserve the Diamond Storage Array, send the command, release the Diamond Storage Array.

Initiator/Host Diamond Storage Array

Goal

: reserve the Diamond Storage Array for an in-band CLI command

SCSI cdb: Reserve LUN 0 =>

Goal

: retrieve the Diamond Storage Array temperature via in-band CLI

1. Issue the command:

SCSI cdb: WriteBuffer LUN 0, bid=’AA’, “get Temperature\n” => places “Temperature=28C\n\r” into

2. Retrieve the results:

SCSI cdb: ReadBuffer LUN 0, bid=’AA’ =>

Goal

: release the Diamond Storage Array for other in-band users

SCSI cdb: Release LUN 0 =>

ID, Mode, Length or Buffer Offset. The Mode is always Data (0x2), the Buffer ID is always 0 and the Buffer Offset is always 0.

<= SCSI success

the read-data buffer

<= SCSI success

<= Returns “Temperature=28C\n\r”

from the read-data buffer

<= SCSI success

In-band CLI

5.2 Serial Port Access

The Diamond Storage Array provides remote service operations over the RS-232 serial port using standard terminal emulation software available with most systems.

The Diamond Storage Array supports service operations over the RS-232 serial port using standard terminal emulation software available with most systems.

1 Connect a DB-9 null modem serial cable

between the port marked RS-232 on the front of the array and one of the computer’s serial ports. A gender changer or DB-9 to DB-25 converter may be needed depending on the cables you are using. The cable must be no longer than three meters

2 Boot the computer terminal or terminal

emulator.

3 Start a terminal emulator program such as

Windows HyperTerminal. Set the emulator to use the COM port with cable attached, then the following settings:

• 115200 baud

• 8 bit ASCII

• no parity

• ASCII terminal type

• 1 stop bit

• flow control none

• echo typed characters locally

4 Turn on the array. Upon successful power on

and diagnostics, the unit should display the POST (power on self test) information found in Exhibit 5.0-1.

5 You should now be in the CLI mode. To see a

list of available commands, type help at the Ready prompt or refer to this manual’s Index.

6 Use the CLI to configure the unit as a JBOD,

RAID Level 0, RAID Level 1, RAID Level 10 or RAID Level 5 array with or without hot spare sleds as described later in this manual.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Serial port access

5.3 Ethernet Access: Telnet and SNMP Protocols

The optional 10/100 BaseT Ethernet port provides Telnet- or SNMP-based monitoring and management.

The 10/100 Base T Ethernet management services card provides Telnet-based monitoring and management, including firmware update using FTP.

Remote system monitoring is also available using Simple Network Management Protocol (SNMP). An agent resides in the array which takes information from the Array and translates it into a form compatible with SNMP. If certain conditions arise, the agent sends asynchronous notifications (traps) to a client.

To connect to the Ethernet port

1 Connect a cross-over cable (for a direct

connection to a PC) or regular network cable from a network device to the optional RJ45 Ethernet port on the Ethernet management card on the front of the array.

2 If using a direct connection, power on and boot

up the host computer.

3 Attach a DB-9 null modem serial cable (the

cable must be no longer than three meters) from the RS-232 port of the array to a host computer and open a terminal emulator program on the host to set the Ethernet parameters. Turn on the array.

4 First time use: Upon successful power up and

diagnostics, set the host computer with the appropriate settings.

The host computer must have appropriate network settings to allow it to communicate with the array. Please see your system administrator for more information.

• Port type: telnet

• Terminal type:

3 If you make any changes to the network setting

on the Diamond Array, use the SaveConfiguration and FirmwareRestart commands.

4 Username/password: You will be prompted for

a username and password, up to eight characters each, case insensitive. Only one username/password combination is available per array.

• You may change the telnet username and/or password after entering a CLI session using the commands

set TelnetUsername [username] set TelnetPassword [password]

• Or change the telnet username and/or password using the ExpressNAV interface Configuration page.

• RestoreConfiguration default sets the telnet username and password to the default values.

• The username default is telnet.and password default is diamond.

5 You should now be in the CLI. To see a list of

available commands, type help at the prompt or refer to this manual’s Index. Or, access the ExpressNAV interface (refer to

ATTO ExpressNAV: Browser-based Interface

page 25).

6 Configure the unit as a JBOD, RAID Level 0,

RAID Level 1, RAID Level 10 or RAID Level 5 array with or without hot spare sleds as described later in the manual.

vt100

Ready

To use Telnet

1 If you have not already done so, change the IP

address from the default by first accessing the serial connection and changing it using the CLI.

Change the IP address to a network specific value or, if the local network uses DHCP, you may enable automatic IP addressing (set IPDHCP enabled) using the CLI.

2 Open a Telnet session on the host computer.

• IP address

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

To use SNMP

1 Enter the CLI through the serial port or

Ethernet.

2 Change the IP address to a network specific

value or, if the local network uses DHCP, you may enable automatic IP addressing.

3 Set the number of trap client addresses by

typing

set SNMPTrapAddress [1-6] [IPAddress] [Level]

4Type

set SNMPUpdates enabled

5Type

set SNMPTraps enabled

6Type SaveConfiguration restart to reboot the

array.

7 Install SNMP management software on each

client you wish to receive traps (messages).

8 Call technical support (refer to

Technology, Inc.

appropriate MIB file for your array.

9 For each client, copy the MIB file to the

directory containing the SNMP management software.

10 From within the SNMP management software,

compile the file the software’s procedures.

on page viii) to get the

attodmnd-mib.mib

Contact ATTO

according to

11 Unload any default MIBs.

12 Load the Diamond MIB ATTODIAMOND

13 When requested, enter the array’s IP address

as the Remote SNMP Agent.

14 The SNMP management software will contact

the agent in the array. The screen will reply with system information.

15 Status will be monitored and reported through

the SNMP management software.

Traps are generated for the following situations

• Temperature status changes in any of the sensors located on the array mid-plane.

• A drive or a sled is physically removed from the array or put into the array.

• The power supply is turned on or off.

• The fan is physically stopped.

Ethernet access

5.4 ATTO ExpressNAV: Browser-based Interface

The easiest way to communicate with the Diamond Storage Array is to use ATTO ExpressNAV, a userfriendly GUI interface accessed through a browser to control the most common capabilities of the array.

Access ATTO ExpressNAV from any browser that supports the latest standards for XHTML 1.0 and CSS1. To take full advantage of the ExpressNAV interface you should have Java script enabled through your browser.

Browser compatibility

All pages are written in pure XHTML 1.0 and CSS1 to be compatible with the latest versions of Internet Explorer, Netscape, Mozilla (including K-Meleon, Camino, Mozilla Firefox, Epiphany and Galeon), and KHTML (including Konqueror and Safari).

Minimum requirement is for Internet Explorer 5.5 and Netscape 6.2.

To optimize ExpressNAV in Internet Explorer

1 Go to the browser toolbar and click on Tools

2Click on Internet Options

3 Click on the Security tab

4 Click on the Custom Level button.

5Click on Microsoft VM, Java permissions

6 Ensure Disable Java is not selected.

7 Click on the Miscellaneous tab.

8Click on Metarefresh

Opening an ExpressNAV session

1 Obtain the IP address of the array.

2 Type the IP address of the array into the

browser address window.

3 The ExpressNAV interface splash screen is

displayed. Click on Enter.

4 Enter the username and password set

previously in

Interface

The

•

• The default password is Password

5 The product faceplate display appears. Click

the component you want to manage on the lefthand side menu or go to the Advanced screen to use the CLI.

ATTO ExpressNAV: Browser-based

on page 25.

default username is Diamond

Navigating ExpressNAV

All pages are accessible by clicking on their titles on the left side of the page. You may also go back one page or go to the

Home

page via the titles on

the left side of the page.

Clicking on any of the red option names will bring up a help window. After making changes on a page, click the

Submit button at the bottom.

Clicking this button is equivalent to typing in all the CLI commands and then the command

saveconfiguration norestart.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Exhibit 5.4-1 Atypical page in the ATTO ExpressNAV configuration tool.

Sidebar:

select the

item you

wish to

view

Information

Parameter name

Red print links to

help text

Configure choices: red type links to

another page to change configuration

options unavailable

because of previous

Submit button

same as typing all

CLI commands and

saveconfiguration

Reset button

return to previous

setting without

making any changes

norestart

choice

ExpressNAV

5.4.1 ExpressNAV Pages

Each page in the ATTO ExpressNAV interface provides information and/or configuration parameters based on a specific topics. Each page can be reached through the menu on the left hand side of each page.

Status

Contains general array information.

• Unit Information including Vendor ID, Product ID, Firmware Revision and Serial Number

• Environmental Information

• World Wide Identifiers including Node Names and Port Names

• Host Interface Card Status

• Fan Status

• Power Supply Status

• Storage Status including ATA Disk Errors and Logical Unit Conflicts

Ethernet

Configures the Ethernet port.

• IP Address

• IP Gateway

• IP Subnet Mask

• Ethernet Speed

• Use DHCP

Refer to on page 23 and

Ethernet Access: Telnet and SNMP Protocols

Ethernet Commands

on page 35 for

details.

SNMP

Remote system monitoring is available using Simple Network Management Protocol (SNMP).

• Updates

• Traps

• Trap Recipient IP Addresses

Refer to

Access: Telnet and SNMP Protocols

Ethernet Commands

on page 35 and

on page 23 for

Ethernet

details on each parameter.

Serial Port

Contains the necessary information to configure the serial port including baud rate and echo.

Refer to

Configuration Commands

Serial Port Access

on page 21 and

Serial Port

on page 39 for more

information on each parameter.

Fibre Channel

Contains parameters and information to manage the Fibre Channel port.

• Data rate

• Full duplex mode

• Connection mode

Refer to and

Connecting a Fibre Channel Array

Fibre Channel Configuration Commands

on page 11

on page

37 for more information.

Storage Management

Shows information about the drives and their status. Information includes

• Sled Number

• Capacity of each sled

• Number of errors

• Configuration type (JBOD, RAID5, etc.)

• Virtual ID

You may place sleds on or off line by selecting or de-selecting a check box.

Refer to page 15 and

Determining Drive and Sled Designations

Drive Configuration Commands

on page

45 for more information.

Once you open this page, several other configuration pages are available on the menu on the left hand side of the page.

RAID

Contains the necessary information and parameters to configure RAID groups. Information provided includes

• Type of configuration (RAID5, JBOD, etc.)

• Virtual Disk ID

• State of sleds

• Capacity of sleds

• Interleave values

You may change the configuration of these parameters

• Configuration Type (JBOD, RAID5)

• Number of Groups (when applicable)

• Interleave

• Striping method (sled or drive)

• Rebuild priority

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Refer to

Configuration Commands

Configuring Drives

information.

on page 49 and

Drive

on page 45 for more

Advanced

Allows you to input any CLI command available through the array.

Clear Data

Allows you to

• view the status of any Clear Data commands in progress

• view the rebuild state of each sled

• initialize a RAID Level 5 Clear Data command

• set a rebuild state for each sled

• change the RAID Interleave.parameter

• Enable/disable AutoRebuild

Refer to

Drive Configuration Commands

on page 45

for more information.

Logical Units

Gives information on the logical units which have been configured on the array and allows you to change the status from online or offline or degraded.

Refer to page 15 and

Determining Drive and Sled Designations

Drive Configuration Commands

on page

45 for more information.

Rebuild

Shows the current status of rebuilds on the array and allows you to halt, resume or initiate rebuilds.

To use the advanced page CLI commands

1 After the page opens and the Ready prompt

appears, type in the CLI command

2 Click the Submit button: this is equivalent to

typing the CLI command into a telnet or serial port CLI session.

A text field beneath the box will list the most recent commands issued to the array through this page.

3Type saveconfiguration norestart

4 Click the Submit button. Your changes will be

implemented.

5 To keep the changes through the next power

cycle, type firmwarerestart or go to the Restart page and click Restart.

Restart

Allows you to implement a firmware restart of the array. Access is via the

Restart

link on the left

side of the page.

Note

Restarting the firmware may take a few minutes.

Refer to and

Drive Configuration Commands

Rebuilding RAID Configurations

on page 45

on page 65for

more information.

Configuration

Contains the necessary information to manage the array. Configurable options are

• User name

• Password (old password, new password, confirm password)

• Minimum operating temperature

• Maximum operating temperature

• Identify Diamond

• Restore defaults

Refer to

Maintenance Commands

on page 43 for

details.

1 Click the Restart button.

A box displays a message to wait until the counter gets to 0 and then the browser will refresh.

2 If the browser does not refresh after the counter

gets to 0, click the link to refresh it manually.

Help

Gives help information about CLI commands and troubleshooting tips.

Provides links to pages with help text for each category of options and one link to the

Troubleshooting Tips

and

FAQs

page on the ATTO website. Contact information for getting in touch with ATTO technical support is on the right.

When you click a red text box on any page, ExpressNAV asks for your login information, then a dialog box opens with help text.

ExpressNAV

5.5 Command Line Interface Use and Guidance

The command line interface (CLI) provides access to Diamond Storage Array services through a set of ASCII-based commands. CLI commands may be entered while in CLI mode or through the Advanced CLI page in the ExpressNAV interface.

• CLI commands are context sensitive and generally follow a standard format

[Get | Set] Command [Parameter 1 | Parameter 2]

followed by the

• CLI commands are not case sensitive: you may type all upper or all lower case or a mixture, no matter what the definition either in Help pages states. Upper and lower case in this manual and the only.

• Commands generally have three types of operation: get, set and immediate as summarized below.

The

get

form returns the value of a parameter or

return

help

enter

key

or these

screen are for clarification

setting and is an informational command.

Responses to get commands are specified in the

Results

field for each command, followed by

Ready.

The

set

form is an action that changes the value of

a parameter or configuration setting. It may

require a

SaveConfiguration

command and a system restart before changes are implemented. The restart can be accomplished as part of the

SaveConfiguration FirmwareRestart

command or by using a separate

command. A number of set

commands may be issued before the

SaveConfiguration

command.

Responses to set commands are either an error message or

SaveConfiguration

Set

commands which do not require a

Ready. *,

which indicates a

command is required.

SaveConfiguration command, defined as Immediate

commands, are immediately

implemented.

Responses to Immediate commands are either an error message or

Ready.

Symbols, typefaces and abbreviations used to indicate functions and elements of the command line interface used in this manual are listed below.

Command Line Interface actions and responses

Set commands configure the Diamond Storage Array and display what you have changed after

completing the task. Commands which require a implementation will return Ready command are immediately implemented.

Get commands display information about the configuration of the Diamond Storage Array. Responses to get commands are followed by Ready

Screen messages, also called returns, may be either terse, with just the current information, or verbose, with labels and the current information. Default is verbose.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

. *

. Set commands which do not require a SaveConfiguration

SaveConfiguration

command to complete their

Symbols, typefaces and abbreviations used to indicate functions and elements of the CLI

Symbol Indicates

[ ] Required entry

< > Optional entry

| pick one of

… Ellipses, repetition of preceding item

\n end of line

- a range (6 – 9 = 6, 7, 8, 9)

Boldface words must be typed as they appear

Italicized words Arguments which must be replaced by whatever they represent

fp Fibre Channel port number (0 <= fp <= 1)

fl Fibre Channel LUN (logical unit number) (0 <= fl <= 24), where 0 represents the

Diamond Storage Array unit, and 1-24 represent the disk drives.

sb SCSI bus number (0<= sb <= 3)

sl SCSI LUN (logical unit number) (0 <= sl <= 7)

st SCSI target ID (0 <= st <= 15)

Command Line Interface

5.5.1 Summary of CLI Commands

A summary of the Command Line Interface commands, their defaults, an example of how they might be used, and where you can find the specifics of the command. Commands which have no default values associated with them have a blank entry in that column of the table.

Command Defaults Example Page

AtaDiskState Online Set AtaDiskState 6 1 offline 45

AudibleAlarm Disabled Set AudibleAlarm disabled 41

AutoRebuild Disabled Set AutoRebuild enabled 45

ClearDiskReservedArea ClearDiskReservedArea 8 2 45

DiamondModel Diamond Get DiamondModel 41

DiamondName “

DriveCopy DriveCopy 1 1 2 2 45

DriveCopyHalt DriveCopyHalt 2 2 45

DriveCopyResume DriveCopyResume 2 2 45

DriveCopyStatus DriveCopyStatus 41

DriveInfo DriveInfo 3 2 41, 45

DriveSledPower On Set DriveSledPower 9 1 off 45

DriveWipe DriveWipe 2 2 46

EthernetSpeed Auto Set EthernetSpeed 100 35

FcConnMode Loop Get FcConnMode 37

FcDataRate Auto Set FcDataRate 2 gigabit 37

FcFairArb Enabled Get FcFairArb 37

FcFrameLength 2048 Get FcFrameLength 37

FcFullDuplex Enabled Set FcFullDuplex enabled 37

FcHard Disabled Get FcHard 37

FcHardAddress 0x03, 0x04 Get FcHardAddress 0 37

FcPortInfo Get FcPortInfo 37

FcPortList FcPortList 41, 38

FcPortName Get FcPortName 1 41

FcWWName Get FcWWName 0 38

FirmwareRestart FirmwareRestart 33, 43

FTPPassword diamond Set FTPPassword barbw52 35

Help Help DriveInfo 33

IdentifyDiamond Disabled Get IdentifyDiamond 41

IdeTransferRate 4 Set IdeTransferRate 4 46

Info Info 41

IPAddress 10.0.0.1 Set IPAddress 10.0.0.02 35

IPDHCP Disabled Set IPDHCP enabled 35

IPGateway 0.0.0.0 Set IPGateway 200.10.22.3 35

...............

” Get DiamondName 41

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Command Defaults Example Page

IPSubnetMask 255.255.255.0 Set IPSubnetMask 255.255.255.0 35

LUNInfo LUNInfo 1 41, 46

LUNState Online Get LunState 1 46

MaxEnclTempAlrm 47 Get MaxEnclTempAlrm 43

MinEnclTempAlrm 5 Set MinEnclTempAlrm 10 43

PowerAudibleAlarm Enabled Set PowerAudibleAlarm disabled 43

QuickRAID0 sled Set QuickRAID0 6 46, 53

QuickRAID1 Set QuickRAID1 45, 57

QuickRAID10 Set QuickRAID10 2 46, 63

QuickRAID5 Set QuickRAID5 4 46

RAID5ClearData RAID5ClearData 47

RAID5ClearDataStatus RAID5ClearDataStatus 47

RAIDHaltRebuild RAIDHaltRebuild 3 47

RAIDInterleave 128 Get RAIDInterleave 47, 65

RAIDManualRebuild RAIDManualRebuild 2 3 47

RAIDRebuildState set RAIDRebuildState 2 OK 47

RAIDRebuildStatus get RAIDRebuildStatus 47

RAIDResumeRebuild RAIDResumeRebuild 5 47

RestoreConfiguration RestoreConfiguration default 33

RestoreModePages RestoreModePages 47

SaveConfiguration SaveConfiguration restart 33

SerialNumber Get SerialNumber 33

SerialPortBaudRate 115200 Set SerialPortBaudRate 9600 39

SerialPortEcho Disabled Get SerialPortEcho 39

SerialPortHandshake None Get SerialPortHandshake 39

SerialPortStopBits 1 Set SerialPortStopBits 1 39

SledFaultLED Set SledFaultLED 9 on 42, 48

SMARTData SMARTData 2 1 41

SNMPTrapAddress 10.0.0.1 Set snmptrapaddress 1

255.555.555.555 All

SNMPTraps Disabled Get snmptraps 35

SNMPUpdates Disabled Get snmpupdates 35

TelnetPassword diamond Set TelnetPassword 123ABC 36

TelnetTimeout Disabled Set TelnetTimeout 360 36

TelnetUsername telnet Set TelnetUsername diamond1 36

Temperature Get Temperature 43

VerboseMode Enabled Get VerboseMode 33

Zmodem Zmodem receive 43

5.5.2 General Use Commands

The following commands, listed alphabetically, describe or perform general functions.

FirmwareRestart

Causes a warm restart of the Diamond Storage Array.

Immediate: FirmwareRestart

Help

Displays a list of available commands. Type

help

followed by a command name to display detailed command-specific information.

Immediate: Help [Command Name]

RestoreConfiguration

Restores configuration to either the default configuration or the configuration last saved into non-volatile memory. The saved option will undo any changes made since the last save.

Immediate: RestoreConfiguration [Default|Saved]

SaveConfiguration

If the restart option is selected, the Diamond will cycle its power. The norestart option will save changes without restarting. Please note: certain modifications require a SaveConfiguration command and a system restart. If required, the return

Ready

. * will be displayed after the

return for the modification. You may make several changes through commands and SaveConfiguration before implementing a restart, but once you have restarted the Diamond, all the command changes created before the restart and saved will be implemented. Restart or no Restart parameter is optional

Immediate: SaveConfiguration <Restart| NoRestart>

SystemSN

Stores the Diamond Storage Array serial number. The serial number may be one to 16 characters.

Set syntax: set SystemSN [n] Get syntax: get SystemSN Requires a SaveConfiguration command

VerboseMode

Specifies the detail of verbal feedback for the Command Line Interface. Disabling this option removes parameter names from ‘get’ commands and removes descriptions from ‘help’ commands.

Default: enabled (returns have parameter information) Set syntax: set VerboseMode [enabled | disabled] Get syntax: get VerboseMode

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

General CLI commands

5.5.3 Ethernet Commands

Ethernet configuration commands configure the Ethernet and TCP/IP parameters for the Diamond Storage Array with an optional Ethernet management services card.

EthernetSpeed

Specifies the speed of the Ethernet Network to which the Diamond Storage Array is connected.

Choices are: [10|100 | Auto] where 10= 10 baseT, 100= 100 baseT Default: auto Set syntax: set EthernetSpeed [10 | 100| Auto] Requires a SaveConfiguration command Get syntax: get EthernetSpeed

FTPPassword

Specifies a password of up to 32 characters for an FTP session.

Default: diamond Set syntax: set FTPPassword Requires a SaveConfiguration Restart command

IPAddress

Specifies the IP Address of the Diamond Storage Array on the Ethernet network. If DHCP is enabled, the assigned address of the Diamond will be displayed. Setting this value always modifies the internal NVRAM value of the IP address.If IPDHCP is enabled (see below), get command reports current IP address assigned by DHCP server

Default IP Address: 10.0.0.1 Set syntax: set IPAddress xxx.xxx.xxx.xxx Requires a SaveConfiguration command Get syntax: get IPAddress

above), get command reports current IP gateway assigned by DHCP server. Must conform to AAA.BBB.CCC.DDD standard network IP addressing.

Default: 0.0.0.0 Set syntax: set IPGateway AAA.BBB.CCC.DDD Requires a SaveConfiguration command Get syntax: get IPGateway

IPSubnetMask

Specifies the IP Subnet Mask for the Diamond Storage Array on the Ethernet network. If DHCP is enabled, the assigned subnet mask for the unit will be displayed. Setting this value always modifies the internal NVRAM value of the IP Subnet Mask.If IPDHCP is enabled (see above), get command reports current subnet mask assigned by DHCP server.

Default: 255.255.255.0 Set syntax: set IPSubnetMask AAA.BBB.CCC.DDD Requires a SaveConfiguration command Get syntax: get IPSubnetMask

SNMPTrapAddress

Sets up IP trap address and trap level.

Default: 10.0.0.1 Set syntax: set SNMPTrapAddress [Index:1-6] [Address: XXX.XXX.XXX.XXX] [Critical | Warning | All |

None]....................................

Requires a SaveConfiguration Restart command Get syntax: get SNMPTrapAddress

IPDHCP

Selecting DHCP allow the Diamond Storage Array to request an IP address from the network. It requires that the Diamond be attached to a network with at least one DHCP server. The network must have at least one DHCP server.

Default: disabled Set syntax: set IPDHCP [enabled | disabled] Requires a SaveConfiguration command Get syntax: get IPDHCP

IPGateway

Specifies the IP Gateway for the Diamond Storage Array on the Ethernet network. If IPDHCP is enabled (see

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

SNMPTraps

Enables or disables SNMP traps.

Default: disabled Set syntax: set SNMPTraps [enabled|disabled] Requires a SaveConfiguration Restart command Get syntax: get SNMPTraps

SNMPUpdates

Enables or disables the SNMP Management Information Base (MIB) database.

Default: disabled Set syntax: set SNMPTraps [enabled|disabled] Requires a SaveConfiguration Restart command Get syntax: get SNMPUpdates

TelnetPassword

Specifies password for telnet session. Only one username/password combination is available per Diamond Storage Array. RestoreConfiguration default sets the telnet username and password to the default values. The password is case insensitive, with 1 to 8 characters.

Default: diamond Set syntax: set TelnetPassword [password] Requires a SaveConfiguration Restart command

TelnetTimeout

Specifies the number of minutes of inactivity which elapses before a telnet session automatically times out.

Choices: 1-1440 minutes Default: disabled Set syntax: set TelnetTimeout [1-1440 | disabled ] Requires a SaveConfiguration Restart command Get syntax: get TelnetTimout

TelnetUsername

Specifies username for telnet session. Only one username/password combination is available per Diamond Storage Array. RestoreConfiguration default sets the telnet username and password to the default values.Username is case insensitive, 1 to 8 characters

Default: telnet Set syntax: set TelnetUsername [username] Requires a SaveConfiguration Restart command

The Telnet and SNMP protocols also use CLI commands.

Ethernet CLI commands

5.5.4 Fibre Channel Configuration Commands

The Fibre Channel ports are configured with default settings but may be customized to your specifications using the CLI commands in this section.

FcConnMode

Specifies the Fibre Channel topology for the Diamond Storage Array. Options are loop only (loop), point-topoint only (ptp), loop preferred (loop-ptp) or point-topoint preferred (ptp-loop). Refer to

Channel Array

on page 11 for more information on Fibre

Connecting a Fibre

Channel topology. Applies to both Host Interface Cards.

Default: loop Set syntax: set FcConnMode [loop | ptp| loop-ptp| ptploop] Requires a SaveConfiguration command Get syntax: get FcConnMode

FcDataRate

Specifies the Fibre Channel data rate at which the Diamond will operate. Applies to both Host Interface Cards.

Default: auto Set syntax: set FcDataRate [1gb | 2gb | auto] Get syntax: get FcDataRate Requires a SaveConfiguration command

FcFairArb

Enabling this feature causes the Diamond Storage Array to follow the arbitration fairness rules on the FC-AL. Applies to both Host Interface Cards.

Default: enabled, enabling arbitration fairness Set syntax: set FcFairArb [enabled | disabled] Get syntax: get FcFairArb Requires a SaveConfiguration command

FcFrameLength

Sets the frame length of a command. If not specified in the set command, current frame length is displayed. Applies to both Host Interface Cards.

Default: 2048 Set syntax: set FcFrameLength [512 | 1024 | 2048] Get syntax: get FcFrameLength Requires a SaveConfiguration command

FcFullDuplex

Enable to allow full duplex Fibre Channel communication between the Diamond Storage Array and host devices. Disable FcFullDuplex causes half duplex mode.Applies to both Host Interface Cards.

Default: enabled Set syntax: set FcFullDuplex [enabled | disabled] Requires a SaveConfiguration command Get syntax: get FcFullDuplex

FcHard

If hard addresses are enabled, the Diamond Storage Array will try to use its internal hard address as its address on the loop. Under soft addressing, the Diamond Storage Array loop address is assigned during loop initialization. Use FcHardAddress (described below) if you enable hard addressing. Applies to both Host Interface Cards.

Default: disabled Set syntax: set FcHard [enabled | disabled] Get syntax: get FcHard Requires a SaveConfiguration command

FcHardAddress

This hexadecimal value represents the address the Diamond Storage Array will try to use if hard addressing is enabled. When an optional address is not present, the current value is displayed. Each port has individual hard address value.

Default: 0x03 for port 0, 0x04 for port 1 Set syntax: set FcHardAddress [fp |[address]] Get syntax: get FcHardAddress [fp] Requires a SaveConfiguration command

FcNodeName

Returns the Fibre Channel node name stored in NVRAM for this Fibre Channel port: the same as the World Wide Name for port 0.

Get syntax: get FcNodeName

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

FcPortInfo

Retrieves information about the current state of each Fibre Channel port. The status field indicates ‘disabled’ when a port has been internally disabled.

Immediate command: FcPortInfo

FcPortList

Lists the status of all available Fibre Channel ports.

Get syntax: FcPortList: [line count]

FcPortName

Returns the Fibre Channel port name stored in NVRAM for this Fibre Channel port.

Get syntax: get FcPortName [port number]

FcWWName

Sets or view the Word Wide Name (WWN) of the referenced Fibre Channel port. The WWN is a unique 8byte number that identifies the port on a Fibre Channel network. Only the three least significant bytes of the WWN can be modified. Fabric and loop operations are unpredictable if duplicate WWNs are assigned.

Default: 20 00 0x where x is for 0 for port 0, 1 for port 1 Set syntax: set FcWWName [PortNumber [0 | 1] Requires a SaveConfiguration command Get syntax: get FcWWName [PortNumber [0 | 1]

Fibre Channel CLI commands

5.5.5 Serial Port Configuration Commands

The serial port configuration may be customized by using the following commands.

SerialPortBaudRate

Sets the baud rate the Diamond Storage Array will use for its terminal interface. Choices are 2400, 9600, 19200, 38400, 57600, 10.

Default: 115200 Set syntax: set SerialPortBaudRate [2400|9600|19200|3840057600|115200] Requires a SaveConfiguration command Get syntax: get SerialPortBaudRate

SerialPortEcho

Controls whether the Diamond Storage Array echoes characters on its RS-232 port. Local ASCII terminal (or terminal emulator) echo settings should be set to disabled while in serialportecho enabled

Default: disabled Set syntax: set SerialPortEcho [enabled | disabled] Requires a SaveConfiguration command Get syntax: get SerialPortEcho

SerialPortHandshake

Describes which handshaking method the Diamond Storage Array will use for its terminal interface (hardware, Xon/Xoff or none)

Default: none Set syntax: set SerialPortHandshake [hard| Xon| none] Requires a SaveConfiguration command Get syntax: get SerialPortHandshake

SerialPortStopBits

Configures/reports the number of stop bits per character for the Diamond Storage Array RS -232 serial port. The number of data bits per character is fixed at 8 with no parity. Choices are 1 or 2.

Default: 1 Set syntax: set SerialPortStopBits [1 | 2] Requires a SaveConfiguration command Get syntax: get SerialPortStopBits

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Serial port CLI commands

5.5.6 Diagnostic Commands

AudibleAlarm

Enables or disables the audible alarm in the Diamond Storage Array. When enabled, an alarm will sound when the Fault LED on the front panel blinks.

Default: disabled Set syntax: set AudibleAlarm [enabled | disabled] Get syntax: get AudibleAlarm

DiamondModel

Returns specific Diamond Storage Array model information including firmware release and date.

Get syntax: get DiamondModel

DiamondName

Used to identify this Diamond over its Fibre Channel and Ethernet networks. You may customize the name of each Diamond Storage Array enclosure to distinguish it from other units. Maximum eight characters.

Set syntax: set DiamondName [name] Requires a SaveConfiguration command Get syntax: get DiamondName

FcPortList

Lists the status of all available Fibre Channel ports.

Immediate command: FcPortList

FcPortName

Returns the Fibre Channel port name stored in NVRAM for this Fibre Channel port.

Get syntax: get FcPortName [port number]

Help

Displays a list of available commands. Type

help

followed by a command name to display detailed command-specific information.

Immediate command: Help [Command Name]

IdentifyDiamond

Enable this option to identify the current Diamond Storage Array. The fault LED on its front panel will blink. Disable to cancel the ‘blink code.’

Set syntax: set IdentifyDiamond [enabled|disabled] Get syntax: get IdentifyDiamond

DriveCopyStatus

Displays the status of a DriveCopy or DriveWipe operation

Immediate command: DriveCopyStatus

DriveInfo

Displays information about all disk drives or detailed information about a specific disk drive. Detailed information about an individual drive is obtained by supplying a drive identifier. For output examples, refer to

Information Commands Results

Immediate command: DriveInfo [sled ID] [drive ID]

on page ii.

FcNodeName

Returns the Fibre Channel node name stored in NVRAM for this Fibre Channel port: the same as the World Wide Name for port 0.

Get syntax: get FcNodeName

Info

Displays version numbers and other key information about the Diamond Storage Array including data rate, connection mode, WorldWideName, Diamond Storage Array name.

Immediate command: Info [all]

LUNInfo

Displays information about all LUNs (logical unit numbers) or detailed information about a specific LUN. Do not specify a LUN to get information about all LUNs. Specify a LUN to get detailed information about that individual LUN.

Immediate command: LUNInfo [LUN]

.........................................

SerialNumber

View the serial board number, a 10-character field. The first four alphanumeric characters are an abbreviation representing the product name. The remaining six digits are the individual unit’s number.

Get syntax: get SerialNumber

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

SledFaultLED

Changes the state of the selected sled LED to the indicated state. Enter sled number (1-12) (all)

Temperature

Returns the current internal temperature of this Diamond Storage Array in degrees Celsius. The value is read only.

Default: off Set syntax: set SledFaultLED [all|1|2|3|4|5|6|7|8|9|10|11|12] [on| off]|

SMARTData

Displays the current SMART Data (Self-Monitoring Analysis and Reporting Technology) for the specified drive.

Immediate command: SMARTData [Sled#] [Drive#]

Get syntax: get Temperature

VirtualDriveInfo

Displays the named Virtual Drive definitions.

Immediate command: VirtualDriveInfo [active | planned] [Virtual Drive ID]

Diagnostic CLI commands

5.5.7 Maintenance Commands

Maintenance commands allow updating and maintenance of the Diamond Storage Array.

FcScsiBusyStatus

You may set the Diamond to report busy or queue full when it is unable to accept the command.

Default: Busy Set syntax: set FcScsiBusyStatus [busy|qfull]

FirmwareRestart

Causes a warm restart of the Diamond Storage Array.

Immediate command: FirmwareRestart

MaxEnclTempAlrm

Sets/displays the maximum enclosure temperature alarm of the Diamond Storage Array in degrees Celsius. Valid entries are between 5 and 52 degrees and above the current minimum enclosure temperature alarm.

Default: 47 Set syntax: set MaxEnclTempAlrm [5-52] Requires a SaveConfiguration command Get syntax: get MaxEnclTempAlrm

MinEnclTempAlrm

Sets/displays the minimum enclosure temperature alarm of the Diamond Storage Array in degrees Celsius.Valid entries are between 5 and 47 degrees and below the current maximum enclosure temperature alarm

Temperature

Returns the current internal temperature of this Diamond Storage Array in degrees Celsius. The value is read only.

Get syntax: get Temperature

Zmodem

Use the Zmodem protocol to transfer a file to or from the Diamond Storage Array. The filename is required if the

send

option is specified.Available only through the RS-

232 interface

CAUTIONCAUTION

After a firmware image is downloaded to the Diamond Storage Array, the image is placed into flash memory. During this time (about 90 seconds), DO NOT remove power to the Diamond Storage Array or the flash may become corrupted. Power should not be removed until the READY prompt appears.

Immediate command: Zmodem [Send filename | Receive] Requires a FirmwareRestart

Default: 5 Set syntax: set MinEnclTempAlrm [5-47] Requires a SaveConfiguration command Get syntax: get MinEnclTempAlrm

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Maintenance CLI commands

5.5.8 Drive Configuration Commands

The Diamond Storage Array ATA drives may be monitored or configured through the CLI using the commands listed below.

ATADiskState

Sets the ATA disk to the specified state.

CAUTIONCAUTION

In a Hot Spare configuration, a drive sled should only be taken offline if there is absolutely no activity on that drive. If there is any activity, the rebuild of the Hot Spare sled may be flawed.

Choices: enter sled number (1-12), drive number (1-2) and online or offline Default: online Set syntax: set AtaDiskState [sled number] [drive number] [online| offline] Get syntax: get AtaDiskState [sled number] [drive number]

AutoRebuild

If enabled, initiates an automatic rebuild of a “degraded” RAID group when a sled is replaced by a new sled. If disabled, you must manually rebuild the RAID configuration for the new sled by using RAIDManualRebuild.

Default: disabled Set syntax: set AutoRebuild [enabled|disabled] Requires a SaveConfiguration Restart command Get syntax: get AutoRebuild

ClearDiskReservedAreaData

Clears the data in the disk’s reserved area. Restarting the Diamond Storage Array is required for these settings to take effect. Omitting ReservedAreaIndex clears the entire reserved area.

Choices: enter sled number (1-12), drive number (1-2) Immediate command: ClearDiskReservedAreaData [sled number] [drive number]

DriveCopy

Copies a drive from the source disk to the destination disk. Parameters are the sled and drive numbers of the source and destination drives. The destination drive must be offline: use the ATADiskState command to determine if the disks are offline. If you choose the same source and destination drive, this command performs a DriveWipe.

Set syntax: DriveCopy [Source Sled] [Source Drive] [Destination Sled] [Destination Drive] Immediate informational command: DriveCopyStatus

DriveCopyHalt

Stops a DriveCopy operation in progress.

Set syntax: DriveCopyHalt [Destination Sled] [Destination Drive]

DriveCopyResume

Resumes a DriveCopy operation that had been stopped.

Set syntax: DriveCopyResume [Destination Sled] [Destination Drive]

DriveCopyStatus

Displays the status of a DriveCopy, DriveWipe or RAID5ClearData operation

Immediate informational command: DriveCopyStatus

DriveInfo

Information Commands Results

Immediate informational command: DriveInfo [sled ID] [drive ID]

on page ii.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

DriveSledPower

Gets/sets power to the specified drive sled. Sled must be offline

Default: on Set syntax: set DriveSledPower [sled number] [on|off] Get syntax: get DriveSledPower [sled number]

DriveWipe

Initializes a drive: wipes it of all data. Drive must be offline

Set syntax: DriveWipe [Destination Sled] [Dest Drive] Requires a SaveConfiguration Restart command

IdeTransferRate

Sets the DMA mode transfer rate for all devices.

Choices: 0, 1, 2, 3, 4

Default: 4

Set syntax: set IdeTransferRate [0 | 1 | 2 | 3 | 4] Requires a SaveConfiguration Restart command Get syntax: get IdeTransferRate

LUNInfo

Commands Results

Get syntax: get LUNInfo [LUN]

on page ii.

...

Information

LUNState

Sets the LUN to the specified state. May be used to facilitate removal and insertion of sleds and RAID groups during power up/power down of sleds.

Default: online Set syntax: set LUNState [LUN number] [online|offline] Get syntax: get LUNState [LUN number]

QuickRAID0

Specifies the RAID Level 0 configuration for the system. The default 0, or no RAID groups, configures the Diamond Storage Array in a JBOD configuration. DRIVE indicates drives on one side of the array are adjacent members of the same stripe group while SLED indicates the two drives on the same sled are adjacent members of a stripe group.

Choices: 0, 1, 2, 3, 4, 6,12 Set syntax: set QuickRAID0 [0|1|2|3|4|6|12][drive| sled] Requires a SaveConfiguration Restart command

QuickRAID1

Specifies the RAID Level 1 configuration for the system. Sets the system to a mirrored array of spanned drives. Causes the Configuration Manager to “stamp” the new configuration onto the drives to take effect at the next system startup. Setting QuickRAID0 0 removes all RAID configurations and creates a JBOD.

Set syntax: set QuickRAID1 <Number Hot Spare sleds> Requires a SaveConfiguration Restart command

QuickRAID5

Specifies the RAID Level 5 configuration for the system. Sets the system to spanned drives with parity information. Causes the Configuration Manager to “stamp” the new configuration onto the drives to take effect at the next system startup.

Setting QuickRAID5 0 removes all RAID configurations and creates JBOD.

Setting QuickRAID5 ALL creates one group that includes all contiguous sleds (minus Hot Spare sleds if applicable).

To complete RAID Level 5 setup, the RAID5ClearData command must be issued after the Diamond Storage Array has been restarted: DO NOT remove power from the array during this operation.

Choices: 0, 1, 2, 3, 4, all Set syntax: set QuickRAID5 [1 | 2 | 3 | 4 | all] <Number Hot Spare sleds> Requires a SaveConfiguration Restart command Get syntax: DriveInfo

QuickRAID10

Specifies the RAID Level 10 configuration for the system; sets the system to a mirrored array of spanned drives and causes the Configuration Manager to “stamp” the new configuration onto the drives to take effect at the next system startup. Setting QuickRAID10 0 removes all RAID configurations and creates a JBOD.

Choices: 0, 1, 2, 3 Set syntax: set QuickRAID10 [1|2|3] <Number Hot Spare sleds> Requires a SaveConfiguration Restart command

Configure drives with CLI commands

RAID5ClearData

To Zero all drives and parity to make newly created RAID5 groups ‘coherent’. The parameter ALL clears data on all RAID5 groups present in the system. Must be used at initial configuration to ensure parity is valid for all drives in the RAID Level 5 group by setting all data and parity to zero. Takes all LUNs offline automatically, then brings them online. The operation takes 3-6 hours, depending on drive capacity. Do not interrupt this process. This is a destructive operation: all information on these drives is lost. Applies to RAID5 only

Set syntax: RAID5ClearData [ALL | LUN] Get syntax: DriveCopyStatus or RAID5ClearDataStatus

RAIDRebuildState

Sets the RAID Level 1, Level 5 or Level 10 rebuild status of the specified sled to OK, degraded or faulted.

Set syntax:. set RAIDRebuildState [Sled Number] [Degraded | OK | Faulted]

RAIDRebuildStatus

Displays the RAID1, RAID 5 or RAID 10 Rebuild Status Summary. If no RAID groups are defined, the header information is displayed with no data. The status summary contains the RAID1, RAID 5 or RAID 10. The RAIDRebuildStatus command has no effect on the state of any rebuild in progress.

RAID5ClearDataStatus

Displays the status of RAID5 Clear Data processing. S represents the sled number, D represents the drive number. Applies to RAID5 only

Set syntax: RAID5ClearDataStatus

RAIDInterleave

Specifies the Interleave size (in 512 byte blocks) between members of a RAID group. SPAN indicates that the interleave size between drives in the group is the minimum drive size of all members in the group. RAIDInterleave options are 16, 32, 64, 128, 256 blocks and span. Span is not available for RAID Level 5.

Choices all but RAID Level 5: SPAN Choices RAID 5: Default: 128 Set syntax: set RAIDInterleave [1-256] [SPAN] Requires a SaveConfiguration Restart command Get syntax: get RAIDInterleave

16, 32, 64, 128, 256

RAIDHaltRebuild

Stops a RAID Level 1, 5 or 10 rebuild that is in progress.

Immediate command: RAIDHaltRebuild [Sled Number]

RAIDManualRebuild

Initiates a manual rebuild of a RAID Level 1, 5 or 10 LUN. An error message is returned if the specified LUN is not a RAID Level 1, Level 5 or 10 LUN or if the sled number is not available; no rebuild takes place. Applies to RAID Levels 1, 5 and 10 only

Set syntax:. RAIDManualRebuild. [LUN]. [Sled Number]

Member index and the following fields in tabular form: Status, Sled Number, Current LBA, Maximum LBA, Status.

RAIDResumeRebuild

Resumes a RAID Level 1, Level 5 or Level 10 rebuild which had been previously stopped.

Immediate command: RAIDResumeRebuild [Sled Number]

RebuildPriority

Sets the priority of a RAID Level 1, Level 5 or Level 10 rebuild. If you select High priority, rebuild I/O requests are implemented before system I/O requests. I f you select Low priority, rebuild I/O requests executes only when there are no pending I/O requests. If you select Same priority, rebuild I/O and system I/O receive equal consideration.

Set syntax: set RebuildPriority [high | low | same] Requires a SaveConfiguration Restart command Get syntax: get RebuildPriority

ResolveLUNConflicts

Re-numbers any conflicting Logical Unit numbers that exist in the Diamond. Conflicts may occur when a unit is taken from one Array and inserted into another Array.

Set syntax: ResolveLUNConflicts

RestoreModePages

Restores all mode pages to the factory set default. Restarting the Diamond Storage Array is required for settings to take effect.

Set syntax: RestoreModePages

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

SledFaultLED

Changes the state of the selected sled LED to the indicated state.

Choices: enter sled number 1-12 or all, on or off Default: off Set syntax: set SledFaultLED [all|sled number] [on| off]

VirtualDriveInfo

Displays the named Virtual Drive definitions. For examples, refer to

Information Commands Results

page ii.

Immediate command: VirtualDriveInfo [active|planned] [Virtual Drive ID]

Configure drives with CLI commands

6.0 Configuring Drives

The Diamond Storage Array can be configured as a JBOD, RAID Level 0, RAID Level 1, RAID Level 10 or RAID Level 5 array. Its default is JBOD. RAID is a storage configuration which uses multiple disk drives to increase capacity, performance and/or reliability.

You may configure your Diamond Storage Array in several different ways depending on your needs although the Diamond makes some choices for your. The following elements must be considered when you are configuring your Diamond.

• RAID level

• Interleave

• Hot Spares option

Using the ExpressNAV browser-based interface is the easiest way to set up your Diamond. You may also use the Command Line Interface commands.

CAUTIONCAUTION

Changing these parameters causes all previous drive data on the Diamond Storage Array to be erased. Make sure you back up all information before setting up a different configuration.

JBOD (Just a Bunch of Disks)

JBOD (Just a Bunch of Disks) configuration, the default for the Diamond Storage Array, allows many individual disk drives to be available for normal storage operations. A JBOD configuration allows you to access each disk drive in the array independently. Any action you can do to a normal disk drive can be performed on any disk in the JBOD.

RAID Level 0

RAID Level 0 (striping) is based on the fact that increased disk performance can be achieved by simultaneously accessing data across multiple disk drives in an array. This arrangement increases data transfer rates while reducing average access time by overlapping drive seeks. RAID Level 0 groups provide data that is striped across several drives. RAID Level 0 is pure striping, without redundancy, meaning there is no

data protection. If one disk fails, all data within that stripe set is lost.

RAID Level 0 is used by applications requiring high performance for non-critical data.

The

QuickRAID0

command, accessed through the CLI, allows a simple, fast, out-of-the-box setup of the array into evenly-sized RAID Level 0 stripe groups.

RAID Level 1

RAID Level 1 (mirroring) ensures the security of data by writing the exact same data simultaneously to two or more different drives. This application is for users with critical data which cannot be lost or corrupted due to the failure of a single drive.

With RAID Level 1, the host will see what it believes to be a single physical disk of a specific size: it does not know or care about the mirrored pair. The RAID controller manages where data is written and read, allowing one disk to fail without the host knowing it has failed. The array will send notification of the failure over the serial or Ethernet port and the fault LED will be illuminated. Service personnel can then replace the failed drive and initiate a rebuild.

RAID Level 1 is used in applications containing mission critical data. The

QuickRAID1

command, accessed through the Command Line Interface, allows a simple, fast, out-of-the-box setup of the array into RAID Level 1 mirrored groups.

RAID Level 10

RAID Level 10 (mirroring with striping) increases data transfer rates while ensuring security by writing the exact same data simultaneously to two or more different drives.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

RAID Level 10 is used in applications requiring high performance and redundancy, combining the attributes of RAID Levels 1 and 0.

The

QuickRAID10

command, accessed through the Command Line Interface, allows a simple, out-of-the-box setup of RAID Level 10 groups.

The array will operate in degraded mode if a drive fails unless you have enabled Hot Spare sleds.

RAID Level 5

RAID Level 5 increases reliability while using fewer disks than mirroring by employing parity redundancy. Distributed parity on multiple drives provides the redundancy to rebuild a failed drive from the remaining good drives. Parity data is added to the transmitted data at one end of the transaction, then the parity data is checked at the other end to make sure the transmission has not had any errors.

In the array, transmitted data with the added parity data is striped across disk drives. A hardware XOR engine computes parity, thus alleviating software processing during reads and writes.

The array will operate in degraded mode if a drive fails unless you have enabled Hot Spare sleds.

file transfer size is large, the interleave value should be large, and vice versa.

The value entered for the

RAIDInterleave

command refers to blocks of data: one block is equivalent to 512 bytes of data.

Valid entries are 16, 32, 64, 128, 256 and SPAN. SPAN, not available in RAID Level 5, indicates that interleave size between the drives in the group will be the minimum drive size of all members in the group.

Hot Spare sleds

In most configurations, if a member of a virtual device becomes degraded, you must swap out the faulted sled as defined in

Instructions on page 71

AutoRebuild

, you must also start a manual rebuild.

Hot Swap Operating

. If you have not enabled

For four configurations, however, Hot Spare sleds may be designated as replacements for faulted sleds without intervention by you or a host.

Each configuration requires a certain number of Hot Spare sleds. These sleds, once designated as Hot Spares, are not available for other use.

The following configurations will support optional Hot Spare sleds

Interleave

The interleave size sets the amount of data to be written to each drive in a RAID group. This is a tunable parameter which takes a single stream of data and breaks it up to use multiple disks per I/O interval.

The CLI command

RAIDInterleave

allows you to change the size of the sector interleave between RAID groups. The value will depend upon the normal expected file transfer size. If the normal

RAID Level 1:

RAID Level 10:

RAID Level 5:

2 Hot Spare sleds

1 group, 2 Hot Spare sleds

1 group, 1 Hot Spare sled

2 groups, 2 Hot Spare sleds

Enhancing performance

SpeedWrite, enabled by the CLI command

SpeedWrite

, improves the performance of

WRITE commands

Configuring drives

6.1 JBOD

The Diamond Storage Array is set up in a JBOD (Just a Bunch of Disks) configuration as default. JBOD configuration allows for many individual disk drives to be available for normal storage operations.

CAUTIONCAUTION

Selecting JBOD configuration causes all previous drive data on the Diamond Storage Array to be erased. Make sure all of your information is backed up before setting up the Diamond Storage Array in a JBOD configuration.

A JBOD configuration allows you to access each of the possible 24 disk drives in the Diamond Storage Array independently. In this configuration, any action you can do to a normal disk drive can be performed on any disk in the JBOD.

To set up the JBOD configuration

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

Command Line Interface.

1 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

on page 17). You should now be in

ExpressNAV: Browser-based Interface

Accessing

ATTO

on page 25).

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

2Type set QuickRAID0 0. The command will

configure the Diamond Storage Array in the JBOD configuration.

3 Information will display on the screen while the

Diamond Storage Array updates NVRAM. The Ready* prompt displays.

4Type SaveConfiguration to save the JBOD

configuration.

5Type FirmwareRestart to reboot the Diamond

Storage Array.

6 Reboot the host computer.

The Diamond Storage Array is now configured as a set of individual disk drives.

7 To verify the configuration, type DriveInfo

any sleds are missing, the sled slots will be reported as offline.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

JBOD

6.2 RAID Level 0

The Diamond Storage Array can be set up into RAID Level 0 (striping) groups to allow it to read and store data at a higher transfer rate. QuickRAID0, a CLI command, allows you to set up the system as if it were a single drive instead of separate drives.

CAUTIONCAUTION

Setting RAID configuration causes all previous drive data on the Diamond Storage Array to be erased. Make sure all information is backed up before setting up RAID groups.

RAID Level 0 groups provide data that is striped across several drives. The

QuickRAID0

command, accessed through the CLI or the ExpressNAV interface, sets up the Diamond Storage Array into evenly-sized RAID Level 0 stripe groups. Each stripe group is a virtual drive named with its own LUN (logical unit number).

With a fully populated array, RAID 0 may be configured as 1, 2, 3, 4, 6, or 12 LUNs. If you set up RAID0 1, all 24 drives are configured as a single stripe group. You may also configure two LUNs of 12 drives each, three LUNs of eight drives each, four LUNs of six drives each, six LUNs of four drives each and 12 LUNS of two drives each. (See Exhibit 6.2-1). The

QuickRAID0

command assumes there are 24

drives available.

Sled-based versus disk-based

RAID Level 0 can be configured one of two ways, sled-based or disk-based. The default is sledbased.

Sled-based

Use sled-based if an external RAID controller controls the Diamond Storage Array to ensure that both drives on a sled will be members

of the same RAID group (LUN). Removing one sled will not affect other LUNs.

Drive-based

QuickRAID0

designates each drive on a sled as a either partner 1 or partner 2. Stripe groups are made by combining all partners designated as 1 together, and all partners designated as 2 together. Removing one sled will affect more than one LUN because each partner will belong to a different LUN.

To set up RAID Level 0 groups

1 Connect to the array via the RS-232 port or the

optional Ethernet management services card (refer to should now be in CLI.

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

ExpressNAV: Browser-based Interface

3 QuickRAID0 options are 0, 1, 2, 3, 4, 6 and 12

RAID groups, sled-based or drive-based. The number indicates the number of RAID groups.

Sled or drive indicates the way you want the array striped. The QuickRAID0 command will divide the total number of drives in the same array equally by the number called out in the command. Type

set QuickRAID0 [0|1|2|3|4|6|12] [DRIVE|SLED]

Accessing the Array

on page 17). You

ATTO

on page 25).

Note

Because ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

An example of RAID 0 configuration (QuickRAID0 4): 2 or more physical drives into 1 or m ore virtual drives

RAID 0 subsystem example:

4 virtual devices

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

24 physical drives

4 The array will update

NVRAM. The Ready displays.

5 Type

SaveConfiguration.

6 Type

FirmwareRestart to reboot the array.

7 Reboot the host computer.

8 If sled(s) or drive(s) are physically missing from

the array, the entire stripe group (LUN) containing the drive(s) will be unavailable. To determine which drives would be unavailable in various configurations, see Exhibit 6.2-1 and Exhibit 6.2-2.

To remove RAID Level 0 groups

(refer to

Accessing the Array

on page 17). You

should now be in CLI.

1 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

ExpressNAV: Browser-based Interface

ATTO

on page 25).

2Type set QuickRAID0 0 to configure the array

in JBOD mode.

3 Continue with steps 4 through 7 of

RAID Level 0 groups

on page 53.

To set up

1 Connect to the array via the RS-232 port or the

optional Ethernet management services card

Exhibit 6.2-1 Sled-based QuickRAID0 stripe groups with LUN designations in a fully populated Array set up as

QuickRAID0 6 sled

Physical Sled123456789101112

Physical Drive121212121212121212121212

JBOD1 111111111111111111111111

2 111111111111222222222222

3 111111112222222233333333

4 111111222222333333444444

6 11112222

12112233445566778899101011111212

QuickRAID0 parameters

. If sled 6 were to be withdrawn from the array, LUN 3 (grayed boxes) would be unavailable.

3 3 3 3444455556666

LUNs: virtual drive numbers

Exhibit 6.2-2 Drive-based QuickRAID0 stripe groups with LUN designations in a fully populated Array set up as

QuickRAID0 6 Drive

Physical Sled123456 7 8 9 10 11 12

Physical Drive121212121212121212121212

JBOD 0 113214315416517618719820921102211231224

1 111111111111111111111111

2 121212121212121212121212

3 121212121313131323232323

4 131313131313242424242424

6 14141414

QuickRAID0 parameter

12 1 7 1 7 2 8 2 8 3 9 3 9 410410511511612612

. If sled 6 were to be withdrawn from the array, LUNs 2 and 5 would be unavailable.

2 5 2 5 2 5 2 536363636

LUNs: virtual drive numbers

Diamond Storage Array

Exhibit 6.2-3 Configurations of a fully populated Diamond Storage Array in RAID Level 0.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

6.3 RAID Level 1

The Diamond Storage Array can be set up into RAID Level 1 (mirrored) groups to provide greater reliability by simultaneously writing data to two sleds. Each sled partnered through QuickRAID1, a CLI command, has the same data as its partner.

CAUTIONCAUTION

Selecting RAID configuration causes all previous drive data on the Diamond Storage Array to be erased. Make sure all information is backed up before setting up RAID groups.

The configuration of RAID Level 1 performs the same operations on two partnered sleds at the same time, providing an automatic backup of data. The operating system sees the two sleds as one virtual sled with its own LUN (Logical Unit Number).

QuickRAID1

The to be set into mirrored drives. The

command allows the array

RAID 1 in a fully populated Diamond Storage Array: 6 sleds, each partnered with another sled, with each partner a mirror image of the other

command first spans two drives on a sled together, then partners two sleds to be a RAID Level 1 group designated by a LUN (logical unit number).

To set up RAID Level 1 groups

1 Connect to Diamond Storage Array

services via the RS-232 port or the optional Ethernet management services card (refer to should now be in CLI.

2 Continue with the CLI or access the

of the ExpressNAV interface (refer to

ExpressNAV: Browser-based Interface

3 QuickRAID1 has no options: the command sets

up each sled and its mirror image. Type set QuickRAID1. The Ready* displays.

4 The array will update NVRAM. The Ready*

displays.

5Type SaveConfiguration.

Accessing the Array

on page 17). You

RAID page

ATTO

on page 25).

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

6Type FirmwareRestart to reboot the array.

7 Reboot the host computer.

8 In a fully populated array, your operating

system displays six drives.

Note

In a less than fully populated array, if both partners of a LUN are missing, the LUN will not exist. If only one partner is missing, the LUN will exist, but it will be degraded (unprotected). See Exhibit 6.3-2 to determine which LUNs would be affected.

Exhibit 6.3-1 Drive sleds, LUNs and mirror partners in a RAID Level 1 configuration.

2 spanned drives per sled

each sled a mirror image of its

6 virtual devices

To remove RAID groups

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

CLI.

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

on page 17). You should now be in

ExpressNAV: Browser-based Interface

3Type set QuickRAID0 0. This will configure the

array in JBOD mode.

4 The array will update NVRAM. The Ready*

displays.

5Type SaveConfiguration.

6Type FirmwareRestart to reboot the array.

7 Reboot the host computer.

8Type DriveInfo when in CLI or access the

Storage Management page of the

partner

Accessing

ATTO

on page 25).

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

ExpressNAV interface to determine the setup of the array.

the CLI or enter ATTO ExpressNAV browser interface Advanced CLI Configuration page.

2Type set QuickRAID 1 2

Hot Spare sleds

To set up RAID Level 1 with Hot Spare sleds

3Type SaveConfiguration Restart

4 The array is configured into one RAID Level 1

group with two Hot Spare sleds.

1 Connect to Diamond Storage Array services

(refer to

Accessing the Array

on page 17) and use

Exhibit 6.3-2 Mirrored stripe groups with LUN designations in a fully populated Diamond Storage Array. If sled 12 were

removed, the drives marked LUN 6 would be available but degraded (unprotected by mirroring). If both sleds 11 and 12 were missing, LUN 6 would be unavailable.

Sled123456789101112

Drive121212121212121212121212

LUN111122223333444455556666

Mirror Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2

Exhibit 6.3-3 Configuration of sleds in RAID Level 1: five LUNs with two Hot Spare sleds

RAID Level 1

6.4 RAID Level 5

CAUTIONCAUTION

Selecting RAID configuration causes all previous drive data on the Diamond Storage Array to be erased. Make sure all information is backed up before configuring RAID groups.

In RAID Level 5, parity data is added to the transmitted data at one end of the transaction, then the parity data is checked at the other end to make sure the transmission has not had any errors.

In the Diamond Storage Array, transmitted data with the added parity data is striped across disk drives. A hardware XOR engine computes parity, thus alleviating software processing during reads and writes.

The array uses parity declustering, a special case of RAID Level 5. Parity information is spread across each LUN, not concentrated on one drive or sled.

When you initially set up RAID groups using the

QuickRAID

command, groups are synchronized automatically because there is no pre-existing data on the drives. However, drives may display as “degraded, and you need to set all LUNs to OK status. Refer to

Rebuilding RAID Level Configurations

on page 63.

Configuring a fully-populated array

You may set up a fully-populated Diamond (12 sleds) into one, two, three or four RAID Level 5 groups, with or without hot spare sleds, using the QuickRAID 5 command.

To set up RAID Level 5 groups

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

CLI.

on page 17). You should now be in the

Accessing

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

ExpressNAV: Browser-based Interface

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

3 Decide how many RAID Level 5 groups you

want (1, 2, 3 or 4)..

4Type set QuickRAID5 [0|1|2|3|4]

0 returns the array to JBOD

5Type SaveConfiguration Restart to save the

RAID Level 5 configuration.

6A Ready prompt displays. You must zero all

drives and parity to make all RAID Level 5 drives coherent. Type RAID5ClearData al

CAUTIONCAUTION

This is a destructive operation: all information stored on these drives is lost.

DO NOT interrupt power until the RAID5ClearData operation has completed (three to six hours).

7 Information displays on the screen while the

array updates NVRAM ending with a Ready

8Type SaveConfiguration.

9Type

10 Reboot the host computer connected via Fibre

11 The array is now configured. To verify the

To set up one RAID Level 5 group with one Hot Spare sled

1 Connect to Diamond Storage Array services

FirmwareRestart to reboot the array.

Channel or SCSI to the array.

configuration, type DriveInfo. If any sleds are missing, the sled slots are reported as

(refer

Accessing the Array

the Command Line Interface or enter ATTO ExpressNAV browser interface Advanced CLI Configuration page.

on page 17) and use

ATTO

on page 25).

offline

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

2Type set QuickRAID5 1 1

3Type SaveConfiguration Restart

The array is configured into one RAID Level 5 group with one Hot Spare sled.

the CLI or enter ATTO ExpressNAV browser interface Advanced CLI Configuration page.

2Type set QuickRAID5 2 2

3Type SaveConfiguration Restart

The array is configured into two RAID Level 5

To set up two RAID Level 5 groups with two Hot Spare sleds

groups with two Hot Spare sleds. Refer to Exhibit 6.4-1

1 Connect to Diamond Storage Array services

(refer to

Accessing the Array

on page 17) and use

Exhibit 6.4-1 LUNs are set up using the drives and sleds shown here. Parity information is spread throughout each LUN.

Sled123456789101112

Drive121212121212121212121212

QuickRAID5 1

QuickRAID5 2

QuickRAID5 3

QuickRAID5 4

LUN 1 LUN 2

LUN 1 LUN 2 LUN 3

LUN 1 LUN 2 LUN 3 LUN 4

LUN 1

Exhibit 6.4-2 Drives are striped, and parity information is interspersed among the sleds.

Physical Sled123456789101112

Physical Drive121212121212121212121212

Step 1 Virtual Drive

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

Step 2 Striping

Step 3 Created LUN

LUN 1 LUN 2 LUN 3 LUN 4

Exhibit 6.4-3 Configuration of sleds in RAID Level 5 in a fully populated array (12 sleds).

Sled123456789101112

Drive12121212121212121212121

QuickRAID5 1 1

QuickRAID5 2 2

LUN 1 LUN 2 Hot

LUN 1 Hot

Spare

Hot

Spare

RAID Level 5

Configuring a partially-populated array

The simplest way to attain RAID Level 5 in a partially-populated array (an array with three or more sleds but less than 12 sleds) is to create one RAID Level 5 group, with or without hot spare

sleds, encompassing all the available sleds by using the CLI command

QuickRAID5 ALL

To set up a partially populated arrays, you must have at least three sleds filling contiguous slots as shown in Exhibit 6.4-4, beginning with the slot closest to the management card.

Exhibit 6.4-4 Examples of RAID Level 5 configurations in partially-populated array with at least six sleds. Hot Spare sleds are in the highest slot numbers.

Sled

QuickRAID5 4 3 contiguous sleds in slots 1-3 and slots 4-6 with no Hot Spare sled

QuickRAID5 3 4 contiguous sleds in slots 1-4 and slots 5-8 with no Hot Spare sled

QuickRAID5 2 6 contiguous sleds in slots 1-6 with no Hot Spare sled

QuickRAID5 2 1 5 contiguous sleds in slots 1-5 with up to 2 Hot Spare sleds

QuickRAID5 ALL 3-12 contiguous sleds with no Hot Spare sleds

QuickRAID5 ALL 1 3-11 contiguous sleds with 1 Hot Spare sled

To set up RAID Level 5 groups

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

CLI.

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

on page 17). You should now be in the

ExpressNAV: Browser-based Interface

1 2 3

LUN1

LUN 1

Accessing

ATTO

on page 25).

4 5 6

LUN 2

LUN 1

3 Decide the RAID Level 5 configuration you

want based on the number of sleds you are using and Exhibit 6.4-4 above.

4Type set QuickRAID5 [2 | 3 | 4 | ALL]

5Type SaveConfiguration Restart to save the

RAID Level 5 configuration.

6A Ready prompt displays. You must zero all

drives and parity to make all RAID Level 5 drives coherent. Type RAID5ClearData al

7 8 9

LUN 2

10 11 12

Hot

Spare

Hot

Spare

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

CAUTIONCAUTION

This is a destructive operation: all information stored on these drives is lost.

DO NOT interrupt power until the RAID5ClearData operation has completed (three to six hours).

7 Information displays on the screen while the

array updates NVRAM ending with a Ready

8Type SaveConfiguration.

9Type

10 Reboot the host computer connected via Fibre

11 The array is now configured. To verify the

To set up one RAID Level 5 group with one Hot Spare sled

1 Connect to Diamond Storage Array services

2Type set QuickRAID5 ALL 1

3Type SaveConfiguration Restar

FirmwareRestart to reboot the array.

Channel or SCSI to the array.

configuration, type DriveInfo. If any sleds are missing, the sled slots are reported as

(refer to the Command Line Interface or enter ATTO ExpressNAV browser interface Advanced CLI Configuration page.

Accessing the Array

on page 17) and use

The array is configured into one RAID Level 5 group with one Hot Spare sled.

offline

Removing RAID groups

To remove RAID groups

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

CLI.

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

on page 17). You should now be in

ExpressNAV: Browser-based Interface

3Type set QuickRAID5 0. This configures the

array in JBOD mode.

4 Information displays on the screen while the

array updates NVRAM ending with a Ready

5Type SaveConfiguration.

6Type

7 Reboot the host computer connected via Fibre

8 The array is now configured. To verify the

FirmwareRestart to reboot the array.

Channel or SCSI to the array.

configuration, type DriveInfo. If any sleds are missing, the sled slots are reported as

Accessing

ATTO

on page 25).

offline

RAID Level 5

6.5 RAID Level 10

The Diamond Storage Array can be set up into RAID Level 10 (striped and mirrored) groups to provide greater reliability by simultaneously writing data to two sleds. Each sled partnered through QuickRAID10, a CLI command, has the same data as its partner.

CAUTIONCAUTION

Selecting RAID configuration causes all previous drive data on the Diamond Storage Array to be erased. Make sure all information is backed up before configuring RAID groups.

The configuration of RAID Level 10 stripes information across several mirrored drives, performing the same operations on two partnered sleds at the same time thus providing an automatic backup of data. The

QuickRAID10

command first creates six mirrored groups, then stripes them into groups of one, two or three RAID 10 groups (see Exhibit 6.5-2).

To set up RAID Level 10 groups

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

CLI.

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

on page 17). You should now be in

ExpressNAV: Browser-based Interface

Accessing

ATTO

on page 25).

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

3 Decide how many RAID10 groups you want (0,

1, 2 or 3).

4Type set QuickRAID10 [0 |1 | 2| 3].

Note

will return the array to JBOD.

5 The array will update NVRAM. The Ready*

displays. 6Type SaveConfiguration. 7Type FirmwareRestart to reboot the array. 8 Reboot the host computer.

To remove RAID groups

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet

management services card (refer to

the Array

CLI.

on page 17). You should now be in

Accessing

An example of RAID 10 (QuickRAID10 2) in a fully populated Diamond Storage Array: 6 sleds, each partnered with another sled, with each partner a mirror image of the other, configured into two stripe groups

2 virtual devices

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

2 spanned drives per sled

each sled a mirror image of its partner

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

ExpressNAV: Browser-based Interface

ATTO

on page 25).

3Type set QuickRAID0 0 to configure the array

in JBOD mode.

4 The array will update NVRAM. The Ready*

displays. 5Type SaveConfiguration. 6Type FirmwareRestart to reboot the array. 7 Reboot the host computer. 8Type DriveInfo when in CLI or access the

Hot Spare sleds

To set up RAID Level 10 with Hot Spare sleds

1 Connect to Diamond Storage Array services

(refer to the Command Line Interface or enter ATTO ExpressNAV browser interface Advanced CLI

Configuration page. 2Type set QuickRAID10 1 2 3Type SaveConfiguration Restart 4 The array is configured into one RAID Level 10

group with two Hot Spare sleds.

Accessing the Array

on page 17) and use

Storage Management page of the ExpressNAV interface to determine the setup of the array.

Exhibit 6.5-1 Configuration of sleds in RAID Level 10: five mirrored pairs of sleds in one Virtual Drive and two Hot Spare non-mirrored sleds

1 Virtual Drive

2 Hot Spare sleds

Exhibit 6.5-2 striping across the mirrored groups

Stripe

QuickRAID10 1 = LUN 1

Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Par tner 1 Partner 2 Partner 1 Partner 2

Sled123456789101112

Drive121212121212121212121212

Stripe

QuickRAID10 2 = LUN 1 QuickRAID10 2 = LUN 2

Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2

Sled

Drive

Stripe

QuickRAID10 3 = LUN 1 QuickRAID10 3 = LUN 2 QuickRAID10 3 = LUN 3

Partner 1 Partner 2 Partner 1 Partner 2

Sled1234

Drive12121212

QuickRaid10

first spans drives across sleds, then partners sleds into mirrored groups, then enables

Mirrored group 1 Mirrored group 2 Mirrored group 3 Mirrored group 4 Mirrored group 5 Mirrored group 6

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

1 2 1 2 1 2 1 2 1 2 1 2121212121212

Mirrored group 1 Mirrored group 2 Mirrored group 3 Mirrored group 4 Mirrored group 5 Mirrored group 6

Partner 1 Partner 2 Par tner 1 Partner 2 Partner 1 Partner 2 Partner 1 Partner 2

5 6 7 8 9 10 11 12

1 2 1 2 1 2 1 212121212

All configurations will not be available if the array has fewer than 24 physical drives in 12 physical sleds. For example, QuickRAID 10 1 will only work with a fully populated array. QuickRAID10 3 will work if sleds 9-12 are removed: LUNs 1 and 2 will be available but LUN 3 will not be available.

RAID Level 10

6.6 Rebuilding RAID Configurations

If a sled must be removed and a new sled inserted into the Diamond Storage Array while it is configured in a RAID Level 1, 5 or 10, you must rebuild the RAID Level using CLI commands or the ExpressNAV interface.

CAUTIONCAUTION

Selecting RAID parameters causes all previous drive data on the Diamond Storage Array to be erased. Make sure all of your information is backed up before setting up RAID groups.

When you initially set up RAID groups using the

QuickRAID

command, groups are synchronized automatically because there is no pre-existing data on the drives. However, drives may display as “degraded, and you need to set all LUNs to OK status.

The simplest method to check RAID group status is to access the interface. Refer to

Interface

To reset LUN status

1 Display the status of the array by typing

2 Set the sleds which are listed as degraded to a

set RAIDRebuildState [sled number] OK

on page 25

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

RAIDRebuildStatus

rebuild state of OK by entering

RAID

page of the ExpressNAV

ATTO ExpressNAV: Browser-based

Note

Drive rebuilding reduces performance. You may want to leave

disabled

hours, or use the described below.

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

CLI.

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

and manually rebuild during off-peak

on page 17). You should now be in

ExpressNAV: Browser-based Interface

3Type AutoRebuild enabled.

4 Information displays on the screen while the

array updates NVRAM ending with a Ready*

5Type SaveConfiguration.

6Type

7 Reboot the host computer connected via Fibre

8 To verify the configuration, type

9 If a rebuild is necessary, the array will

Rebuild priority

FirmwareRestart to reboot the array.

Channel or SCSI to the array.

RAIDRebuildStatus

automatically rebuild drives.

AutoRebuild

RebuildPriority

at the default

command

Accessing

ATTO

on page 25).

Drive rebuilding reduces performance. Use the

RebuildPriority

RAID

page of the ExpressNAV interface to

command through the CLI or on the

customize when your rebuilds will occur.

To synchronize mirrored drives automatically

If mirrored drives are removed for more than 15 to 30 seconds and then re-inserted or replaced, the replaced drives are labeled “degraded” when you check the Array’s status by typing

RAIDRebuildStatus

. If you enable array rebuilds the degraded drives automatically when a new drive is inserted.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

AutoRebuild.

the

Choices are

• If you select High priority, rebuild I/O requests are implemented before system I/O requests.

• If you select Low priority, rebuild I/O requests execute only when there are no pending I/O requests.

• If you select Same priority, rebuild I/O and system I/O receive equal consideration.

High, Low

and

Same

You must use a

SaveConfiguration Restart

command to implement the rebuild priority command.

3If AutoRebuild has not been disabled, type

set AutoRebuild disabled saveconfiguration restart.

To synchronize mirrored drives manually

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

CLI.

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

on page 17). You should now be in

ExpressNAV: Browser-based Interface

Accessing

ATTO

on page 25).

4 After the array reboots and completes its

diagnostics, enter the CLI and type RAIDManualRebuild L S where L is the LUN and S

is the sled to be rebuilt.

This procedure may take a few hours, depending on the size of the LUN.

5 To check the rebuild status, type

RAIDRebuildStatus

6.7 RAID Interleave Parameters

The interleave size sets the amount of data to be written to each drive in a RAID Level group. This is a tunable parameter which takes a single stream of data and breaks it up to use multiple disks per I/O interval.

CAUTIONCAUTION

Changing this parameter causes all previous drive data on the Diamond Storage Array to be erased. Make sure you back up all information before setting up different interleave sizes.

The default sector interleave set by the

QuickRAID

command is 128 blocks (64k). The CLI command,

RAIDInterleave

, allows you to change the size of the sector interleave between RAID groups. The value depends upon the normal expected file transfer size. If the normal file transfer size is large, the interleave value should be large, and vice versa.

The value entered for the RAIDInterleave command refers to blocks of data: one block is equivalent to 512 bytes of data.

To change the RAID Interleave parameter

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

Command Line Interface.

on page 17). You should now be in

Accessing

2 Continue with the CLI or access the RAID page

of the ExpressNAV interface (refer to

ExpressNAV: Browser-based Interface

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

RAIDInterleave options are 16, 32, 64, 128, 256 blocks and span. Span is not available for RAID Level 5.

3 In all RAID levels except RAID Level 5, type

ATTO

on page 25).

set RAIDInterleave [16|32|64|128|256| span]

For RAID Level 5 type

set RAIDInterleave [16|32|64|128|256]

4 Information displays on the screen while the

array updates NVRAM ending with a Ready

5Type SaveConfiguration.

6Type

7 Reboot the host computer connected via Fibre

8 The array is now configured. To verify the

FirmwareRestart to reboot the array.

Channel or SCSI to the array.

configuration, type

get

RAIDInterleave.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

RAID interleave

7.0 Hardware Maintenance

The disk drive sleds, blower assemblies, power supplies, host interface cards, and system management card may be replaced with identical or upgraded parts.

WARNING

The only way to completely de-energize the unit is

to turn off both power

Power Switch Positions

On Stand-by

off power to the array; it is not an AC on-off switch. Power may still be in the unit through the other power supply.

supplies and unplug both power cords from the back of the unit.

Turning the power switch to the Stand-by position on one power supply does not completely turn

xhibit 7-1: The management card may be accessed via a

erial port DB-9 connector or an optional Ethernet

connection.

CAUTIONCAUTION

All modular components must be replaced by qualified personnel only.

Do not leave empty openings in the Diamond Storage Array. Empty openings affect airflow and may cause the unit to overheat and shut down. Components are electrostatic sensitive. Use proper grounding methods when working with or around the array. Always store spare components in proper ESD containers when not in use.

The power supply and blower assembly may be replaced while the unit is running. (Refer to

Operating Instructions on page 71

• Host interface cards and management cards may only

be replaced when the Diamond Storage Array is off. Backup the unit fully before replacing these components.

• You may remove a disk drive sled while the array is

powered on. Refer to the instructions in

Operating Instructions on page 71

To remove a management card

1 Power down both power supplies.

2 Loosen the screws holding the card in place. 3 Pull out the assembly and replace it with

another.

4 Securely tighten all screws after replacing the

component.

To remove a disk drive sled (Exhibit 7-2)

1 Loosen the screws on either side of the assembly

2 Pull on the assembly’s handle and carefully

slide it out of its bay.

)

for details.

Hot Swap

3 Replace the disk drive sled with another sled.

To remove a Host Interface Card

1 Power down both power supplies

2 Remove any cable attached to the port. 3 Loosen the retaining screws 4 Pull the Host Interface Card out of the unit. 5 To replace the card, push it back into the unit

and tighten the retaining screws.

To remove the power supply (Exhibit 7-4),

1 Press the Stand-by power switch to the off position

2 Remove the power cord 3 Using a No. 1 Phillips screwdriver, loosen the

screws holding the assembly in place.

4 Pull out the assembly and replace it with

another.

5 Securely tighten all screws after replacing the

component.

To remove a blower assembly (Exhibit 7-4),

1 Using a No. 1 Phillips screwdriver, loosen the screws

holding the assembly in place.

2 Pull out the assembly and replace it with

another.

3 Securely tighten all screws after replacing the

component.

WARNING

Do not leave empty openings on the front or rear of the Diamond Storage Array under any circumstances.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Exhibit 7-3: The Fibre Channel or SCSI Host

Interface Card may be replaced by shutting power down, removing any cable attached to the port, removing the SFP according to manufacturer's instructions, loosening the screws at the top and bottom of the card, then carefully pulling out the unit.

xhibit 7-2: Above, disk drive sled partially pulled out of the Diamond

Storage Array. Bottom left, top of disk drive sled. Bottom right, underside of disk drive sled showing individual drives

WARNING

Do not leave empty openings on the front or rear of the Diamond Storage Array under any circumstances.

Exhibit 7-4: A power supply pulled out from a rackmount Diamond Storage Array: do not leave an empty opening while the Diamond Storage Array is operating. Access the blower assembly and the power supply by loosening the screws on either side of the component, then pulling out the part. If a blower or power supply stops working, keep it in place until another component is installed. (Note: host interface configuration pictured is not supported)

Maintenance

7.1 Hot Swap Operating Instructions

To maintain array up time, individual disk drive sled assemblies, power supplies and blower assemblies can be replaced with the unit fully operational. Special instructions need to be followed to perform these operations.

Disk drives

The following cautions must be observed when hot swapping disk drives.

• Individual disk drive sled assemblies may be

replaced while the array is operating with no other intervention only if there is absolutely no activity on that drive. Failure to ensure no activity may destroy any data on that drive and possibly stop the entire array operation. Follow the instructions below using CLI commands to replace drives.

• Removing a drive sled that is part of a RAID

Level 0 group will result in the loss of all data in the groups associated with that sled. Follow your backup procedures when removing sleds.

• Do not leave empty openings on the front or rear

of the array under any circumstances. Empty openings affect airflow and may cause the unit to overheat and shut down.

• All modular components must be replaced by

qualified personnel only. Components are electrostatic sensitive. Use a static wriststrap when handling any of the cards inside the array. Use proper grounding methods. Always store spare components in proper ESD containers when not in use.

Follow your backup procedures before removing a sled. You may copy drives using the appropriate CLI commands (Refer to page 45).

The following method is the safest way to perform a hot swap of a drive.

The general approach: identify the disk drive sled to be replaced, take it offline using the appropriate CLI commands, turn off its power using CLI commands, remove and replace with a new disk drive sled, power up the new sled, and place back on line.

1 Connect to Diamond Storage Array services

(refer to enter the Command Line Interface mode.

2 The fault LED should be blinking on the disk

drive sled to be replaced. If it is not, type set SledFaultLED [n] on. The LED of the drive you want [n] will light.

Drive Configuration Commands

Accessing the Array

on page 17) and

3 Determine the Drive Sled Number. The disk

drive sled closest to the management card is always the number 1 disk drive sled

The Drive Sled Number identified in step 2 is the SledNum variable to be entered into the CLI commands. Take the disk drive sled offline by entering the following CLI commands at the control computer Ready prompt

Set AtaDiskState (SledNum, 1, OFFLINE) Set AtaDiskState (SledNum, 2, OFFLINE) Set DriveSledPower (SledNum, OFF)

CAUTIONCAUTION

In a Hot Spare configuration, a drive sled should only be taken offline if there is absolutely no activity on that drive. If there is any activity, the rebuild of the Hot Spare sled may be flawed.

4 Wait 30 seconds for the disk drive sled to spin

down and complete any remaining I/O activity.

5 Unscrew the two screws on the disk drive sled

with the appropriate tool.

6 Carefully pull the disk drive sled out of chassis

using its handle.

7 Mark or tag the disk drive sled with the array

serial number, the date removed, and its slot number.

8 Place the disk drive sled in an appropriate ESD

container or bag.

9 Install the replacement disk drive sled into the

array chassis using proper ESD control steps and with the proper orientation (The disk drive sled assembly is keyed and can only be inserted one way).

10 Tighten the two screws on the disk drive sled

with the appropriate tool.

11 If you do not have

the disk drive sled number identified in step 2, on the control computer Ready

1 2 3 4 5 6 7 8 9 10 1 1 12

Management Card

AutoRebuild enabled

prompt type

Drive Sleds

, using

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

ClearDiskReservedArea (SledNum, 1) ClearDiskReservedArea (SledNum, 2) Set AtaDiskState (SledNum, 1, ONLINE) Set AtaDiskState (SledNum, 2, ONLINE)

12 The new disk drive sled will be available for

system use 10 to 15 seconds after the disk drives spin up and communication is reestablished.

Power supplies

CAUTIONCAUTION

Do not leave empty openings on the front or rear of the Diamond Storage Array under any circumstances. Empty openings affect airflow and may cause the unit to overheat and shut down.

WARNING

Hazardous voltage and stored energy hazard when removing power supplies.

In a system with at least one operational power supply, the other power supply can be successfully removed and replaced without powering the system down and with no loss of array functionality. The green activity LED on the front of the system management card identifies the operational status of each power supply (Green means the power supply is operating correctly).

Note

System command overlap is discontinued across drives when only one power supply is operational.

Labels on the rear of the Diamond Storage Array point to the A and B power supplies.

1 Identify the power supply to be swapped on the

rear of the unit.

2 Turn off the power supply on/off switch.

3 Disconnect AC line cord.

4 Unscrew the two screws on the power supply

with the appropriate tool.

5 Pull power supply out of chassis using the

power supply module handle.

6 Install new power supply in chassis with proper

orientation (Power supply is keyed and can only be inserted one way).

7 Tighten the two screws on the power supply

with the appropriate tool.

8 Connect AC line cord

9 Turn on power supply switch.

10 Verify correct operation by observing that the

green light on the rear of the power supply is lit and that the appropriate power supply light on the system management card on the front of the unit is lit.

Note

It takes up to 30 seconds for the system to recognize the insertion or removal of a power supply and change the LED on the system management board. The CLI will issue messages about the change (refer to

Reporting

on page 79).

System Monitoring and

Blower assemblies

CAUTIONCAUTION

Do not leave empty openings on the front or rear of the array under any circumstances. Empty openings affect airflow and may cause the unit to overheat and shut down.

The array contains two blower assemblies. The blowers are critical to proper array cooling operation. However, the array can operate with only one functional blower within certain ambient temperatures. The blowers are electronically connected to the power supplies and a power supply will not run without its corresponding blower assembly operating. This means if the blower adjacent to power supply A is removed, the ‘A’ power supply will shut down, turning off the corresponding LED on the system management card. The CLI will issue a message about the power supply being turned off.

To replace a blower assembly

1 Unscrew the two screws on the blower

assembly with the appropriate tool.

2 Pull the blower assembly out of chassis.

3 Install new blower assembly in chassis with

proper orientation (Blower assembly is keyed and can only be inserted one way).

4 Tighten the two screws on the blower assembly

with the appropriate tool.

5 Verify correct operation by observing that the

green light is lit on the rear of the power supply, and that the appropriate power supply green light on the system management card on the front of the unit is also on.

• It takes up to 30 seconds for the system to recognize the insertion or removal of a power supply and change the LED on the system

Hot swapping

7.2 Optional Hot Spare Sled

To maintain array up time with minimal risk of data loss, individual sleds which fail may be replaced with a spare sled in some configurations.

In most configurations, if a member of a virtual device becomes degraded, you must swap out the faulted sled as defined in

Instructions on page 71

AutoRebuild

, you must also start a manual rebuild.

Hot Swap Operating

. If you have not enabled

For four configurations, however, Hot Spare sleds may be designated as replacements for faulted sleds without intervention by you or a host.

Each configuration requires a certain number of Hot Spare sleds. These sleds, once designated as Hot Spares, are not available for other use.

The following configurations will support optional Hot Spare sleds

• RAID Level 1: 2 Hot Spare sleds (see Exhibit 7.2-1)

• RAID Level 10: 1 group, 2 Hot Spare sleds (see Exhibit 7.2-2)

• RAID Level 5:1 group, 1 Hot Spare sled (see Exhibit 7.2-3)

• RAID Level 5: 2 groups, 2 Hot Spare sleds (see Exhibit 7.2-3)

Note

All arrays using RAID Level 10 and Hot Spares options must be fully populated.

If a sled becomes degraded and a Hot Spare sled has been designated

• the Diamond will replace the degraded sled with the Hot Spare sled, simulating a hot swap of a sled with AutoRebuild enabled, without intervention

• any sled with a Faulted drive will not be used. Faulted sleds maintain their faulted status until they are removed and re-inserted or if the system is restarted.

• Hot Spares are handled as sleds, not as individual drives

• the Hot Spare sled replaces the Faulted sled in the Virtual Device

• a rebuild will automatically start after the hot spare is switched into the Virtual Device, even if

AutoRebuild

• the DriveInfo command will list the number of Hot Spare sleds currently in the system

• you can replace the faulted drive or sled later. However, faulted is a non-persistent state. After a power cycle, the faulted sled will display on the DriveInfo screen as have no type, i.e, the type field will be blank. The sled cannot be accessed but must be removed and replaced. If it is replaced, it will be assigned in this order

1 if a Virtual Device is missing a sled or you are

hot swapping a sled, the new sled becomes part of the existing Virtual Device.

2 if a Hot Spare sled is missing, the replacement

sled will become a Hot Spare sled.

CAUTIONCAUTION

In a Hot Spare configuration, a drive sled should only be taken offline if there is absolutely no activity on that drive. If there is any activity, the rebuild of the Hot Spare sled may be flawed.

To set up RAID Level 1 with Hot Spare sleds

1 Connect to Diamond Storage Array services

(refer to the CLI or enter ATTO ExpressNAV Advanced CLI Configuration page.

2Type set QuickRAID1 2

3Type SaveConfiguration Restart

The array is configured into one RAID Level 1 group with two Hot Spare sleds.

To set up RAID Level 10 with Hot Spare sleds

1 Connect to Diamond Storage Array services

(refer to the CLI or enter ATTO ExpressNAV interface Advanced CLI Configuration page.

2Type set QuickRAID10 1 2

3Type SaveConfiguration Restart

The array is configured into one RAID Level 10 group with two Hot Spare sleds.

is disabled.

Accessing the Array

on page 17) and use

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

To set up one RAID Level 5 group with one Hot Spare sled

1 Connect to Diamond Storage Array services

(refer to

Accessing the Array

on page 17) and use the CLI or enter ATTO ExpressNAV interface Advanced CLI Configuration page.

2Type set QuickRAID5 1 1

3Type SaveConfiguration Restart

The array is configured into one RAID Level 5

To set up two RAID Level 5 groups with Hot Spare sleds

1 Connect to Diamond Storage Array services

and use the CLI or enter ATTO ExpressNAV interface Advanced CLI Configuration page.

2Type set QuickRAID5 2 2

3Type SaveConfiguration Restart

The array is configured into two RAID Level 5 groups with two Hot Spare sleds.

group with one Hot Spare sled.

Exhibit 7.2-1 Configuration of sleds in RAID Level 1: five LUNs with two Hot Spare sleds

Exhibit 7.2-2 Configuration of sleds in RAID Level 10: five mirrored pairs of sleds in one Virtual Drive and two Hot Spare non-mirrored sleds

1 Virtual Drive

2 Hot Spare sleds

Exhibit 7.2-3 Configuration of sleds in RAID Level 5

Sled123456789101112

Drive12121212121212121212121

QuickRAID5 1

QuickRAID5 2

LUN 1 LUN 2

LUN 1

Hot

Spare

Hot

Spare

Hot

Spare

Hot Spares

8.0 Copying Drives

Copying drives using the

DriveCopy

CLI command may be necessary on drives in the JBOD

configuration.

DriveCopy can create a backup of a drive onto another drive sled. If one drive fails, DriveCopy may be used to copy the data off the remaining drive on the sled to another drive on another sled. Once completed, the sled containing the failed drive may be replaced.

CAUTIONCAUTION

Make sure the destination backup drive sled does not contain any important data because it will be completely overwritten by the

DriveCopy

command.

Procedure

Use the

Configuration Commands

Advanced CLI Configuration

DriveCopy

CLI command (refer to

on page 45) or use the

page in the

Drive

ExpressNAV interface to copy a drive from the source disk to the destination disk. Parameters are the sled and drive numbers of the source and destination drives.

The destination drive must be offline: use the

ATADi skState

command to determine if the

destination disk is offline. If you choose the same

source and destination drive, this command will perform a

DriveCopyHalt

progress. operation that had been stopped. displays the status of a

DriveWipe

DriveCopyResume

stops a

DriveCopy

operation in

resumes a

DriveCopyStatus

DriveWipe

DriveCopy

operation

CAUTIONCAUTION

JBOD configurations

You may want to copy drives for backup. However, since several configurations may be present on a Diamond Storage Array, you must be confident you are accessing and overwriting the appropriate drives and volumes.

CAUTIONCAUTION

RAID Level 0 systems

DriveCopy

generation of a backup of an entire RAID Level 0 LUN.

may be used to coordinate the

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Copy drives

9.0 Updating Firmware

Engineers, technicians and/or system administrators/integrators may update the firmware of the Diamond Storage Array using the Command Line Interface (CLI) or the Advanced CLI Configuration page of the ExpressNAV interface (Refer to RS-232 serial port or the optional Ethernet management services card.

ATTO ExpressNAV: Browser-based Interface

on page 25) via the

Updating firmware using the RS-232 serial port

To update the firmware via a connection to the RS-232 serial port, you will need

• a host computer with a terminal emulation program such as HyperTerminal in Windows

• binary information file (“...”.ima) available from technical support or on our website (refer to

Contact ATTO Technology, Inc.

• a null modem serial cable with a DB-9 connector

To update firmware

1 Connect to Diamond Storage Array services via

the RS-232 port (refer to page 17). You should now be in the Command Line Interface mode.

2 Copy the latest Diamond Storage Array image

file (“...”.ima) onto the host computer.

3Type

ZModem Receive at the Ready prompt.

The terminal program on the host should be in Zmodem only mode, with no other parameters.

on page viii)

Accessing the Array

4 On the terminal program, choose

Transfer Send File.

5 In the Send File box, enter the current Diamond

flash (“...”.ima) filename or click the browse button to find it.

6Click Send File

7 The Diamond should acknowledge receiving

the file and display a message not to interrupt power for 90 seconds.

CAUTIONCAUTION

Do not interrupt the flash process. If the process is interrupted, the Diamond Storage Array will become inoperable and will have to be returned to the factory for repair.

Do not turn off the Diamond Storage Array until the display returns the Ready prompt.

8 When the flash procedure is complete, cycle

power on the array.

Updating firmware using the optional Ethernet card

To update the firmware via the optional Ethernet management services card, you will need

• the optional Ethernet management services card installed in your array

• a host computer with a network card or a network-connected device such as a hub

• binary information file (“...”.ima) available from technical support or on our website (refer to

Contact ATTO Technology, Inc.

• a crossover network cable for a direct connect or standard network cable if attached to a network device

• a valid IP address

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

on page viii)

Procedure

1 Connect a cross-over cable (for a direct

connection to a PC) or regular network cable from a network device to the optional RJ45 Ethernet port on the Ethernet management card on the front of the Diamond Storage Array.

Or, attach a DB-9 null modem serial cable from the RS-232 port of the Diamond Array to a host computer and open a terminal emulator program on the host to set the Ethernet parameters.

2 Power on and boot up the host computer.

3 Turn on the Diamond Storage Array.

4 Copy the latest Diamond Storage Array image

file (“...”.ima) onto the host computer and note its directory

The Diamond should acknowledge receiving the

file and display a message not to interrupt power for 90 seconds.

c:\diamond\flash\

5 Change directories to the place where you

copied the “...”.ima file

cd c:\diamond\flash

6 Open an FTP session using a user-defined IP

address or the default (10.0.0.1) if it has not changed. The IP address must be a valid address for your network.

7 At the FTP login prompt, type sysadmin as the

userID.

8 Press return at the password prompt.

9Type put “...”.ima

“...

”.ima

CAUTIONCAUTION

Do not interrupt the flash process. If the process is interrupted, the Diamond Storage Array will become inoperable and will have to be returned to the factory for repair.

Do not turn off the Diamond Storage Array until the display returns the prompt.

10 If upgrading the firmware from versions older

than 2.5.3, follow the procedures outlined in

Resetting Defaults

11 When the flash procedure is complete, cycle

power on the array.

on page 85.

Ready

10.0 System Monitoring and Reporting

RS-232 monitoring port and CLI

One tool to support a Diamond Storage Array is a host computer with an RS-232 port and terminal emulation software. Using this computer, you can hook up a null modem serial cable to the RS-232 port on the Diamond Storage Array management card and control the array via the Command Line Interpreter (CLI) management software built into the array. The CLI can be used to configure the unit, modify key parameters and read back key system information. (refer to

Accessing the Array

page 17)

Ethernet monitoring port and CLI

If the optional Ethernet management system card has been installed, you may use the RJ45 Ethernet port and Telnet to access the CLI software on the Diamond Storage Array. Use the same CLI commands or the ExpressNAV browser-based interface to configure, modify or read key systems information (refer to

based Interface

on page 25). If you restart the array,

ATTO ExpressNAV: Browser-

the Telnet session will end and the session cannot be re-established until the Array completes its Power On Self Test.

Power On Self Test (POST)

Each time the Diamond Storage Array array is powered up, it will perform a series of internal tests called POST (Power on Self Test). The POST sequence takes from 1 to 2 minutes to complete.

The Diamond Storage Array’s LEDs will blink at various points in the test sequence and, if the RS232 system management port is connected, a list of tests and test results will scroll across the screen. When the POST is nearly complete, all the LEDs on the array will flash simultaneously twice in a row. If the system is fully operational, the

ready LED will be lit and the unit is available. If the POST fails, the System Fault LED will be lit and an error code will be flashed.

Ready LED

The Ready LED (RDY) indicates the operational status of the Diamond Storage Array.

At power up, the green Diamond Array Ready (RDY) LED is disabled until successful completion of the POST (Power On Self Test) indicating the array is available for normal operation. If a host computer is hooked to either Fibre Channel or SCSI port, the Ready LED will momentarily blink and then stay lit if the connection is good. If that host is rebooted, the Ready LED will blink again and stay on once the connection is reestablished.

Audible alarm

The Diamond Storage Array audible alarm warns of potential problems or faults. It beeps twice at power up to indicate it is operational and will beep repeatedly when the System Fault (FLT) light on the system management card is activated. The alarm will turn off when the fault condition is cleared or the alarm can be disabled by issuing a

PowerAudibleAlarm

CLI as described in

AudibleAlarm

Diagnostic Commands

command via

on page

41. Disabling the alarm for a specific error will not silence the alarm for other errors.

Thermal monitoring

The Diamond Storage Array provides excellent advance warning of problems in temperature through a variety of warning mechanisms (visual, audible and software) and through an automatic system which protects the disk drives under abnormal operating or ambient conditions.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

The Diamond Storage Array will function under normal external operating temperatures of 5°C and 40°C. Temperature sensors provide data to the software to trip a temperature warning alarm and will, at excessively high or low temperatures, disable disk drive activity to protect the drives. The Diamond reports the temperature via SES, CLI and the ExpressNAV interface as well as the

state of the temperature warning alarm (

Present, OK, Warning, Critical

). The temperature

warning alarm will report OK during normal operating conditions.

In an abnormal operating condition, such as blower failure, under high ambient temperatures when the Diamond internal midplane temperature reaches 47°C, the temperature alarm will report

Warning

and activate the audible alarm and fault LED on the front of the unit. If the internal midplane temperature were to reach 52°C, the temperature warming alarm will report

Critical

, the array will be taken off line, and all disk drive activity will be disabled. The exact external ambient conditions which the Array will be taken off line depends on data activity, but will occur between 38 and 42°C.

Exhibit 10.0-1 Typical Diamond Storage Array operation over a range of external ambient temperatures. Alarms will be audible, visual, CLI and SES cues as described above.

Temp eratu re

Condition

Read/Writes normal normal normal normal normal normal normal

Alarms OK OK OK OK OK

25°C

blowers OK

25°C

1 blower fails

32°C

blowers OK

32°C

1 blower fails

35°C

blowers OK

35°C

1 blower fails

WARNING OK CRITICAL

40°C

blowers OK

40°C

1 blower fails

Array offline

Power supply monitoring

The Diamond Storage Array monitors the operation of its power supplies and blower assemblies every 30 seconds. If the status changes, the system reports it visually with an LED and a message in the CLI.

If a power supply or blower fails, the management system sends a CLI message and turns off the corresponding power supply LED.

The blower assembly directly adjacent a power supply must be functioning properly for the power supply to work properly. If a blower assembly fails, the power supply will shut down and the management system will send a CLI message. The corresponding power supply LED on the system management card will also turn off.

If a good replacement power supply or blower assembly is inserted into the array, the management system will send a CLI message and the corresponding system management card LED will go on. It may take up to 30 seconds for the system to note these changes.

System fault LED and error codes

If a serious hardware or software error occurs in the Diamond Storage Array, the System Fault LED will display a series of flashes or blink codes. Error information will be reported via the Command Line Interface if it is operational.

At power up, the green Diamond Array Ready (RDY) LED turns on after successful completion of the POST (Power On Self Test) indicating the array is available for normal operation.

During operation, the Ready LED will stay lit even if the amber Fault (FLT) LED lights.

If a serious error occurs in the array, the amber Fault LED will flash repeatedly in a blink code pattern

• an initial series of blinks indicating the system

problem

Monitors, reports

Number of blinks Problem area

1, 2 or 3 processor or memory

4 Fibre Channel interface

5 SCSI interface

7 Fibre Channel connection

8 general internal processing

9 SCSI Enclosure Services

• a two second pause

• another series of blinks providing more detailed information for technical personnel.

• a four-second pause

• the blink code sequence will repeat from step 1 until the error is cleared.

In general, any fault will require notification of Diamond Array technical personnel for resolution or for further debug instructions. When you report an error code, please provide both the first and second blink code values.

drive activity is present. As the disk drives are accessed, the green LEDs will flash. If the disk drives are heavily accessed the green activity LEDs will appear to flash at a high rate or may even appear to be completely turned off.

The amber Disk Fault LED will be off under normal operation. If either of the disk drives on a dual disk drive assembly reports a disk error of any kind, the amber Disk Fault LED will come on. The Disk Fault LED can be activated by minor issues such as a disk drive writing to a bad sector (which is usually corrected by the disk drive the next time it writes) or major issues such as a head crash or complete drive failure.

When the Disk Fault LED is turned on, the system will issue a detailed message via the CLI. These messages are not written permanently to the error log file but should be recorded to help assess the disk problem.

During a fault condition, more detailed information about the fault may be available via the CLI over the RS-232 interface port or the optional Ethernet port. These error messages should be reported to technical personnel to assist in debugging the problem.

The blink codes are also saved internally by the array to NVRAM (Non Volatile Random Access Memory) and will be displayed at power up if the power to the array is recycled.

Disk drive activity and disk fault LEDs

Each dual disk drive sled assembly contains two green activity LEDs and an amber Disk Fault LED. Once the system has successfully powered up and passed POST, the green activity LEDs are full on and the Disk Fault LED off.

The Drive 1 and Drive 2 activity LEDs will stay full on when the system is operational and no disk

If the disk drive error is a non fatal error and the drive is still functional, the array will continue to read and write data to the disk drive but the Disk Fault LED will remain on. If you repeat a drive command or action and it completes successfully, the Disk Fault LED may have been set by an anomaly in the disk drive. You can clear the Disk Fault LED by either power cycling the array or issuing the SledFaultLED command in CLI as per

Diagnostic Commands

on page 41

If you repeat a disk command or action and the Disk Fault LED remains on, the disk error may be serious. You should write down the error message issued by the CLI and contact technical support via the means easiest for you for assistance (refer to

Contact ATTO Technology, Inc.

on page viii). If you choose to replace the suspected faulty dual disk drive sled assembly, follow the appropriate procedures.

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

10.1 Troubleshooting

The Diamond Storage Array provides a number of visual, audible and computer system-generated indicators to identify the operational status of the array. System status and error information is readily available. If your situation is not defined here or elsewhere in the manual, if these solutions do not help, or if you have any questions or concerns about any aspect of operating the array, please contact technical support.

Windows 2000 special instructions

When using Windows 2000, the

New Hardware

screen pops up and asks for a driver when the array is first booted up. While a driver is not necessary for operation, you should install our dummy driver to eliminate the New Hardware screen’s appearance. Download

AttoDM2k.zip

from our website, www.attotech.com, unzip it, and install the driver instructions in

AttoDM2k.inf

AttoDM2k.pdf

according to the

Error messages

System Fault LED

If a serious hardware or software error occurs in the array, the System Fault LED will display a series of flashes or blink codes in a pattern.

• an initial series of blinks indicating the system

problem

Number of blinks Problem area

1, 2 or 3 processor or memory

4 Fibre Channel interface

5SCSI interface

7 Fibre Channel connection

8 general internal processing

9 SCSI Enclosure Services

• a two second pause

• another series of blinks providing more detailed

information for technical personnel.

• a four-second pause

• the blink code sequence will repeat from step 1

until the error is cleared.

In general, any fault will require notification of Diamond Array technical personnel for resolution or for further debug instructions. When you report

an error code, please provide both the first and second blink code values.

During a fault condition, more detailed information about the fault may be available via the CLI over the RS-232 interface port or the optional Ethernet port. The blink codes are also saved internally by the array to NVRAM (Non Volatile Random Access Memory) and will be displayed at power up if the power to the array is recycled.

Command Line Interface messages

ERROR. Wrong/Missing Parameters

Check

Help

for the correct input and retype

command

ERROR. Invalid Command. Type ‘help’ for command list

Check

Help

to find a list of all commands which are available. Contact array technical support via the means easiest for you. Refer to

Technology, Inc.

on page viii for additional

Contact ATTO

information.

ERROR. Command Not Processed.

The array did not accept the command you requested. Check

Help

for a list of commands or check this manual for the function you wish to access. If you cannot accomplish what you want to do with the commands listed, contact array technical support via the means easiest for you (refer to

Contact ATTO Technology, Inc.

on page viii)

for more information.

Audible Alarm

The array audible alarm will warn of potential problems or faults. It will beep repeatedly when the System Fault (FLT) light on the system management card is activated. The

Troubleshooting

alarm will turn off when the fault condition is cleared or the alarm can be disabled by issuing a

PowerAudibleAlarm

CLI as described in

AudibleAlarm

command via

Diagnostic Commands

on page

41. Disabling the alarm for a specific error will not silence the alarm for other error conditions.

arrays with two operational power supplies and blower assemblies will support Command Overlap: the array will automatically disable the Command Overlap feature if a power supply fails and notify the system administrator through the audible alarm and CLI.

If you cannot read or write to the array

Specific situations and suggestions

For all problems, first check the pages of the ExpressNAV interface appropriate for the problem especially the

Management

or use the appropriate CLI

Status and Storage

commands

If a drive fails to respond

• Determine which drive has failed by observing the amber fault LED on the drive sled or connect to the CLI and type driveinfo drives and their status.

• For specific information on a particular drive in CLI, type number]. Record all errors.

• Determine if the drive is configured in JBOD or RAID Level 0.

• Follow the instructions in the appropriate chapters of this manual for removing, replacing and reconfiguring the drive (begin with

driveinfo [sled number] [drive

Operating Instructions on page 71

• You may copy drives by using Command Line interface commands. (Refer to

Configuration Commands

If a power supply fails

• Verify the power cord is correctly plugged and there is power at the power receptacle.

• If there is power, the cord is secure and the blower and power supply LEDs are off, replace the blower unit (refer to

Instructions on page 71

• If there is power, the cord is secure and the power supply LED is off, but the blower LED is on, replace the power supply (refer to

Operating Instructions on page 71

• Command Overlap: The array contains special software to take advantage of the Command Overlap feature offered in some high performance disk drives. Because all 24 disk drives in a Diamond Array may be seeking at the same time under Command Overlap, only

for a list of all

Hot Swap

)

Drive

on page 45).

Hot Swap Operating

Hot Swap

)

• You may have lost connection to the host via the Host Interface Card. The Host Interface Card LED on the back of the array should be lit green. The amber LED should go out and the green LED should light when the connection is complete.

• Also check the host bus adapter (HBA) in the host machine for proper functioning and drivers.

• Verify that the connector and both ends of the cable are completely seated.

• Try connecting directly to the host, bypassing any hubs or switches.

To determine if the problem exists with the HIC

• Swap the cable from one HIC to the other HIC.

• If the LED is now green, swap the cable back to the original HIC. If the LED is amber, the HIC is defective.

• Swap the defective HIC (refer to

Operating Instructions on page 71

If you can’t access the CLI via Ethernet

• Verify there are three or fewer concurrent sessions using Telnet or FTP. You may be the fourth session or someone in another session has entered a command which requires a Save Configuration command.

• Verify you are using a crossover cable for direct connection, or a network cable for a network connection.

• Verify the array is set to expect an IP address will be assigned by the network and DHCP is an option available on the network.

• Verify the IP address is compatible with the host machine.

• Verify the system is functioning and accessible via in-band inquires such as Disk Management.

• Try setting your terminal emulator with a different baud rate, starting at 2400, then 9600, 19200, 38400, 57600, and 115200.

IPDHCP enabled

Hot Swap

)

if you

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

• If you perform a hardware restart, the Ethernet connection is dropped. You must re-establish the Ethernet connection.

• If you enter a command that requires a

Configuration

interface window or the Ethernet connection window, you will not be able to access the inactive window until the command is complete in the active window.

command in either the serial

Save Configuration

Save

• If you still fail to communicate with the array, swap out the management card and try to connect using default settings.

If LUNs do not display on the host

• Ensure any configuration changes are appropriate

• Type FirmwareRestart

• Re-start the host computer

• Verify all drives associated with the missing LUN(s) are inserted properly and powered up.

Troubleshooting

10.2 Resetting Defaults

Resetting the Diamond Storage Array to defaults will not alter the RAID configuration, IP configuration or Telnet information. However, resetting the Diamond Storage Array to factory defaults is a last-ditch effort to recover from corrupt configurations or complete failure. All data will be lost.

Default

If you need to return to the default settings of the Diamond Storage Array but do not want to lose data or Ethernet settings, use the CLI command

RestoreConfiguration default

in the CLI or in the

Advanced CLI Configuration page in the

ExpressNAV interface.

To return to Default settings

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

Command Line Interface.

2 Continue with the CLI or access the Advanced

CLI page of the ExpressNAV interface (refer to

on page 17). You should now be in

ATTO ExpressNAV: Browser-based Interface

page 25).

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

3Type RestoreConfiguration default to reset

the system configuration (See Exhibit 10.2-1 for a list of configurations which will be changed).

4Type FirmwareRestart

Diamond Storage Array.

5 Reboot the host PC after the Diamond Storage

Array completes its power on cycle.

or cycle power of the

Accessing

Factory Default

Resetting the Diamond Storage Array to factory defaults is a last-ditch effort to recover from corrupt configurations or complete failure. All data will be lost.

CAUTIONCAUTION

Data will be lost if you follow these procedures. Make sure you have no other choice before resetting the Diamond Storage Array to factory defaults.

To reset to factory defaults firmware version 2.5.3 or higher

1 Connect to Diamond Storage Array services via

the RS-232 port or the optional Ethernet management services card (refer to

the Array

Command Line Interface.

2 Continue with the CLI or access the Advanced

CLI Configuration page of the ExpressNAV interface (refer to

based Interface

3Type RestoreConfiguration factorydefault to

reset the system configuration. (See Exhibit 10.2-1 for a list of configurations which will be changed).

4Type FirmwareRestart or cycle power.

5 Reboot the host PC after the Diamond Storage

Array completes its power on cycle.

on page 17). You should now be in

ATTO ExpressNAV: Browser-

on page 25).

Note

Because the ExpressNAV pages take you through this process easily, the following instructions are based on the CLI commands. Use these instructions as a guide in ExpressNAV.

Accessing

ATTO Technology Inc. Diamond Storage Array Installation and Operation Manual

Exhibit 10.2-1 Configurations which will be changed during a

RestoreConfiguration

command

Command Default

AudibleAlarm Disabled

AutoRebuild Disabled

DiamondName “

EthernetSpeed Auto

FcConnMode Loop

FcDataRate Auto

FcFairArb Enabled

FcFrameLength 2048

FcFullDuplex Enabled

FcHard Disabled

FcHardAddress 0x03

IdentifyDiamond Disabled

IdeTransferRate 4

IPAddress 10.0.0.1

IPDHCP Disabled

IPGateway 0.0.0.0

IPSubnetMask 255.255.255.0

MaxEnclTempAlrm 47

MinEnclTempAlrm 5

PowerAudibleAlarm Enabled

QuickRAID0 0

RAIDInterleave 128

SerialPortBaudRate 115200

SerialPortEcho Disabled

SerialPortHandshake None

SerialPortStopBits 1

TelnetPassword diamond

TelnetTimeout Disabled

TelnetUsername telnet

VerboseMode Enabled

............

”

Reset if default chosen

✔✔

✔✔ ✔✔ ✔✔ ✔✔ ✔✔ ✔✔ ✔✔ ✔✔ ✔✔

✔✔ ✔✔ ✔✔

✔✔ ✔✔ ✔✔ ✔✔

✔✔

Reset during

factorydefault

✔ ✔ ✔

✔ ✔ ✔ ✔

✔ ✔

✔ ✔ ✔

Appendix A ATA Disk Technology

ATA is the dominant disk drive technology today and will be for the foreseeable future. It offers all the cost advantages of a mass produced, consumer-driven technology which is rapidly being driven forward. Today ATA also offers all the performance and reliability features needed to create high performance ATA-based disk storage arrays.

We view ATA as the perfect disk drive technology for “in-the-box” disk storage array solutions. Aggregated Data Interface Technology (ADXT

Fabric (DRF) and ATA Interface Engines (AIE) smoothly and efficiently organizes Fibre Channel data streams for storage or retrieval on ATA disk drives. ADXT provides end users with the power and sophisticated data control needed to take moderately priced ATA disk drives combine them in a disk storage array and power them to the performance levels of SCSI or Fibre Channel disk arrays

Unlike other storage arrays which use expensive SCSI or Fibre Channel disk drives to achieve performance, the Diamond Series uses lower cost ATA disk drives combined with an intelligent midplane to create a storage array with exceptional price and performance characteristics.

ATA disk drives are used in the enormous PC marketplace and an estimated 85% of all disk drives sold today contain an ATA interface. Shipments of ATA drives are expected to grow rapidly over the next five years with increasing consumer and business demand. In general, ATA disk drives are the “driver” of all disk storage technology as the costs, volumes and demands of the PC marketplace are the primary focus of the world’s largest disk drive manufacturers.

ATA disk drives today offer capacities comparable to SCSI and FC disk drives – although at a tremendously lower cost. Advances in platter and head technology will push drive capacity along a very steep curve in the next five years. Interface advances are also planned for ATA drives with an Ultra100 ATA interface expected in 2001. This interface will provide low cost, simple and reliable communication just as ATA66 does, but at a much higher data rate.

Exhibit A-1 summarizes ATA and other disk interface standards. It clearly indicates that ATA drive technology offers characteristics that ideally suit it for “in the box” usage in high capacity, high performance disk arrays.

Exhibit A-1 Disk features and standards

Feature IDE ATA SCSI Fibre Channel

Connectivity Max Devices Cable Length

Performance Bandwidth Bandwidth/drive Multi-threaded I/O

Max Initiators 1 1 <16 125...millions Topology Bussed Bussed Bussed Loop/Fabric Error Detection None Yes Yes Yes Protocol “weight” very light light medium to heavy heavy Cost

Interface Drives

Manageability none low to medium medium medium+

46 cm

16 MB/sec.

8 MB/sec.

n/a n/a

™

) with its built-in Virtual Device Manager (VDM), Data Routing

46 cm

66 MB/sec. 33 MB/sec.

Yes

lowest lowest

25 m

160 MB/sec.

10 MB/sec.

Yes

medium to high medium to high

126...millions 10 km

200 MB/sec.

1.6 MB/sec. Yes

highest

medium to high

ATTO Technology Inc.

Appendix B Information Commands Results

The following shows some of the output from diagnostic commands available through the CLI or the Advanced page of the ExpressNAV interface.

DriveInfo [sled ID] [drive ID]

Displays information about all disk drives or detailed information about a specific disk drive. Detailed information about an individual drive is obtained by supplying a drive identifier.

Driveinfo for a JBOD configuration, verbose result

=========== Drive 1 ========== ========= Drive 2 ============== Sled Capacity Errors TYPE ...VDID Capacity Errors TYPE VDID === ====== ..===== ===== ===== ====== ===== =====. ===== 1 76292MB 0 JBOD 1 76292MB 0 JBOD 13 2 76292MB 0 JBOD 2 76292MB 0 JBOD 14 3 76292MB 0 JBOD 3 76292MB 0 JBOD 15 4 76292MB 0 JBOD 4 76292MB 0 JBOD 16 5 76292MB 0 JBOD 5 76292MB 0 JBOD 17 6 76292MB 0 JBOD 6 76292MB 0 JBOD 18 7 76292MB 0 JBOD 7 76292MB 0 JBOD 19 8 76292MB 0 JBOD 8 76292MB 0 JBOD 20 9 76292MB 0 JBOD 9 76292MB 0 JBOD 21 10 76292MB 0 JBOD 10 76292MB 0 JBOD 22 11 76292MB 0 JBOD 11 76292MB 0 JBOD 23 12 73307MB 0 JBOD 12 73307MB 0 JBOD 24

VD ID is the Virtual Disk ID

DriveInfo 7 2

JBOD VDID 19 Sled = 7, Drive = 2 Drive Model: Maxtor 98196H8 S/N: V80D15GC F/W Rev: ZAH814Y0 ATA Rev: 007e Capacity: 76292 MB (09502790 blocks) PIO: 03 MDMA: 07 UDMA: 07 QD: 00 CMDSet: 7c69 4309

Operations Blocks Read 00000000 00000000 Write 00000000 00000000 Errors 00000000

DriveInfo RAID0 6

========= Drive 1 =========== ========= Drive 2 ========== Sled Capacity Errors TYPE VDID Capacity Errors TYPE VDID === ======= ==== ==== ====== ======= ==== ===== ===== 1 76292MB 0 RAID0 26 76292MB 0 RAID0 26 2 76292MB 0 RAID0 26 76292MB 0 RAID0 26 3 76292MB 0 RAID0 27 76292MB 0 RAID0 27 4 76292MB 0 RAID0 27 76292MB 0 RAID0 27 5 76292MB 0 RAID0 28 76292MB 0 RAID0 28 6 76292MB 0 RAID0 28 76292MB 0 RAID0 28 7 76292MB 0 RAID0 29 76292MB 0 RAID0 29 8 76292MB 0 RAID0 29 76292MB 0 RAID0 29 9 76292MB 0 RAID0 30 76292MB 0 RAID0 30 10 76292MB 0 RAID0 30 76292MB 0 RAID0 30 11 76292MB 0 RAID0 31 76292MB 0 RAID0 31 12 73307MB 0 R]AID0 31 73307MB 0 RAID0 31

Diamond Storage Array Installation and Operation Manual

DriveInfo 3 1

RAID0 VDID 27 State: ONLINE RAID0 Member Index 0 of total members 4 disks RAID0 Interleave Size 128 Blocks Sled = 3, Drive = 1 Drive Model: Maxtor 98196H8 S/N: V80DTESC F/W Rev: ZAH814Y0 ATA Rev: 007e Capacity: 76292 MB (09502790 blocks) PIO: 03 MDMA: 07 UDMA: 07 QD: 00 CMDSet: 7c69 4309

Operations Blocks Read 00000000 00000000 Write 00000000 00000000 Errors 00000000

Info [all]

Displays version numbers and other key information about the Diamond Storage Array including data rate, connection mode, WorldWideName, Diamond Storage Array name.

Info

Device = Diamond Array Serial Number = MIDP100197 Device Version = 016G Device Build = 016G Build Date = "Feb 19 2003" 15:52:01 NVRAM Revision = 0 CLI Revision = 0.21 Interface 0 Type = "1Gb Fibre Channel" FC0 Firmware Revision = 2.01.26 FC0 Data Rate = 1Gb FC0 Connection Mode = ptp Interface 1 Type = None Diamond Array Name = " " World Wide Name 0 = 20 00 00 10 86 10 02 dc World Wide Name 1 = 20 00 00 10 86 10 02 dc Active Configuration = ATTO DiamondClass =(V)86

Info all

iii

ATTO Technology Inc.

ATTO Technology Diamond Array S Installation And Operating Manual

Specifications and Main Features

Frequently Asked Questions

User Manual