imageRAID VT-100 Text-based Interface Software Guide€¦ · SOFTWARE GUIDE VT-100 imageRAID® Series Storage System (Text-based User Interface) VT100_Wahoo-TUI.book Page 1 Tuesday,

VT100_Wahoo-TUI.book Page 1 Tuesday, September 23, 2003 5:26 PM

S O F T W A R E G U I D E

VT-100i m a g e R A I D® S e r i e s S t o r a g e S y s t e m

( T e x t - b a s e d U s e r I n t e r f a c e )



VT-100i m a g e R A I D® S e r i e s S t o r a g e S y s t e m

( T e x t - b a s e d U s e r I n t e r f a c e )

S O F T W A R E G U I D E


Fujitsu Europe Limited

Restricted Rights and Liability

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

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

Copyright

Fujitsu Europe LimitedHayes Park CentralHayes End RoadHayes, Middlesex, England UB4 8FE

Copyright © 2003 Fujitsu Europe Limited. All Rights Reserved.

imageRAID and the imageRAID logo are registered trademarks of Fujitsu Europe Limited, Fujitsu is a registered trademark of Fujitsu Limited.

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

Contents

VT100_Wahoo-TUI.book Page i Tuesday, September 23, 2003 5:26 PM

About this Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vWelcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vNumbering Convention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viTypographical Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi

1 Accessing the Configuration Utility Software . . . . . . . . . . . . . . . . . 1Using the VT-100 Terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Navigating the Configuration Software . . . . . . . . . . . . . . . . . . . . 3Menu System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Controller Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Troubleshooting Assistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2 Creating Disk Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12Optimization and Drive Selection for RAID 5 Arrays . . . . . . . . . . 13

Creating Arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14Configuring Array Writeback Cache . . . . . . . . . . . . . . . . . . . . . . 23Configuring Array Readahead Cache . . . . . . . . . . . . . . . . . . . . . 26Assigning Hot Spare Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28Deleting an Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31Expanding an Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33Trust an Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

3 Host LUN Assignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Accessing the Host LUN Configuration Menu . . . . . . . . . . . . . . . 40Viewing Unassigned Free Space . . . . . . . . . . . . . . . . . . . . . . . . . 41Creating a Host LUN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42Set Host LUN Availability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45Tips on Setting LUN Availability . . . . . . . . . . . . . . . . . . . . . . . . . 47

Modifying Mapped LUN Number . . . . . . . . . . . . . . . . . . . . . . . . 53

i

Table of Contents

ii

VT100_Wahoo-TUI.book Page ii Tuesday, September 23, 2003 5:26 PM

4 SAN LUN Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

Accessing the SAN LUN Mapping Configuration Menu . . . . . . . . 56Viewing SAN LUN Mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . 57Creating a SAN LUN Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . 58Delete SAN LUN Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62Modify SAN LUN Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63Viewing Connected Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

5 Advanced Configuration and Management . . . . . . . . . . . . . . . . . . 67Hot Spare Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67View Hot Spare Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67Add Hot Spare Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68Delete Hot Spare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70Enable/Disable Auto Rebuild . . . . . . . . . . . . . . . . . . . . . . . . . . 72Auto Hot Spare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Rebuild Priority . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75UPS Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76Ethernet Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

6 Additional Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Host Port Addressing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77Setting the Controller LUN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80Modify Configuration Name . . . . . . . . . . . . . . . . . . . . . . . . . . . 82Modify Configuration WWN . . . . . . . . . . . . . . . . . . . . . . . . . . . 84Setting Single Controller Mode . . . . . . . . . . . . . . . . . . . . . . . . . 87Background Verification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89Controller SES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91

7 Monitor Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93Using the Monitor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94Monitor Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

Updating the RAID Controller Firmware . . . . . . . . . . . . . . . . . . . 99

8 Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103Offline Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105Online Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111Controller Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112Drive Maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114RAID 5/50 Parity Check . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121Command Tracing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

Table of Contents

VT100_Wahoo-TUI.book Page iii Tuesday, September 23, 2003 5:26 PM

Resetting the Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125Shutting Down the Controllers . . . . . . . . . . . . . . . . . . . . . . . . . . 126Diagnostics Dmp . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

9 Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129Access Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129Readahead Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133Command Cluster Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135Miscellaneous Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

10 Event Logs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137Accessing Event Logs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137Controller Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140Drive Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145Host Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151SES Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153Failed Drives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

A Optimizing RAID 5 Write Performance . . . . . . . . . . . . . . . . . . . . . . 157Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157Sequential Access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158Number of Outstanding Commands . . . . . . . . . . . . . . . . . . . . . . 158Access Size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159Access Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159RAID 5 Sub-Array . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160Multiple Drive Failures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161Faster Rebuild . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163

iii

Table of Contents

iv

VT100_Wahoo-TUI.book Page iv Tuesday, September 23, 2003 5:26 PM

Preface

VT100_Wahoo-TUI.book Page v Tuesday, September 23, 2003 5:26 PM

About this Manual

Welcome

Congratulations on the purchase of your new imageRAID™ Series Storage System

from Fujitsu Europe Limited. This software guide provides the operational and

reference information to configure and manage the imageRAID Controllers

installed in your storage system. The controllers are accessed through a VT-100

terminal via a text-based user interface connected directly to the controller.

Although the target audience is experienced system administrators who are

familiar with the principles and conventions of Small Computer System Interface

(SCSI), Fibre Channel Interface, and Redundant Array of Independent Disk

(RAID) technology, you will find step-by-step procedures to perform each of the

functions of the configuration utility software onboard the controllers and the

enclosure monitoring system.

v

About this Manual

vi

VT100_Wahoo-TUI.book Page vi Tuesday, September 23, 2003 5:26 PM

Numbering Convention

The capacity of most disk drives is reported from the Controller using decimal GB

(1 GB = 1,000,000,000 bytes). Many operating systems use the definition of 1 GB

= 1,073,741,824 bytes. This can cause a discrepency of approximately 7%

between what the controller reports and what the operating system reports.

Therefore, a drive labeled 73 GB has a true physical capacity of 68 GB (binary).

NOTE: The RAID Controller also reserves some capacity for configuration data.

Typographical Conventions

The following typographical conventions are used in the user’s guide:

� Menu items are displayed in the format: “Array Configuration menu,

choose View Unused Drives.”

� Code font will indicate literal text used in examples.

� Italic code font indicates a replaceable or variable item in code.

� Italic text indicates the item that is selected or chosen.

� Key strokes are enclosed in brackets, e.g., <Esc>, <K>, or <Enter>.

Numbering Convention

Chapter 1


Accessing the Configuration Utility Software

The text-based user interface can be used to control and manage the imageRAID

Controllers. You can:

� Create and manage disk arrays

� View detailed information about each controller

� Manage drive spares

� Configure the controller

� Monitor system status

� Manage disk drives and enclosures

� Perform Online and Offline Diagnostics

� Review Event Logs

You access the controllers using a VT-100 terminal connected to the RS-232 serial

port on the controller’s face plate. Configure your terminal using the parameters

described on the following page.

NOTE: Access to each RAID Controller is provided through independent

controller serial ports.

1

Chapter 1 - Accessing the Configuration Utility Software

2


Using the VT-100 Terminal

You access the imageRAID Controller via its onboard Configuration Utility software

using a VT-100 terminal and the RS-232 serial ports on the controller.

A null-modem serial cable is used to connect the terminal to the serial port on the

controller.

Configure your host system or terminal RS-232 port to use the following settings

Set the communications parameters for the terminal program as follows

To access the imageRAID Controllers

1 Power on the imageRAID Series Storage System.

2 From the computer or terminal connected to the RAID Controller, start your

terminal emulation software.

Be sure that your terminal emulation software is set to use the correct COM

port on your computer.

Setting Value

Terminal Emulation ANSI

Function Terminal Keys

Fonts Terminal

Translations None

Columns 80

Setting Value

Baud Rate 9,600 - 115,200*

Data Bits 8

Stop Bits 1

Parity None

Flow Control None

Connector COM1 (typically)

*The default baud rate is 19,200, however it is recommended to use 115,200 for optimum results.

Using the VT-100 Terminal



3 From the terminal window, press the <Control>+<W> keys.

Main Menu Screen

You can now perform all of the functions described in the following chapters. All

steps start from the Main Menu.

If an event or error occurs, you can review the log of the problem by

examining the event logs.

Navigating the Configuration Software

You can navigate the Configuration utility software menu system using the keys

on your keyboard. The table below describes the primary navigation and

selection methods.

To Do this

Select the Main menu. Press <Control>+<W>

Return to the previous menu or screen (Exit).

Type <0> and press the <Enter> key.

Select a menu option. Press the specific numeric key for the menu choice.

Navigating the Configuration Software 3


4


Menu System

Configuration Software Utility Menu Chart 1 of 3

Main Menu

ConfigurationDiagnosticsController InformationStatisticsEvent LogsRS232 Settings

Configuration Menu

imageRAID Main Menu

Array ConfigurationHost LUN ConfigurationAdditional Host ConfigurationSAN LUN Mapping ConfigurationRebuild ConfigurationUPS ConfigurationEthernet ConfigurationAdditional ConfigurationView ConfigurationClear Configuration

Controller Information

View Controller 0View Controller 1



Statistics

Event Logs

RS232 Setting

1152005760038400192009600

View ArrayCreate ArrayDelete ArrayExpand ArrayArray Cache ConfigurationView Unused DrivesAdvanced Array Configuration

Array Configuration

Host LUN Configuration

Additional Host Configuration

View Host LUNCreate Host LUNDelete Host LUNExpand Host LUNSet Host LUN AvailabilityView Unassigned Free SpaceAdvanced Host LUN Configuration

Host Port Hard AddressingController LUN ConfigurationHost Port Connection OptionsHost Port Data RateHost Port Different Node Name

SAN LUN Mapping Configuration

View SAN LUN MappingsCreate SAN LUN MappingDelete SAN LUN MappingModify SAN LUN MappingView Connected Hosts

Rebuild ConfigurationView Hot Spare DrivesAdd Hot Spare DrivesDelete Hot Spare DrivesEnable/Disable Auto Rebuild Set Rebuild PriorityEnable/Disable Auto Hot Spare

Diagnostic

Offline DiagnosticsOnline DiagnosticsReset ControllerShutdown ControllerDiagnostic Dump

Menu System



Menu System (continued)


Configuration Menu

Array ConfigurationHost LUN ConfigurationAdditional Host ConfigurationSAN LUN Mapping ConfigurationRebuild ConfigurationUPS ConfigurationEthernet ConfigurationAdditional ConfigurationView ConfigurationClear Configuration

Controller A UPS SetupController B UPS Setup

Set Controller IP AddressRemote Mirror ConfigurationAccept IP Address ConfigurationEmail Notification Configuration

UPS Configuration (not implemented)

Ethernet Configuration (not implemented)

View Configuration

Clear Configuration

Additional Configuration

Modify Configuration NameModify Configuration WWNSet Single Controller ModeEnable/Disable Background VerificationEnable/Disable Controller SES

Array Configuration

Advanced Host LUN Configuration

Create Array

Initialize ArrayTrust Array

Displays Array Information

Displays Unused Drives

Select Array

Select ArrayDelete Array

Select ArrayExpand Array

Readahead CacheWriteback Cache

Select RAID LevelManual ConfigureChunk SizeSelect Drives

View ArrayCreate ArrayDelete ArrayExpand ArrayArray Cache ConfigurationView Unused DrivesAdvanced Array Configuration

Menu System 5


6


Menu System (continued)


Displays Controller Parameters

View All StatisticsView Access StatisticsView Readahead StatisticsView Command Clustering StatisticsView Miscellaneous StatisticsClear StatisticsView Access Statistics On Each LUNClear Access Statistics On Each LUN

View All Events <4096>View New Events <22>View Last 50 EventsView Last Hour EventsView Last 24 Hour EventsView Drive Loop EventsView Host Loop EventsView Environment EventsView Failed Drive ListClear Event Log





Statistics

Event Logs

Main Menu

ConfigurationDiagnosticsController InformationStatisticsEvent LogsRS232 Settings

imageRAID Main Menu

Menu System




Detailed information about the imageRAID Controller is available through the

VT-100 interface and includes:

� Controller WWN

� Firmware Information

� Run-time Information

� Loop Information

� Temperature and Voltage Information

To view controller information:

1 After the controller(s) have completed booting, enter the Main menu by

pressing the <Control>+<W> keys.

2 Choose “Controller Information” type <3> and press <Enter>.

Main Menu Screen

3 Select the controller you wish to examine. Type <1> or <2> and press

<Enter>. If only one controller is present, only one selection is listed.

Controller Information Menu

Controller Information 7


8


Below is an example of the controller information screen.

Controller Information Display Screen

The following is a brief summary of the information displayed.

Error Description

Controller WWN An 8-byte unique World Wide Name the controller uses for identification. This is unique to each controller and programmed at manufacture.

Configuration WWN An 8-byte World Wide Name the controller uses for identification to hosts. This may be identical to the Controller WWN, or may be different if the configuration was initially not created with this particular controller. This remains constant even if controllers are replaced. It is possible to modify this name to use the WWN of either controller or another controller.




Troubleshooting Assistance

To assist you when troubleshooting problems with your system, the Events

Chapter provides a list of all the events along with a complete description, and a

suggested cause and action. When events occur, you can refer to the specific

event in that chapter review a complete description of the event then determine

the possible cause or possible causes and the suggested action to take.

Config Name A 64-byte ASCII name used to identify a specific controller configuration.

Firmware Revision The firmware version in use by the controller.

Firmware Date/Time Date and time of when the firmware was created.

Total Power on Count The number of times this controller has been power cycled.

Total Power on Time The total length of time this controller has been powered on.

Current Power on Time The length of time since this controller has last been powered on or reset.

Current Controller Time This is the current time on the controller.

Last Configuration Time This is the time the controller was last configured.

Last Configuration Time This is the time the controller was last configured.

Host Loop x Status The current status of the specified host loop.

Drive Loop x Status The current status of the specified host loop.

Onboard Temperature This is the current temperature as measured by the controller. If this value exceeds predefined limits, an event will be written into the event log.

Controller Voltage Current readings for various controller and coprocessor voltages.

Error Description

Troubleshooting Assistance 9


10


Troubleshooting Assistance

Chapter 2


Creating Disk Arrays

Configuring a imageRAID system requires some planning to ensure that you

define the correct RAID level and array options. It may be helpful to refer to the

Topology chapter in the imageRAID Series User’s Guide for the specific topology

setup during hardware installation.

This chapter will step you through the process to configure your arrays. You are

then directed to the next chapter for the procedures to define the Host LUNs

which makes the drive array(s) available to the operating system.

This manual assumes you have a basic understanding of RAID concepts and

terminologies.

Arrays

You can create an array at any time. The table below describes the drive

requirements for each RAID level.

Drive Requirements by RAID Level

RAID Level Minimum No. of Drives Maximum No. of Drives

0 1 16

1 2 16

5 3 16

50 6 16

10 (Mirrored) 4 16

11

Chapter 2 - Creating Disk Arrays

12


NOTE: Before you create more than one array, you must be sure that your host

operating system supports multiple Logical Unit Numbers (LUNs). Most

operating systems do support multiple LUNs, or have an option for it. If

your operating system does not support multiple LUNs, the host will only

be able to see one array at the first disk LUN.

Terminology

The following describes the terminology used when creating and managing array’s.

Term Description

Array A group of disk drives that are combined together to create a single large storage area. Up to 64 arrays are supported, each containing up to 16 drives per array. There is no capacity limit for the arrays.

Chunk Size This is the amount of data that is written on a single drive before the controller moves to the next drive in the stripe.

Stripe Size This is the number of data drives multiplied by the chunk size.

Cache Flush Array This is the array that is used to automatically flush cache data in the situation where power has failed to some of the drives.

Initialization RAID 5/50 arrays must have consistent parity before they can be used to protect data. Initialization writes a known pattern to all drives in the array. If the user chooses not to initialize an array, this is known as a “trusted array” and any drive failure will result in data corruption. It is possible to later perform a parity rewrite, which recalculates the parity based on the current data, thus ensuring the data and parity are consistent.

Reserved Capacity In order to allow drives from a different family or manufacturer to be used as a replacement for a drive in an array, it is recommended that a small percentage of the drive’s capacity be reserved when creating the array. This is user selectable, from 0 to 10 percent.

RAID Level 0 RAID 0 is defined as disk striping where data is striped or spread across one or more drives in parallel. RAID 0 is ideal for environments in which performance (read and write) is more important than fault tolerance or you need the maximum amount of available drive capacity in one volume. Drive parallelism increases throughput because all disks in the stripe set work together on every I/O operation. For greatest efficiency, all drives in the stripe set must be the same capacity. Because all drives are used in every operation, RAID 0 allows for single-threaded I/O only (i.e., one I/O operation at a time). Environments with many small simultaneous transactions (e.g., order entry systems) will not get the best possible throughput.

Terminology



Optimization and Drive Selection for RAID 5 Arrays

Typical RAID 5 implementations require a number of steps to write the data to

the drives. In order to optimize your system performance based on the type of

writes you expect in your operation, we have provided detailed information of

optimizing the performance using full strip write operations in an appendix.

If you intend to setup a RAID 5 array and wish to consider optimum

performance, you will need to consider the number of drives, parity drives,

chunk size, and should review the information provided in “Optimizing RAID 5

Write Performance” on page 157. Additional information is provided at the

appropriate step during configuration.

RAID Level 1 RAID 1 is defined as Disk mirroring where one drive is an exact copy of the other. RAID 1 is useful for building a fault-tolerant system or data volume, providing excellent availability without sacrificing performance. However, you lose 50 percent of the assigned disk capacity. Read performance is somewhat higher than write performance because Windows NT reads data off the drive whose head is closest to the desired sector (called locality of reference); all write operations are made to both disks simultaneously.

RAID Level 5 RAID 5 is defined as disk striping with parity where the parity data is distributed across with parity all drives in the volume. Normal data and parity data are written to drives in the stripe set in a round-robin algorithm. RAID 5 is multi threaded for both reads and writes because both normal data and parity data are distributed round-robin. This is one reason why RAID 5 offers better overall performance in server applications. Random I/O benefits more from RAID 5 than does sequential I/O, and writes take a performance hit because of the parity calculations. RAID 5 is ideal for database applications.

RAID Level 10 RAID 10 is defined as mirrored stripe sets or also known as RAID 0+1. You can build RAID 10 either directly through the RAID controller (depending on the controller) or by combining software mirroring and controller striping, or vice versa (called RAID 01).

RAID Level 50 This RAID level is a combination of RAID level 5 and RAID level 0. Individual smaller RAID 5 arrays are striped, to give a single RAID 50 array. This can increase the performance by allowing the controller to more efficiently cluster commands together. Fault tolerance is also increased, as one drive can fail in each individual array.

Sub-array In RAID 50 applications, this is the name given to the individual RAID 5 arrays that are striped together. Each sub-array has one parity drive.

Optimization and Drive Selection for RAID 5 Arrays 13


14


Creating Arrays

Configuring the controllers for arrays involves a few simple steps. Each controller

only stores its RS-232 settings internally, all other configuration parameters

(i.e., array specific settings) are stored on all disk drives that are attached to the

controller(s). This makes it possible to remove and replace controllers without

requiring any configuration changes. The configurations can be viewed or

modified through either controller’s RS-232 port.

No changes are made until the configuration is saved, so it is possible to quit at

any time without affecting the current configuration as long as you don’t save

those changes. Configuration can be performed while the system is active.

To create an array:

1 From the Main menu, type <1> for the “Configuration” menu and press

<Enter>.

Main Menu

2 From the Main Configuration menu, type <1> for the “Array Configuration”

menu and press <Enter>.

The Array Configuration menu allows the Arrays to be reviewed, created,

deleted, or otherwise managed. Before creating a new array, the user may

view the current arrays (if any exists) and the available unused drives.

Creating Arrays



Main Configuration Menu

3 Choose “View Array” by typing <1> and pressing <Enter>.

Array Configuration Menu

The example below shows one array with ten drives. This array uses RAID

level 50, with two sub-arrays (denoted by ‘8+2’). The reserved capacity is 1%.

Writeback cache is enabled and detailed drive information is shown for each

drive in the array. Refer to the table on the following page for explanations

of each field.

Creating Arrays 15


16


View Array Screen

Next, you will most likely wish to view the unused drives before creating a

new array.

4 From the Array Configuration menu, choose “View Unused Drives” by typing

<6> and pressing <Enter>.

View Array Field Explanation

Field Description

No. Drive number in the array.

World Wide Name 8-byte Node Name, which is unique for every drive.

Product ID Drive model number.

F/W Drive firmware version.

Sav Indicates whether a copy of the drive’s firmware is saved on the media.

Ser. No. Drive serial number.

Cap. Drive capacity expressed in GBs.

IDx Loop ID of the specified drive loop.

HID Hard addresss of the drive.

EN, SL The enclosure and drive slot number, used for drive identification.

ST The drive status, will either be “OK” or “FL.”

Creating Arrays



View Unused Drives Screen

5 Create a new array.

From the Array Configuration menu, choose “Create Array” by typing <2>

and pressing <Enter>.


6 Choose the appropriate RAID level. Type the <number> for your selection

and press <Enter>.

NOTE: The menu’s that will follow are mostly identical for all RAID levels,

however RAID 5/50 is used in this example.

Creating Arrays 17


18


RAID Level Screen

7 Select “Manual Configuration”, type <1> and press <Enter>.

If all the drives attached are of the same type, the controller may make some

configuration suggestions. If one of these suggestions is satisfactory, you may

choose the associated number to select that choice instead of manual

configuration. (If you choose a suggested configuration, skip to step 13 then

refer to “Configuring Array Writeback Cache” on page 23.)

Manual Configuration Screen

8 Specify the chunk size for the array. Type the <number> for the selected

chunk size and press <Enter>.

The available chunk sizes are 64K, 128K, and 256K. This is the amount of

data that is written on a single drive before the controller moves to the next

drive in the stripe.

Creating Arrays



To achieve optimum RAID 5 write performance you should consider setting

the chunk size based the specified number of drives for a Full Stripe Write

when configuring RAID 5/50 arrays. Refer to “Optimizing RAID 5 Write

Performance” on page 157 for detailed information.

Chunk Size Screen

For RAID 5/50, the primary aim of setting a chunk size is to try to set a stripe

size that allows for full stripe writes. The stripe size is determined by the

number of data drives multiplied by the chunk size, (8 data drives * 64K

chunk size = 512 stripe size).

For maximum performance with RAID 5/50 arrays, you want to do as many

full stripe writes as possible. Typically, Microsoft® Windows NT/

Windows®2000 accesses at 64K, therefore a stripe size of up to 1 MB would

mean the controller has to cluster 16 commands to perform a full stripe write

(actually 17 because of alignment). If you were to use a larger stripe size,

then you run the risk of not being able to cluster sufficiently for the

application.

In cases where you are performing larger writes to the controller, then you

could go up to 2 MB for a stripe size, since you have more data to cluster.

Never exceed 2 MB for a stripe size, since the controller cannot cluster over

this size.

It is recommended to keep the stripe size to 1 MB or less for general use,

perhaps increasing it for specific applications. This stripe size is actually the

substripe size in RAID 50 cases. A 4+1 array (4 data and 1 parity drives) with

a 256K chunk has a 1 MB stripe size, as does an 8+1 array with 128K stripe

size, and an 8+2 array with 256K chunk size.

Creating Arrays 19


20


Although 8+1 gives an even stripe size, this does not really matter for a

operating system (OS) that writes in significantly smaller chunks. So, 8+1,

10+1, or 11+1 with a chunk size of 64K would be fine for an OS that does

64K writes. If the OS does much larger writes, then you may want to increase

the chunk size. With writeback cache enabled, then the controller can cache

data and do a full stripe write.

For a single enclosure example - best all around performance:

11+1 with 64K chunk size would yield a 704K stripe (11*64=704)

10+2 with 128K chunk size would yield a 640K stripe (5*128=640)

When using more drives, 14+2 (64K), or 12+3 (128K) should be as good

as 12+4. However, it is recommended to keep to a minimum of 4 data

drives for a subarray.

For most sequential accesses, the difference may not be too noticeable for

different configurations. However, for large block random writes, it can help

if the stripe size is similar to the I/O size. This allows the controller to

perform a partial full strip writes, where it has most of the data for a full

stripe, and can just read some from the drives to complete the stripe.

While Microsoft® Windows NT/Windows®2000 does 64K accesses, these are

not aligned. If the chunk size is 64K, every access crosses a chunk boundary

and so involves two read/modify/write commands instead of one. However,

since the controller can cluster, this problem is somewhat offset since the

controller usually can cluster sufficiently to do full stripe writes. If it is

completely random 64K access on Microsoft® Window NT/Windows®2000,

then a 64K chunk is not the best, rather 128K or 256K is better to minimize

the number of commands that cross chunk.

Larger chunk sizes should be used if the OS is writing large blocks, or with

large sequential writes where the controller can cluster sufficiently. Always

keep the strip size below 2 MB.

9 Select the drives for the array.

The controller will list drives one at a time, and you can choose whether to

add each drive to the array. Select the displayed drive by typing <Y> and

pressing <Enter>, or just pressing <Enter> (default selects Yes).

NOTE: Choosing “Yes” or “No” will always prompt the message whether

you wish to continue selecting drives. Pressing the <Enter> key will

automatically assume you wish to continue selecting drives.

Creating Arrays



Select Drives Screen

NOTE: When a drive has previously been marked as failed, then the default

is “No.” This prevents bad drives from being used by mistake. The

user must type “Y” to add a previously failed drive.

10 Continue selecting the desired number of drives. When you have completed

the drive selection, type <N> and press <Enter>.

Press any key to continue.

NOTE: To skip a drive, type <N> and press <Enter>. The screen will prompt

if you wish to continue adding drives, type <Y> to continue selecting

drives. Be sure to read the screen message, if you go too far or

answer incorrectly, you will need to start over again since the

controller assumes you have finished selecting drives.

11 Choose the drive capacity to reserve. Type the correct <value> and press

<Enter>.

This determines how much drive capacity to reserve for future capacity

fluctuations of replacement drives. (This is not applicable to RAID 0

configurations.) The default is 1%.

Drive Capacity Reserve Screen

Creating Arrays 21


22


12 If applicable, choose the number of required sub-arrays.

For RAID 5/50 configurations, if there are sufficient drives selected to create

more than one subarray, the controller then presents a final menu to allow

the subarray count to be chosen. In this example, there are ten drives added

to the array, so the options are for either one or two sub-arrays.

Sub-Array Selection Screen

13 Create the disk array.

For RAID 5/50 configurations the array will be initialized after saving the

configuration. This is accomplished by pressing any key, then typing <0>

and pressing <Enter> as you back out of each screen until you are prompted

to save. The array will be initialized at this point. You can monitor the

initialization by selecting View Array from the Array Configuration menu. You

must periodically return to the Array Configuration menu and then View

Array to get updated percent complete information.

Normally there will be a few other steps to complete setting up the system,

when these steps have been completed and you exit the menu, it will save

and start the initialization.

For RAID 0/1/10 configurations, the array is created when you complete the

drive selections and press any key to continue. Initialization is not required.

Continue with setting up the Array Writeback Cache and Hot Spare drives

options. Then go to the next chapter and configure the Host LUN assignment

before exiting and saving.

Creating Arrays



Configuring Array Writeback Cache

The following steps will guide you through configuring the writeback cache

option for each array. Each array has a different writeback cache setting. You will

need to set this option for each array you have created.

In a writeback cache operation, data is sent to the controller from the host and

the controller responds to the host confirming the data was received and written

to the media. The host may then send more data. This can significantly increase

performance for host systems that only send a low number of commands at a

time. The controller caches the data, and if more sequential data is sent from the

host, it can cluster the writes together to increase performance further. If

sufficient data is sent to fill a stripe in RAID 5/50 configurations, the controller

can perform a Full Stripe Write, which significantly reduces the write overhead

associated with RAID 5/50.

Disabling writeback cache ensures that the data is written to the drives before

status is returned to the host. With writeback cache enabled, if a short term

power failure occurs, the battery back-up unit provides adequate power to

ensure that cache is written to disk when the power is restored. In duplex

operations, the cache is mirrored to both controllers which provides further

redundancy in the event of a single controller failure.

Mirrored cache is designed for absolute data integrity. The cache in each storage

processor contains both primary cached data for the disk groups it owns, and a

copy of the primary data of the other storage processor. Mirrored cache ensures

that two copies of cache exist on both storage processors, before confirming to

the operating system that the write has completed.

Below is a table from the imageRAID Series User’s Guide, which list the hold-up

times for data for the battery backup unit.

BBU Battery Hold-Up Times

ConfigurationMemory Vendor and

Part NumberMeasured

Current DrawAbsolute Maximum

Backup TimeExpected Safe Backup Time

Main board only w/512 MB

Kingston KVR100X72C2/512

27.9mA 41.2 hours 20.6 hours

Main board w/512 MB and Coprocessor w/512 MB

Kingston KVR100X72C2/512

48.3mA 23.8 hours 11.9 hours

Configuring Array Writeback Cache 23


24


Normally write-intensive operations will benefit from the higher performance

when writeback cache is enabled on that array. Read-intensive operations, such

as a streaming server, may not benefit from writeback cache.

To configure the array’s writeback cache option:

1 From the Array Configuration menu, choose “Array Cache Configuration” by

typing <5> and pressing <Enter>.


2 Choose the array to which you wish to change the cache options. Type <x>

and press <Enter>, where ‘x’ is the number of your array.

Choose Array Screen

3 Choose the Writeback Cache option by typing <2> and pressing <Enter>.

Configuring Array Writeback Cache



Array Cache Configuration Menu

4 Choose to enable or disable writeback cache for this array. Type “Y” (or just

press <Enter> to enable or “N” to disable, and press <Enter>.

When the information first appears it displays the current parameter settings.

If you want to leave those settings as is, just press the <Enter> key and step

through pressing <Enter> for each parameter. You may change any one of

those parameter when it is displayed. If you made no changes after the last

parameter, you will see a message indicated no change was made.

Writeback Cache Screen

If you made any changes to the parameter settings, the message will indicate

the parameters were modified.

5 Save the configuration changes. Type <0> and press <Enter>.

6 Again, type <0> and press <Enter>.

7 And, type <0> and press <Enter> to Save and Exit. Answer “Yes” to save the

changes. Afterwards press any key to return to the Main menu.

Configuring Array Writeback Cache 25


26


Configuring Array Readahead Cache

1 From the Array Configuration menu, choose “Array Cache Configuration” by

typing <5> and pressing <Enter>.


2 Choose the array to which you wish to change the cache options. Type <x>

and press <Enter>, where ‘x’ is the number of your array.

Choose Array Screen

3 Choose the Readahead Cache option by typing <1> and pressing <Enter>.

Array Cache Configuration Menu

Configuring Array Readahead Cache



4 Choose an setting for your Readahead Cache setting, type <x> and press

<Enter>, where ‘x’ is one of the menu numbers displayed.

The choices are automatic, disabled, and four pre-determined sizes. Select

Automatic, which is the the default and the recommended setting. It allows

the controller to determine the optimum setting. Disabled will turn off the

Read-Ahead Cache. Choose one of the pre-determined sizes to optimize the

read performance based on your data patterns.

Readahead Cache Screen

If you made any changes to the parameter settings, the message will indicate

the parameters were modified.

5 Save the configuration changes. Type <0> and press <Enter>.

6 Again, type <0> and press <Enter>.

7 And, type <0> and press <Enter> to Save and Exit. Answer “Yes” to save the

changes. Afterwards press any key to return to the Main menu.

Next you will configure your hot spare drives, refer to “Assigning Hot Spare

Drives” on page 28.

Configuring Array Readahead Cache 27


28


Assigning Hot Spare Drives

The process of configuring fault tolerant arrays includes assigning drives for global or dedicated hot spares. In the event of a drive failure, the controller will use a global hot spare to replace the failed drive in any array or if a dedicated spare is assigned to the specific array, then that array will use its dedicated hot spare.

This step is accomplished through the Rebuild Configuration menu.

1 From the Configuration menu, type <5> for the “Rebuild Configuration”



2 From the Rebuild Configuration Menu, type <2> for the “Add Hot Spare

Drives” option and press <Enter>.

Rebuild Configuration Menu




3 From the Add Hot Spare Drives menu, select either <1> Add All Unused

Drives or <2> Select Drive to Add, and press <Enter>.

Add Hot Spare Configuration Menu

To add all unused drives as hot spares:

a Type <1> for All Unused Drives and press <Enter>.


NOTE: If a drive has previously failed, perform a full drive maintenance

prior to attempting to re-use the drive. Manually add the drive as a

hot spare, then if you would like it available, remove it as a hot

spare. The drive’s serial number is remember by the controller as

having been failed. The above process will clear that state.

The total number of drives added as hot spares is displayed. To view the

list of hot spare drives return to the previous menu and select “View Hot

Spare Drives.”

If you had selected this option, proceed now to “Host LUN Assignment”

on page 39.

Assigning Hot Spare Drives 29


30


To add a specific drive as hot spare:

a Type <2> to “Select a Drive to Add” and press <Enter>.

b Select the drives for use as a hot spare.

The drives will be listed one at a time, and you can choose whether to

add each drive as a hot spare or skip to another. Select the displayed

drive by typing <Y> and pressing <Enter>, or just press <Enter> (default

selects Yes).

Choosing “No” <N> will prompt the message whether you wish to

continue selecting drives. Press the <Y> key to display the next drive.

Select Drive for Hot Spare Screen

c After you have selected your drives to be used as hot spares, press the

<N> key to the prompt “Continue adding disk drives.” It will display the

number of drives added as hot spares.

d Return to the Main Configuration menu. Type <0> and press <Enter>.

Proceed now to “Host LUN Assignment” on page 39.




Deleting an Array


<Enter>.

Main Menu

2 From the Configuration menu, type <1> for the “Array Configuration” menu

and press <Enter>.


3 Choose “Delete Array” by typing <3> and pressing <Enter>.

Deleting an Array 31


32



4 Enter the number of the array to delete and press <Enter>.

Delete Array Screen

5 Type “Yes” <Y> and press <Enter> to confirm, or “No” <N> to cancel the

deletion.

Delete Array Confirm Screen

Deleting an Array



Expanding an Array

This feature is used to expand an existing array by adding drives to the array. An

array can be expanded up to a maximum of 16 drives. Only one array can be

expanded at a time, and no changes can be made to the arrays or LUNs while an

expansion operation is in progress.

During the expansion process data is re-striped across a new set of data drives in

the sub-array and new parity is calculated. If the array is a fault tolerant array,

such as RAID level 1, 10, 5, or 50, it will remain fault tolerant during the expand

operation. Should a disk drive fail in a fault tolerant array during the expansion,

the expand operation will continue as normal where it will flag the drive as failed

and use the data and parity information to create the new data and parity stripe.

After the expansion process is complete, and if you have a hot spare designated,

the automatic rebuild operation will commence, bringing the non-fault tolerant

expanded array back to a fault tolerant condition. Note that if a second drive

failure does occur, that condition is not recoverable. You may wish to consider

backing up the data prior to expanding an array, however there is a level of

protection during this operation without the backup.

1 From the Main menu, type <1> for the Configuration Menu and press

<Enter>.

Main Menu


and press <Enter>.

Expanding an Array 33


34



3 Choose “Expand Array” by typing <4> and pressing <Enter>.


4 Select the array to expand by entering the number of the array and pressing

<Enter>.

Expand Array Screen

Expanding an Array



5 Select the drives to be used for the array expansion.

The controller will list the number of drives added and following that, the

total number of drives in the array. New drives are displayed one at a time,

and you can choose whether to add each drive to the expansion. Select the

displayed drive by typing <Y> and pressing <Enter>, or just press <Enter>

(default selects Yes).

NOTE: Choosing “Yes” or “No” will always prompt with the message

whether you wish to continue selecting drives. Pressing the <Enter>

key will automatically assume you wish to continue selecting drives.

NOTE: You can add disk drives of equal or greater capacity size drives that

comprise the array being expanded. If you use drives larger than those

in the array, the larger drive will be marked as a drive equal to the

array drives. For example, if you have a three drive array comprised of

36 GB drives and expand the array using a 146 GB capacity drive, the

capacity of that 146 GB drive will be limited to 36 GBs.

Select Drives Screen

NOTE: When a drive has previously been marked as failed, then the default

is “No.” This prevents bad drives from being used by mistake. The

user must type “Y” to add a previously failed drive.

Continue selecting the desired number of drives to expand the array with,

and when you have completed the drive selection, type <N> and press

<Enter>. Press any key to continue.

NOTE: To skip a drive, type <N> and press <Enter>. The screen will prompt

you to continue adding drives, type <Y> to continue selecting drives.

Be sure to read the screen message, if you go too far or answer

incorrectly, you will need to start over again since the controller

assumes you have finished selecting drives.



36


6 For RAID 5/50 configurations and if prompted, choose the number of

sub-arrays.

If you are adding an equal number of disk drives that comprise the original

array, you will be presented with a menu that allows you to select a new data

drive/parity drive or sub-array configuration. For example, if you have an

array that contains 3 drives (2 data and 1 parity), you would have configured

one sub-array for the array (default choice). If you add three drives or

perhaps six drives, you will now have more choices for sub-arrays.

Sub-Array Selection Screen

7 Expand the disk array. Do this by answering the prompt to expand the array,

type <Y> to expand or <N> to cancel. The expansion will begin immediately.

You can monitor the expansion by selecting View Array from the Array

Configuration menu. You must periodically return to the Array Configuration

menu and then View Array to get updated percent complete information.

NOTE: Because the entire data structure must be re-striped across all the

drives, this operation will take a significant amount of time. So be

sure to plan for this type of operation.

Expanding an Array



Trust an Array

When you create a RAID 5 or 50 array, you have the option to trust the array.

This option should only be used in environments where the user fully

understands the consequences of the function. Trust array option is provided to

allow immediate access to an array for testing application purposes only.

Trust array does not calculate parity across all drives and therefore there is no

known state on the drives. As data is received from the host parity is calculated as

normal, but it occurs on a block basis. There is no way to guarantee that parity

has been calculated across the entire drive. The parity data will be inconsistent

and so a drive failure within a trusted array will cause data loss.

1 From the Main menu, choose “Configuration” <1> and press <Enter>.


and press <Enter>.

3 From the Array Configuration menu, choose “Advanced Array Configuration”

<7> and press <Enter>.

4 Choose “Trust Array” by pressing the <2> key and pressing <Enter>.

The available arrays will be displayed.

Trust Array Menu Screen

5 Enter the array number to trust and press <Enter>.

Trust an Array 37


38


6 You are prompted to trust the array. Answer Yes (Y) to trust or No (N) to

cancel. A message appears announcing that the array will not be initialized

after you save the configuration.

7 Press any key to continue.

Type <0> and press <Enter> as you back out of each screen until you are

prompted to save.

Trust an Array

Chapter 3


Host LUN Assignment

Configuration

A Host LUN, also known as a Disk LUN, is defined as a region or combination of

regions of storage that can be made available to one or more host systems. Up to

512 LUNs are supported. After an array has been created, this region is first

marked as unassigned. One or more LUNs can be created in this region or

existing LUNs can be expanded using this region. A LUN can be created or

expanded in 1 GB increments with a maximum total size per LUN of 2198 GBs.

This corresponds to the SCSI 32 bit addressing limitation of 2 TB.

NOTE: For Microsoft® Windows NT there is a limitation of 231 LUNs. A hot fix is

available from Microsoft®. Refer to the Microsoft® Knowledge Base

Article - 245637.

Terminology

The following table describes the terminology relating to Host LUNs.

Term Description

LUN Segmentation Any Host LUN (Disk LUN) can be expanded into any free region, so it is possible to easily add capacity at any time. There is no requirement that any additional space be contiguous. LUN Segmentation is completely transparent to the host systems.

39

Chapter 3 - Host LUN Assignment

40


Accessing the Host LUN Configuration Menu

1 From the Main menu, type <1> for the Configuration Menu and press

<Enter>.

2 From the Configuration menu, type <2> for the “Host LUN Configuration”


Configuration Menu

LUN Availability To accommodate hosts with multiple ports and multiple host systems, it is possible to restrict a LUN’s availability to a particular controller or port. Access can be enabled or disabled for each host port of each controller.

Mapped LUN Number Each LUN is presented to the host systems with a unique LUN Number. In certain cases (such as after deleting another LUN) it may be desirable to change the number that a LUN is presented as. This can be done at any time, bearing in mind that any attached host systems may need to be rebooted or reconfigured to maintain access.

Unassigned Free Space The controller keeps a map of all the space that is unassigned to any LUN. This is available for LUN creation or expansion. Each unassigned region is individually listed.

Accessing the Host LUN Configuration Menu



The Host LUN Configuration menu provides the options for viewing,

creating, deleting, and managing Host LUNs.

Host LUN Configuration Menu

Viewing Unassigned Free Space

Prior to creating a Host LUN, you may wish to review the available unassigned

free space. This will help you to identify the unused regions or segments for use

during the creation of your LUNs. Normally with a first time configuration this is

not necessary, since all of the space is unassigned.


<Enter>.



3 From the Host LUN Configuration menu, type <6> for the “View Unassigned

Free Space” menu and press <Enter>.

4 View the free space regions and make a note of them for later use.

Viewing Unassigned Free Space 41


42


Creating a Host LUN

To complete the process of configuring your arrays, you will need to create one

or more LUNs. Creating a LUN from the available free space regions presents the

LUN to the host operating system as a drive. Referring to the example below, we

will use this example to document the procedures for creating the Host LUNs.

Sample Array and LUN Layout


<Enter>.



3 From the Host LUN Configuration menu, type <2> for the “Create Host LUN”



Array 1RAID 50

2400 GB

LUN 3:0 700 GB

LUN 2:0 400 GB

LUN 1:0 1000 GB

LUN 0:0 300 GB

Creating a Host LUN



4 Choose the option to Manually Select Unused Free Space, type <1> and press

<Enter>.

NOTE: If you are creating just one single LUN, you may choose option 2, All

Unsegmented Free Space. This creates one LUN using all the free

space.

Select Region for LUN Menu

5 Enter the number of the unused region, type <0> and press <Enter>.

Since this is a new configuration, the unused region will be “0” for the entire

contiguous region.

Select Unused Region Screen

6 Enter the number of LUNs to create, type <1> and press <Enter>.

Creating a Host LUN 43


44


If you are creating several LUNs that will have equal sizes, you can enter the

total number of LUNs at one time.

Select Number of LUNs to Create Screen

7 Enter the size of LUNs to create, type <xxx> and press <Enter>, where ‘xxx’

is the size you desire.

Referring to our original example, enter “300.”

NOTE: If you selected multiple LUNs from the previous step, enter the size

for the multiple LUNs here. Remember they will all be the same size

using this batch method.

Select Size for the LUNs to Create Screen

Following along with the Array and LUN layout example, these sub-steps will

repeat the process for demonstration purposes to complete the number of

LUNs. After entering the size and pressing <Enter>, the LUN is created.

Creating a Host LUN



a Choose the option to “Manually Select Unused Free Space,” type <1> and

press <Enter>.

b Enter the number of the unused region, type <0> and press <Enter>.

c Enter the number of LUNs to create, type <1> and press <Enter>.

d Enter the size of LUNs to create, type <xxx> and press <Enter>.

Referring to our example, enter “1000.”

e Repeat steps a) through d) for the remaining two LUNs. So that you end

up with LUN 0:300, LUN 1:1000, LUN 2:400, and LUN 3:700.

This completes assigning Host LUNs.

NOTE: If you wish to change a LUN number, refer to “Modifying Mapped LUN

Number” on page 53.

Set Host LUN Availability

Be default, the availability is set to both ports of both controllers for all LUNs. It is

recommended to limit the availability to only the ports that will be used to access

this LUN, as this may help to avoid unnecessary inter-controller access

contention.

1 From the Host LUN Configuration menu, type <5> for the “Set Host LUN

Availability” menu and press <Enter>.


Set Host LUN Availability 45


46


2 Choose a LUN to set its availability, type <x> and press <Enter>, where ‘x’ is

the LUN number.

Set Host LUN Availability Menu

NOTE: A Controller Host Port is also known as the Controller Port. The

controllers and their ports are sometimes referred as C0P0, C0P1,

C1P0, and C1P1.

If you are using a dual host system, you will probably want to set Controller

0/1 Host Port 0 enabled for one LUN and set Controller 0/1 Host Port 1 to

another LUN. In the example screen above Host Port 0 is being made

available to the chosen LUN and not available to Host Port 1. The next LUN

will be set the opposite, i.e., Controller 0/1 Host Port 0: Disabled for the next

LUN and Controller 0/1 Host Port 1: Enabled for that same next LUN. Further,

if there were two LUNs created, LUN 0 would be available to Host Port 0 on

both controllers and LUN 1 would be available to Host Port 1 on both

controllers.

NOTE: Following our example setup of one array and four LUNs, one

would determine which host requires access to which LUNs. Then

set the availability for the specific LUNs as required for your storage

configuration.

3 Repeat the above step for each LUN you created.

Set Host LUN Availability



Tips on Setting LUN Availability

The following information will assist you when creating LUNs and setting their

availability for each of the supported imageRAID Series Storage System topologies.

You will need to have both, the imageRAID Series User’s Guide and this manual

available for reference. Locate the topology you set up on the following pages and

review the information on its host LUN availability.

imageRAID IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port One Host)

Because of the specific Host I/O card switch settings for this configuration, all

LUNs are visible on all Host Ports, if they are available to either controller port.

� LUN Availability should be set to only one controller port for each LUN and

the user should balance the load between the two controller ports (C0P0 &

C0P1) for best performance.

NOTE: If the LUN Availability is set to both ports the host will see a given

array twice unless it does special processing to detect duplicate paths

to the array, which would result in potential data overwrites. A host

device driver may provide this functionality.

imageRAID IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port One Host with Dual Connection)

The user should balance the load between the two controller ports for best

performance. This can be accomplished using one of two methods.

� LUN Availability can be set to both controller ports and the user can then

map LUNs to the different HBA WWN.

� LUN Availability can be set to one port for each LUN and alternate

controller ports to balance the load.

LEFT H0Host I/O Card

Controller Ports

Port 0

Port 1

Ports 0 & 1

LEFT H1 RIGHT H0 RIGHT H1

LUN LUN LUN LUN

LUN LUN LUN LUN

LUN LUN LUN LUN

Tips on Setting LUN Availability 47


48


If setup requires Upstream Failover, LUN Availability should be set to both ports.

A host device driver or third party software which provides upstream failover

functionality will do special processing to detect duplicate paths to array.

imageRAID IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port Multi Host Connection)

When this configuration is set up, you will have the option for dedicated LUNs

for each host, or clustered servers.

� For dedicated LUNs with two host systems, set LUN Availability to the

controller port to which the respective host is attached.

� For dedicated LUNs with more than two host, set LUN Availability to both

and the user can use LUN Mapping to a specific HBA WWN to restrict

access between hosts.

� For clustered servers, enable LUN Availability to both ports which will

cause the same LUN to be reported to all attached Hosts.

NOTE: Protected LUN sharing between two host systems is not supported

without third party software which will provide the necessary file

share/file lock capabilities.

LEFT H0

Port 0

Port 1

Ports 0 & 1


N/A N/A LUN LUN

LUN LUN N/A N/A

LUN LUN LUN LUN

Host I/O Card

Controller Ports

LEFT H0

Port 0

Port 1

Ports 0 & 1


N/A N/A LUN LUN

LUN LUN N/A N/A

LUN LUN LUN LUN

Host I/O Card

Controller Ports

imageRAID IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port Multi Host Connection)



imageRAID IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port Dual Host with a Quad Connection)

When this configuration is set up, you will have the option for dedicated LUNs

for each host, or clustered servers.

� Set LUN Availability to both controller ports.




imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx ((Duplex: Multi Port Mirrored One Host)

When this configuration is set up, the user should balance the load between the

two controller ports for best performance.

� Set LUN Availability to only one controller port for each LUN.

� If LUN Availability is set to both ports the host will see a given array twice

unless it does special processing to detect duplicate paths to the array. This

may result in potential data overwrites. A host device driver may provide

this functionality.

LEFT H0

Port 0

Port 1

Ports 0 & 1


N/A N/A LUN LUN

LUN LUN N/A N/A

LUN LUN LUN LUN

Host I/O Card

Controller Ports

LEFT H0

Port 0

Port 1

Ports 0 & 1


LUN N/A LUN LUN

LUN LUN N/A LUN

LUN x2 LUN LUN LUN x2

Host I/O Card

Controller Ports

imageRAID IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port Dual Host with a Quad Connection) 49


50


imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored One Host Dual Connection)

When this configuration is set up, the user should balance the load between the

two controller ports for best performance. This can be accomplished in one of

two ways:

� Set LUN Availability to both controller ports. Use LUN Mapping to restrict

access to different HBA WWNs.

� Set LUN AVailability to one port for each LUN and alternate controller ports

availability between LUNs.

NOTE: Set the LUN Availability and Mapping for Upstream Failover

functionality to both ports. A host device driver or third party

software which provides the upstream failover will do special

processing to detect duplicate paths to the array.

imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored Multi Host)

When this configuration is set up, you can choose from dedicated LUNs for each

host, or clustered servers.

� For two host systems with dedicated LUNs, set LUN Availability to the

controller port to which the respective host is attached.

� For more than two host systems with dedicated LUNs, set the LUN

Availability to both ports and the user can use LUN Mapping to a specific

HBA WWN to restrict access between hosts.

� For clustered servers set the LUN Availability to both ports. This will cause

the same LUN to be reported to all attached hosts.

LEFT H0

Port 0

Port 1

Ports 0 & 1


N/A N/A LUN LUN

LUN LUN N/A N/A

LUN LUN LUN LUN

Host I/O Card

Controller Ports

imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored One Host Dual Connection)






imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored Dual Host with Quad Connections)

When this configuration is set up, you are creating a system level fault-tolerant,

high-performance shared storage with clustering.





imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored SAN Single Switch)

When this configuration is set up, you can choose from dedicated LUNs for each

host, or clustered servers.

� For dedicated LUNs, set LUN Availability to only one controller port for

each LUN and the user should balance the load between the two controller

ports for best performance.

LEFT H0

Port 0

Port 1

Ports 0 & 1


N/A N/A LUN LUN

LUN LUN N/A N/A

LUN LUN LUN LUN

Host I/O Card

Controller Ports

LEFT H0

Port 0

Port 1

Ports 0 & 1


LUN LUN N/A N/A

N/A N/A LUN LUN

LUN LUN LUN LUN

Host I/O Card

Controller Ports

imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored Dual Host with Quad Connections) 51


52


� Use LUN Mapping to a specific HBA WWN to restrict access between hosts

attached to the same controller port.

� For clustered servers set the LUN Availability to both ports. This will cause

the same LUN to be reported to all attached hosts.




imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored SAN Dual Switches)

When this configuration is set up, you set the LUN Availability for clustered

servers.





LEFT H0

Port 0

Port 1

Ports 0 & 1


N/A N/A LUN LUN

LUN LUN N/A N/A

LUN LUN LUN LUN

Host I/O Card

Controller Ports

LEFT H0

Port 0

Port 1

Ports 0 & 1


N/A N/A LUN LUN

LUN LUN N/A N/A

LUN LUN LUN LUN

Host I/O Card

Controller Ports

imageRAID IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored SAN Dual Switches)



Modifying Mapped LUN Number

This option allows you to change the mapped or assigned LUN number after the

LUN has been created.


<Enter>.



3 From the Host LUN Configuration menu, type <7> for the “Advanced Host

LUN Configuration” menu and press <Enter>.


4 Select a Host LUN number to modify, type in the number and press <Enter>.

Host LUN Configuration Menu - Select LUN Screen

Modifying Mapped LUN Number 53


54


You are presented with the current mapping.

5 Enter the new LUN number that the Host LUN will be mapped to, type <xxx>

and press <Enter>.

Be sure to use the full three digits, for example for LUN 3 enter 003.

Host LUN Configuration Menu - Enter New LUN Screen

You are presented with the LUN number change that you will be confirming.

6 Type <Y> to confirm the change or <N> to cancel.

Host LUN Configuration Menu - Confirm LUN Change Screen

7 Exit the menu, type <0> and press <Enter>. Continue typing <0> and

pressing <Enter>, until you reach the main menu at which time you will be

prompted to save the changes, select Yes <Y> to save your changes.

Modifying Mapped LUN Number

Chapter 4


SAN LUN Mapping

When attaching many host systems in a SAN environment, it may be necessary to

more precisely control which hosts have access to which LUNs. In addition to

controlling LUN availability on a controller port by port basis, it is also possible to

further restrict access to a specific host system or single adapter in a host system.

Up to 512 SAN LUN Mappings are supported.

Terminology

The following table describes the terminology relating to SAN LUNs.

Term Description

Node Name This is an eight byte hexadecimal number, uniquely identifying a single host system. It incorporates the World Wide Name and two additional bytes which are used to specify the format. In a system with multiple FC ports, all adapters will typically use the same Node Name, but unique Port Names.

Port Name This is an eight byte hexadecimal number, uniquely identifying a single host port. It incorporates the World Wide Name and two additional bytes which are used to specify the format and indicate the port number.

Mapping Name A 28 character name that can be used to help identify the host system.

Exclusive Access A LUN is presented to one and only one host system. It is not available for any other host systems.

Read/Write Access A host may read and write to the LUN.

Read Only Access A host may only read from the LUN.

Mapped LUN Number This is the LUN number that a specific LUN responds to when accessed by a host. It is not necessary for this to bear any relation to the LUN number.

Mapping Availability Which controller ports the mapping is valid for.

55

Chapter 4 - SAN LUN Mapping

56


Accessing the SAN LUN Mapping Configuration Menu


<Enter>.

Configuration Menu

2 From the Configuration menu, type <4> for the “SAN LUN Mapping

Configuration” menu and press <Enter>.

SAN LUN Mapping Configuration Menu

The SAN LUN Mapping Configuration menu provides the options for

viewing, creating, deleting, and managing SAN LUNs, and for viewing the

host systems attached to the storage network.

Accessing the SAN LUN Mapping Configuration Menu



Viewing SAN LUN Mappings

To view existing LUN mappings, choose this option. The mappings will be listed

in the following format:

View SAN LUN Mapping Screen

In the example above, there are two mappings created. They have an identical

node name but different port names, indicating they are both in the same host

system. The port name is used for the mapping, and access is only allowed on

Port 0 of the controllers. Each host port can access one controller LUN,

exclusively.

Viewing SAN LUN Mappings 57


58


Creating a SAN LUN Mapping

1 From the SAN LUN Mapping Configuration Menu, type <2> for the “Create

SAN LUN Mapping” menu and press <Enter>.


2 Enter a name for this mapping and press <Enter>.

You can use from 1 to 32 characters for the unique mapping name.

Selecting Mapping Name Screen

3 Enter the Node and Port names for manual setup, or select from the list of

connected host systems.

To setup manually, it is necessary to enter both the node name and the port

name. (You should be able to get this information from the option “View

Connected Hosts.”)




To select from the list of connected hosts, all that is necessary to enter is the

corresponding host number.

Choosing the Method to Enter Node/Port Names Menu

After selecting the host system, the following is displayed.

Node/Port Names Screen

Here, it is possible to restrict the access to a host based on either the host’s

node name or port name. Typically, a host with multiple adapters or ports

will use a single node name and multiple port names.

Creating a SAN LUN Mapping 59


60


Select the desired option based on the specific requirements of the

installation.

4 Choose the Set Mapping options for one or both controller ports, type <1-3>

and press <Enter>.

Set SAN LUN Mapping Screen

Here, the availability of this mapping can be set. You can further restrict

access, particularly in a switch environment.

In the next step, you will select which LUNs are made to this particular host,

and whether they are exclusive or shared between multiple hosts.

Select Controller LUN List Screen

If a LUN is already part of another Mapping and is defined there as exclusive,

it will not be available here for selection.




If a LUN is part of another Mapping but is not mapped exclusively to that

host, then it will be available for selection here, however only for shared (S)

access.

NOTE: Microsoft® Windows NT®, Microsoft® Windows® 2000 Microsoft®

and Windows® 2003 and do not support a “Read Only” file system.

5 Select the controller LUN number that is to be used and press <Enter>.

If it is not used elsewhere, the option to map it exclusively (E) will be

presented.

LUN Availability Screen

After entering the mapped LUN number, the list of LUNs is again presented.

6 Continue to add LUNs for this particular host system by entering the LUN

number. Follow steps 5 through 6.

You may also choose to Quit <Q> which exits without saving the changes, or

End <E> which ends the sessions and adds the mapping.

Creating a SAN LUN Mapping 61


62


Delete SAN LUN Mapping

This option allows you to delete a SAN LUN Mapping.

1 From the SAN LUN Mapping Configuration Menu, type <3> for the “Delete



2 Enter a LUN Mapping number for the map to be deleted and press <Enter>.

Delete SAN LUN Mapping Screen

3 You will receive a warning and confirmation. Answer Yes <Y> to the prompts

which will delete the SAN LUN map.

NOTE: SAN LUN Mappings can be deleted in any order without affecting

the other mappings.

4 When you have completed the task, type <0> and press <Enter> to exit.

Delete SAN LUN Mapping



Modify SAN LUN Mapping

It is sometimes necessary to modify some parameters of a SAN LUN Mapping. It

is possible to change the Mapping Name, Node Name, Port Name, whether to

use the node or port name and the controller port availability of the mapping.

This may be necessary if, for example, a host system has an adapter failure and a

new adapter with different Node and Port names is installed.

1 From the SAN LUN Mapping Configuration Menu, type <4> for the “Modify



2 Enter the number of the SAN LUN map to be modified and press <Enter>.

Modify SAN LUN Mapping Screen

After entering the map number, each changeable parameter is listed one at a

time allowing them to be modified.

Modify SAN LUN Mapping 63


64


Viewing Connected Hosts

Choose this option to view all connected host ports.

1 From the SAN LUN Mapping Configuration Menu, type <5> for the “View

Connected Host” menu and press <Enter>.


The following screen is displayed. In this example two connected host ports

are shown.

Connected Host Ports Screen

The table on the following page provides an explanation of each of the fields

shown in the display.




Field Description

No. Host number assigned by the controller. This is used when creating a SAN LUN Mapping and selecting a host from the list.

Node Name This is the Host Port’s node name.

Port Name This is the Host Port’s port name.

ID-xy ID used by this host for Controller x Host Port y. (L) indicates Arbitrated Loop, and (F) indicates fabric connect.

Viewing Connected Hosts 65


66



Chapter 5


Advanced Configuration and Management

This chapter covers the advanced configuration and management options, which

includes viewing your current hot spares drives, adding or removing hot spare

drives, configuring automatic hot spares, and fine tuning rebuild parameters.

These options are accessed from the Rebuild Configuration menu.

Hot Spare Drives

View Hot Spare Drives



2 From the Rebuild Configuration Menu, type <1> for the “View Hot Spare



67

Chapter 5 - Advanced Configuration and Management

68


All currently configured hot spare drives are displayed.

View Hot Spare Drives Screen


Add Hot Spare Drives



Configuration Menu

2 From the Rebuild Configuration Menu, type <2> for the “Add Hot Spare


Add Hot Spare Drives




To add all unused drives as hot spares:

a Type <1> for All Unused Drives and press <Enter>.


The total number of drives added as hot spares is displayed. To view the

list of hot spare drives return to the previous menu and select “View Hot

Spare Drives.”

b Press any key to continue and return to the previous menu.

To add a specific drive as hot spare:

a Type <2> to Select a Drive to Add and press <Enter>.

Add Hot Spare Drives 69


70


b Select the drives for use as a hot spare.

The drives will be listed one at a time, and you can choose whether to

add each drive as a hot spare or skip to another. Select the displayed

drive by typing <Y> and pressing <Enter>, or just press <Enter> (default

selects Yes).

Choosing “No” <N> will prompt with the message asking if you wish to

continue selecting drives. Press the <Y> key to display the next drive.

Select Drive for Hot Spare Screen

c After you have selected your drives to be used as hot spares, press the

<N> key to the prompt “Continue adding disk drives.” It will display the

number of drives added as hot spares.

d Press any key to continue and return to the previous menu.

e Return to the Main Configuration menu. Type <0> and press <Enter>.

Delete Hot Spare

NOTE: This option will delete all assigned hot spare drives. After deletion if you

want to continue having one or more hot spares for your fault tolerant

arrays, you will need to Add Hot Spare Drives again.



Delete Hot Spare



Configuration Menu

2 From the Rebuild Configuration Menu, type <3> for the “Delete Hot Spare



3 Enter “Yes” <Y> to delete all hot spare drives.

Delete All Hot Spare Drives Menu


Delete Hot Spare 71


72


Enable/Disable Auto Rebuild

Whenever a hot spare is added, the automatic rebuild feature is enabled.

However, there may be occasions where it is necessary to manually enable or

disable this option.

When enabled, auto rebuild will begin the rebuild cycle when a redundant array

is configured and a hot spare drive(s) are available.


Menu and press <Enter>.

Configuration Menu

2 From the Rebuild Configuration Menu, type <4> for the “Enable/Disable Auto

Rebuild” option and press <Enter>.


Enable/Disable Auto Rebuild



3 Enter “Yes” <Y> to enable Auto Rebuild and press <Enter>.

To disable auto rebuild, enter “No” <N> and press <Enter>.

Configure Auto Rebuild Menu

4 Type <0> and press <Enter> to return to the previous menu.

If you have a fault tolerant or redundant array defined, and you do not have a hot

spare drive assigned, no rebuild will begin. Once you replace the failed drive

with a known good drive, you must assign that replacement drive as a hot spare

drive. Once this has been accomplished, the rebuild will automatically begin.

Remember that the Auto Rebuild function will be enabled when you assign a hot

spare drive.

Auto Hot Spare

This option allows the user to designate upon a drive failure that a hot spare

drive is used to automatically rebuild data when a new drive is inserted into the

drive slot of the failed drive. With this feature you are not required to designate

the drive as a hot spare, or manually add it as a hot spare. Other than physically

replacing the drives, no further user intervention is required.



2 From the Rebuild Configuration Menu, type <6> for the “Enabled/Disable

Auto Hot Spare” option and press <Enter>.

Auto Hot Spare 73


74



3 Press <Y> to enabled auto hot spare or <N> to disable this feature, and <Enter>.

The current “Enabled” or “Disabled” status is displayed.

Auto Hot Spare Menu

4 Press any key to continue. You can verify the change to the configuration by

Viewing the Configuration (option 9 under the Main Configuration menu.

Auto Hot Spare



Rebuild Priority

The option allows the user to specify the percentage of internal resources

allocated to rebuild operations, while the controller is also processing host

commands.



2 From the Rebuild Configuration Menu, type <5> for the “Set Rebuild Priority”

option and press <Enter>.


3 Enter the desired value <x> and press <Enter>. Then press <0> and <Enter>

for the previous menu.

The values range from 10% through 90%.


Rebuild Priority 75


76


UPS Configuration

At this time, the UPS Configuration feature is not implemented. This option will

be supported in a future release.

Ethernet Configuration

At this time, the Ethernet Configuration feature is not implemented. This option

will be supported in a future release.

UPS Configuration

Chapter 6


Additional Configuration

In this chapter you will find information about setting the host port addresses,

changing the Configuration Name, WWN Name, setting the Controller LUN ID,

setting background verification and enabling the controller SES logging option.

Host Port Addressing

For host systems that rely on hard addressing, it is necessary to provide the

controller with a unique hard address for each port. To ensure hardware failure,

hard addresses must be established for the controller. The ID you provide for the

controller ports will be dependent on what other devices are on that bus. In this

example we will use ID 4 and 5 for the controller’s ports.

1 From the Main menu, type <1> for the “Configuration” menu and press <Enter>.

Main Menu

2 From the Configuration menu, type <3> for the “Additional Host

Configuration” option and press <Enter>.

77

Chapter 6 - Additional Configuration

78


Configuration Menu

3 Choose “Host Port Hard Addressing” by typing <1> and pressing <Enter>.

Additional Host Configuration Menu

4 Choose the first controller port and set the hard address. Type <1> and press

<Enter>.

Host Port Hard Addressing Screen

5 Enter the new hard address for the controller port. Be sure to use the full

three digits, (e.g., 001, 002, 003, etc.).

Host Port Addressing




6 Choose the second controller port and set the hard address. Type <2> and

press <Enter>.


7 Enter the new hard address for the controller port and press <Enter>. Be sure

to use the full three digits, (e.g., 001, 002, 003, etc.).



pressing <Enter> until you reach the Main menu at which time you will be


Host Port Addressing 79


80


Setting the Controller LUN

The purpose of this option is to allow the controller to respond as a processor

device at a different LUN number. By default it is set to LUN 0. Under most

normal configurations the default setting is fine, however it is important that a

disk LUN not be defined to the same LUN number as the Controller LUN. The

disk LUN has priority and will cause a loss of communication from StorView to

the Controller. If you are using the default be sure to avoid using LUN 0 for a disk

LUN, if you have a requirement for the disk LUN to be set to LUN 0 such as with

a boot device, you will need to assign a new LUN number for the Controller.

1 From the Main menu, type <1> for the Configuration Menu and press <Enter>.

Main Menu

2 From the Configuration menu, type <3> for the “Additional Host Configuration”


Configuration Menu

Setting the Controller LUN



3 Choose “Controller LUN Configuration” by typing <2> and pressing <Enter>.


4 Enter the LUN number for the controller. It requires a three decimal value

(000, 001, etc.).

If you wish to disable the Controller LUN, press the <D> key. Disabling the

Controller LUN will cause the Controller to not be available at any particular

LUN number, (except for zero, which is always enabled). If it is set to any

other value (0...511), then the Controller will always respond to that LUN as a

processor device.

Controller LUN Configuration Menu




NOTE: Under Microsoft® Windows NT, the LUN can be checked using the

“SCSI Adapters” function. It will be reported as a scanner device (this

is not an error).

Setting the Controller LUN 81


82


Modify Configuration Name

To help identify a particular controller configuration, it is possible to assign a 64

byte ASCII name to it.


<Enter>.

Main Menu

2 From the Configuration menu, type <8> for the “Additional Configuration”


Configuration Menu

3 From the Additional Host Configuration menu, choose “Modify Configuration

Name” option. Type <1> and press <Enter>.

Modify Configuration Name




4 Type in the new name and press <Enter>. (You may keep the current name

by simply pressing <Enter>.

Modify Configuration Name Menu


pressing <Enter> until you reach the main menu at which time you will be


Modify Configuration Name 83


84


Modify Configuration WWN

When the controllers are manufactured, every controller is given a unique WWN.

When a configuration is created for the first time, the WWN for this configuration

will be the WWN of one of the controllers present at that time. Then, both

controllers will always use this WWN on the host loops. This is to facilitate

failover/failback. In the event that one or both controllers are changed, the new

controllers will still use the configuration WWN, meaning that there is no need to

reconfigure the controllers after failure, they can just be plugged in.

There are a few situations that will require modifying the WWN.

In the event that the controller which was removed is ever used to create another

configuration in another enclosure, there is the potential for two systems to

report the same WWN. This is not allowed, so in that case it is necessary to force

the original system to now use the WWN of one of the two controllers presently

installed. This is achieved by the “Modify Configuration WWN” feature. It may

also be necessary to change some host parameters too, depending on how the

host systems access the controller. If the host expects to see a particular WWN,

then this needs to be changed on the host.

In another case when a system is pre-configured for the customer, for whatever

reason, and the actual shipping controllers are not available. The configuration

may be setup using another controller that is not shipped with the system. Then

when the user receives their system, the controller installed does not have the

WWN of the controller that created the configuration. This generally happens

when the manufacturer or distributors are pre-configuring systems before

shipping.


Main Menu






Configuration Menu

3 From the Additional Host Configuration menu, choose the “Modify

Configuration WWN” option. Type <2> and press <Enter>.

Additional Configuration Menu

4 Select one controllers’ WWN by typing <1> or <2> and pressing <Enter>.

You can use either Controller's WWN, the primary focus is to change it from

the WWN of another Controller that has been removed and may now be

used elsewhere. For consistency use the WWN of Controller 0 if both are

present, unless advised otherwise by Technical Support.

NOTE: A reset of the controllers is required after changing this parameter.

Modify Configuration WWN 85


86


Modify WWN Name Menu




6 After you have saved the changes, press <2> for the “Diagnostics” menu and

press <Enter>.

Diagnostic Menu Screen

7 Reset the controllers, type <3> and press <Enter>.

8 Select a Controller that you wish to reset or choose both, and press <Enter>.

9 Press <Y> to confirm resetting the controller and press <Enter>.




Setting Single Controller Mode

In situations where only one controller is present, it is possible to configure the

controller for this operating mode. This prevents warnings that the other

Controller has failed or is missing. Select this option, and the following is

presented:


<Enter>.

Main Menu




Setting Single Controller Mode 87


88


3 From the Additional Host Configuration menu, choose the “Set Single

Controller Mode” option. Type <3> and press <Enter>.

Modify Additional Configuration Menu

4 Select <Y> to enable or <N> to disable, and press <Enter>.

Single Controller Mode Screen

Enabling single controller mode will cause the controller not to light the

partner fail LED when only one controller is present, and no errors will be

reported.




Setting Single Controller Mode



Background Verification

This option is used to automatically verify the media on all drives in the

background, when there is no host activity. If a media error is detected, the

controller can automatically re-write the data, providing that the array is in fault

tolerant mode.


Main Menu




3 From the Additional Host Configuration menu, choose the “Enable/Disable

Background Verification” option. Type <4> and press <Enter>.

Background Verification 89


90



4 Type <Y> to enable background verification or <N> to disable, and press

<Enter>.

Background Verification Configuration Menu

A message is displayed indicating the new state of the verification process.

Background Verification Change Complete

5 Press any key to continue and return to the Additional Configuration screen.

Background Verification



Controller SES

This option is used to enable or disable the controller’s ability to log SES (SCSI

Enclosure Services) events. When enabled, the controller logs events for enclosure

error events; such as power supply failure, fan failure, or temperature errors.


Main Menu




3 From the Additional Host Configuration menu, choose the “Enable/Disable

Controller SES” option. Type <5> and press <Enter>.

Controller SES 91


92



4 Type <Y> to enable Controller SES logging or <N> to disable, and press

<Enter>.

Controller SES Configuration Menu

A message is displayed indicating the new state of the SES logging process.

Controller SES Change Complete

5 Press any key to continue and return to the Additional Configuration screen.

Controller SES

Chapter 7


Monitor Interface

The imageRAID Controller has a comprehensive set of setup, diagnostic and

testing features which are accessed through the VT-100 interface. This mode is

know as the Monitor mode.

You can perform the following operations from the Monitor mode:

� Automatically set BAUD rate

� Erase flash memory

� Download and program operational firmware

� Low level board test

� Set date and time

� Display detailed voltage and temperature information

� Test controller memory

� Test controller signals and LEDs

NOTE: Several of the functions available in the Monitor mode are unique to

manufacturing and production. They are not designed for normal

controller use.

93

Chapter 7 - Monitor Interface

94


Using the Monitor

To access these functions, it is first necessary to enter the Monitor mode. Ensure

that the VT-100 terminal is connected to the controller. Refer to “Using the

VT-100 Terminal” on page 2 for the correct communication settings.

1 Power on the controller/enclosure and after approximately 12 seconds the

“Loop-Up” and “Loop-Activity” LEDs on the controller face plate will start to

flash, after approximately 12 seconds, the “Loop-Up” and “Loop-Activity”

LEDs will flash to indicate that it is in “Monitor Wait” mode. This will occur

for 10 seconds.

2 During this time, press the <Enter> key a number of times to enter the

monitor mode. This also sets the BAUD rate for the controller to whatever

BAUD rate the terminal is set to.

NOTE: Nothing is displayed on the VT-100 before or during this 12 second

wait.

This will invoke the Monitor mode for the controller and displays the

following.

Monitor Mode Screen

The initial message displays information about the revision, amount of

memory installed, BAUD rate, and whether a coprocessor is present. If a

coprocessor is present, it will have the same memory available as the main

processor. To get a list of the available commands, type “he” or “?.”

Using the Monitor



Monitor Option Display Screen

Monitor Commands

The commands available with the Monitor mode allow basic testing and setup of

the controller. Note that no higher level operational parameters (arrays, LUNs,

etc.) are set here. This interface is used to test and configure the controller

hardware. The following are explanations of each command:

ef: Erase Flash

The controller contains an 8 MB flash memory for storing operational firmware.

This command can be used to completely erase this flash. Note that this does not

affect the Monitor, since it resided in a separate 512K flash. Thus, it is safe to

erase and reprogram the flash at any time, without having to be concerned about

power failures.

Monitor Commands 95


96


do: Download Firmware

This command causes the controller to wait for an Xmodem transfer of the

firmware file. The transfer protocol can be either Xmodem or 1K Xmodem.

1K Xmodem is preferred due to its superior error detection (CRC as opposed to

checksum). After issuing this command, the user should start the transfer. Refer to

“Updating the RAID Controller Firmware” on page 99 for more details.

bt: Board Test

This command is used to test the controller memory and internal PCI busses. It

can be run indefinitely or for a specific number of loops. This is typically used in

a production test environment.

rs: Reset Controller

This resets the controller, and is identical to performing a power cycle. It causes

the controller to re initialize.

date: Set the Controller Date

This command is used to set the date on the controller, in the mm-dd-yyyy

format. The date is preserved when power is turned off.

time: Set the Controller Time

This command is used to set the time on the controller, in the hh:mm:ss 24-hour

format. The time is preserved when power is turned off. This should be set to

the local time where the controller is installed.

temp: Temperature and Voltage

The controller has a number of on-board temperature sensors, and voltage

monitoring. These are displayed in the following format:

do: Download Firmware



Monitor Temperature and Voltage Display

The temperature sensors are placed at various locations on the printed circuit

board. These allow the controller to monitor for any temperature problems, and

to alert the user and/or shutdown as necessary.

fill: Fill Memory with Data

This is used during production testing to ensure the Flash, NVRAM and SDRAM

are functioning correctly. This command fills the specified memory region with a

known pattern.

compare: Compare Memory

This is used during production testing to ensure the Flash, NVRAM and SDRAM

are functioning correctly. This command compares the specified memory region

with a known pattern, and if a mis-compare occurs it is output to the terminal.

This is used in conjunction with the fill command.

fill: Fill Memory with Data 97


98


iotest: Test Controller Backplane Signals and LEDs

This test is typically used when connected to a production test backplane. It

allows all external controller signals to be tested. Additionally, some internal

signals are tested, as are the LEDs.

iotest: Test Controller Backplane Signals and LEDs



Updating the RAID Controller Firmware

The following information describes the procedures to upload new Wahoo RAID

Controller firmware.

1 Ensure that the enclosure is properly powered down.

2 Connect one end of a null-modem RS-232 cable to one of the controller

RS-232 ports (either controller will work for duplex mode), and the other end

to the host system’s serial communication port.

Set your host system communication protocol to 115,200 Baud, 8 Data bits, 1

Stop bit, No parity, and Flow control off. Refer to “Using the VT-100

Terminal” on page 2 for the correct communication settings.

3 Power on the enclosure.

After approximately 12 seconds, the “Loop-Up” and “Loop-Activity” LEDs will

flash to indicate that it is in “Monitor Wait” mode. This will occur for 10

seconds. During this time, press the <Enter> key a number of times to enter

the command line monitor. This also sets the BAUD rate for the controller to

whatever BAUD rate the terminal is set to.

NOTE: Nothing is displayed on the VT-100 before or during this 12 second

wait.

This will invoke the Monitor mode for the controller.

Monitor Mode Screen

4 View the list of options. Type <Shift>+<?> and press <Enter>.

Updating the RAID Controller Firmware 99


100


Monitor Option Display Screen

5 To start the firmware download, type <do> and press <Enter>.

The screen will display a series of “C” characters.

6 When the “C’s” appear, click the Transfer pull-down menu selection and

choose Send File. Then select the “1K Xmodem” protocol.

7 Click the browse button and locate the new Firmware file and click “Send.”

The firmware file will be named similar to “W_t006.bin.”

NOTE: Ensure that the protocol “1K Xmodem” is selected.

From the send screen you can monitor the progress of the download. You

can safely stop the transfer without affecting your existing firmware any time

during the transfer until it has been completed. The upload does not

overwrite the firmware during the upload process, it writes the new code into

RAM until completed, then copies the new firmware code to the EEPROM

after verification.

If you elect to stop a download in progress, ensure that the stop (abort)

command was completed by clicking the Cancel button.




Xmodem Transfer Screen

8 After the transfer screen disappears, press the <Enter> key.

9 Restart the controller(s). Type <rs> and press <Enter>.

Monitor Restart Screen

The RAID Controller(s) will restart with the new firmware.

NOTE: In duplex mode (dual controllers), the second controller’s firmware is

automatically updated during the restart procedure.

Updating the RAID Controller Firmware 101


102



Chapter 8


Diagnostics

The imageRAID controller has a comprehensive VT-100 Diagnostics menu. The

following operations may be performed from the diagnostics menu:

� Drive Loop Integrity Test

� Download and Program Operational Firmware

� Download and Program Drive Firmware

� RAID 5/50 Parity Check

� Command Tracing

� Resetting the Controllers

� Shutting down the Controllers

To enter diagnostics:

1 After the controllers have completed booting enter the Main menu by

pressing <Control>+<W>.

Main Menu Screen

103

Chapter 8 - Diagnostics

104


2 Choose “Diagnostics”, type <2> and press <Enter>.

Diagnostic Menu Screen

From this menu you can enter the offline or online diagnostics, or reset or

shutdown the controller(s).

Offline diagnostics can only be performed when there is no activity and requires

that the controller be disconnected from the host system(s). Refer to “Offline

Diagnostics” on page 105.

Online diagnostics may be performed while there is no activity in progress with

the enclosure still connected to the host with no effect on the controller’s

operating status. Refer to “Online Diagnostics” on page 111.



Offline Diagnostics

Before performing this diagnostic, you must disconnect the host system(s) from

the controller.

1 From the Main menu type <2> and press <Enter> for the Diagnostics menu,

then from the Diagnostics Menu, select “Offline Diagnostics” <1> and press

<Enter>.

Diagnostic Menu

2 Select “Drive Loop Integrity Test”, type <1> and press <Enter>.

Offline Diagnostic Menu

This test is used to test the drive loops in a configured system. The test is

non-destructive. It can give an indication of whether there are any signal

integrity issues, or whether there are marginal components such as drives,

cables or enclosures components.

Offline Diagnostics 105


106


3 Setup the test data pattern. Type <4> and press <Enter>.

Drive Loop Integrity Test Menu

4 Select the appropriate test pattern by typing the appropriate key and press

<Enter>.

There are three possible choices. The data pattern can be modified, but for

normal operations the default of alternating pattern is sufficient.

Offline Test Pattern Setup Menu

Alternating data pattern provides a mix of different patterns including

random and zero patterns.

Random data provides a complete random pattern.

User defined data provides for entering a hexadecimal value of either 8, 16,

32, 64, or 128 bits, which is used in the repeating pattern.

Offline Diagnostics



If you choose User Defined Data Pattern, the following screen will appear,

enter the value for the test pattern. It will display the hex value.

User Defined Data Pattern Screen

5 Press any key to continue. You are returned to the Drive Loop Integrity Test

menu.

6 Choose the Loop Test, type the appropriate number and press <Enter>.

Drive Loop Integrity Test Menu

It is possible to set up the test for one or both loops. Typically, both drive

loops would be tested simultaneously, unless a specific problem on one loop

is being isolated.



108


7 Choose the type of loop test. Type the appropriate number “Read Test” <1>,

“Write Test” <2>, or “Write/Read Test” <3>, and press <Enter>.

Test Type Menu

The mixed test, “Read/Write” is recommended.

8 Enter the test duration in minutes. Enter the value and press the <Enter> key.

Test Duration Screen

If possible, it is advisable to run this test for an extended period of time, such

as overnight. After selecting the test duration, the test starts. First, a map of

the connected drives is displayed along with the drive IDs. This is helpful to

assist with isolating an error to a specific area.

Offline Diagnostics



Test Display Screen

The display indicates the data patterns used to test the loops. These data

patterns are selected to show up typical loop integrity issues. The test can be

cancelled at any time by pressing the <ESC> key. After the test completes,

that status is shown.

Test Results Screen

If errors occur during the test, they are displayed. If there are errors greater

than 250, the test automatically aborts.



110


The following is a brief summary of the errors.

For a properly configured system, errors should not occur. If any errors occur

during testing, it is necessary to find and replace the defective components. This

utility can be used after first installing a system, to ensure all cables and drives are

correctly installed, or whenever a system is moved or changed.

Error Description

Drive Loop x LIP Occurred A LIP has occurred on the specified drive loop. This can indicate a drive has been inserted or re initialized. It also can indicate a problem with the loop.

Drive Loop x Parity Error The drive has detected a CRC error while writing to it. This is typically due to a loop integrity issue before the drive in question.

Drive Loop x Timeout Error The specified drive has timed out. This error does not indicate a location for the error, since it may be before or after the drive.

Drive Loop x Underrun Error The controller has detected a CRC error while reading from a drive. This is typically due to a loop integrity issue after the drive in question.

Drive Loop x Reset Error The indicated drive has detected a reset on the loop. This error does not indicate a location for the error, since it may be before or after the drive.

Drive Loop x Internal Error An internal error has occurred while transferring data. This is not related to the loop, but indicates a problem with the controller.

Drive Loop x Loop (yyyy) Error An error has occurred indicated by the number yyyy. This type of error should be reported to nStor for further analysis.

Offline Diagnostics



Online Diagnostics

While the controller is connected to the host, it is possible to perform these

diagnostics. Activity should be suspended to the controller.

Open the Online Diagnostics menu.

1 From the Main menu, select “Diagnostics”, type <2> and press <Enter>.

Main Menu

2 Select “Online Diagnostics”, type <2> and press <Enter>.

Diagnostic Menu

Online Diagnostics 111


112


Controller Maintenance

You may need to download the firmware file from the support web site or have it

stored on a disc or floppy disk.



3 Select “Controller Maintenance”, type <1> and press <Enter>.

Online Diagnostics Menu

4 Select “Download and Program Controller Firmware”, type <1> and press

<Enter>.

Firmware Menu

The screen will display a series of “C” characters.

Controller Maintenance






The firmware file will be named similar to “W_2_10.bin.”

NOTE: Ensure that the protocol “1K Xmodem” is selected. The “1K

Xmodem” protocol is preferred due to its superior error detection

(CRC vs. checksum).

From the send screen you can monitor the progress of the download. You

can safely stop the transfer without affecting your existing firmware any time

during the transfer until it has been completed. The upload does not

overwrite the firmware during the upload process, it writes the new code into

RAM until completed, then copies the new firmware code to the EEPROM

after verification.

If you elect to stop a download in progress, ensure that the stop (abort)

command was completed by clicking the Cancel button.

Xmodem Send File Screen

The imageRAID Controller(s) will restart with the new firmware.

NOTE: In duplex mode (dual controllers), the second controller’s firmware is

automatically updated during the restart procedure.

Controller Maintenance 113


114


Drive Maintenance

CAUTION: It is recommended to backup data from any drives before attempting

a download. During the download process, it is essential that no

drives are inserted or removed, and that power is not switched off.

If a drive firmware download is interrupted, it is possible that the

drive will become unusable. For this reason, the controller will not

download drive firmware in parallel to all drives, but one drive at a

time. In this way, even if a power failure occurs, at most only one

drive will be damaged.

The controller has the ability to download firmware to the attached disk drives.

Due to the large number of disk drives on the market, it is essential to use only

an nStor approved drive firmware file. Different drive manufacturers have

different methods for downloading drive firmware, sometimes requiring vendor

unique utilities to perform the download. nStor rigorously tests drive firmware

and creates a single download file that can be used through the controller. This

eliminates the problems associated with host based utilities, and also the

possibility of permanently damaging a disk drive by downloading incorrect

firmware. When first selecting the Drive Maintenance option from the Main

Diagnostics menu, the following will appear:

Drive Firmware Menu

If this is the first time in this menu, there will be no valid firmware in the buffer.

The firmware can be read from one of two places, either from the host or from a

drive, if it was previously saved there.

Drive Maintenance



Download the Firmware from the Host



3 Select “Drive Maintenance”, type <2> and press <Enter>.


4 Select “Download Drive Firmware to Buffer”, type <1> and press <Enter>.

Drive Firmware Menu



Download the Firmware from the Host 115


116



The firmware file will be named something similar to “hitachi_ggn3.bin.”

Xmodem Send File Screen

After the download is complete, the following is displayed:

Transfer Complete Screen

If the file is a valid firmware file, then the Drive Firmware Menu will appear

similar to the example on the following page.

Download the Firmware from the Host



Valid Drive Firmware Menu

Here, the drive vendor and supported models are listed. This prevents the

incorrect firmware from being downloaded to a drive. The firmware version

is also displayed. It is also possible to read the firmware from a drive, if it

has previously been saved there. The controller reserves a buffer at the end

of the drives to store a copy of the drive’s firmware. This allows a new drive

to be updated in the field without requiring the firmware to be downloaded

again.



118


Read Firmware from a Drive to the Buffer

1 From the Main menu, select “Diagnostics” <2> and press <Enter>.

2 Select “Online Diagnostics” <2> and press <Enter>.

3 Select “Drive Maintenance” <2> and press <Enter>.


4 Choose “Read Firmware From Drive to Buffer” <2> and press <Enter>.

Drive Firmware Menu

If there is at least one valid firmware file on a drive, a menu similar to the

following screen sample will appear.




5 Select the appropriate firmware that is displayed <x> and press <Enter>.

Select Drive Firmware Menu

You will be returned to the Drive Firmware menu.

6 Choose “Program and Save Drive Firmware” <3> and press <Enter>.

7 Choose one of the available options <1-3> and press <Enter>.

Drive Type Firmware Menu

Here, it is possible to download to all supported drives, all supported unused

or hot spare drives, or all supported drives that are not already at this

firmware level.

If any used drives are to be updated, the warning screen will appear. Refer to

the illustration on the following page:

Read Firmware from a Drive to the Buffer 119


120


Drive Firmware Warning Screen

CAUTION: If a drive firmware download is interrupted, it is possible that the

drive will become unusable. For this reason, the controller will

not download drive firmware in parallel to all drives, but one

drive at a time.

When downloading drive firmware, the controller first checks the vendor and

product ID of each drive to determine whether the firmware should be

updated. Then, it updates the drive. If necessary, the mode pages are set

after the download. The following is a typical sequence of drive firmware

download, with mixed drives:

Drive Firmware Progress Screen

8 Power cycle the enclosure after a drive firmware download. This will ensure

the drives use the correct firmware.




RAID 5/50 Parity Check

If an array is not initialized during creation (i.e., trusted) or if both controllers are

replaced after the array is powered off with write operations in progress, it is

necessary to verify the parity data on the array. It may also be desirable to

perform this check from time to time.


Main Menu


Diagnostic Menu

3 Select “RAID 5/50 Parity Check”, type <3> and press <Enter>.

RAID 5/50 Parity Check 121


122



4 Select whether all or a specific RAID 5/50 array is to be checked. Enter the

selection <x> and press <Enter>.

Parity Check Menu

Here, it is possible to perform a parity check, rewrite, or check and rewrite.

These are defined as follows:

Option Description

Check Parity This option reads all the data and parity, calculates the XOR of the data, and compares it to the parity. If there is an error, it is displayed.

Rewrite Parity This option reads all the data, calculates the XOR of the data, and writes this out as the new parity. This is the fastest to complete, since it does not have the overhead of a comparison.

RAID 5/50 Parity Check



Type Parity Check Menu

5 Choose a parity check option, type <1, 2, or 3> and press <Enter>.

After starting a parity check, the RAID 5/50 Parity Check menu changes to

the following:

Modified Parity Check Menu

Here, it is possible to cancel a parity check and to check the progress. If an

error occurs during the check or rewrite, it is displayed on the terminal.

Additionally, an event is entered in the event log upon start and completion.

Check and Rewrite Parity This option reads all the data and parity, calculates the XOR of the data, and compares it to the parity. Then, if there is a discrepancy, it writes this out as the new parity. This is the slowest to complete, since it has the overhead of a comparison as well as a rewrite.

Option Description



124


Command Tracing

It is possible to monitor the commands that a host is sending to the controller,

and this has a variety of uses.

It can be used to verify that a host is communicating with the controller, or as a

method of debugging a system. All commands that are sent to the controller are

displayed on the terminal. Select Enable Target Command Trace from the Main

Diagnostics menu:


All incoming commands will be displayed on the terminal until the trace mode is

disabled. The format is similar to the following:

Trace Format Screen

Here, the important information presented is that the command is on port 1 (T1)

from the host with ID 124 (ID:7c) for LUN 0. The command is a SCSI inquiry

command (CDB: 12).

Command Tracing



Resetting the Controllers

1 From the Main menu type <2> and press <Enter> for the Diagnostics menu.

Diagnostic Menu

2 Select “Reset Controllers”, type <3> and press <Enter>.

Reset Controllers Screen

3 Type the number of the Controller you wish to reset and press <Enter>.

4 Answer Yes <Y> to confirm the reset operation.

Resetting the Controllers 125


126


Shutting Down the Controllers


Diagnostic Menu

2 Select “Shutdown Controllers”, type <4> and press <Enter>.

Shutdown Controllers Screen

3 Type the number of the Controller you wish to shutdown and press <Enter>.

4 Answer Yes <Y> to confirm the shutdown operation.

Shutting Down the Controllers



Diagnostics Dmp

During the process of troubleshooting a problem, it may be helpful to capture the

diagnostics data. The information capture to the VT-100 screen can then be saved

to an external file for later analysis.


Diagnostic Menu

NOTE: If you desire to capture all the information to a file, be sure to

enable the Transfer>Capture Text function of your terminal program

prior to the next step.

2 Select “Diagnostics Dump”, type <5> and press <Enter>.

The diagnostic data is immediately sent to VT-100 window.

Diagnostics Dmp 127


128


Diagnostics Dmp

Chapter 9


Statistics

Access Statistics

The imageRAID Controller monitors all incoming commands and calculates

various statistics. The statistics monitored include:

� Command Count

� Command Alignment

� Command Size

� Readahead Statistics

� Write Clustering Statistics

� RAID 5/50 Write Statistics

The controller maintains individual access statistics for each LUN and controller

port. These can be used to help balance the load from the host. Identical statistics

are maintained for both reads and writes.

1 After the Controllers have completed booting enter the Main menu by


2 From the Main menu, choose “Statistics”, type <4> and press <Enter>.

129

Chapter 9 - Statistics

130


Statistics Menu Screen

3 Select the controller you wish to examine. Choose <1> or <2> and press


Statistics View Menu Screen

The Statistics groups can be viewed all together or individually, as well as

statistics can be viewed on a LUN or Port basis.

Every time statistics are viewed, the controller first outputs the current time and

the time since it was last reset. The statistics can be reset at any time. This is

useful in determining the access pattern for a particular test or period of time.

Statistic Description

Total Read Commands This is a count of the total number of read accesses (for all ports and LUNs) that have occurred since the controller was first powered on. This value is never reset, and indicates the number of commands that are sent to the controller from when it was manufactured

Access Statistics



Total Read Transfer This is a count of the total number of 512 byte blocks (for all ports and LUNs) that have been read since the controller was first powered on. This value is never reset, and indicates the amount of data transferred by the controller from when it was manufactured.

Total Read Commands Since Reset

This is a count of the total number of read accesses that have occurred since these statistics were reset, or the controller was last powered on.

Total Read Transfer Since Reset This is a count of the total number of 512 byte blocks that have been read since these statistics were reset, or the controller was last powered on.

Alignment This is the percentage of commands whose address is aligned on the specified address boundary. The alignment of a command from a host system is determined by the command’s address. In an optimal system, a write of one chunk of data would reside exactly within a chunk on one disk. However, if this is not the case, this write will be split up into two separate writes to two different data drives. This of course will have a negative effect on performance. To overcome these problems, the user can, with more sophisticated operating systems, set the access size and alignment to an optimal value. These statistics can help the user to tune the operating system.

Access Size This is the percentage of commands whose size is as specified. On occasions, an application may specify a particular access size, but the operating system may reduce this. This can lead to performance problems. For example, on Microsoft® Windows NT/2000 the maximum command size that the operating system will allow is 1MB. Additionally, some HBAs further reduce this value by default. However, some benchmark programs allow a larger access size to be specified, resulting in misleading results. By viewing the statistics on the controller, it is possible to determine the exact size of commands sent by the host, and so determine whether the results are appropriate.

The access size, in conjunction with the alignment, gives an indication of how many drives are involved in an access. For example, consider a RAID 5/50 array with a chunk size of 64K. In this case, a 64K access with an alignment of 8K will actually involve 2 data drives, since it needs to access some data in the first drive, and the remaining data in the next drive. This is clearly inefficient, and could be improved by setting the alignment to 64K on the operating system. If that is not possible, using a larger chunk size can help, as this reduces the number of accesses that span chunks. Aligning an access on the same value as the access size will improve performance, as it will ensure that there are not multi-chunk accesses for commands that are smaller than a chunk size.


Access Statistics 131


132


The following is an example of Read Access Statistics:

Read Access Statistics Screen

Access Statistics



Readahead Statistics

If sequential read commands are sent to the controller, it assumes that the

following commands may also be sequential. It can then go and perform a read

of the data, before the host requests it. This improves performance, particularly

for smaller reads. The size of the readahead is calculated based on the original

command size, so the controller does not read too much data. The controller

maintains statistics for all readahead commands performed.

The following is an example of the readahead statistics:

Readahead Statistics Screen


Readahead Command Hit Rate This is the percentage of read command hits versus the total number of read commands that have been issued. This gives an indication of the sequential nature of the data access pattern from the host.

Readahead Command Efficiency

This is the percentage of the number of read command hits versus the projected number of readahead command hits. This is a measure of the efficiency of the readahead algorithm. A low value means that much of the data that the controller reads in the readahead command is not actually requested by the host.

Readahead Statistics 133


134


Sequential Command Interval In determining whether to perform a readahead or not, the controller will search back in the command queue whenever it receives a new read command that is not satisfied by an existing readahead cache buffer. In a multi threaded operating system, commands from one thread may be interspersed with commands from another thread. This requires that the controller not just check the immediately previous command. The controller will search back for a number of commands, to see if the new command is exactly sequential to any one of these previous commands. If it is, then the controller determines that the data access pattern is sequential, and so performs a readahead. These statistics record the average number of commands the controller must search back for when it finds a sequential command match, the maximum number, and also the percentage for each one of these values. These give an indication of the multi threaded nature of the host.


Readahead Statistics



Command Cluster Statistics

To increase performance, the controller can cluster sequential write commands

together to create a larger write command. This results in less commands being

sent to the disk drives. Additionally, if sufficient data is clustered by the controller,

then it can perform a a full stripe write for RAID5/50 arrays. This significantly

improves performance. In cases where the host does not send a sufficient number

of outstanding writes, writeback cache can be used to delay the write to disk,

increasing the likelihood of clustering more data.

The following is an example of the command cluster statistics:

Command Cluster Statistics Screen


Write Cluster Rate This is the percentage of the number of write commands that are part of a cluster versus the total number of write commands that have been issued. This gives an indication of the sequential nature of the data access pattern from the host, and of the performance of the writeback cache.

Command Cluster Statistics 135


136


Miscellaneous Statistics

To aid in tuning performance these additional statistics are provided for internal

development use.

RAID 5/50 Partial Command Cluster Rate

This is the percentage of the number of clustered commands that wrap around the end of a stripe versus the total number of clustered commands. This gives an indication of the alignment of the operating system.

RAID 5/50 Full Stripe Write Rate This is the percentage of the amount of data that is written as a full stripe write versus the total amount of data written. This gives an indication of the sequential nature of the data access pattern from the host, and of the performance of the writeback cache, for RAID 5/50 drive ranks.

Command Cluster Interval In determining whether to cluster write commands or not, the controller will search back in the command queue whenever it receives a new write command. In a multi threaded operating system, commands from each thread may be interspersed with commands from another thread. This requires that the controller not just check the immediately previous command. The controller will search back for a number of commands, to try to determine if the new command is exactly sequential to any one of these previous commands. If it is, then the controller determines that it can cluster these commands.

These statistics record the average and maximum number of commands the controller must search back for when it finds a sequential command match, and also the percentage for each one of these values.

Command Cluster Count When the controller clusters a write command, it may cluster a large number of them together. These statistics record the average and maximum number of commands the controller clusters, and also the percentage for each one of these values.


Miscellaneous Statistics

Chapter 10


Event Logs

Accessing Event Logs

The imageRAID Controller has a comprehensive non-volatile event log, that can

be used for monitoring controller events and error conditions. The following

event types are logged:

� Drive Failures and Errors

� Controller Failures and Errors

� Temperature and Voltage Errors

� Fibre Loop Events and Errors (Drive and Host)

� SES Enclosure Events

The event log size is 4096 entries, and older events are overwritten as necessary.

Some repetitive events such as LIPs are appended to previous events, so entries

are not used up unnecessarily. A time stamp accompanies each event, so the

user can easily determine when the event happened. The time stamp of both

controllers is synchronized to make the troubleshooting task easier.

1 After the controllers have completed booting enter the Main menu by


2 From the Main menu, choose “Event Logs” <5> and press <Enter>.

137

Chapter 10 - Event Logs

138


Events Log Menu Screen

3 Select the controller you wish to examine. Choose <1> or <2> and press


Events Menu Screen

The Events can be viewed all together or for a specific time period or device.

First, the current time and controller number are displayed. Then the

required events are listed. It is possible to view a certain type of events, all

events, or all events in a certain timeframe. It is advisable to capture the

output to a log file, since the number of events may exceed the VT-100

buffer.

Accessing Event Logs



These options are as follows:

A typical event log will look similar to the following:

Typical Event Log Screen

Event Log Description

View All Events This option lists all events. The number of events is indicated.

View New Events This lists all events that have occurred since the event log has last been viewed.

View Last 50 Events This option lists the last 50 events.

View Last Hour/24 Hour Events This lists the events that occurred in the specified time.

View Drive Loop Events This lists all the events that have occurred on the drive loops. These include LIPs, drive failures, etc.

View Controller Port Loop Events

This lists all the events that have occurred on the host side of the controller port loops. These include LIPs, hosts logging in, etc.

View Environmental Events This lists events such as temperature warnings, voltage level errors, battery errors, etc.

View Failed Drive List This is a list of drives that the controller has marked as having failed. These drives have either failed in operation, or been removed.

Accessing Event Logs 139


140


Controller Events

Following is a brief description of events that relate to the controller and

configuration.

Controller Event Messages Description Cause Action

Array xx Cache Disabled [Reasons]

The controller has disabled the writeback cache for the indicated reasons, as follows:

In Preparation For Shutdown:The controller has received a command to disable the writeback cache in preparation for shutdown.

Partner: The controller has disabled the cache because the partner controller has failed.

Battery: The controller has disabled the cache because the battery is either not present or fully charged.

Critical: The controller has disabled the cache because the array has become critical.

UPS: The controller has disabled the cache because a configured UPS has indicated power has been removed or communications has been lost with the UPS.

Disabling of writeback cache for the indicated reasons:

• A “prepare for shutdown” command has been received by the controller.

• The partner controller has failed.

• The battery is not charged or present.

• The array has become critical.

• The UPS has failed.

• No action necessary.

• Replace the failed controller.

• Charge the backup battery or re-install the battery.

• Resolve the array issue and rebuild the array.

• Replace the failed UPS.

Array xx Cache Re-Enabled The controller has re-enabled the writeback cache.

Re-enabling of writeback cache.

No action necessary.

Battery Failure This indicates that the onboard cache backup battery (if present) has failed.

Battery failure. Replace battery in the controller.

Battery OK This indicates that the onboard cache backup battery (if present) is fully charged.

Battery charging complete.


Battery Charging This indicates that the onboard cache backup battery is charging.

Battery charging started.


Cached Data Lost xxx Entries, yyy MB

If the controller is powered off with writeback cache still present, it is necessary to flush this data when power is restored. If power is removed for an extended time, the onboard battery will deplete, and the cached data will be lost.

Failure of power for an extended time with writeback cache present.

Check the file system.

Controller Events



Cntrl x Temp Exceeded: yy The current internal controller temperature is indicated in degrees Celsius. The controller’s internal temperature has exceeded the allowable limit, and the controller will shut down to prevent damage to internal components.

Blocked fan.

Failing fan.

Elevated ambient temperature.

Check enclosure for sufficient air flow.

Check for a failed fan, if found replace cooling fan module.

Check the ambient temperature of the environment, decrease the local ambient temperature.

Cntrl x Temp Warning: yy Shutdown will occur at zzC

The current internal controller temperature is indicated in degrees Celsius. The controller’s internal temperature is approaching the maximum allowable limit.

Blocked fan.

Failing fan.

Elevated ambient temperature.

Check enclosure for sufficient air flow.

Check for a failed fan, if found replace cooling fan module.

Check the ambient temperature of the environment, decrease the local ambient temperature.

Controller x Failed/Removed The specified controller has been removed, or has failed. Message is generated by the partner controller.

Failure or removal of one controller (partner) in an Active-Active configuration.

Replace the controller.

Controller x Timeout The specified controller has not responded in the allotted time. Message is generated by the partner controller.

Failure or removal of one controller (partner) in an Active-Active configuration.


Controller x Present The specified controller has been inserted.

Partner controller has been inserted.


Controller x Valid The specified controller has passed it’s selftest and is now ready (failback).

Partner controller is ready to fail back.


Controller Failover Started The process of failing over to the remaining controller has started.

Failure or removal of the partner controller.


Controller Failover Completed The process of failing over to the remaining controller has completed.

Completion of failover process.



Controller Events 141


142


Controller Failback Started The process of failing back to a replacement controller has started.

Partner controller started failback.


Controller Failback Completed The process of failing back to a replacement controller has completed.

Completion of failback process.


Controller Powered On The controller has been powered on or reset.

The controller was powered on.


Controller Powered Off The controller has been powered off or removed.

Removal of controller or power.


Controller Reset The controller has been reset either locally or remotely.

User initiated a controller reset.


Controller Shutdown The controller has been shutdown either locally or remotely.

User initiated a controller shutdown.

The controller temperature was exceeded and the controller shut itself down.


Check for a failed fan, replace as needed. Check for blocked air flow, correct as needed. Check for high ambient temperature, reduce the environments ambient temperature.

Controller Selftest Passed The controller has completed it’s power on selftest.

Selftest completion on startup.


Controller Selftest Failed The controller has failed it’s power on selftest.

Selftest failure on startup.


Controller Firmware Upgraded to version xxxx

The controller’s firmware has been upgraded to the indicated version.

User upgraded the controller firmware.


Configuration Changed The configuration has been changed.

A change in the configuration has occurred.

If you are using the Save Configuration feature, re-save your configuration information - it no longer matches, otherwise no action is necessary.

DMA Error This indicates an internal error in the controller.

Internal hardware failure.

Replace the controller. Contact Technical Support.

Event Log Cleared The controller’s event log has been erased.

The user has cleared the event log.



Controller Events



Fatal Watchdog Error The controller watchdog timer has detected a fatal hardware or firmware error.

Internal hardware or firmware failure.


Fatal SDRAM ECC The controller has detected unrecoverable ECC errors on the SDRAM, either a multiple bit error or uncorrectable single bit error.

Fault SDRAM or damaged internal bus.


Fatal Coprocessor Error The coprocessor has failed. Internal hardware or firmware failure on the coprocessor.


Fatal Host FC Channel x Error The controller has detected a fatal error on the indicated host channel.

Memory or bus error on the indicated channel.


Fatal Drive FC Channel x Error The controller has detected a fatal error on the indicated drive channel.

Memory or bus error on the indicated channel.


Flush Mirrored Cache Started, xxx Entries, yyy MB

After failover, mirrored writeback cache needs to be flushed to the drives. This indicates the start of that process.

Failure or removal of the partner controller.


Flush Mirrored Cache Completed

The flushing of mirrored writeback cache after a controller failure has completed.

Completion of mirrored cache flushing.


Flush Cache Started, xxx Entries, yyy MB

If the controller is powered off with writeback cache still present, it is necessary to flush this data when power is restored. This indicates the start of that process.

Failure of power with writeback cache present.


Flush Cache Completed The flushing of writeback cache after a power failure has completed.

Completion of cache flushing.


Recovered SDRAM ECC Error, Address: xxxxxxxx Bit: yy

The controller has detected and corrected a single bit ECC error in the SDRAM. The address and error bit are indicated.

SDRAM error. If it repeats, replace the controller.

Resync Started A stripe synchronization of a RAID 5/50 set has started. This will be done when a controller fails or when powered off during a write.

A controller fails or is powered off during a RAID 5/50 write operation.





144


Resync Completed A stripe synchronization of a RAID 5/50 set has completed.

A controller fails or is powered off during a RAID 5/50 write operation.


UPS Failure This indicates that an attached UPS (if configured) has failed.

UPS has been disconnected or communications with the UPS has failed.

Check the UPS monitoring connections. Check the UPS, if necessary replace it.

UPS OK This indicates that an attached UPS (if configured) is functioning correctly.

UPS is connected. No action necessary.

Voltage Error: Monitored Voltage: xx.yyV

This indicates that the specified voltage has exceeded limits. This is a serious error, and should be reported to technical support.

Voltage regulator hardware failure.

Enclosure 5V or 12V problem in the power supply.


Replace the defective power supply.

Controller Version Mismatch Indicates the version of the controllers is different.

The cause of this is both controllers have different drive (Disk I/O Card) or host loop (Host I/O Card) operation parameters. The controllers need to be individually checked by looking at the controller information on the respective controller. (i.e., to check Controller 0, look on the VT-100 connected to Controller 0). Only the enabled drive and host loops will have information.

The action to be taken depends on how this occurred.

Insert matching controllers.

Upgrade the host loop (Host I/O Card) to dual loop. This could occur if a customer bought a 4311F first, then another controller to make it a 4321F. They would need to update the host port (Host I/O Card) of the original controller using the key provided by technical support.

If the user plugs a single loop controller into a standard dual loop system, purchase a licence or key to upgrade both loops (Disk I/O Card and Host I/O Card).


Controller Events



Drive Events

These events are related to the drive loops and arrays.

Drive Event Messages Description Cause Action

Array xx Critical This indicates the specified array is critical.

Drive removal or failure.

Replace the disk drive and rebuild the array.

Array xx Expansion Started Expansion of the indicated array has started.

Expansion has started.


Array xx Expansion Restarted Expansion of the indicated array has restarted, after a power cycle or failover/failback.

Expansion has restarted.


Array xx Expansion Complete Expansion of the indicated array has completed.

Expansion has completed.


Array xx Initialization Started Initialization of the indicated array has started.

Initialization has started.


Array xx Initialization Complete Initialization of the indicated array has completed.

Initialization has completed.


Array xx Parity Check/Rewrite/Check and Rewrite Started

These events indicate that a RAID 5/50 Parity Check, Rewrite or Check and Rewrite has started.

Parity check started. No action necessary.

Array xx Parity Check/Rewrite/Check and Rewrite Completed, yy Errors

These events indicate that a RAID 5/50 Parity Check, Rewrite, or Check and Rewrite has completed. Any errors are indicated.

Parity check completed.


Array xx Parity Check/Rewrite/Check and Rewrite Aborted, yy Errors

These events indicate that a RAID 5/50 Parity Check, Rewrite, or Check and Rewrite has been stopped. Any errors up to this point are indicated.

Parity check canceled by the user.


Controller LIP: Drive Loop x The controller has initiated a LIP on the specified loop. This typically occurs during failover/failback or if a loop indicates problems.

Controller initiated a LIP.


Drive Events 145


146


Data Underrun WWN:xx xx xx xx xx xx Py OP:zz

The controller has detected a data underrun from the indicated drive. This is caused by the controller detecting a bad CRC in a frame, and usually indicates a link problem, either with cabling or an enclosure. The meaning of each field is as follows:

WWN: Drive World Wide Name

P: Drive port number

OP: SCSI command code

Loop error. Check cabling, Disk I/O card, and ensure that the disk drive is properly seated.

Drive Busy WWN:xx xx xx xx xx xx Py OP:zz

The drive indicated has returned a Busy status to a command. The meaning of each field is as follows:




Drive is busy and cannot accept any more commands.


Drive CC WWN:xx xx xx xx xx xx Py OP:zz SN:aa EXT:bb QL:cc

The drive indicated has returned a Check Condition status to a command. The meaning of each field is as follows:




SN: SCSI sense key

EXT: SCSI extended sense key

QL: SCSI extended sense key qualifier

Typically due to a non-recoverable media error, hardware error, or loop error.


Drive CC Failure WWN:xx xx xx xx xx xx A:yy D:zz

The drive listed has failed due to an unrecoverable error. The meaning of each field is as follows:


A: Array number

D: Drive number in array

Typically due to a non-recoverable media error or hardware error.

Replace the disk drive.

Drive Inserted WWN:xx xx xx xx xx xx xx ID;yyy

The drive indicated has been inserted. The meaning of each field is as follows:


ID: Drive hard ID

Drive was inserted. No action necessary.


Drive Events



Drive Invalid Block Size (Size) WWN:xx xx xx xx xx xx Py

This indicates that the specified drive has an invalid block size, and so cannot be used. A low level format is required. The meaning of each field is as follows:



Size: Current block size (in hexadecimal). The controller requires 512 Bytes.

The drive has an invalid block size.


Drive Missing WWN:xx xx xx xx xx xx A:yy D:z

The drive listed has been marked as failed due to being removed from the disk drive enclosure. The meaning of each field is as follows:


A: Array number


Drive has been removed or bypassed by the user, or has a serious hardware error.

Removal of cables connecting the enclosures.

Removal of power to daisy-chained enclosure(s).


Replace the cables.

Restore power to the daisy-chained enclosure(s).

Drive Rebuild Failure WWN:xx xx xx xx xx xx A:yy D:zz

Rebuild has failed due to an unrecoverable error on another drive in the array. The meaning of each field is as follows:


A: Array number


Typically due to a non-recoverable media error or hardware error.

Backup all data and restore to a new array.

Drive Status aa WWN:xx xx xx xx xx xx Py OP:zz

The drive indicated has returned an unknown status to a command. The meaning of each field is as follows:




Unknown status returned by the disk drive.

Contact Technical Support and provide them with a copy of the event log.

Drive Task Full WWN:xx xx xx xx xx xx Py OP:zz

The drive indicated has returned a Task Full to a command. The meaning of each field is as follows:




Drive is busy and cannot accept any more commands.



Drive Events 147


148


Drive Timeout WWN:xx xx xx xx xx xx Py OP:zz

The drive indicated has timed out for the specified command. The meaning of each field is as follows:




Drive hardware error or loop error.

Check cabling, Disk I/O cards, and ensure the disk drives are properly seated.

Drive Timeout Failure WWN:xx xx xx xx xx xx A:yy D:zz

The drive listed has failed because of a timeout. The meaning of each field is as follows:


A: Array number


Drive error. Replace the disk drive.

Excessive Errors on Drive Loop x, Loop Disabled

The indicated loop has exceeded the allowable error count. The controller will not use this loop for data transfers. After two hours have elapsed, the loop will be re-enabled.

Damaged cable.

Faulty Disk I/O card (LRC failure).

Check/replace cable.

Check/replace Disk I/O card.

LIP: Drive Loop x (Repeated y times)

A LIP has occurred on the indicated drive loop.

Loop disruption or drive insertion.


Loop Up: Drive Loop x The indicated drive loop is now up.

Loop becoming ready.


Loop Down: Drive Loop x The indicated drive loop is now down.

Loop has gone down.

Check/replace cable.

Check/Replace Disk I/O card.

Loop Error On Drive Loop x The loop indicated is down. This means that the loop is no longer functional and no commands are being passed on this loop until a “Loop Up” is reported in the log again. This can indicate a problem with the loop, such as an LRC card or cable.

Loop is going down. Check/replace cable.

Check/Replace Disk I/O card.

New Drive Rebuild Failure WWN:xx xx xx xx xx xx A:yy D:zz

Rebuild has failed due to an unrecoverable error on the new drive. The meaning of each field is as follows:


A: Array number


Typically due to a non-recoverable media error, or hardware error.

Replace new drive and initiate a rebuild.


Drive Events



Rebuild AbortedA:xx D:yy

A rebuild has been stopped by the user on the indicated drive. The meaning of each field is as follows:

A: Array number


A rebuild was canceled by the user.


Rebuild Started WWN:xx xx xx xx xx xx A:yy D:zz

A rebuild has started on the indicated drive. The meaning of each field is as follows:


A: Array number


A rebuild has started. No action necessary.

Rebuild Restarted WWN:xx xx xx xx xx xx A:yy D:zz

A rebuild has restarted on the indicated drive. The meaning of each field is as follows:


A: Array number


A rebuild has started. No action necessary.

Rebuild Complete A:xx D:yy A rebuild has completed on the indicated drive. The meaning of each field is as follows:

A: Array number


A rebuild has completed.


SES Initialized: SES LEVEL 1. Verify the Configuration

This event indicates that there is a problem with the SES processor, no drives are in the specified SES slots, or there is an address conflict with the enclosures.

No drives installed in slots 1 or 7.

Enclosure ID conflict.

Insert a disk drive in either or both drive slots 1 and 7.

Check the enclosure ID on all enclosure(s).

SES Initialized: SES LEVEL 3, Access: Critical

This event indicates that only one SES drive has been found in one or more of the attached enclosures.

No drives installed in slots 1 or 7.

Insert a disk drive in either or both drive slots 1 and 7.

SES Initialized: SES LEVEL 3, Access: Fault Tolerant

This event indicates that both SES drives have been found in one or more of the attached enclosures.

Drives are installed in slots 1 or 7.


Some Or All Drives Not Available On Loop x

There is a problem with some drives on the indicated loop, but the loop is still up and functional.

Disk I/O card removal or failure.

Cable disconnected or failed.

Check/replace Disk I/O card.

Check/replace data cable.


Drive Events 149


150


WWN:xx xx xx xx xx xx ID:aaa FW Download Start Rev:yyyy

The drive listed has started a firmware upgrade to the indicated revision. The meaning of each field is as follows:


ID: Drive Target ID

Rev: New firmware version

A firmware upgrade has started.


WWN:xx xx xx xx xx xx ID:aaa FW Download Complete

The drive listed has completed a firmware upgrade. The meaning of each field is as follows:


ID: Drive Target ID

The firmware upgrade is complete.


WWN:xx xx xx xx xx xx ID:aaa Reallocate LBA yyyyyyyy

The controller has reallocated blocks at the indicated LBA in the drive. This is due to a verify failure. The meaning of each field is as follows:


ID: Drive Target ID

LBA: Address of reallocated blocks (hexadecimal)

Disk scrubbing detected an error.



Drive Events



Host Events

These events are related to the host side loops (Controller Port loops).

Host Event Messages Description Cause Action

Controller LIP: Host Loop x The controller has initiated a LIP on the specified loop. This typically occurs during failover/failback.

Controller initiated a LIP.


CxPx Detected Power-on/Reset at ID:yyy LUN:zzz

The specified host has accessed a LUN for the first time, or for the first time following a reset. The meaning of each field is as follows:

CxPx: Controller and host port number

ID: Loop ID of the host

LUN: LUN number accessed

First access by a particular host after a LIP or reset.


Host Port x Incorrect Address, Should Be:yyy Is:zzz

The host port indicated has not taken the Loop ID specified in the Hard Address setting. It is important to power cycle the controllers after changing this setting in the configuration.

Address conflict with either host adapter or other device on the same loop.

Resolve address conflict.

Host Px WWN: yy yy yy yy yy yy Logged in at ID:zzz

The indicated host system has logged into the controller. These events will only be listed for hosts that have SAN mappings created.

P: Host port number

WWN: Host World Wide Name

ID: ID of Host on the loop

Host systems logs into the controller.


LIP: Host Loop x (Repeated y times)

A LIP has occurred on the indicated host loop.

Loop disruption or host booting.


Loop Down: Host Loop x The indicated host loop is now down.

Loop is down. Check/replace cable.

Check/Replace Host I/O card.

Loop Up: Host Loop x The indicated host loop is now up.

Loop is ready. No action necessary.

Host Events 151


152


CC to Host ID:xxx LUN:yyy Pz OP:aa SN:bb EXT:cc QL:dd

This indicates that an unrecoverable drive error has occurred for a particular command. This may be due to a drive error in a non fault tolerant mode, such as RAID 0, or when the array is already in a degraded mode. The controller will pass the status from the drive back to the host system, to allow the host recovery mechanisms to be used. The meaning of each field is as follows:

ID: Host ID

LUN: LUN requested

P: Host port number


SN: SCSI sense key

EXT: SCSI extended sense key

QL: SCSI extended sense key qualifier

Data is not recoverable.

Backup all data, and restore to a new array.

Host Event Messages Description Cause Action

Host Events



SES Events

These events are related to the SES monitored enclosure components.

SES Event Messages Description Cause Action

Encl.yy Power supply zz OK WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated power supply is functioning correctly.

Normal condition reported.


Encl.yy Power supply zz Critical WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated power supply is not functioning.

The specific power supply has failed.

The specific power supply is powered off.

Replace the power supply.

Ensure that the specific power supply On/Off button is in the On position ( l ).

Encl.yy Power supply zz Not Present WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated power supply is not present.

The power supply was removed.

Re-insert the power supply, connect the AC power cord and power on the power supply.

Encl.yy Fan zz OK WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated fan is functioning correctly.

Normal condition reported.


Encl.yy Fan zz Critical WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated fan is not functioning.

A specific fan failure.

Total fan failure.

Cooling fan module was removed.

Replace the cooling fan module.

Replace the cooling fan module.

Re-inset the cooling fan module.

Encl.yy Temp. zz <aa>C OK WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated temperature sensor is within limits.

Temperature sensors are reporting normal temperatures in the enclosure.

No action required.

Encl.yy Temp. zz <aa>C Warning WWN: xx xx xx xx xx xx xx xx

(continued next page)

The controller has detected that the indicated temperature sensor has reached a warning limit.

Temperature sensors are reporting enclosure temperatures have reached the threshold of 70°C.

Automatic system shutdown will begin. In Active-Active controller configurations, one controller will shutdown its partner and shutdown the drives, then itself.

SES Events 153


154


Encl.yy Temp. zz <aa>C Warning WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated temperature sensor has reached a warning limit.


Ensure that both cooling fans are operating normally. (Replace if needed.)

If the fans are set to automatic speed control, place the jumper on the Cooling fan module circuit board to force the fans to high speed.

If the environment ambient temperature is high, reduce the ambient temperature.

Ensure that the airflow is not blocked or restricted on the enclosure.

Encl.yy Temp. zz <aa>C Critical WWN: xx xx xx xx xx xx xx xx

The controller has detected that the indicated temperature sensor has reached a critical limit.


If the fans are set to automatic speed control, place the jumper on the Cooling fan module circuit board to force the fans to high speed.

If the environment ambient temperature is high, reduce the ambient temperature.

Ensure that the airflow is not blocked or restricted on the enclosure.

Encl.xx Alarm zz is OFF WWN: yy yy yy yy yy yy

All conditions are normal or the alarm is off.

No condition being reported.

Alarm silenced.


User pressed the Alarm Silence button on the front panel.

Encl.xx Alarm zz is ON WWN: yy yy yy yy yy yy

Alarm is on. A condition caused the alarm to sound.

Press the Alarm Silence button on the front panel and isolate the cause of the alarm.

SES Event Messages Description Cause Action

SES Events



Failed Drives

The controller maintains a list of failed drives. Drives are listed in the following

format:

Failed Drive:xx WWN: yy yy yy yy yy yy

Reason Code

The reason code may be one of the following:

Reason Code Reason Action

Drive Timeout The drive has either timed out or been removed.

Re-insert the disk drive.


Command: xx Sense Key: yy Ext Sense: zz

The drive has failed for the specified command, with the indicated SCSI sense key and extended sense key.


Failed Drives 155


156


Failed Drives

Appendix A


Optimizing RAID 5 Write Performance

Introduction

With a typical RAID 5 implementation, there are a number of steps necessary to

write the data to the media. Every write from the host system will typically

generate two XOR operations and their associated data transfers, to two drives. If

the accesses are sequential, the parity information will be updated a number of

times in succession. However, if the host writes sufficient data to cover a

complete stripe, the parity data does not need to be updated for each write, but it

can be recalculated instead. This operation takes only one XOR operation per

host write, compared to two for a standard RAID 5 write. The number of data

transfers necessary are also reduced, increasing the available bandwidth. This

type of write access is termed a “Full Stripe Write.”

Distribution of Data and Parity in a RAID 5 with Five Drives

P (20-23) C20 C21 C22 C23 Stripe 5

P (16-19)C17 C18 C19 Stripe 4

P (12-15)C13 C14 C15 Stripe 3

P (8-11)C9 C10 C11 Stripe 2

P (4-7)

C16

C12

C8

C4 C5 C6 C7 Stripe 1

P (0-3) C0 C1 C2 C3 Stripe 0

157

Appendix A - Optimizing RAID 5 Write Performance

158


The illustration on the previous page displays the distribution of data chunks

(denoted by Cx) and their associated parity (denoted by P(y-z)) in a RAID 5 array

of five drives. An “array” is defined as a set of drives, on which data is

distributed. An array will have one RAID level. A “chunk” is the amount of

contiguous data stored on one drive before the controller switches over to the

next drive. This parameter is adjustable from 64K to 256K, and should be

carefully chosen to match the access sizes of the operating system. A Stripe is a

set of disk chunks in an array with the same address. In the above example,

Stripe 0 consists of C0, C1, C2, and C3 and their associated parity.

Maximum performance will be achieved when all drives are performing multiple

commands in parallel. To take advantage of a Full Stripe Write, the host has to

send enough data to the controller. This can be accomplished in two ways.

Firstly, if the host sends one command with sufficient data to fill a stripe, then the

controller can perform a Full Stripe Write. Alternatively, if the host sends

multiple sequential commands, smaller than a stripe size (typically matching the

chunk size), the controller can internally combine these commands to get the

same effect. In the above example, if a 256K chunk size is used, then the stripe

size is 1MB (4 chunks * 256K). So, for maximum performance, the host could

either send 5 * 1 MB write commands, or 20 * 256K write commands.

The effectiveness of the controller’s ability to perform a Full Stripe Write depends

on a number of parameters:

Sequential Access

If the commands sent from the host are not sequential, the controller will not be

able to cluster them together. So, unless each individual access is sufficient to fill

a stripe, a Full Stripe Write will not occur.

Number of Outstanding Commands

For the controller to successfully cluster commands, there has to be a number of

write commands sent simultaneously. Setting the host to send up to 64

commands should prove adequate. Alternatively, enabling writeback cache will

have a similar effect, as the controller can then cluster sequential commands even

if the host only sends a small number of commands at a time.

Sequential Access



Access Size

With very small accesses, it is necessary to have a large number of commands to

cluster together to fill up a full stripe. So, the larger the access size the better. It

is best to use an access size that will fill a chunk. Of course, even if a stripe is

not filled up, small sequential writes will still benefit from command clustering.

Access Alignment

The alignment of a command from a host system is determined by the

command’s address. In an optimal system, a write of one chunk of data would

reside exactly within a chunk on one disk. However, if this is not the case, this

write will be split up into two separate writes to two different data drives. This

will have a negative effect on performance. To overcome these problems, the

user can, with more sophisticated operating systems, set the access size and

alignment to an optimal value.

As can be seen from the figure below, to get the highest performance from this

system, it is necessary to have a number of stripes being written in parallel. As

the array expands, with more and more drives, the number of commands (and

amount of sequential data) necessary to do this increases.

Distribution of Data and Parity in a RAID 5 with Eight Drives

P (56-62) C56 C57 C58 C59 C60 C61 C62 Stripe 8

C53 C54 C55 Stripe 7





P (49-55)C50 C51 C52

P (42-48)C43 C44 C45

P (20-23)C36 C37 C38

P (28-34)C29 C30 C31

P (21-27)C22 C23

C17 C18 C19 C20 Stripe 2P (14-20)C15 C16



P (7-13)

C49

C42

C35

C28

C21

C14

C7 C8 C9

P (0-6) C0 C1 C2

Access Size 159


160


In the figure on the previous page, we can see that seven chunks of sequential

data are necessary to fill a stripe. To have multiple commands active for all disk

drives, this requires more data than for the case with five drives. As can be seen,

this number will increase as the number of drives increases. If a large number of

drives are used, it may get difficult to achieve maximum performance, as it

becomes more difficult to cluster a large number of commands to achieve a Full

Stripe Write.

RAID 5 Sub-Array

This difficulty in realizing the maximum performance possible introduces the

concept of a Sub-Array. Suppose an array consisted of two RAID 5 sets, as

described in the figure on page 157. If these are then striped, the resulting array

would appear as shown in the figure below. In this case, in order for a Full

Stripe Write to be performed, it is still only necessary to cluster four write

commands together, as opposed to the seven necessary in the figure below. The

array of drives appears as two separate sub-arrays, each with it’s own rotating

parity.

Distribution of Data and Parity in a RAID 5 with Ten Drives and Two Sub-Arrays

It can be seen that the more sub-arrays used, the more likely it is for a Full Stripe

Write to occur, and hence the higher the performance. It is recommended to use

either four or five drives in a sub-array, for best performance. On the following

page the figure shows that even with 15 drives, it is still possible to perform Full

Stripe Writes, by clustering together 4 chunks of data.

P (40-43) P (44-47)C40 C41 C42 C43 C44 C45 Stripe 5

C36 C37 Stripe 4

C27 C28 C30

C38

Stripe 3

C19 C21 P (20-23) Stripe 2

C12 P (12-15) C13 Stripe 1

C3 P (4-7) C4 C5 Stripe 0

P (32-35)C33 C34 C35

C29C25 C26 P (24-27)

C20C17 P (16-19) C18

C11P (8-11) C9 C10

C2C0 C1

C46

P (28-31)

C22

C14

C6

C39

C47

C31

C23

C15

C7

P (36-39)C32

C24

C16

C8

P (0-3)

RAID 5 Sub-Array



Distribution of Data and Parity in a RAID 5 with Fifteen Drives and Three Sub-Arrays

As well as the performance advantage gained from using multiple sub-arrays,

there are a number of other advantages:

Multiple Drive Failures

In a configuration with multiple sub-arrays, it is possible for the array to sustain

multiple drive failures, provided that there is only one failure in each sub-array.

Faster Rebuild

A rebuild operation must read data and calculate parity from all the remaining

drives in the RAID set. If multiple sub-arrays are used, this means that it is only

necessary to read the data from the remaining drives in the sub-array, not all of

the drives in the array. This increases both the rebuild speed and the speed of

access to missing data, which also has to be recreated from the remaining drives.

Summary

In summary, for maximum performance using RAID 5, it is recommended to use

four or five drives in a sub-array. If there are more than five drives in a sub-array,

it is better to use a smaller chunk size, say 64K or 128K, as this will lead to more

Full Stripe Writes.

P PC60 C61 C62 C63 C64 C65

C52 C53

C39 C40 C42

C54

C27 C29 P

C16 P C17

C3 P C4 C5

PC49 C50 C51

C41C37 C38 P

C28C25 P C26

C15P C13 C14

C2C0 C1

C66

P

C30

C18

C6

C55

C67

C43

C31

C19

C7

P

P C68

C56

C45

C57

C32 C33

C20 P

P C8

C44

C69

C46

P

C21

C9

C58

C70

P

C34

C22

C10

C59

Stripe 5

Stripe 4

Stripe 3

Stripe 2

Stripe 1

Stripe 0

C71

C47

C35

C23

C11

PC48

C36

C24

C12

P

Multiple Drive Failures 161


162


Summary

Index


AAccess Alignment 159

Access Size 159

Accessing Configuration Utility 1

Additional Configuration 77

Assigning Hot Spare 28

BBackground Verification 89

BBU Battery Hold-Up Times 23

CCheck Parity 122

Chunk Size 19

Command Cluster Statistics 135

Command Tracing 124

Communications Parameters 2

Configuration UtilityAccessing 1

Configuring Array Cache Parameters 23

Configuring Array Readahead Cache 26

Controller EventBattery Charging 140Battery Failure 140Battery OK 140Cache Disabled 140Cache Re-Enabled 140Cached Data Lost 140Cntrl Temp Exceeded 141Cntrl Temp Warning 141Configuration Changed 142Controller Failback Completed 142Controller Failback Started 142Controller Failed 141Controller Failover Completed 141Controller Failover Started 141

Controller Firmware Upgraded 142Controller Powered Off 142Controller Powered On 142Controller Present 141Controller Removed 141Controller Reset 142Controller Selftest Failed 142Controller Selftest Passed 142Controller Shutdown 142Controller Timeout 141Controller Valid 141DMA Error 142Encl Temp Critical 154Encl Temp OK 153Encl Temp Warning 153, 154Event Log Cleared 142Fan Critical 153Fan OK 153Fatal Coprocessor Error 143Fatal Drive FC Channel Error 143Fatal Host FC Channel Error 143Fatal SDRAM ECC 143Fatal Watchdog Error 143Flush Cache Completed 143Flush Cache Started 143Flush Mirrored Cache 143Flush Mirrored Cache Started 143Power Supply Critical 153Power Supply Not Present 153Power Supply OK 153Recovered SDRAM ECC Error 143Resync Completed 144Resync Started 143UPS Failure 144UPS OK 144Voltage Error 144


Controller Information 7

Controller Information Display 8

163

164


Controller LUN 80

Controller LUN Configuration Menu 81

Controller Maintenance 112

Controller SES 91

Creating a Host LUN 42

Creating Arrays 11

DDeleting an Array 31

Diagnostics 103

Diagnostics Dump 127

Disk LUN 80


Drive Capacity Reserve 21

Drive EventArray Critical 145Array Expansion Complete 145Array Expansion Restarted 145Array Expansion Started 145Array Initialization Complete 145Array Initialization Started 145Array Parity Check Canceled 145Array Parity Check Completed 145Array Parity Check Started 145Controller LIP 145Data Underrun 146Drive Busy 146Drive CC 146Drive CC Failure 146Drive Inserted 146Drive Invalid Block Size 147Drive Missing 147Drive Rebuild Failure 147Drive Status 147Drive Task Full 147Drive Timeout 148Drive Timeout Failure 148Excessive Errors on Drive Loop 148FW Download Complete 150FW Download Start 150LIP-Drive Loop 148Loop Down 148Loop Error 148Loop Up 148New Drive Rebuild Failure 148Reallocate LBA 150Rebuild Aborted 149Rebuild Complete 149Rebuild Restarted 149Rebuild Started 149

SES Initialized Level 1 149SES Initialized Level 3 149

Drive Events 145Not Available On Loop 149

Drive Loop Integrity 106

Drive Maintenance 114

EEnable/Disable Auto Rebuild 72

Ethernet Configuration 76

Event Logs 137


FFailed Drives 155

Faster Rebuild 161

HHost Event

CC to Host ID 152Controller LIP 151Detected Power-on/Reset 151Host Port Incorrect Address 151Logged in at ID 151Loop Down 151Loop Up 151

Host Events 151

Host LUN Availability 45, 46

Host Port Addressing 77

Hot Spare DrivesAdding 68Deleting 70Viewinig 67

LLoop Information 7

LUN Assignment 39

MManual Configuration 18

Menu System 4

Modify Configuration Name 82

Modify Configuration WWN 84

Modifying Mapped LUN Number 53

Monitor Commands 95bt 96compare 97date 96

Index


do 96ef 95fill 97iotes 98rs 96temp 96time 96

Monitor Mode 94

NNavigating 3

Numbering Convention vi

OOffline Diagnostics 105

Offline Test Pattern 106

Online Diagnostics 111

RRAID 5 Sub-Array 160

RAID 5 Write Performance 157


RAID Level 18

Read Firmware from Drive 118

Readahead Statistics 133

Rebuild Configuration 28

Rebuild Priority 75

Resetting the Controllers 125

Rewrite Parity 122

RS-232 Port settings 2

Run-time Information 7

SSAN LUN Mapping 55

SAN LUN MappingsCreating 58Deleting 62Modify 63Viewing 57

Sequential Access 158

SES EventAlarm is ON 154Encl Alarm is OFF 154

SES Events 153

SES Logging 91

Set Rebuild Priority 75

Setting Single Controller Mode 87

Shutting Down the Controllers 126

Statistics 129

Sub-array Selection 22, 36

Summary of the errorsOffline 110

TTemperature Information 7

TerminologyArrays 12LUNs 39SAN LUNs 55

Test DurationOffline Diagnostics 108

Tips on Setting LUN Availability 47

Troubleshooting Assistance 9

Trust Array 37

UUnassigned Free Space 41

Updating RAID Enclosure Firmware 99

UPS Configuration 76

User Defined Data 107

Using the Monitor 94

VVerification 89

View Unused Drives 17

Viewing Connected Hosts 64

Voltage Information 7

WWWN controller 7

Index 165

166


Index



www.fel.fujitsu.com 91-9-94627001 (A)August 2003

Fujitsu Europe LimitedHayes Park CentralHayes End RoadHayes, Middlesex, England UB4 8FE44.208.573.4444

Documents

imageRAID VT-100 Text-based Interface Software Guide€¦ · SOFTWARE GUIDE VT-100 imageRAID® Series Storage System (Text-based User Interface) VT100_Wahoo-TUI.book Page 1 Tuesday,