SOFTWARE GUIDE VT-100. imageraid Series Storage System. (Text-based User Interface)

Transcription

1 SOFTWARE GUIDE VT-100 imageraid Series Storage System (Text-based User Interface)

2

3 SOFTWARE GUIDE VT-100 imageraid Series Storage System (Text-based User Interface)

4 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 Limited Hayes Park Central Hayes End Road Hayes, 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.

5 Contents About this Manual v Welcome v Numbering Convention vi Typographical Conventions vi 1 Accessing the Configuration Utility Software Using the VT-100 Terminal Navigating the Configuration Software Menu System Controller Information Troubleshooting Assistance Creating Disk Arrays Arrays Terminology Optimization and Drive Selection for RAID 5 Arrays Creating Arrays Configuring Array Writeback Cache Configuring Array Readahead Cache Assigning Hot Spare Drives Deleting an Array Expanding an Array Trust an Array Host LUN Assignment Configuration Terminology Accessing the Host LUN Configuration Menu Viewing Unassigned Free Space Creating a Host LUN Set Host LUN Availability Tips on Setting LUN Availability Modifying Mapped LUN Number i

6 Table of Contents 4 SAN LUN Mapping Terminology Accessing the SAN LUN Mapping Configuration Menu Viewing SAN LUN Mappings Creating a SAN LUN Mapping Delete SAN LUN Mapping Modify SAN LUN Mapping Viewing Connected Hosts Advanced Configuration and Management Hot Spare Drives View Hot Spare Drives Add Hot Spare Drives Delete Hot Spare Enable/Disable Auto Rebuild Auto Hot Spare Rebuild Priority UPS Configuration Ethernet Configuration Additional Configuration Host Port Addressing Setting the Controller LUN Modify Configuration Name Modify Configuration WWN Setting Single Controller Mode Background Verification Controller SES Monitor Interface Using the Monitor Monitor Commands Updating the RAID Controller Firmware Diagnostics Offline Diagnostics Online Diagnostics Controller Maintenance Drive Maintenance RAID 5/50 Parity Check Command Tracing ii

7 Table of Contents Resetting the Controllers Shutting Down the Controllers Diagnostics Dmp Statistics Access Statistics Readahead Statistics Command Cluster Statistics Miscellaneous Statistics Event Logs Accessing Event Logs Controller Events Drive Events Host Events SES Events Failed Drives A Optimizing RAID 5 Write Performance Introduction Sequential Access Number of Outstanding Commands Access Size Access Alignment RAID 5 Sub-Array Multiple Drive Failures Faster Rebuild Summary Index iii

8 iv Table of Contents

9 Preface 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

10 About this Manual 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>. vi Numbering Convention

11 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

12 Chapter 1 - Accessing the Configuration Utility Software 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 Setting Value Terminal Emulation Function Fonts Translations ANSI Terminal Keys Terminal None Columns 80 Set the communications parameters for the terminal program as follows Setting Value Baud Rate 9, ,200* Data Bits 8 Stop Bits 1 Parity Flow Control None None Connector COM1 (typically) *The default baud rate is 19,200, however it is recommended to use 115,200 for optimum results. 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. 2 Using the VT-100 Terminal

13 Chapter 1 - Accessing the Configuration Utility Software 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. Return to the previous menu or screen (Exit). Select a menu option. Press <Control>+<W> Type <0> and press the <Enter> key. Press the specific numeric key for the menu choice. Navigating the Configuration Software 3

14 Chapter 1 - Accessing the Configuration Utility Software Menu System imageraid Main Menu Main Menu Configuration Diagnostics Controller Information Statistics Event Logs RS232 Settings Configuration Menu Array Configuration Host LUN Configuration Additional Host Configuration SAN LUN Mapping Configuration Rebuild Configuration UPS Configuration Ethernet Configuration Additional Configuration View Configuration Clear Configuration Diagnostic Offline Diagnostics Online Diagnostics Reset Controller Shutdown Controller Diagnostic Dump Controller Information View Controller 0 View Controller 1 Statistics View Controller 0 View Controller 1 Array Configuration View Array Create Array Delete Array Expand Array Array Cache Configuration View Unused Drives Advanced Array Configuration Host LUN Configuration View Host LUN Create Host LUN Delete Host LUN Expand Host LUN Set Host LUN Availability View Unassigned Free Space Advanced Host LUN Configuration Additional Host Configuration Host Port Hard Addressing Controller LUN Configuration Host Port Connection Options Host Port Data Rate Host Port Different Node Name SAN LUN Mapping Configuration View SAN LUN Mappings Create SAN LUN Mapping Delete SAN LUN Mapping Modify SAN LUN Mapping View Connected Hosts Event Logs View Controller 0 View Controller 1 Rebuild Configuration View Hot Spare Drives Add Hot Spare Drives Delete Hot Spare Drives Enable/Disable Auto Rebuild Set Rebuild Priority Enable/Disable Auto Hot Spare RS232 Setting Configuration Software Utility Menu Chart 1 of 3 4 Menu System

15 Chapter 1 - Accessing the Configuration Utility Software Menu System (continued) Configuration Menu Array Configuration Host LUN Configuration Additional Host Configuration SAN LUN Mapping Configuration Rebuild Configuration UPS Configuration Ethernet Configuration Additional Configuration View Configuration Clear Configuration UPS Configuration (not implemented) Controller A UPS Setup Controller B UPS Setup Ethernet Configuration (not implemented) Set Controller IP Address Remote Mirror Configuration Accept IP Address Configuration Notification Configuration View Configuration Clear Configuration Additional Configuration Modify Configuration Name Modify Configuration WWN Set Single Controller Mode Enable/Disable Background Verification Enable/Disable Controller SES Array Configuration View Array Create Array Delete Array Expand Array Array Cache Configuration View Unused Drives Advanced Array Configuration Displays Array Information Select Array Delete Array Select Array Expand Array Select Array Displays Unused Drives Create Array Select RAID Level Manual Configure Chunk Size Select Drives Readahead Cache Writeback Cache Advanced Host LUN Configuration Initialize Array Trust Array Configuration Software Utility Menu Chart 2 of 3 Menu System 5

16 Chapter 1 - Accessing the Configuration Utility Software Menu System (continued) imageraid Main Menu Main Menu Configuration Diagnostics Controller Information Statistics Event Logs RS232 Settings Controller Information View Controller 0 View Controller 1 Displays Controller Parameters Statistics View Controller 0 View Controller 1 Event Logs View Controller 0 View Controller 1 View All Statistics View Access Statistics View Readahead Statistics View Command Clustering Statistics View Miscellaneous Statistics Clear Statistics View Access Statistics On Each LUN Clear Access Statistics On Each LUN View All Events <4096> View New Events <22> View Last 50 Events View Last Hour Events View Last 24 Hour Events View Drive Loop Events View Host Loop Events View Environment Events View Failed Drive List Clear Event Log Configuration Software Utility Menu Chart 3 of 3 6 Menu System

17 Chapter 1 - Accessing the Configuration Utility Software Controller Information 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

18 Chapter 1 - Accessing the Configuration Utility Software Below is an example of the controller information screen. Controller Information Display Screen The following is a brief summary of the information displayed. Error Controller WWN Configuration WWN Description An 8-byte unique World Wide Name the controller uses for identification. This is unique to each controller and programmed at manufacture. 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. 8 Controller Information

19 Chapter 1 - Accessing the Configuration Utility Software Error Description Config Name Firmware Revision Firmware Date/Time Total Power on Count Total Power on Time Current Power on Time Current Controller Time Last Configuration Time Last Configuration Time Host Loop x Status Drive Loop x Status Onboard Temperature Controller Voltage A 64-byte ASCII name used to identify a specific controller configuration. The firmware version in use by the controller. Date and time of when the firmware was created. The number of times this controller has been power cycled. The total length of time this controller has been powered on. The length of time since this controller has last been powered on or reset. This is the current time on the controller. This is the time the controller was last configured. This is the time the controller was last configured. The current status of the specified host loop. The current status of the specified host loop. 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. Current readings for various controller and coprocessor voltages. 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. Troubleshooting Assistance 9

20 Chapter 1 - Accessing the Configuration Utility Software 10 Troubleshooting Assistance

21 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 (Mirrored)

22 Chapter 2 - Creating Disk Arrays 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 Chunk Size Stripe Size Cache Flush Array Initialization Reserved Capacity RAID Level 0 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. This is the amount of data that is written on a single drive before the controller moves to the next drive in the stripe. This is the number of data drives multiplied by the chunk size. This is the array that is used to automatically flush cache data in the situation where power has failed to some of the drives. 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. 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 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. 12 Terminology

23 Chapter 2 - Creating Disk Arrays RAID Level 1 RAID Level 5 RAID Level 10 RAID Level 50 Sub-array 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 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 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). 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. 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 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. Optimization and Drive Selection for RAID 5 Arrays 13

24 Chapter 2 - Creating Disk Arrays 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. 14 Creating Arrays

25 Chapter 2 - Creating Disk 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

26 Chapter 2 - Creating Disk Arrays View Array Screen View Array Field Explanation Field No. World Wide Name Product ID F/W Sav Ser. No. Cap. IDx HID EN, SL ST Description Drive number in the array. 8-byte Node Name, which is unique for every drive. Drive model number. Drive firmware version. Indicates whether a copy of the drive s firmware is saved on the media. Drive serial number. Drive capacity expressed in GBs. Loop ID of the specified drive loop. Hard addresss of the drive. The enclosure and drive slot number, used for drive identification. The drive status, will either be OK or FL. 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>. 16 Creating Arrays

27 Chapter 2 - Creating Disk Arrays View Unused Drives Screen 5 Create a new array. From the Array Configuration menu, choose Create Array by typing <2> and pressing <Enter>. Array Configuration Menu 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

28 Chapter 2 - Creating Disk Arrays 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. 18 Creating Arrays

29 Chapter 2 - Creating Disk 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

30 Chapter 2 - Creating Disk Arrays 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 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. 20 Creating Arrays

31 Chapter 2 - Creating Disk 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

32 Chapter 2 - Creating Disk Arrays 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. 22 Creating Arrays

33 Chapter 2 - Creating Disk 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 Configuration Memory Vendor and Part Number Measured Current Draw Absolute Maximum Backup Time Expected Safe Backup Time Main board only w/512 MB Main board w/512 MB and Coprocessor w/512 MB Kingston KVR100X72C2/512 Kingston KVR100X72C2/ mA 41.2 hours 20.6 hours 48.3mA 23.8 hours 11.9 hours Configuring Array Writeback Cache 23

34 Chapter 2 - Creating Disk Arrays 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>. Array Configuration Menu 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>. 24 Configuring Array Writeback Cache

35 Chapter 2 - Creating Disk Arrays 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

36 Chapter 2 - Creating Disk Arrays Configuring Array Readahead Cache 1 From the Array Configuration menu, choose Array Cache Configuration by typing <5> and pressing <Enter>. Array Configuration Menu 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 26 Configuring Array Readahead Cache

37 Chapter 2 - Creating Disk Arrays 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

38 Chapter 2 - Creating Disk Arrays 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 menu and press <Enter>. Main Configuration Menu 2 From the Rebuild Configuration Menu, type <2> for the Add Hot Spare Drives option and press <Enter>. Rebuild Configuration Menu 28 Assigning Hot Spare Drives

39 Chapter 2 - Creating Disk Arrays 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: Add Hot Spare Configuration Menu 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

40 Chapter 2 - Creating Disk Arrays To add a specific drive as hot spare: a b Type <2> to Select a Drive to Add and press <Enter>. 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 d 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. Return to the Main Configuration menu. Type <0> and press <Enter>. Proceed now to Host LUN Assignment on page Assigning Hot Spare Drives

41 Chapter 2 - Creating Disk Arrays Deleting an Array 1 From the Main menu, type <1> for the Configuration menu and press <Enter>. Main Menu 2 From the Configuration menu, type <1> for the Array Configuration menu and press <Enter>. Main Configuration Menu 3 Choose Delete Array by typing <3> and pressing <Enter>. Deleting an Array 31

42 Chapter 2 - Creating Disk Arrays Array Configuration Menu 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 32 Deleting an Array

43 Chapter 2 - Creating Disk Arrays 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 2 From the Configuration menu, type <1> for the Array Configuration menu and press <Enter>. Expanding an Array 33

44 Chapter 2 - Creating Disk Arrays Main Configuration Menu 3 Choose Expand Array by typing <4> and pressing <Enter>. Array Configuration Menu 4 Select the array to expand by entering the number of the array and pressing <Enter>. Expand Array Screen 34 Expanding an Array

45 Chapter 2 - Creating Disk Arrays 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. Expanding an Array 35

46 Chapter 2 - Creating Disk Arrays 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. 36 Expanding an Array

47 Chapter 2 - Creating Disk Arrays 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>. 2 From the Configuration menu, type <1> for the Array Configuration menu 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

48 Chapter 2 - Creating Disk Arrays 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. 38 Trust an Array

49 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 Terminology The following table describes the terminology relating to Host LUNs. Term LUN Segmentation Description 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

50 Chapter 3 - Host LUN Assignment LUN Availability Mapped LUN Number Unassigned Free Space 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. 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. 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 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 menu and press <Enter>. Configuration Menu 40 Accessing the Host LUN Configuration Menu

51 Chapter 3 - Host LUN Assignment 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. 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 menu and press <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

52 Chapter 3 - Host LUN Assignment 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. LUN 0:0 300 GB Array 1 RAID GB LUN 1: GB LUN 2:0 400 GB LUN 3:0 700 GB Sample Array and LUN Layout 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 menu and press <Enter>. 3 From the Host LUN Configuration menu, type <2> for the Create Host LUN menu and press <Enter>. Host LUN Configuration Menu 42 Creating a Host LUN

53 Chapter 3 - Host LUN Assignment 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

54 Chapter 3 - Host LUN Assignment 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. 44 Creating a Host LUN

55 Chapter 3 - Host LUN Assignment a b c d Choose the option to Manually Select Unused Free Space, type <1> and press <Enter>. Enter the number of the unused region, type <0> and press <Enter>. Enter the number of LUNs to create, type <1> and press <Enter>. Enter the size of LUNs to create, type <xxx> and press <Enter>. Referring to our example, enter 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>. Host LUN Configuration Menu Set Host LUN Availability 45

56 Chapter 3 - Host LUN Assignment 2 Choose a LUN to set its availability, type <x> and press <Enter>, where x is the LUN number. NOTE: Set Host LUN Availability Menu 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. 46 Set Host LUN Availability

57 Chapter 3 - Host LUN Assignment 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 LUN LUN LUN LUN LUN LUN LUN LUN LUN LUN LUN LUN 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. Tips on Setting LUN Availability 47

58 Chapter 3 - Host LUN Assignment 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 N/A N/A LUN LUN LUN LUN N/A N/A LUN LUN LUN LUN 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 N/A N/A LUN LUN LUN LUN N/A N/A LUN LUN LUN LUN 48 imageraid IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port Multi Host Connection)

59 Chapter 3 - Host LUN Assignment 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. 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 N/A N/A LUN LUN LUN LUN N/A N/A LUN LUN LUN LUN 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 LUN N/A LUN LUN LUN LUN N/A LUN LUN x2 LUN LUN LUN x2 imageraid IRF-1Sxx-xx/IRF-1Dxx-xx (Simplex: Dual Port Dual Host with a Quad Connection) 49

60 Chapter 3 - Host LUN Assignment 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 N/A N/A LUN LUN LUN LUN N/A N/A LUN LUN LUN LUN 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. 50 imageraid IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored One Host Dual Connection)

61 Chapter 3 - Host LUN Assignment 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 N/A N/A LUN LUN LUN LUN N/A N/A LUN LUN LUN LUN 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. Set LUN Availability to both controller ports. 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 LUN LUN N/A N/A N/A N/A LUN LUN LUN LUN LUN LUN 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. imageraid IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored Dual Host with Quad Connections) 51

62 Chapter 3 - Host LUN Assignment 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. 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 N/A N/A LUN LUN LUN LUN N/A N/A LUN LUN LUN LUN 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. Set LUN Availability to both controller ports. 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. Host I/O Card Controller Ports LEFT H0 LEFT H1 RIGHT H0 RIGHT H1 Port 0 Port 1 Ports 0 & 1 N/A N/A LUN LUN LUN LUN N/A N/A LUN LUN LUN LUN 52 imageraid IRF-2Sxx-xx/IRF-2Dxx-xx (Duplex: Multi Port Mirrored SAN Dual Switches)

63 Chapter 3 - Host LUN Assignment Modifying Mapped LUN Number This option allows you to change the mapped or assigned LUN number after the LUN has been created. 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 menu and press <Enter>. 3 From the Host LUN Configuration menu, type <7> for the Advanced Host LUN Configuration menu and press <Enter>. Host LUN Configuration Menu 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

64 Chapter 3 - Host LUN Assignment 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. 54 Modifying Mapped LUN Number

65 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 Node Name Port Name Mapping Name Exclusive Access Read/Write Access Read Only Access Mapped LUN Number Mapping Availability Description 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. 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. A 28 character name that can be used to help identify the host system. A LUN is presented to one and only one host system. It is not available for any other host systems. A host may read and write to the LUN. A host may only read from the LUN. 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. Which controller ports the mapping is valid for. 55

66 Chapter 4 - SAN LUN Mapping Accessing the SAN LUN Mapping Configuration Menu 1 From the Main menu, type <1> for the Configuration menu and press <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. 56 Accessing the SAN LUN Mapping Configuration Menu

67 Chapter 4 - SAN LUN Mapping 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

68 Chapter 4 - SAN LUN Mapping 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>. SAN LUN Mapping Configuration Menu 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. ) 58 Creating a SAN LUN Mapping

69 Chapter 4 - SAN LUN Mapping 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

70 Chapter 4 - SAN LUN Mapping 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. 60 Creating a SAN LUN Mapping

71 Chapter 4 - SAN LUN Mapping 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

72 Chapter 4 - SAN LUN Mapping 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 SAN LUN Mapping menu and press <Enter>. SAN LUN Mapping Configuration Menu 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. 62 Delete SAN LUN Mapping

73 Chapter 4 - 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 SAN LUN Mapping menu and press <Enter>. SAN LUN Mapping Configuration Menu 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

74 Chapter 4 - SAN LUN Mapping 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>. SAN LUN Mapping Configuration Menu 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. 64 Viewing Connected Hosts

75 Chapter 4 - SAN LUN Mapping Field No. Node Name Port Name ID-xy Description Host number assigned by the controller. This is used when creating a SAN LUN Mapping and selecting a host from the list. This is the Host Port s node name. This is the Host Port s port name. ID used by this host for Controller x Host Port y. (L) indicates Arbitrated Loop, and (F) indicates fabric connect. Viewing Connected Hosts 65

76 Chapter 4 - SAN LUN Mapping 66 Viewing Connected Hosts

77 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 1 From the Configuration menu, type <5> for the Rebuild Configuration menu and press <Enter>. 2 From the Rebuild Configuration Menu, type <1> for the View Hot Spare Drives option and press <Enter>. Rebuild Configuration Menu 67

78 Chapter 5 - Advanced Configuration and Management All currently configured hot spare drives are displayed. 3 Press any key to continue. View Hot Spare Drives Screen Add Hot Spare Drives 1 From the Configuration menu, type <5> for the Rebuild Configuration menu and press <Enter>. Configuration Menu 2 From the Rebuild Configuration Menu, type <2> for the Add Hot Spare Drives option and press <Enter>. 68 Add Hot Spare Drives

79 Chapter 5 - Advanced Configuration and Management Rebuild Configuration Menu To add all unused drives as hot spares: a Type <1> for All Unused Drives and press <Enter>. Add Hot Spare Configuration Menu 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

80 Chapter 5 - Advanced Configuration and Management 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 d e 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. Press any key to continue and return to the previous menu. 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. 1 From the Configuration menu, type <5> for the Rebuild Configuration menu and press <Enter>. 70 Delete Hot Spare

81 Chapter 5 - Advanced Configuration and Management Configuration Menu 2 From the Rebuild Configuration Menu, type <3> for the Delete Hot Spare Drives option and press <Enter>. Rebuild Configuration Menu 3 Enter Yes <Y> to delete all hot spare drives. 4 Press any key to continue. Delete All Hot Spare Drives Menu Delete Hot Spare 71

82 Chapter 5 - Advanced Configuration and Management 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. 1 From the Configuration menu, type <5> for the Rebuild Configuration Menu and press <Enter>. Configuration Menu 2 From the Rebuild Configuration Menu, type <4> for the Enable/Disable Auto Rebuild option and press <Enter>. Rebuild Configuration Menu 72 Enable/Disable Auto Rebuild

83 Chapter 5 - Advanced Configuration and Management 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. 1 From the Configuration menu, type <5> for the Rebuild Configuration menu and press <Enter>. 2 From the Rebuild Configuration Menu, type <6> for the Enabled/Disable Auto Hot Spare option and press <Enter>. Auto Hot Spare 73

84 Chapter 5 - Advanced Configuration and Management Rebuild Configuration Menu 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. 74 Auto Hot Spare

85 Chapter 5 - Advanced Configuration and Management 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. 1 From the Configuration menu, type <5> for the Rebuild Configuration menu and press <Enter>. 2 From the Rebuild Configuration Menu, type <5> for the Set Rebuild Priority option and press <Enter>. Rebuild Configuration Menu 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 Configuration Menu Rebuild Priority 75

86 Chapter 5 - Advanced Configuration and Management 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. 76 UPS Configuration

87 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

88 Chapter 6 - Additional Configuration 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.). 78 Host Port Addressing

89 Chapter 6 - Additional Configuration Host Port Hard Addressing Screen 6 Choose the second controller port and set the hard address. Type <2> and press <Enter>. Host Port Hard Addressing Screen 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.). Host Port Hard Addressing Screen 8 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. Host Port Addressing 79

90 Chapter 6 - Additional Configuration 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 option and press <Enter>. Configuration Menu 80 Setting the Controller LUN

91 Chapter 6 - Additional Configuration 3 Choose Controller LUN Configuration by typing <2> and pressing <Enter>. Additional Host Configuration Menu 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 ( ), then the Controller will always respond to that LUN as a processor device. Controller LUN Configuration Menu 5 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. 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

92 Chapter 6 - Additional Configuration Modify Configuration Name To help identify a particular controller configuration, it is possible to assign a 64 byte ASCII name to it. 1 From the Main menu, type <1> for the Configuration menu and press <Enter>. Main Menu 2 From the Configuration menu, type <8> for the Additional Configuration option and press <Enter>. Configuration Menu 3 From the Additional Host Configuration menu, choose Modify Configuration Name option. Type <1> and press <Enter>. 82 Modify Configuration Name

93 Chapter 6 - Additional Configuration Additional Host Configuration Menu 4 Type in the new name and press <Enter>. (You may keep the current name by simply pressing <Enter>. Modify Configuration Name Menu 5 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. Modify Configuration Name 83

94 Chapter 6 - Additional Configuration 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. 1 From the Main menu, type <1> for the Configuration menu and press <Enter>. Main Menu 84 Modify Configuration WWN

95 Chapter 6 - Additional Configuration 2 From the Configuration menu, type <8> for the Additional Configuration option and press <Enter>. 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

96 Chapter 6 - Additional Configuration Modify WWN Name Menu 5 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. 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>. 86 Modify Configuration WWN

97 Chapter 6 - Additional Configuration 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: 1 From the Main menu, type <1> for the Configuration menu and press <Enter>. Main Menu 2 From the Configuration menu, type <8> for the Additional Configuration option and press <Enter>. Main Configuration Menu Setting Single Controller Mode 87

98 Chapter 6 - Additional Configuration 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. 5 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. 88 Setting Single Controller Mode

99 Chapter 6 - Additional Configuration 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. 1 From the Main menu, type <1> for the Configuration menu and press <Enter>. Main Menu 2 From the Configuration menu, type <8> for the Additional Configuration option and press <Enter>. Main Configuration Menu 3 From the Additional Host Configuration menu, choose the Enable/Disable Background Verification option. Type <4> and press <Enter>. Background Verification 89

100 Chapter 6 - Additional Configuration Additional Configuration Menu 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. 90 Background Verification

101 Chapter 6 - Additional Configuration 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. 1 From the Main menu, type <1> for the Configuration menu and press <Enter>. Main Menu 2 From the Configuration menu, type <8> for the Additional Configuration option and press <Enter>. Main Configuration Menu 3 From the Additional Host Configuration menu, choose the Enable/Disable Controller SES option. Type <5> and press <Enter>. Controller SES 91

102 Chapter 6 - Additional Configuration Additional Configuration Menu 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. 92 Controller SES

103 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

104 Chapter 7 - Monitor Interface 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?. 94 Using the Monitor

105 Chapter 7 - Monitor Interface 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

106 Chapter 7 - Monitor Interface 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: 96 do: Download Firmware

107 Chapter 7 - Monitor Interface 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

108 Chapter 7 - Monitor Interface 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. 98 iotest: Test Controller Backplane Signals and LEDs

109 Chapter 7 - Monitor Interface 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

110 Chapter 7 - Monitor Interface 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. 100 Updating the RAID Controller Firmware

111 Chapter 7 - Monitor Interface 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

112 Chapter 7 - Monitor Interface 102 Updating the RAID Controller Firmware

113 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

114 Chapter 8 - Diagnostics 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

115 Chapter 8 - Diagnostics 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

116 Chapter 8 - Diagnostics 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. 106 Offline Diagnostics

117 Chapter 8 - 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. Offline Diagnostics 107

118 Chapter 8 - Diagnostics 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. 108 Offline Diagnostics

119 Chapter 8 - 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. Offline Diagnostics 109

120 Chapter 8 - Diagnostics The following is a brief summary of the errors. Error Drive Loop x LIP Occurred Drive Loop x Parity Error Drive Loop x Timeout Error Drive Loop x Underrun Error Drive Loop x Reset Error Drive Loop x Internal Error Drive Loop x Loop (yyyy) Error Description 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. 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. 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. 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. 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. An internal error has occurred while transferring data. This is not related to the loop, but indicates a problem with the controller. An error has occurred indicated by the number yyyy. This type of error should be reported to nstor for further analysis. 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. 110 Offline Diagnostics

121 Chapter 8 - 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

122 Chapter 8 - Diagnostics 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. 1 From the Main menu, select Diagnostics, type <2> and press <Enter>. 2 Select Online Diagnostics, type <2> and press <Enter>. 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. 112 Controller Maintenance

123 Chapter 8 - Diagnostics 5 When the C s appear, click the Transfer pull-down menu selection and choose Send File. Then select the 1K Xmodem protocol. 6 Click the browse button and locate the new Firmware file and click Send. 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

124 Chapter 8 - Diagnostics 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. 114 Drive Maintenance

125 Chapter 8 - Diagnostics Download the Firmware from the Host 1 From the Main menu, select Diagnostics, type <2> and press <Enter>. 2 Select Online Diagnostics, type <2> and press <Enter>. 3 Select Drive Maintenance, type <2> and press <Enter>. Online Diagnostics Menu 4 Select Download Drive Firmware to Buffer, type <1> and press <Enter>. Drive Firmware Menu 5 When the C s appear, click the Transfer pull-down menu selection and choose Send File. Then select the 1K Xmodem protocol. Download the Firmware from the Host 115

126 Chapter 8 - Diagnostics 6 Click the browse button and locate the new Firmware file and click Send. 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. 116 Download the Firmware from the Host

127 Chapter 8 - Diagnostics 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. Download the Firmware from the Host 117

128 Chapter 8 - Diagnostics 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>. Online Diagnostics Menu 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. 118 Read Firmware from a Drive to the Buffer

129 Chapter 8 - Diagnostics 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

130 Chapter 8 - Diagnostics 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. 120 Read Firmware from a Drive to the Buffer

131 Chapter 8 - Diagnostics 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. 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 3 Select RAID 5/50 Parity Check, type <3> and press <Enter>. RAID 5/50 Parity Check 121

132 Chapter 8 - Diagnostics Online Diagnostics Menu 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 Rewrite 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. 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. 122 RAID 5/50 Parity Check

133 Chapter 8 - Diagnostics Option Check and Rewrite Parity Description 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. 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. RAID 5/50 Parity Check 123

134 Chapter 8 - Diagnostics 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: Online 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). 124 Command Tracing

135 Chapter 8 - Diagnostics 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

136 Chapter 8 - Diagnostics Shutting Down the Controllers 1 From the Main menu type <2> and press <Enter> for the Diagnostics menu. 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. 126 Shutting Down the Controllers

137 Chapter 8 - Diagnostics 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. 1 From the Main menu type <2> and press <Enter> for the Diagnostics menu. 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

138 Chapter 8 - Diagnostics 128 Diagnostics Dmp

139 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 pressing <Control>+<W>. 2 From the Main menu, choose Statistics, type <4> and press <Enter>. 129

140 Chapter 9 - Statistics Statistics Menu Screen 3 Select the controller you wish to examine. Choose <1> or <2> and press <Enter>. If only one controller is present, only one selection is listed. 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 Total Read Commands Description 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 130 Access Statistics

141 Chapter 9 - Statistics Total Read Transfer Total Read Commands Since Reset Total Read Transfer Since Reset Alignment Access Size Statistic Description 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. 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. 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. 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. 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

142 Chapter 9 - Statistics The following is an example of Read Access Statistics: Read Access Statistics Screen 132 Access Statistics

143 Chapter 9 - 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 Statistic Readahead Command Hit Rate Readahead Command Efficiency Description 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. 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

144 Chapter 9 - Statistics Statistic Sequential Command Interval Description 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. 134 Readahead Statistics

145 Chapter 9 - 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 Statistic Write Cluster Rate Description 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

146 Chapter 9 - Statistics Statistic RAID 5/50 Partial Command Cluster Rate RAID 5/50 Full Stripe Write Rate Command Cluster Interval Description 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. 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. 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 To aid in tuning performance these additional statistics are provided for internal development use. 136 Miscellaneous Statistics

147 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 pressing <Control>+<W>. 2 From the Main menu, choose Event Logs <5> and press <Enter>. 137

148 Chapter 10 - Event Logs Events Log Menu Screen 3 Select the controller you wish to examine. Choose <1> or <2> and press <Enter>. If only one controller is present, only one selection is listed. 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. 138 Accessing Event Logs

149 Chapter 10 - Event Logs These options are as follows: View All Events Event Log View New Events View Last 50 Events View Last Hour/24 Hour Events View Drive Loop Events View Controller Port Loop Events View Environmental Events View Failed Drive List Description This option lists all events. The number of events is indicated. This lists all events that have occurred since the event log has last been viewed. This option lists the last 50 events. This lists the events that occurred in the specified time. This lists all the events that have occurred on the drive loops. These include LIPs, drive failures, etc. This lists all the events that have occurred on the host side of the controller port loops. These include LIPs, hosts logging in, etc. This lists events such as temperature warnings, voltage level errors, battery errors, etc. This is a list of drives that the controller has marked as having failed. These drives have either failed in operation, or been removed. A typical event log will look similar to the following: Typical Event Log Screen Accessing Event Logs 139