This chapter explains how to update device drivers and apply hotfix.

MegaRAID SAS User's Guide Areas Covered Before Reading This Manual This section explains the notes for your safety and conventions used in this manual. Chapter 1 Overview This chapter explains an overview and configuration precautions for the disk array, and an explanation of the array configuration flow. Chapter 2 Array Configuration and Management [WebBIOS] This chapter explains the WebBIOS setup procedures. WebBIOS is a basic utility to set up and manage array controllers. Chapter 3 Chapter 4 Chapter 5 Chapter 6 Appendix Preparations This chapter explains how to update device drivers and apply hotfix. Array Configuration and Management [ServerView RAID] This chapter explains an overview of and product requirements for ServerView RAID, and describes how to install and use the program. Array Configuration and Management [GAM] This chapter explains an overview of and product requirements for GAM, and describes how to install and use the program. Replacing Hard Disk This chapter explains maintenance related issues, such as hard disk replacement. This section explains event codes for ServerView RAID and GAM, and how to replace batteries. 1

Before Reading This Manual This manual is a guide for using the following array controllers: MegaRAID SAS 300-8E ROMB MegaRAID SAS 8300XLP MegaRAID SAS 8344ELP MegaRAID SAS 8408E MegaRAID SAS 8880EM2 RAID 5/6 SAS based on LSI MegaRAID Remarks Symbols Symbols used in this manual have the following meanings. These sections explain prohibited actions and points to note when using this software. Make sure to read these sections. These sections explain information needed to operate the hardware and software properly. Make sure to read these sections. This mark indicates reference pages or manuals. Key descriptions / operations Keys are represented throughout this manual in the following manner. E.g.: [Ctrl] key, [Enter] key, [ ] key, etc. The following indicate the pressing of several keys at once. E.g.: [Ctrl] + [F3] key, [Shift] + [ ] key, etc. CD/DVD drive descriptions In this manual, both CD-ROM and DVD-ROM drives are described as CD/DVD drives. Select a proper drive depending on your environment. Entering commands (keys) Command entries are written in the following way. In the spaces indicated with the " " mark, press the [Space] key once. In the example above, the command entry is written in lower case, but upper case is also allowed. CD/DVD drive names are shown as [CD/DVD drive]. Enter your drive name according to your environment. [CD/DVD drive]:\setup.exe 2

MegaRAID SAS User's Guide Screen shots and figures Screen shots and figures are used as visual aids throughout this manual. Windows, screens, and file names may vary depending on the OS, software, or configuration of the server used. Figures in this manual may not show cables that are actually connected for convenience of explanation. Consecutive operations Consecutive operations are described by connecting them with a dash ( ). Example: For the operation to click the [Start] button, point to [Programs], and click [Accessories] Click [Start] [Programs] [Accessories]. PRIMERGY Startup Disc "PRIMERGY Startup Disc" referred in this manual is supplied with a PRIMERGY server. Depending on your PRIMERGY type, "PRIMERGY Document & Tool CD" may be supplied instead. Then "PRIMERGY Startup Disc" also means "PRIMERGY Document & Tool CD" in this manual. RAID 1+0(10) description In this manual, RAID 1+0(10) is referred to as RAID 10. Server Core installation option The environment in which the Windows Server 2008 operating system is installed by the Server Core installation option is written as Server Core in this manual. 3

Abbreviations The following expressions and abbreviations are used throughout this manual. Product name MegaRAID SAS 300-8E ROMB MegaRAID SAS 8300XLP MegaRAID SAS 8344ELP MegaRAID SAS 8408E MegaRAID SAS 8880EM2 RAID 5/6 SAS based on LSI MegaRAID table: Abbreviation of product name Expression and abbreviation the array controller, this array controller Microsoft Windows Server 2008 Standard Microsoft Windows Server 2008 Enterprise Microsoft Windows Server 2008 Standard without Hyper- V Microsoft Windows Server 2008 Enterprise without Hyper- V Microsoft Windows Server 2003, Standard Edition Microsoft Windows Server 2003, Enterprise Edition Microsoft Windows Server 2003 R2, Standard Edition Microsoft Windows Server 2003 R2, Enterprise Edition Microsoft Windows Server 2003, Standard x64 Edition Microsoft Windows Server 2003, Enterprise x64 Edition Microsoft Windows Server 2003 R2, Standard x64 Edition Microsoft Windows Server 2003 R2, Enterprise x64 Edition Microsoft Windows 2000 Server Microsoft Windows 2000 Advanced Server Windows Server 2008, Windows Server 2008 (32-bit), Windows Server 2008 (64-bit) Windows Server 2003 Windows Server 2003 x64 [*1] Windows 2000 Server Windows Microsoft Windows Vista Business Windows Vista Microsoft Windows Vista Ultimate Microsoft Windows Vista Enterprise Microsoft Windows XP Professional Windows XP Microsoft Windows 2000 Professional Windows 2000 Professional Red Hat Enterprise Linux 5 (for x86) Red Hat Linux Linux RHEL5(x86) Red Hat Enterprise Linux 5 (for Intel64) RHEL5(Intel64) Red Hat Enterprise Linux AS (v.4 for x86) RHEL-AS4(x86) Red Hat Enterprise Linux ES (v.4 for x86) RHEL-ES4(x86) Red Hat Enterprise Linux AS (v.4 for EM64T) RHEL-AS4(EM64T) Red Hat Enterprise Linux ES (v.4 for EM64T) RHEL-ES4(EM64T) Red Hat Enterprise Linux AS (v.3 for x86) RHEL-AS3(x86) Red Hat Enterprise Linux ES (v.3 for x86) RHEL-ES3(x86) SUSE Linux Enterprise Server 9 SUSE Linux *1: Unless otherwise noted, Windows Server 2003 can also mean Windows Server 2003 x64. 4

MegaRAID SAS User's Guide Reference Information Information for PRIMERGY For the latest information on PRIMERGY, update modules, drivers and the software, refer to the Fujitsu PRIMERGY website. http://www.fujitsu.com/global/services/computing/server/ia/driver/ Regarding BIOS and FW, contact to Fujitsu Support Office. Refer to the website (http://www.fujitsu.com/global/contact/computing/prmrgy_index.html). Trademarks Microsoft, Windows, Windows Server, Windows Vista, and Hyper-V are trademarks or registered trademarks of Microsoft Corporation in the USA and other countries. Linux is a trademark or registered trademark of Linus Torvalds in the USA and other countries. Red Hat and all Red Hat-based trademarks and logos are trademarks or registered trademarks of Red Hat, Inc. in the USA and other countries. SUSE is a registered trademark of SUSE LINUX AG, a Novell business. LSI, Global Array Manager (GAM), and MegaRAID are trademarks or registered trademarks of LSI Corp. All other hardware and software names used are trademarks or registered trademarks of their respective manufacturers. Other product names are copyrights of their respective manufacturers. All Rights Reserved, Copyright FUJITSU LIMITED 2008 Screen shot(s) reprinted with permission from Microsoft Corporation. 5

6

MegaRAID SAS User's Guide Contents Chapter 1 Overview 1.1 Array Controller Work Flow............................. 14 1.2 Overview of Disk Array Configuration.................... 17 1.2.1 Array Controller Specifications..................................17 1.2.2 What Is Disk Array?..........................................18 1.2.3 RAID Levels................................................18 1.2.4 Disk Group and Logical Drive..................................24 1.2.5 Checking Hard Disk Status....................................26 1.3 Disk Array Configuration Features....................... 28 1.3.1 Write Policy.................................................28 1.3.2 Logical Drive Initialization......................................30 1.3.3 Rebuild....................................................33 1.3.4 Make Data Consistent........................................35 1.3.5 Expand Capacity............................................36 1.3.6 Hard Disk Failure Prediction Function (PFA / S.M.A.R.T.).............38 1.3.7 HDD Check Scheduler........................................38 1.3.8 Battery Backup Unit (BBU)....................................39 1.4 Notes before Operation................................ 41 1.4.1 Notes on Installation to Server..................................41 1.4.2 Notes on Hard Disk to Be Used.................................42 1.4.3 Cautions for OS Installation....................................42 1.5 Notes on Operation................................... 43 1.5.1 Notes for Using Array Controller................................43 1.5.2 Message during POST........................................43 1.5.3 When the Server Is Not Shut Down Properly......................43 1.5.4 Using ServerView RAID under the Windows 2000 Server Environment.44 1.5.5 Notes on WebBIOS Startup when Installing Multiple MegaRAID SAS Controllers......................................................45 1.5.6 Hard Disk Failure LED during Operation..........................46 Chapter 2 Array Configuration and Management [WebBIOS] 2.1 Starting and Exiting WebBIOS........................... 48 2.1.1 Starting Up WebBIOS........................................48 2.1.2 Exiting WebBIOS............................................52 2.2 Checking Each Status [WebBIOS]....................... 53 2.2.1 Checking Hard Disk Status....................................53 2.2.2 Checking Logical Drive Status..................................57 2.2.3 Checking Progress of Background Task..........................61 2.2.4 Confirming Array Controller Information..........................63 7

2.3 Creating/Deleting Disk Array Configuration............... 66 2.3.1 Creating Disk Array Configuration............................... 66 2.3.2 Deleting Disk Array Configuration............................... 85 2.4 Setting/Releasing Spare Disk........................... 87 2.4.1 Setting Spare Disk........................................... 87 2.4.2 Releasing Spare Disk........................................ 88 2.5 Operations on Logical Drive............................ 89 2.5.1 Initializing Logical Drive....................................... 89 2.5.2 Fast Initialization of Logical Drive............................... 90 2.5.3 Make Data Consistent of Logical Drive........................... 91 2.5.4 Expand Capacity of Logical Drive............................... 92 2.5.5 Deleting Logical Drive........................................ 95 2.6 Rebuilding Hard Disk.................................. 97 2.7 Reusing Hard Disk.................................... 98 2.7.1 Cautions when Reusing Hard Disk.............................. 98 2.7.2 Clearing Information from Hard Disk............................. 98 Chapter 3 Preparations 3.1 Updating Device Driver............................... 106 3.1.1 Updating Driver (Windows Server 2008)......................... 106 3.1.2 Creating Driver Disk (Windows Server 2003 / Windows 2000 Server).. 108 3.1.3 Updating Driver (Windows Server 2003)......................... 108 3.1.4 Updating Driver (Windows 2000 Server)......................... 110 3.2 Applying Hotfix...................................... 112 Chapter 4 Array Configuration and Management [ServerView RAID] 4.1 Overview of and Requirements for ServerView RAID....... 114 4.1.1 Overview of ServerView RAID................................ 114 4.1.2 Requirements for ServerView RAID............................ 115 4.1.3 Access Privileges to ServerView RAID.......................... 117 4.1.4 Using ServerView RAID in Linux Environment.................... 118 4.1.5 Operations via Network...................................... 118 4.1.6 When Using Both ServerView RAID and GAM.................... 120 4.2 Installing ServerView RAID [Windows].................. 121 4.2.1 Installing ServerView RAID................................... 121 4.2.2 Uninstalling ServerView RAID................................. 123 4.3 Installing HDD Check Scheduler [Windows].............. 124 4.3.1 Installing HDD Check Scheduler............................... 124 4.3.2 Changing HDD Check Start Time.............................. 125 4.3.3 Uninstalling HDD Check Scheduler............................ 127 4.4 Installing Battery Recalibration Scheduler [Windows]..... 129 8

MegaRAID SAS User's Guide 4.4.1 Installing Battery Recalibration Scheduler.......................129 4.4.2 Changing Battery Recalibration Start Date and Time...............130 4.4.3 Uninstalling Battery Recalibration Scheduler.....................132 4.5 Starting and Exiting ServerView RAID Manager........... 133 4.5.1 Preparations and Precautions for Using ServerView RAID Manager...133 4.5.2 Starting and Logging in ServerView RAID Manager................134 4.5.3 Exiting ServerView RAID Manager.............................136 4.6 ServerView RAID Manager Window Layout............... 137 4.6.1 Startup Window Layout and Functions..........................137 4.6.2 Menu Layout and Functions...................................139 4.6.3 Layout of Tree View.........................................142 4.7 Settings............................................ 147 4.7.1 Changing Array Controller Setting..............................147 4.7.2 Configuring Battery Backup Unit...............................150 4.7.3 Setting Logical Drive........................................151 4.7.4 Setting ServerView RAID.....................................153 4.8 Operating RAID Configuration......................... 156 4.8.1 Creating Disk Array Configuration (Create logical drive).............156 4.8.2 Expanding Capacity of Logical Drive (Migrate logical drive)..........160 4.8.3 Deleting Existing Logical Drive (Delete last logical drive)............163 4.8.4 Setting and Releasing Spare Disk..............................164 4.8.5 Changing Write Policy.......................................166 4.8.6 Deleting Disk Array Configuration Information.....................167 4.8.7 Make Data Consistent.......................................168 4.8.8 Rebuild...................................................169 4.8.9 Recalibration..............................................170 4.9 Checking Each Status [ServerView RAID Manager]........ 171 4.9.1 Checking the Server Condition................................171 4.9.2 Checking Array Controller Status...............................172 4.9.3 Checking Hard Disk Status...................................174 4.9.4 Checking Logical Drive Status.................................176 4.9.5 Checking Progress of Background Task.........................178 Chapter 5 Array Configuration and Management [GAM] 5.1 Overview of and Requirements for GAM................. 182 5.1.1 Overview of GAM...........................................182 5.1.2 Requirements for GAM......................................183 5.1.3 Access Privileges to GAM....................................184 5.1.4 Using GAM in Linux Environment..............................185 5.1.5 Using GAM in Network Environment............................186 5.1.6 When Using Both ServerView RAID and GAM....................188 5.2 Installing GAM [Windows]........................... 189 5.2.1 Installing GAM.............................................189 5.2.2 Uninstalling GAM...........................................192 9

5.3 Installing HDD Check Scheduler [Windows]............. 194 5.3.1 Installing HDD Check Scheduler............................... 194 5.3.2 Changing HDD Check Start Time.............................. 195 5.3.3 Uninstalling HDD Check Scheduler............................. 197 5.4 Installing Battery Recalibration Scheduler [Windows]..... 198 5.4.1 Installing Battery Recalibration Scheduler........................ 198 5.4.2 Changing Battery Recalibration Start Date and Time............... 199 5.4.3 Uninstalling Battery Recalibration Scheduler..................... 201 5.5 Starting and Exiting GAM............................. 202 5.5.1 Starting GAM and Signing On................................. 202 5.5.2 Exiting GAM............................................... 203 5.6 GAM Window Layout................................. 204 5.6.1 Startup Window Layout and Functions.......................... 204 5.6.2 Menu Layout and Functions.................................. 207 5.6.3 Toolbar Icon............................................... 209 5.6.4 Starting Controller View and the Window Layout.................. 210 5.7 Settings............................................ 213 5.7.1 Server group and server settings.............................. 213 5.7.2 Setting and Changing Controller Options........................ 214 5.7.3 Configuring Battery Backup Unit............................... 217 5.8 Operating RAID Configuration......................... 218 5.8.1 Starting RAID Assist Function................................. 218 5.8.2 Creating New Disk Array Configuration (New Configuration)......... 220 5.8.3 Adding Logical Drive to Existing Disk Array Configuration (Add Logical Drive)............................................................. 223 5.8.4 Expanding Capacity of Logical Drive (Expand Array)............... 225 5.8.5 Deleting Existing Logical Drive (Edit Configuration)................ 228 5.8.6 Setting and Releasing Spare Disk (Edit Configuration)............. 230 5.8.7 Changing Write Policy....................................... 231 5.8.8 Saving and Deleting Disk Array Configuration Information........... 232 5.9 Viewing Information [GAM]............................ 234 5.9.1 Events................................................... 234 5.9.2 Viewing Array Controller Information............................ 236 5.9.3 Viewing Hard Disk Information................................ 237 5.9.4 Viewing Logical Drive Information.............................. 240 5.9.5 Checking Progress of Background Tasks........................ 242 5.10 Maintenance Functions.............................. 245 5.10.1 Make Data Consistent..................................... 245 5.10.2 Rebuild.................................................. 246 5.10.3 Recalibration............................................. 247 Chapter 6 Replacing Hard Disk 6.1 How to Replace Hard Disk [ServerView RAID]............ 250 10

MegaRAID SAS User's Guide 6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID]............250 6.1.2 Replacing Failed Hard Disk [ServerView RAID] (With Hot Swap Support)..............................................................251 6.1.3 Replacing Failed Hard Disk [ServerView RAID] (No Hot Swap Support)..............................................................253 6.1.4 Preventive Replacement of Hard Disk [ServerView RAID]...........255 6.2 How to Replace Hard Disk [GAM]....................... 265 6.2.1 Checking Hard Disk to Be Replaced [GAM]......................265 6.2.2 Replacing Failed Hard Disk [GAM].............................266 6.2.3 Preventive Replacement of Hard Disk [GAM].....................268 Appendix A List of ServerView RAID Event Log..................... 276 B List of GAM Event Log............................... 313 C Replacing Battery................................... 336 11

12

Chapter 1 Overview This chapter explains an overview and configuration precautions for the disk array, and an explanation of the array configuration flow. 1.1 Array Controller Work Flow....................... 14 1.2 Overview of Disk Array Configuration................ 17 1.3 Disk Array Configuration Features.................. 28 1.4 Notes before Operation.......................... 41 1.5 Notes on Operation............................. 43 13

Chapter 1 Overview 1.1 Array Controller Work Flow The work flow when using a disk array controller is as follows. For Windows 1. Designing the Disk Array Set the operation pattern of the disk array configuration. Design the RAID level and the logical drive configuration. "1.2 Overview of Disk Array Configuration" "1.3 Disk Array Configuration Features" 2. Installing the Hardware Install the array controller card on the server when using the card as an option. Also, install the hard disk drive on the bay. "1.4 Notes before Operation" "User's Guide" supplied with the server 3. Configuring the Disk Array Configure the disk array using WebBIOS before installing the OS. "Chapter 2 Array Configuration and Management [WebBIOS]" 4. Installing the OS "1.4.3 Cautions for OS Installation" "User's Guide" supplied with the server "3.2 Applying Hotfix" ServerStart will help you set up Step 3 to 5. 5. Installing Management Tools Install the following management tools. ServerView RAID [Note 1] Global Array Manager(GAM) [Note 1] HDD Check Scheduler [Note 2] Battery Recalibration Scheduler [Note 3] (only when the battery backup unit is installed "Chapter 4 Array Configuration and Management [ServerView RAID]" "Chapter 5 Array Configuration and Management [GAM]" "1.3.8 Battery Backup Unit (BBU)" [Note 1]: Either one can be installed. For details, refer to ReadmeEN.html in "Array Controller Document & Tool CD". [Note 2]: HDD check is set to start at noon every day by default. For details, refer to "1.3.7 HDD Check Scheduler". Change the time to start as necessary. [Note 3]: Battery recalibration is set to start at 11:00 on the first day every month by default. For details, refer to "Battery Recalibration Scheduler" in "1.3.8 Battery Backup Unit (BBU)". Change the start time as necessary. (Continued on the next page) 14

MegaRAID SAS User's Guide (Continued from the previous page) 6.Updating Device Drivers Update device drivers to the latest version stored in "Array Controller Document & Tool CD". "3.1 Updating Device Driver" 7.Preparing the Environment for Management Tools Register the user account with the OS to use management tools. (Not required if it is set at the installation of the management tools.) Set the start time for the HDD Check Scheduler. Set the start time for Battery Recalibration Scheduler. "Chapter 4 Array Configuration and Management [ServerView RAID]" "Chapter 5 Array Configuration and Management [GAM]" 8.Perform the Battery Recalibration When a battery backup unit is installed, perform the battery recalibration. For ServerView RAID "4.8.9 Recalibration" For GAM "5.10.3 Recalibration" 15

Chapter 1 Overview For Linux 1. Designing the Disk Array Set the operation pattern of the disk array configuration. Design the RAID level and the logical drive configuration. "1.2 Overview of Disk Array Configuration" "1.3 Disk Array Configuration Features" 2. Installing the Hardware Install the array controller card on the server when using the card as an option. Also, install the hard disk drive on the bay. "1.4Notes before the Operation" "User's Guide" supplied with the server 3. Configuring the Disk Array Configure the disk array using WebBIOS before installing the OS. "Chapter 2 Array Configuration and Management [WebBIOS]" 4. Installing the OS "Installation Guide" 5. Installing Management Tools Install the following management tools. ServerView RAID * Global Array Manager(GAM) * HDD Check Scheduler Battery Recalibration Scheduler (only when the battery backup unit is installed) "Installation Guide" * Either one can be installed. For details, refer to ReadmeEN.html in "Array Controller Document & Tool CD". 6. Preparing the Environment for Management Tools Register the user account with the OS to use the management tools. (Not required if it is set at the installation of the management tools.) Set the start time for the HDD Check Scheduler. Set the start time for the Battery Recalibration Scheduler. 7. Perform the Battery Recalibration When a battery backup unit is installed, perform the battery recalibration. "Installation Guide" For ServerView RAID "4.8.9 Recalibration" For GAM "5.10.3 Recalibration" For Linux, refer to the PRIMERGY page on the Fujitsu website (http://primergy.fujitsu.com) and refer to information about Linux. 16

MegaRAID SAS User's Guide 1.2 Overview of Disk Array Configuration This section contains an overview (RAID levels, disk groups, and logical drives) and a functional description of the disk array. 1.2.1 Array Controller Specifications The specifications of the array controller described in this manual are as follows. Item Product name Whether installed in the server Interface The number of SAS ports Supported RAID level Cache memory and capacity Cache memory battery Supported OS Management tool MegaRAID SAS 8408E Installed as a standard Onboard feature [Note 1] MegaRAID SAS 300-8E ROMB [Note 2] 8 ports (for internal device connection) Available (256 MB) Available table: Specifications MegaRAID SAS 8300XLP PCI-X slot installable card Contents MegaRAID SAS 8344ELP PCI-Express slot installable card SAS (Serial Attached SCSI) 0 ports [Note 4] 4 ports (for optional and internal device connection) [Note 5] MegaRAID SAS 8880EM2 RAID 5/6 SAS based on LSI MegaRAID PCI-Express slot installable card [Note 3] 4 ports (for optional), 2 connectors 8 ports (for internal device connection) RAID 0, 1, 5, and 10 RAID 0, 1, 5, 6, and 10 Available [Note 6] Available / Not available [Note 6] Available (128 MB) Available / Not available Available (256 MB) Available OS supported by the server Available (256 MB / 512 MB) Available / Not available There are two types of tools: (1) WebBIOS which configures disk arrays before installing the OS and (2) ServerView RAID or GAM which monitors or manages array controllers on the OS. Make sure to install a management tool before using array controllers. You must install either ServerView RAID or GAM, but not both. For information about which to use, refer to ReadmeEN.html in Array Controller Document & Tool CD. WebBIOS BIOS utility in the array controller "Chapter 2 Array Configuration and Management [WebBIOS]" (pg.47) ServerView RAID "Chapter 4 Array Configuration and Management [ServerView RAID]" (pg.113) Global Array Manager (GAM) "Chapter 5 Array Configuration and Management [GAM]" (pg.181) [Note 1]: It is pre-installed on the server. 17

Chapter 1 Overview [Note 2]: It is directly installed on the baseboard of the server. [Note 3]: If there is no model name, an array controller is installed in your server as a standard feature. In this case, refer to the specifications of the server. [Note 4]: Since this array controller itself does not have any ports, use a port on the SAS controller installed on the server. [Note 5]: Connection ports for internal devices are not supported. [Note 6]: Refer to the specifications of the server. When you start the server with this product installed in, you may hear the buzzer sound a few times. This is not a warning and you can continue your operation. 1.2.2 What Is Disk Array? A disk array or RAID (Redundant Array of Independent Disks) is a system that uses an array controller and multiple hard disk units to achieve better performance and higher reliability than when using a single hard disk unit. An array controller controls the access to each hard disk. There are different types of control methods that are defined as a RAID level. By using a redundant RAID level, system operation can be continued without data loss in the event that one or even two, for some RAID levels, of the hard disk unit fail. 1.2.3 RAID Levels There are several types of RAID level, with different characteristics. Depending on the RAID level, the number of installable hard disks, available capacity and applicability of redundancy are different. The RAID levels supported by this array controller vary depending on the product model. For details, refer to "1.2.1 Array Controller Specifications" (pg.17). RAID level table: Characteristics of each RAID level The number of hard disk Available total capacity Redundancy units [Note1] RAID 0 1 to 32 Capacity of one hard disk unit the number of hard disk units No RAID 1 2 Capacity of one hard disk unit Yes RAID 5 3 to 32 Capacity of one hard disk unit (the number of hard disk units - 1) Yes RAID 6 3 to 32 Capacity of one hard disk unit (the number of hard disk units - 2) Yes RAID 10 4 to 32 (Even number) Capacity of one hard disk unit (the number of hard disk units / 2) Yes [Note 1]: The maximum number of hard disk units indicates the specifications for an array controller. The actual configurable number of hard disk units depends on the specifications and configuration of the server. You should backup the data as frequently as possible just in case even with redundancy. When using RAID 6, you can continue the operation even when two hard disk units fail. Leaving them in the failed status, however, may increase the possibility of losing data if more hard disk units fail. In a RAID configuration with four or more hard disk units, use RAID 5 if capacity is important, RAID 10 if performance is important, or RAID 6 if redundancy is important. 18

MegaRAID SAS User's Guide Reliability and performance comparison between RAID levels When choosing RAID level, compare reliability and performance between RAID levels to determine the right level. The recommended RAID level is RAID 1. RAID 1 is superior to other RAID levels in data recoverability. RAID 1 can only be configured with two hard disk units and the available capacity will be equal to the capacity of a single hard disk unit. If the logical drive capacity is insufficient, you can increase the capacity by configuring multiple RAID 1 logical drives. RAID level Redundancy [Note 1] Data reliability table: Characteristics of each RAID level Performance Recoverability [Note 2] Write Read Capacity Remarks RAID 0 I I A A S RAID 1 A L A A L RAID 5 A I L A A RAID 6 S I I A L RAID 10 A I A A L Not recommended because no data redundancy is provided. Highly recommended for its excellent data recoverability. The write performance is poorer than other RAID levels. The write performance is poorer than RAID 5. Data is safe with failure of up to two hard disks. Available capacity is smaller than RAID 5, yet the write performance is usually higher than RAID 5. S : Superior A : Average L : Slightly inferior I : Inferior [Note 1]: Data redundancy when hard disks fail or get disconnected. [Note 2]: Data recoverability when two or more hard disk units are disconnected due to a failure other than a hard disk failure. For RAID 1, in which data is not striped, other systems may read one of the constituent hard disks in the RAID 1 configuration. However, even for RAID 1, data cannot always be recovered. 19

Chapter 1 Overview RAID 0 (striping) Striping refers to the function of dividing data and writing them on multiple hard disk units. For example, when writing nine blocks of data (D1 to D9), if three hard disk units are involved, data is written evenly on each hard disk as shown in the figure below. Performance is improved by writing data on three different hard disk units rather than writing all nine data blocks on a single hard disk unit. Data D-1 D-4 D-7 D-1 D-2 D-3 Hard disk 1 Array controller D-2 D-5 D-8 D-4 D-5 D-6 Hard disk 2 D-7 D-8 D-9 D-3 D-6 D-9 Hard disk 3 RAID 0 has no redundancy and all data may be lost in the case of even one hard disk unit failure. System operation will also be interrupted. RAID 1 (mirroring) Mirroring is a function in which identical data is written in two hard disk units in duplicate. When operating in RAID 1 disk array configuration, the system always writes the same data in two hard disk units, using the redundancy feature. Operation continues even if one of the hard disk units fails and its redundancy is lost (critical). RAID 1 always consists of two hard disk units, and the actual available capacity is equal to the capacity of one hard disk unit. Data D-1 D-2 D-3 Array controller D-1 D-2 D-3 Hard disk 1 D-1 D-2 D-3 Hard disk 2 With RAID 1, operation continues even if one of the hard disk units fails (critical). However, data may be lost if both hard disk units fail. If the status of a logical drive becomes critical, replace the failed hard disk unit as soon as possible to restore the redundancy. For how to replace hard disk units, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 20

MegaRAID SAS User's Guide RAID 5 (striping + parity) In addition to striping, which divides the data and distributes them over multiple hard disk units, RAID 5 generates parity data to provide redundancy. Parity data is data obtained through calculations on the data. For example, when three hard disk units compose a RAID 5, data is evenly written to two of the hard disk units. The parity data that is obtained through calculations of the data written to those two hard disk units is then written to the remaining one hard disk unit. The hard disk where parity data is written to is not fixed, but it is used per block in turn by three hard disk units. In a RAID 5 configuration, the size of a single hard disk unit is used for parity data in order to provide redundancy by parity data. Data D-1 D-2 D-3 D-4 Array controller D-1 D-4 Hard disk 1 D-2 Parity D-3 D-4 Hard disk 2 Parity D-5 D-6 D-5 D-5 D-6 Parity D-1 D-2 Hard disk 3 D-3 D-6 With RAID 5, operation continues even if one of the hard disk units fails and its redundancy is lost (critical). However, data may be lost if two or more hard disk units fail. If the status of a logical drive becomes critical, replace the failed hard disk unit as soon as possible to restore the redundancy. For how to replace hard disk units, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 21

Chapter 1 Overview RAID 10 (mirroring + striping) In addition to striping, which divides the data and distributes them over multiple hard disk units, RAID 10 provides redundancy by mirroring while using four or more (even numbers only) hard disk units. For example, when writing six blocks of data (D-1 to D-6), if four hard disk units are involved, each section of data and redundant data are written evenly on each hard disk unit as shown in the figure below. Performance is improved by distributing data over two hard disk units and redundancy is provided by the redundant data. Actually available disk size is half of the total hard disk size. When RAID 10 is configured using the same number of hard disk units, the available disk size is reduced compared to RAID 5, yet performance while writing the data is increased. D-1 D-3 D-5 Hard disk 1 Data D-1 D-2 D-3 D-4 D-5 D-6 Array controller D-1 D-3 D-5 Hard disk 2 D-2 D-4 D-6 Hard disk 3 D-2 D-4 D-6 Hard disk 4 With RAID 10, operation continues even if one of the hard disk units fails and its redundancy is lost (critical). However, data may be lost if two or more hard disk units fail. If the status of a logical drive becomes critical, replace the failed hard disk unit as soon as possible to restore the redundancy. For how to replace hard disk units, refer to "Chapter 6 Replacing Hard Disk" (pg.249). The capacity of a RAID 10 logical drive cannot be expanded. "1.3.5 Expand Capacity" (pg.36) In a disk group of RAID 10, multiple logical drives cannot be defined. Only a single logical drive can be created in a single RAID 10 disk group with the maximum capacity of the group. 22

MegaRAID SAS User's Guide RAID 6 (striping + dual parity) RAID 6, in addition to the striping + parity of RAID 5, improved its redundancy by doubling parity data. Take a case with RAID 6 configured with five hard disk units for instance. The data is written into three hard disk units, and the remaining two have the parity data [P] which was calculated based on the data written to the three disks, and the parity data [Q] which was calculated in a different way from the [P]. Parity data is not written into specific disks but is written into all disks by the block in turn. In RAID 6, for the better redundancy, the capacity equal to two hard disk units are used for the parity data. Data D-1 D-2 D-3 D-4 D-5 D-6 Array controller D-1 D-5 D-9 Hard disk 1 D-2 D-6 Hard disk 2 D-3 Parity-P D-4,D-5 D-6 Hard disk 3 Parity-P D-7,D-8 D-9 Parity-Q D-7,D-8 D-9 D-7 D-8 D-9 Parity-P D-1,D-2 D-3 Parity-Q D-1,D-2 D-3 Parity-Q D-4,D-5 D-6 Hard disk 4 D-4 Hard disk 5 D-7 D-8 RAID 6 keeps redundancy (partially critical) even when a hard disk unit fails. Furthermore, even when another hard disk unit fails resulting in redundancy free status, the operation will continue. The chance, however, of losing data increases since the first hard disk unit failure. It gets higher at the second hard disk unit failure. The third and the subsequent hard disk unit failures result in the data loss. Replace a failed hard disk unit soon after its failure while logical drives are still in partially critical status, and restore its redundancy. When two hard disk units fail (logical drives are in the critical status), their data is almost lost. Replace the failed disks at once to restore its redundancy. For how to replace hard disk units, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 23

Chapter 1 Overview 1.2.4 Disk Group and Logical Drive The management unit that collectively groups the number of hard disk units required for forming RAID is referred to as a disk group and the logical units configured within a disk group are called logical drives. Logical drive 0 Logical drive 1 Logical drive 2 Hard disk 1 Logical drive 3 Logical drive 4 Logical drive 5 Hard disk 2 Hard disk 4 Hard disk 3 Disk group A Hard disk 5 Disk group B Disk group A disk group is a group of physical hard disks that form a disk array. It is not recognized by the OS. A disk group can consist of 1 to 32 hard disk units. Hard disks in a single disk group should be of the same model (with the same capacity and rotation speed). Up to eight disk groups can be configured within a single array controller. Logical drive A logical drive is a logical hard disk space within a disk group. It is recognized as a single hard disk by the OS. Up to eight logical drives can be configured within a single array controller. In a disk group of RAID 10, multiple logical drives cannot be defined. Only a single logical drive can be created in a single RAID 10 disk group with the maximum capacity of the group. Only one type of RAID level can be configured for the logical drives within a disk group. 24

MegaRAID SAS User's Guide Example of disk group and logical drive configuration Assume that two disk groups A and B are created using five hard disk units and six logical drives are created in those groups. In this case, the OS will assume that six hard disk units are connected. The OS assumes that six hard disk units are connected. Logical drive 0 Logical drive 1 Logical drive 2 Hard disk 1 Logical drive 3 Logical drive 4 Logical drive 5 Hard disk 2 Hard disk 4 Hard disk 3 Disk group A Hard disk 5 Disk group B Configuring conditions Disk group A (hard disk 1 to 3) must consist of hard disks of the same type and capacity. Disk group B (hard disk 4 and 5) must consist of hard disks of the same type and capacity. One type of RAID level can be configured for logical drives 0 to 2 of disk group A. One type of RAID level can be configured for logical drives 3 to 5 of disk group B. Status during hard disk failure In the example above, if hard disk 1 fails, the status of all the logical drives from 0 to 2 in the same disk group becomes without redundancy (critical). If another disk (e.g. hard disk 2 or 3) also fails, the status of the logical drives from 0 to 2 becomes unavailable (offline). However, the logical drives from 3 to 5 of disk group B will remain normal operation (online) because a hard disk problem is not present. 25

Chapter 1 Overview 1.2.5 Checking Hard Disk Status Hard disks and logical drives in a disk group should be constantly monitored, and be replaced if there is a failure or indication of a failure. The status differs depending on each management tool. For information about the status, refer to the appropriate description in your management tool's documents. Web BIOS "2.2.2 Checking Logical Drive Status" (pg.57) "2.2.1 Checking Hard Disk Status" (pg.53) ServerView RAID "4.6.3 Layout of Tree View" (pg.142) GAM "5.6.4 Starting Controller View and the Window Layout" (pg.210) When ServerView RAID or GAM operates properly, information is logged as an event if an error occurs in the logical drive or hard disk. "Appendix A List of ServerView RAID Event Log" (pg.276) "Appendix B List of GAM Event Log" (pg.313) Logical drive status A logical drive can be in the following status. Hard disk status A hard disk can be in the following status. table: Logical drive status Status Description Online Operating normally. Critical Operating without redundancy. A hard disk unit (or more) for RAID 1, RAID 5, or RAID 10 logical drives, or two hard disk units for RAID 6 logical drive have been failed. Replace the failed hard disk units as soon as possible, and rebuild to restore the online status. For how to replace and rebuild hard disks, refer to "Chapter 6 Replacing Hard Disk" (pg.249). Partially critical Operating with redundancy. But a hard disk unit of RAID 6 logical drives has been failed. Replace the failed hard disk unit as soon as possible, and rebuild to restore the Online" status. For how to replace and rebuild hard disks, refer to "Chapter 6 Replacing Hard Disk" (pg.249). Offline Not operating. Two or more hard disk units in a disk group have been failed. The data in the logical drive has also been lost. Status Online Unused Failed Offline Hotspare table: Hard disk status Description Operating normally. It is included in the disk array. Operating normally. It is unused because it is not included in the disk array. Unable to read/write data. The hard disk has been failed. Replace and rebuild it. For how to replace hard disks, refer to "Chapter 6 Replacing Hard Disk" (pg.249). Data read/write is disabled by the array controller. You need to rebuild the hard disk for the further usage. The hard disk is configured as a spare disk. 26

MegaRAID SAS User's Guide Status Rebuild Failure Predicted table: Hard disk status The hard disk is currently being rebuilt. Description Currently operating normally. But it may fail in the near future (failure predicted status by the PFA / S.M.A.R.T. function). Refer to "Chapter 6 Replacing Hard Disk" (pg.249), and replace the hard disk as soon as possible. If there is a hard disk with failure status, there may be a hard disk failure and the hard disks may be operating without redundancy. The hard disk must be replaced immediately. For how to replace a hard disk, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 27

Chapter 1 Overview 1.3 Disk Array Configuration Features This section explains the features in disk array configuration. 1.3.1 Write Policy Write Policy is a setting of a data writing mode for logical drives on the array controller and can be configured for each logical drive. There are three Write Policy modes: Write Through, Write Back, and Bad BBU. Changing your Write Policy from Write Through to Write Back (or Bad BBU) improves many software applications in writing performance. If efficiency is essential in your operation, review your Write Policy. However there are things you should keep in mind when using Write Back (or Bad BBU). Before setting up, be sure to read the following contents and know the gist of Write Policy operation. When installing the OS on the logical drive connected to this array controller, set Write Policy of the logical drive to Write Through before the OS installation. If you want to set Write Policy to Write Back (or Bad BBU), change it after the OS installation is complete. Summary of Write Policy operation Write Through When writing data from the OS and application to hard disk (disk array) is issued, the completion of writing is only recognized after the writing to hard disk is finished. Data Write command Array controller receives data and reports on the writing completion to the host after the data writing in the disk is complete. Array Controller Writing of data Hard disk Data Completion report Completion report Logical drive 28

MegaRAID SAS User's Guide Write Back / Bad BBU When writing data from the OS and application to hard disk (disk array) is issued, writing is complete after finishing the writing data to cache memory on the array controller. Writing into hard disk is actually carried out by the array controller after the operation of the OS or application, not at the same time. Data Write command Array controller receives data and reports on the writing completion to the host simultaneously. The multiple data will be written in the disk later. Array Controller Writing of data Hard disk Data Completion report Writing to hard disk is performed mechanically, but writing to cache memory is performed electrically. Consequently, drastic improvement in efficiency can be expected by the OS or application finalizing the control with cache memory, not by directly accessing to hard disk. Write Back (or Bad BBU) mode is an efficient means of bringing a full capacity of high speed cache memory when writing. Notes on Write Back (or Bad BBU) In setting up Write Back (Bad BBU) mode, time lug exists between the time when the OS and software application acknowledge the completion of writing and the time of actual writing into hard disk. Consequently, if power problem to server occurs before writing into a hard disk, data to be written which is left on cache memory may be destroyed, and there may be a gap in data between the one that OS or software application acknowledges and the one that is actually written on hard disk. When using Write Back mode Logical drive When you use Write Back, make sure to install battery backup unit (BBU) which protects the data on the cache memory to minimize the risk described above. Make sure to either utilize array controller which supports battery backup unit, or install battery backup unit option. For the details of battery backup unit, refer to "1.3.8 Battery Backup Unit (BBU)" (pg.39). If you use Write Back with battery backup unit for cache memory installed, the data on the cache memory is protected by the battery at the time of unexpected blackout, and the writing of data from the cache memory to the hard disk is carried on when the system restarts (when the next server is on), reducing the risk of the data loss. If the battery is not fully charged, or when a recalibration is in progress, the data may not be kept on cache memory. To avoid such data loss, the logical drive set on Write Back is automatically changed to Write Through mode. It returns to Write Back automatically when the battery is fully charged or the recalibration is complete. If the server's power-off state continues for a long time, the logical drive with Write Back setting will forcibly operate as Write Through at the next startup due to the discharged battery. The logical drive whose setting is forcibly changed will automatically return to Write Back after the battery is charged enough. It is recommended that you connect UPS to the server, as well as the battery backup unit to avoid accidents caused by blackout. 29

Chapter 1 Overview When using Bad BBU mode When you set up Bad BBU, the Write Back mode is maintained whether a battery is installed or if a battery is charged. Therefore, when no battery backup unit is installed or a battery backup unit is not fully charged, Write Policy is not automatically changed to Write Through mode, and data on cache memory is not saved. Risks to lose data in case of unexpected accidents are higher compared to Write Back. Do not use Bad BBU if the power supply to the server is not secured by UPS or other devices. Otherwise, there is a risk of data loss in case of unexpected blackout. If the battery is fully charged, the data on the cache memory is protected by the battery at the time of unexpected blackout, and the writing of data from the cache memory to the hard disk is carried on when the system restarts (next time the server is on). 1.3.2 Logical Drive Initialization After creating a logical drive, it must be initialized in order to use it properly. There are three methods to initialize a logical drive as described below. Foreground initialization With this method, "0 (zero)" is written in all areas of a logical drive to maintain consistency between the hard disks and to initialize them. This requires a certain amount of time because write operations are performed for the entire space of the hard disks. Foreground initialization is performed from WebBIOS. "2.5.1 Initializing Logical Drive" (pg.89) When the server restarts during foreground initialization, the initialization is stopped and the logical drive is not initialized. Perform the initialization again. If foreground initialization is not performed, background initialization is performed automatically. 30

MegaRAID SAS User's Guide Estimated time for foreground initialization The following table shows the estimated initialization time per 10 GB of logical drive space. table: Estimated time for foreground initialization per 10 GB RAID level Number of hard disks Time per 10 GB RAID 1 2 approx. 3 min. RAID 5 3 approx. 2 min. 6 approx. 2 min. RAID 6 3 approx. 2 min. 6 approx. 2 min. RAID 10 4 approx. 3 min. For example, the capacity of a RAID 5 logical drive consisting of three 73 GB hard disks is 146 GB (= 73 GB (3-1)), which requires approximately 30 minutes (= approx. 2 min./10 GB 146 GB) to initialize. However, the time may differ from the value in the above table depending on the configuration and the hard disk type. Background initialization With this method, initialization of the logical drive is performed in parallel with normal I/O access operations from the host. You can use the logical drive immediately after creating the logical drive without waiting for completion of the initialization process. This can save the time of a foreground initialization, but the following points must be noted. During background initialization, access to the hard disks may occur regardless of normal I/O access. Therefore, I/O processing performance may decrease during background initialization. Do not perform background initialization on hard disks that may contain partition information. If you need to perform background initialization on hard disks that contain data, refer to "2.7 Reusing Hard Disk" (pg.98), and erase the data from the hard disks before configuring the array. Background initialization starts automatically when the array controller checks redundant logical drives (RAID 1, RAID 5, RAID 10, and RAID 6) and detects a drive that is not completely initialized. The array controller checks for logical drives that are not completely initialized every 5 minutes. If the server is reset or turned off before completing the background initialization, the background initialization is interrupted. Background initialization is resumed from where it was stopped the next time. During background initialization, the access lamps of the hard disks that make up the target logical drive remain lit. The execution time of this function can be reduced by changing the BGI Rate and giving priority to this function over server I/O, but I/O performance will be reduced accordingly. Normally, use the default settings. If foreground initialization is not performed, background initialization is performed automatically. Even when RAID 6 logical drives are partially critical, background initialization can be done. When you replace the failed hard disks and the rebuild starts, background initialization is canceled and the rebuild starts. In that case, background initialization starts after the rebuild. 31

Chapter 1 Overview Estimated time for background initialization The following table shows the estimated execution time per 10 GB of logical drive space when there is no server I/O. table: Estimated time for background initialization per 10 GB RAID level Number of hard disks Execution time per 10 GB RAID 1 2 approx. 20 min. RAID 5 3 approx. 15 min. 6 approx. 24 min. RAID 6 3 approx. 15 min. 6 approx. 24 min. RAID 10 4 approx. 24 min. For example, the capacity of a RAID 5 logical drive consisting of three 73 GB hard disks is 146 GB (= 73 GB (3-1)), the execution time is approximately 220 minutes (= approx. 15 min./10 GB 146 GB) to initialize. However, the time may differ from the value in the above table depending on the configuration and the hard disk type. The above execution time is an estimate for when the BGI Rate is the default value. Fast initialization With this method, "0 (zero)" is written to only the first 50KB of the logical drive and the remaining area of the logical drive is initialized with the subsequent background initialization which starts automatically. If partition information remains in the first section of the logical drive, performing fast initialization rather than background initialization deletes the partition information. Therefore, background initialization can be performed without performing initialization on the hard disk in advance. Fast initialization can be performed with WebBIOS. For the detailed procedure, refer to "2.5.1 Initializing Logical Drive" (pg.89). This fast initialization completes in few seconds, but the subsequent background initialization may take some time. For notes about background initialization, refer to " Background initialization" (pg.31). 32

MegaRAID SAS User's Guide 1.3.3 Rebuild Even when a single hard disk unit fails in logical drives with redundancy such as RAID 1, RAID 5, or RAID 10, or two hard disk units fail in RAID 6 which has double redundancy, the logical drives continue to operate in the critical status. If, however, another hard disk in the same logical drive fails, the logical drive becomes unavailable (offline). Rebuild is an operation to recover a logical drive from the critical status to the normal (online) status. The rebuild process is performed in the background in parallel with normal I/O access. During rebuild, if a logical drive becomes critical, the redundancy of the logical drive remains lost. Once the rebuild is complete, it is restored to online. There are two methods for rebuilding as described below. Hot spare rebuild Hot spare rebuild is rebuild that is performed automatically in the event of a hard disk failure, by having a pre-installed spare hard disk. The pre-installed spare hard disk is called a spare disk. When a hard disk failure occurs, rebuild is immediately performed on the spare disk, to improve the reliability of the disk array. You need to have a spare hard disk set as a spare disk in advance. Setting up the spare disk can be performed from WebBIOS or a management tool. A spare disk is used as a substitute for a failed hard disk. Therefore, use a hard disk with the same capacity and speed as the hard disks connected in the disk group. When a failed hard disk is replaced after starting a hot spare rebuild, the newly installed hard disk is automatically configured as the spare disk. When a hot spare rebuild is performed, the bay where the disk group and the spare disk are installed is changed. A hot spare rebuild is not performed for the hard disk with a failure prediction sign. Once preventive replacement is done, the rebuild starts. "1.3.6 Hard Disk Failure Prediction Function (PFA / S.M.A.R.T.)" (pg.38) When configuring spare disks in an environment with multiple disk groups using different types of hard disks, for each disk group, configure at least one spare disk with the same capacity and speed as the hard disk used in each disk group. 73 GB Hard disk 1 73 GB Hard disk 2 Disk group A Spare disk 73 GB Hard disk A 147 GB Hard disk B 147 GB Hard disk 1 147 GB Hard disk 2 Disk group B Manual rebuild Manual rebuild is rebuild that is performed by replacing a failed hard disk with a new hard disk. When no spare disk is available, the disk array needs to be repaired through manual rebuild. 33

Chapter 1 Overview Estimated time for rebuild The following table shows the estimated execution time per 10 GB of logical drive space when there is no server I/O. table: Estimated time for manual rebuild per 10 GB RAID level Number of hard disks Execution time per 10 GB RAID 1 2 approx. 4 min. RAID 5 3 approx. 3 min. 6 approx. 3 min. RAID 6 3 approx. 3 min. 6 approx. 3 min. RAID 10 4 approx. 2 min. For example, the capacity of a RAID 5 logical drive consisting of three 73 GB hard disks is 146 GB (= 73 GB (3-1)), which requires approximately 44 minutes (= approx. 3 min./10 GB 146 GB) to rebuild. However, the time may differ from the value in the table above depending on the configuration and the hard disk type. The execution time above is an estimate when the Rebuild Rate is the default value. The rebuild process is performed in parallel with normal I/O access, therefore I/O performance for the logical drives can be decreased during rebuild. With this array controller and the default settings, the I/O performance decreases about 30% at maximum. The rebuild time can be reduced by setting [Rebuild Rate] with a large value (the maximum is "100") which will give higher priority to the rebuild process. However, I/O performance can be lower than that of the default settings. Normally, use the default settings. [Rebuild Rate] can be set with each management tool. WebBIOS "2.2.4 Confirming Array Controller Information" (pg.63) ServerView RAID "4.7.1 Changing Array Controller Setting" (pg.147) GAM "5.7.2 Setting and Changing Controller Options" (pg.214) If the system restarts or shuts down during rebuild, the rebuild is resumed from where it was stopped the next time. Replacing two failed hard disks from RAID 6 logical drives at the same time will keep both in rebuilding states. The actual rebuild, however, is done each at a time. While a hard disk is rebuilding, another rebuild progress is 0%. 34

MegaRAID SAS User's Guide 1.3.4 Make Data Consistent Make Data Consistent is a function to maintain consistency between data on redundant logical drives (RAID 1, RAID 5, RAID 6, or RAID 10) and mirrored data or parity data. If the server is not shut down properly, the data consistency among hard disks may be lost. In such cases, use this function to make the data consistent. Data sent to the server is never changed by performing Make Data Consistent because correction of mirrored data and parity data is performed as necessary. The Make Data Consistent function is also used to recover from hard disk media errors (different from inconsistency) because it reads the data on the hard disk. If rebuild is performed for RAID 10, Make Data Consistent is automatically performed after the rebuild is complete. The I/O performance will decrease during Make Data Consistent. The execution time of Make Data Consistent can be reduced by changing the CC Rate (MDC Rate for ServerView RAID) and giving priority to this function over system I/O, but then I/O performance is reduced accordingly. Normally, use the default settings. Even when RAID 6 logical drives are partially critical, Make Data Consistent can be done. When you replace the failed hard disks and rebuild starts, Make Data Consistent is canceled and rebuild starts. Estimated time for Make Data Consistent The following table shows the estimated execution time per 10 GB of logical drive space when there is no server I/O. table: Estimated time for Make Data Consistent per 10 GB RAID level Number of hard disks Execution time per 10 GB RAID 1 2 approx. 20 min. RAID 5 3 approx. 15 min. 6 approx. 24 min. RAID 6 3 approx. 15 min. 6 approx. 24 min. RAID 10 4 approx. 24 min. For example, the capacity of a RAID 5 logical drive consisting of three 73 GB hard disks is 146 GB (= 73 GB (3-1)), which requires approximately 220 minutes (= approx. 15 min./10 GB 146 GB) for Make Data Consistent. However, the time may differ from the value in the table above depending on the configuration and the hard disk type. The execution time above is an estimate when CC Rate is the default value. 35

Chapter 1 Overview 1.3.5 Expand Capacity With the Expand Capacity function, you can expand the capacity of the logical drive or change the RAID level by adding a new hard disk to the logical drive while leaving the existing data. The following figure shows an example of adding two hard disks to a disk group which consists of three hard disks. Reconfiguring the RAID of five hard disks will keep the data of the logical drive, and the capacity amount of the new disks will be added to that of the logical drive. Logical drive 0 (after expanding the capacity) Logical drive 0 Hard disk 1 Hard disk 2 Hard disk 3 Hard disk 4 Hard disk 5 Hard disk 1 Hard disk 2 Hard disk 3 Hard disk 4 Hard disk 5 Disk group After Expand Capacity is performed, the RAID level may change as follow. RAID level before Expand Capacity Additional disk table: RAID level after Expand Capacity RAID level after Expand Capacity Migration only Disk group Migration with addition RAID 0 N/A RAID 0, RAID 1, RAID 5, or RAID 6 RAID 1 RAID 0 RAID 0, RAID 5, or RAID 6 RAID 5 RAID 0 RAID 0, RAID 5, or RAID 6 RAID 6 RAID 0, RAID 5 RAID 0, RAID 5, or RAID 6 RAID 10 Unable to expand the capacity Unable to expand the capacity Only the capacity of logical drives is expanded by Expand Capacity. The capacity of OS partitions is not expanded. The execution time of Expand Capacity can be reduced by changing the Reconstruction Rate (More Rate for GAM, and Migration Rate for ServerView RAID) and giving priority to this function over system I/O, but I/O performance is reduced accordingly. Normally, use the default settings. If a logical drive expanded its capacity has RAID level with redundancy (RAID 1, RAID 5, or RAID 6), background initialization will be performed. If Expand Capacity is performed without adding a hard disk, the RAID level of the logical drive is converted to RAID 0 with only few exceptions, and the logical drive loses its redundancy. It is recommended that you add a hard disk when performing Expand Capacity. During Expand Capacity, logical drives with their Write Policy set to Write Back or Bad BBU will temporarily be set to Write Through. 36

MegaRAID SAS User's Guide Expand Capacity is only available when the logical drive is used in a Windows environment and the partition style for the logical drive has been initialized in MBR format. Do not expand the capacity if the partition style is in GUID Partition Table (GPT) format under Windows, or if using Linux. The GPT partition style may be used for Windows Server 2008 or Windows Server 2003. The capacity of RAID 10 logical drive cannot be expanded. If there are multiple logical drives within a disk group, the capacity cannot be expanded for the logical drives in that disk group. Expand Capacity cannot be interrupted while it is in progress. Also, if the server restarts during Expand Capacity, the data within the logical drive will be lost. Make sure to back up data in advance. The I/O load is very high while capacity is being expanded, and operation will slow down. It is recommended that you perform Expand Capacity during non-business hours, or low-business hours. In addition, since the logical drive loses its redundancy during Expand Capacity, all the data on the logical drive will be lost if a hard disk failure occurs. Estimated time for Expand Capacity The following table shows the estimated execution time per 10 GB of logical drive space when there is no server I/O. RAID level before Expand Capacity table: Estimated time of Expand Capacity per 10 GB Execution time per 10 GB RAID level after Expand Capacity Number of hard disks When adding 1 drive When adding 3 drives RAID 0 1 RAID 0 approx. 6 min. approx. 3 min. RAID 0 1 RAID 1 approx. 7 min. N/A RAID 0 1 RAID 5 N/A approx. 4 min. RAID 0 1 RAID 6 N/A approx. 4 min. RAID 1 2 RAID 0 approx. 3 min. approx. 2 min. RAID 1 2 RAID 5 approx. 4 min. approx. 2 min. RAID 1 2 RAID 6 approx. 6 min. approx. 4 min. RAID 5 3 RAID 0 approx. 4 min. approx. 3 min. RAID 5 3 RAID 5 approx. 3 min. approx. 3 min. RAID 5 3 RAID 6 approx. 5 min. approx. 5 min. RAID 6 3 RAID 0 approx. 5 min. approx. 3 min. RAID 6 3 RAID 5 approx. 4 min. approx. 4 min. RAID 6 3 RAID 6 approx. 4 min. approx. 4 min. For example, when you expand capacity of a RAID 5 logical drive consisting of three 73 GB hard disks is 146 GB (= 73 GB (3-1)), the execution time is approximately 44 minutes (= approx. 3 min./10 GB 146 GB) by adding one hard disk. However, the time may differ from the value in the above table depending on the configuration, the hard disk type and the number of added hard disks. Use the table only as a reference. The execution time above is an estimate when the Reconstruction Rate is the default value. 37

Chapter 1 Overview 1.3.6 Hard Disk Failure Prediction Function (PFA / S.M.A.R.T.) The PFA / S.M.A.R.T. function is a failure prediction function for the hard disks that determines the risk of a failure in advance and issues a warning when the risk is high. Although a hard disk will still operate normally even when a risk of a failure is predicted, that hard disk may fail in the near future and should be replaced as soon as possible. For the replacement procedure, refer to "6.1.4 Preventive Replacement of Hard Disk [ServerView RAID]" (pg.255) or "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268). The hard disk with a failure predicted can be identified by a management tool or event logs. 1.3.7 HDD Check Scheduler HDD Check Scheduler is a program to detect or recover media errors on hard disks. HDD Check Scheduler performs media checks to all logical drive areas including drive areas that are not accessed in everyday operation by performing Make Data Consistent periodically. Overview Media errors may occur accidentally on a hard disk, in particular if the drive is exposed to a severe shock during operation. In that case, since this kind of event is not a hard disk failure, the media errors will not be detected until reading the data in the erroneous sector. If the drive is configured in a RAID level with redundancy, this array controller's usual functions will recover the data using the data on other hard disks. However, if a medium error occurs in a file that is rarely accessed, the error will remain unrecovered, and if an additional hard disk should also fail, the data will be lost. Use HDD Check Scheduler properly to perform media checks periodically. Make sure to install HDD Check Scheduler. If you install the OS with ServerStart, HDD Check Scheduler will be automatically installed with a RAID management tool. Manual installation of the OS requires you to install HDD Check Scheduler after RAID management tool installation. For how to install it, refer to the following. When using ServerView RAID "4.2 Installing ServerView RAID [Windows]" (pg.121) "4.3 Installing HDD Check Scheduler [Windows]" (pg.124) When using GAM "5.2 Installing GAM [Windows]" (pg.189) "5.3 Installing HDD Check Scheduler [Windows]" (pg.194) HDD Check Scheduler periodically executes Make Data Consistent using a standard OS taskscheduling function. When you install HDD Check Scheduler, Make Data Consistent is set to start at 12:00 every day by default. For how to change the settings, refer to the following. When using ServerView RAID "4.3.2 Changing HDD Check Start Time" (pg.125) When using GAM "5.3.2 Changing HDD Check Start Time" (pg.195) The HDD checks using HDD Check Scheduler can only be performed on logical drives with redundancy (RAID 1, RAID 5, RAID 6, or RAID 10). The checks cannot be performed on RAID 0 logical drives or spare disks. 38

MegaRAID SAS User's Guide 1.3.8 Battery Backup Unit (BBU) For a model with a battery backup unit installed, unwritten data can be maintained in the cache memory of the array controller even if shutdown occurs due to unintentional damage, power fault, or blackout. When Write Back or Bad BBU is used, if the power is accidentally not supplied while the data unwritten to the hard disk is still in the cache memory, the battery backup unit helps to maintain the data by supplying the power to the cache memory. In such cases, data can be maintained for the following time period since the time of power-off when the battery is fully charged. Up to 72 hours for MegaRAID SAS 300-8E ROMB, MegaRAID SAS 8344ELP, MegaRAID SAS 8408E, MegaRAID SAS8880EM2, RAID 5/6 SAS based on LSI MegaRAID Up to 37 hours for MegaRAID SAS 8300XLP Battery is a lifetime component that degrades over time, so you must replace it periodically. For details, refer to "Appendix C Replacing Battery" (pg.336). The length of the data holding time is an approximate value. The length of the data holding time varies depending on the temperature of the operation environment, the age and cache memory size of the server, and other factors. The battery backup unit maintains data in the cache memory of this product and does not ensure the data in the system memory of the server. The battery backup unit must be used solely for unintentional power fault, so do not use it for any operations (such as power interruption during access) other than its intended purpose. In normal operation, power must not be supplied to the cache memory from the battery. The battery is automatically charged while server is on. When the server is off for long periods of time, the remaining battery capacity may be reduced by discharging naturally even if data does not exist in the cache memory. In such cases, the battery recalibration may automatically be performed when the server is turned on. In the following situations, perform battery recalibration before using the array controller. When you newly purchase a model of the array controller with a battery backup unit installed When you additionally purchase a battery backup unit and it is newly installed When you replace parts including a battery For the battery recalibration, refer to "4.8.9 Recalibration" (pg.170) for ServerView RAID and "5.10.3 Recalibration" (pg.247) for GAM. Battery recalibration requires 12 hours with the server power on. If the server is turned off and on during battery recalibration, the recalibration restarts from the beginning. When you replace array controller cards, battery backup units, or the battery, the recalibration of the battery may automatically start. While the remaining battery capacity is low and the recalibration is performed, a logical drive that is set to Write Back automatically changes to Write Through for safety. When the battery recalibration is complete, Write Policy automatically change back to Write Back. For a logical drive that is set to Bad BBU, write cache is always enabled even during battery recalibration. 39

Chapter 1 Overview Battery Recalibration Scheduler The battery backup unit on MegaRAID SAS 300-8E ROMB, MegaRAID 8344ELP, MegaRAID SAS 8408E, MegaRAID SAS 8880EM2, or RAID 5/6 SAS based on LSI MegaRAID has a battery capacity management function. Periodical recalibration is necessary for the battery backup unit mentioned above, since long-time operation causes a margin of error in the battery capacity managed by the battery backup unit. This array controller can perform periodical recalibration by installing Battery Recalibration Scheduler. When using MegaRAID SAS 300-8E ROMB, MegaRAID SAS 8408E, MegaRAID 8344ELP, or RAID 5/6 SAS based on LSI MegaRAID with a battery backup unit, make sure to install Battery Recalibration Scheduler. If you install the OS with ServerStart, Battery Recalibration Scheduler will be automatically installed with a RAID management tool. Manual installation of the OS requires you to install Battery Recalibration Scheduler after RAID management tool. For how to install it, refer to the following. When using ServerView RAID "4.2 Installing ServerView RAID [Windows]" (pg.121) "4.4 Installing Battery Recalibration Scheduler [Windows]" (pg.129) When using GAM "5.2 Installing GAM [Windows]" (pg.189) "5.4 Installing Battery Recalibration Scheduler [Windows]" (pg.198) Battery Recalibration Scheduler periodically executes recalibration using a standard OS taskscheduling function. When you install Battery Recalibration Scheduler, recalibration is set to start at 11:00 on the 1st day of each month by default. For how to change the settings, refer to the following. When using ServerView RAID "4.4.2 Changing Battery Recalibration Start Date and Time" (pg.130) When using GAM "5.4.2 Changing Battery Recalibration Start Date and Time" (pg.199) Battery Recalibration Scheduler is not necessary for MegaRAID SAS 8300XLP. Periodical recalibration will not be performed even if Battery Recalibration Scheduler is installed. 40

MegaRAID SAS User's Guide 1.4 Notes before Operation Check the following before starting the operation. 1.4.1 Notes on Installation to Server This section explains notes when installing the array controller of a PCI card type on the server. When you install this product on the server, turn off the server, all peripheral devices, and any other connected devices. Also unplug all power cables from the outlet. If you install or remove this product when the power is turned on, it may cause device failure, fire, or electric shock. This product reaches very high temperatures while the server is in operation. To remove this product, first turn off the server, wait until it is cool, and then remove the product from the server. Notes When you install the array controller in a PCI slot, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server, and properly follow the procedure described in the document to install them on the server. Installable PCI slot location is specified by the target server. Make sure the location of the installable PCI slot is correct before installing. Usage may be limited depending on the target server. On a RAID management tool, a controller number is assigned for this array controller. The number to be assigned is different depending on the number of array controllers to be installed and their PCI slot location. For details, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 41

Chapter 1 Overview 1.4.2 Notes on Hard Disk to Be Used The following notes apply to the hard disks to be used. Please check in advance. Usable hard disk All hard disks in a single disk group use the same model (with the same capacity and rotation speed) as a rule. Check that the installed hard disks have the same capacity and rotation speed. Also, be sure to check that the hard disks are installable on the server. Reusing hard disk Hard disks containing data may have partition information or array configuration information. Using such hard disk units without taking the proper measures may cause unexpected problems. When using a previously used hard disk, erase the data by performing low level format on the system that was using the hard disk, before connecting the disk to this array controller. The same caution applies when using hard disks used by this array controller on another system. Refer to "2.7.2 Clearing Information from Hard Disk" (pg.98), and completely erase the information on the hard disk before using it on another system. Removing hard disk When the server is turned on, do not remove the hard disks except for the following situations: When replacing a failed hard disk When replacing a hard disk as a preventive measure Notes on connecting device Do not connect any devices other than Fujitsu-supported hard disks to this array controller. 1.4.3 Cautions for OS Installation When you install an OS on the hard disk under the array controller, there are the following cautions. Please check in advance. About Write Policy When you install an OS, set Write Policy of the logical drive to Write Through. When you use Write Back or Bad BBU, change the Write Policy from a management tool or WebBIOS after the OS installation and hotfix application are complete. For details about applying the hotfix, refer to "3.2 Applying Hotfix" (pg.112). 42

MegaRAID SAS User's Guide 1.5 Notes on Operation This section contains notes concerning system operation when using this array controller. 1.5.1 Notes for Using Array Controller When using Windows in a disk array configuration, the following event may be logged in the event viewer's system log. Source : msas2k3 Type : Warning Event ID : 129 Description: The description for Event ID (129) in Source (msas2k3) cannot be found. (The rest is omitted.) This log entry means that an internal reset has been issued in the device driver, but since the event has been restored by an OS retry, you can continue with the operation. However, if this event occurs repeatedly (about twice or more in every 10 minutes), there is a possibility of a hardware failure. Contact an office listed in the "Contact Information" of "Start Guide", and check the array controller and hard disks. 1.5.2 Message during POST If the following message appears, the hard disk may have a failure. Foreign configuration(s) found on adapter Some configured disks have been removed from your system, or are no longer accessible. Please check your cables and also ensure all disks are present. In that case, refer to "Chapter 6 Replacing Hard Disk" (pg.249) to check the hard disk condition, replace the hard disk, and perform rebuild. 1.5.3 When the Server Is Not Shut Down Properly If the server is turned off without performing a proper shutdown due to blackout, the consistency of the logical drive may be lost. In such cases, perform Make Data Consistent. For details, refer to the following. "1.3.4 Make Data Consistent" (pg.35) "2.5.3 Make Data Consistent of Logical Drive" (pg.91) "4.8.7 Make Data Consistent" (pg.168) (When using ServerView RAID) "5.10.1 Make Data Consistent" (pg.245) (When using GAM) 43

Chapter 1 Overview 1.5.4 Using ServerView RAID under the Windows 2000 Server Environment Under the conditions below, at the restart of Windows 2000 Server, events reporting abnormalities of hard disks or logical drives may be logged. When restarting Windows 2000 Server after installing Windows 2000 Server and ServerView RAID by using ServerStart When restarting Windows 2000 Server after applying the hotfix KB904374 of Windows 2000 Server with ServerView RAID installed in it The details of the logged events are as follows. Note that the [Array Controller Name], Logical drive [%s], Disk [%s], and [Server Name] in the messages are different according to the types of array controller that are installed, the number of created logical drives, the number of installed hard disks, and the setting values of the server name. For events via ServerView Source Fujitsu ServerView Services ID 3 Message ServerView received the following alarm from server [Server Name]: Adapter [Array Controller Name]: Disk [%s] missing after reboot (Server [Server Name]) Description Unable to find the hard disk at the restart of the server. Example ServerView received the following alarm from server PRIMERGY03: Adapter LSI MegaRAID SAS 8408E (0): Disk (0) missing after reboot (Server PRIMERGY03) Source Fujitsu ServerView Services ID 3 Message ServerView received the following alarm from server [Server Name]: Adapter [Array Controller Name]:Logical drive [%s] missing after reboot (Server [Server Name]) Description Unable to find the logical drive at the restart of the server. Example ServerView received the following alarm from server PRIMERGY03: Adapter LSI MegaRAID SAS 8408E (0): Logical drive (0) missing after reboot (Server PRIMERGY03) 44

MegaRAID SAS User's Guide Source: For ServerView RAID events, or ServerView RAID Manager event window Source ServerView RAID ID 10476 Message Adapter %s: Disk (%s) missing after reboot Description Unable to find the hard disk at the restart of the server. Example Adapter LSI MegaRAID SAS 8408E (0): Disk (0) missing after reboot Source ServerView RAID ID 10477 Message Adapter %s: Logical drive %s missing after reboot Description Unable to find the logical drive at the restart of the server. Example Adapter LSI MegaRAID SAS 8408E (0): Logical drive (0) missing after reboot When the above events are logged, check the states of the array using ServerView RAID Manager. If the array is recognized as the normal state, ignore the events. Information that Windows SCSI mini port driver outputs will be changed by applying the hotfix. So, the information is different from what ServerView RAID memorized until then. That is why these events occur. These events may be logged at the first restarting after applying the hotfix. These events will not occur if the state of the array is normal after the next startups. These events are logged when restarting the server after installing Windows 2000 Server since the hotfix is automatically applied if it is installed using ServerStart. 1.5.5 Notes on WebBIOS Startup when Installing Multiple MegaRAID SAS Controllers When you have multiple MegaRAID SAS controllers in your system (for example, RAID5/6 SAS based on LSI MegaRAID for internal hard disk connection, and MegaRAID SAS 8344ELP for external hard disk connection) a POST screen appears for each MegaRAID SAS during POST. The following message to start WebBIOS appears as many times as the number of MegaRAID SAS installed in your system. Press <Ctrl><H> for WebBIOS. The WebBIOS you are using is that of RAID 5/6 SAS based on LSI MegaRAID. And even just one of the following two actions will display only MegaRAID SAS 8344ELP, not RAID 5/6 SAS based on LSI MegaRAID, on the [Adapter Selection] window appeared right after the WebBIOS startup. When pressing the [Ctrl] + [H] keys twice or more on the POST screen. When pressing the [Ctrl] + [H] keys once on the POST screen of RAID 5/6 SAS based on LSI MegaRAID, then pressing the [Ctrl] + [H] keys on the POST screen of MegaRAID SAS 8344ELP that is optionally installed. 45

Chapter 1 Overview If you keep accessing to MegaRAID SAS 8344ELP when you want to access WebBIOS of RAID 5/6 SAS based on LSI MegaRAID and configure an array, press the [Alt] + [Ctrl] + [Del] keys to restart the system. Then, while the POST screen of RAID 5/6 SAS based on LSI MegaRAID is displayed, press the [Ctrl] + [H] keys only once to start WebBIOS. Here are some examples of actual WebBIOS windows. When both RAID 5/6 SAS based on LSI MegaRAID and MegaRAID SAS 8344ELP which is optionally installed appear on the same screen Select the array controller that you want to operate and click [Start]. Then you can operate any array controllers for such functions as array configuration. When only MegaRAID SAS 8344ELP which is optionally installed appears, but not RAID 5/6 SAS based on LSI MegaRAID With MegaRAID SAS 8344ELP, clicking [Start] will allow you such operation as configuring an array. To operate RAID 5/6 SAS based on LSI MegaRAID, restart the system since you cannot operate it at this point. 1.5.6 Hard Disk Failure LED during Operation When you use RAID 5/6 SAS based on LSI MegaRAID, failure LEDs of some hard disks may occasionally be lit for a few seconds from when the monitor is turned off during the OS restart process till when a POST screen appears. If the LEDs go off after the POST screen appearance, there is no problem. If not, even with a successful POST operation, the hard disks may have failed. Check the status of the array with a management tool. 46

Chapter 2 Array Configuration and Management [WebBIOS] This chapter explains the WebBIOS setup procedures. WebBIOS is a basic utility to set up and manage array controllers. 2.1 Starting and Exiting WebBIOS..................... 48 2.2 Checking Each Status [WebBIOS].................. 53 2.3 Creating/Deleting Disk Array Configuration........... 66 2.4 Setting/Releasing Spare Disk..................... 87 2.5 Operations on Logical Drive....................... 89 2.6 Rebuilding Hard Disk............................ 97 2.7 Reusing Hard Disk............................. 98 47

Chapter 2 Array Configuration and Management [WebBIOS] 2.1 Starting and Exiting WebBIOS This section explains how to start up and exit WebBIOS. WebBIOS can be set to start from the BIOS at system startup, regardless of whether the OS has been installed or not on the computer to be used. To use WebBIOS, a mouse needs to be connected. Before starting WebBIOS, make sure that the mouse is connected to the server. If both this array controller and the SCSI array controller (MegaRAID SCSI) are installed in the system, the connected USB mouse may not work when WebBIOS for the SCSI array controller starts up. In this case, start up WebBIOS for this array controller to operate the SCSI array controller. While the POST screen of this array controller appears, press the [Ctrl] + [H] keys to start up WebBIOS. 2.1.1 Starting Up WebBIOS 1 Turn on the server, and press the [Ctrl] + [H] keys while the following messages are displayed on the screen. Press the [Ctrl] + [H] keys while the message "Press <Ctrl> <H> for WebBIOS" is displayed. When two or more of this array controller are installed in your server, only press the [Ctrl] + [H] keys once while a message of the array controller to operate WebBIOS is displayed. The following message appears, and WebBIOS starts up after the system BIOS operations are complete. WebBIOS will be executed after POST completes 48

MegaRAID SAS User's Guide When WebBIOS starts, the [Adapter Selection] window of WebBIOS appears. If other array cards are also installed, two or more array controllers are displayed. When the system has multiple array controllers, the array controller that you want to operate may not appear on the [Adapter Selection] window. In this case, follow the procedure at "1.5.5 Notes on WebBIOS Startup when Installing Multiple MegaRAID SAS Controllers" (pg.45). For this array controller, one of the following is displayed in [Type] in the window shown above. MegaRAID SAS PCI Express(TM) ROMB MegaRAID SAS 8300XLP MegaRAID SAS 8344ELP MegaRAID SAS 8408E MegaRAID SAS 8880EM2 RAID 5/6 SAS based on LSI MegaRAID 2 Select the array controller to be operated, and click [Start]. The main window of WebBIOS will appear. 49

Chapter 2 Array Configuration and Management [WebBIOS] The following screen appears when an abnormal hard disk is detected. Click [Cancel] and continue. Check the status of the hard disk and replace it if failed by referring to "Chapter 6 Replacing Hard Disk" (pg.249). WebBIOS main window Do not click [GuidPreview] or [ClearForeighCfg]. The main window of WebBIOS consists of three areas. Physical Drives view Main menu Virtual Drives view / Configured Drives view 50

MegaRAID SAS User's Guide Main menu WebBIOS has the following menus. Menu name Adapter Properties Scan Devices Virtual Disks Physical Drives Configuration Wizard Adapter Selection Physical View Logical View Events Exit table: WebBIOS main menu Description Allows you to view/change the properties settings of the array controller. Redetects the hard disks connected. Displays a list of the logical drives. Displays a list of the hard disks. You can also check the status of connected hard disks. "2.2.1 Checking Hard Disk Status" (pg.53) Creates and adds/deletes a disk array configuration. Switches array controllers. Displays the [Configured Drives] view, where the disk group can be browsed and operated. The menu name switches to Logical View. Displays the [Virtual Drives] view, where the logical drive can be browsed and operated. The menu name switches to Physical View. Displays the event log on the array controller. This function is not supported. Do not use it. Exits WebBIOS. You can switch between the [Virtual Drives] and [Configured Drives] views by selecting [Logical View] or [Physical View] in the main menu. [Physical Drives] view The [Physical Drives] view displays a list of the hard disks. The status and the capacity of each hard disk are also displayed. For details about the status of hard disks, refer to "2.2.1 Checking Hard Disk Status" (pg.53). [Virtual Drives] view The [Virtual Drives] view displays the relationship between disk groups and logical drives. The RAID level, capacity and status of logical drives are also displayed. For details about the logical drive status, refer to "2.2.2 Checking Logical Drive Status" (pg.57). [Configured Drives] view Selecting [Physical View] in the main menu switches the [Virtual Drives] view to the [Configured Drives] view. The [Configured Drives] view allows you to check the relationship between disk groups and the hard disks that make up the disk groups. The status and the capacity of each hard disk are also displayed. 51

Chapter 2 Array Configuration and Management [WebBIOS] 2.1.2 Exiting WebBIOS 1 Click [Exit] in the main menu. 2 When the message "Exit Application" appears, select [Yes]. WebBIOS exits. 3 When the message "Please Reboot your System" appears, turn off the server or press the [Ctrl] + [Alt] + [Delete] keys to restart the server. 52

MegaRAID SAS User's Guide 2.2 Checking Each Status [WebBIOS] This section explains how to check each status, such as the hard disk status, the logical drive status, and the background task status. Checking Hard Disk Status (pg.53) Checking Logical Drive Status (pg.57) Checking Progress of Background Task (pg.61) Confirming Array Controller Information (pg.63) 2.2.1 Checking Hard Disk Status To check the hard disk status, follow the procedure below. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) In the [Physical Drives] view, a list of connected hard disks and their status are displayed. For MegaRAID SAS 8880EM2 a b c d e f g table: Hard disk information Meaning of item Example a The number of the enclosure and the slot where a hard disk is connected. 17:0 b Hard disk interface form. SAS c Disk group number to which the hard disk belongs. It only appears when the hard disk is included in a disk group. DG0 d Hard disk status. ONLINE e Hard disk capacity. 285568 MB f Hard disk vendor name, or interface form. SEAGATE g Hard disk model name. ST3300655SS 53

Chapter 2 Array Configuration and Management [WebBIOS] For MegaRAID SAS 300-8E ROMB / MegaRAID SAS 8300XLP / MegaRAID SAS 8344ELP / MegaRAID SAS 8408E / RAID 5/6 SAS based on LSI MegaRAID a b c d e f a b Hard disk status (icons) table: Hard disk information Meaning of item Hard disk ID. This is a number to specify each hard disk, not to indicate the actual storage bay number. To see slot numbers and addresses, refer to the detailed information in Step 2. The disk group number to which the hard disk belongs. It only appears when the hard disk is included in a disk group. Hard disk icons The hard disk status is indicated with an icon, as follows. PD0 DG0 Example c Hard disk status. ONLINE d Hard disk capacity. 139488 MB e Hard disk vendor name, or interface form. For SAS hard disks: FUJITSU For SATA hard disks: ATA f Hard disk model name. MAX3147RC table: Indications for each hard disk status Color Message Status Description Green ONLINE Online An array is configured, and the hard disk functions normally. Blue UNCONF GOOD Unused The hard disk is not used and functions normally. Red OFFLINE Offline The hard disk is read/write-protected. Red FAILED Failed The hard disk has a failure. Black UNCONF BAD Failed An array is not configured, and the hard disk has a failure. Brown REBUILD Rebuilding Rebuild is in progress. Pink HOTSPARE or GL HOTSPARE Spare disk The hard disk is set as a spare disk. For hard disks displayed in yellow, PFA (S.M.A.R.T.) is detected. Since hard disks with PFA (S.M.A.R.T.) detected may fail in the near future, they should be replaced immediately. For how to replace hard disks, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 54

MegaRAID SAS User's Guide Depending on the failure condition of the hard disk, a "PD Missing from DGx:Slot y" entry may be added and displayed. This entry means that the hard disk with ID y in the disk group x is in failure condition. Refer to "Chapter 6 Replacing Hard Disk" (pg.249) to replace the hard disk with the displayed ID and perform a rebuild. 2 Click the icon of a hard disk that you want to check detailed information or that you want to operate. The status and the settings of the selected hard disk are displayed. Detailed information Functions Detailed information table: The meaning of the detailed information for each item Item Description Revision Displays the firmware version. Enclosure ID Displays the connected enclosure number. Slot Number Displays the installed slot number. Device Type Displays the device type. Connected Port Displays the connected port number. Media Errors Not supported. Displays the number of media errors. Even if errors are counted, there is no problem because the errors are restored by the firmware. Pred Fail Count If PFA (S.M.A.R.T.) is detected, other than zero is displayed. Hard disks, where PFA (S.M.A.R.T.) is detected may fail in the near future, and should be replaced immediately. For the replacement procedure, refer to "Chapter 6 Replacing Hard Disk" (pg.249). SAS Address Displays the SAS address of the hard disks. Physical Drive State Displays the current status of the hard disks. For details about the status of a hard disk, refer to " Hard disk icons" (pg.54). Coerced Size Displays the hard disk capacity recognized by the array controller. 55

Chapter 2 Array Configuration and Management [WebBIOS] Functions This window allows you to execute functions on the hard disks. The executable functions may vary according to the hard disk status. table: Functions in the function executing area Item Description Executable status Locate Makes the failure LED of the hard disk blink for ten All seconds to tell the location. Make Global HSP Sets a spare disk. "2.4.1 Setting Spare Disk" (pg.87) UNCONF GOOD Make Dedicated HSP Make Unconf Bad Prepare for Removal Undo Prepare for Removal Replace Missing PD Make Drive Offline Make Online Rebuild Drive Mark as Missing Make Unconf Good Remove HOTSPARE Creates a special spare disk. Disabled. Do not use this. Forcibly puts the hard disk into UNCONF BAD status. Disabled. Do not use this. Stops the motor rotation. Disabled. Do not use this. Restarts the stopped motor. Disabled. Do not use this. Allocates a new hard disk instead of the failed disk. Do not execute this function without instructions from our engineer. The data may be lost. Forcibly puts the hard disk into OFFLINE status. Do not execute this function without instructions from our engineer, as the data may be lost. Forcibly puts the hard disk into ONLINE status. Do not execute this function without instructions from our engineer. The data may be lost. Starts a rebuild. "2.6 Rebuilding Hard Disk" (pg.97) Forcibly releases the target hard disk from the disk array configuration. Do not execute this function without instructions from our engineer. The data may be lost. Forcibly puts the hard disk into UNCONF GOOD status. Disabled. Do not use this. Releases a spare disk. "2.4.2 Releasing Spare Disk" (pg.88) UNCONF GOOD UNCONF GOOD UNCONF GOOD UNCONF GOOD UNCONF GOOD ONLINE OFFLINE OFFLINE OFFLINE UNCONF BAD HOTSPARE 3 When the check or operation is finished, click [HOME] or to return to the main window. 56

MegaRAID SAS User's Guide 2.2.2 Checking Logical Drive Status To check the logical drive status, follow the procedure below. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) In the [Virtual Drives] view, the logical drive status appears. If the [Configured Drives] view is displayed, select [Logical View] in the main menu to display the [Virtual Drives] view. a b c d Status of the logical drive (icon) table: Logical drive information Meaning of item Example a Logical drive number. VD0 b RAID level of the logical drive. RAID1 c Logical drive capacity. 139488 MB d Logical drive status. Optimal When a logical drive is configured for RAID 10, as shown below, the same logical drive number (VD y) is allocated to a group of disk groups with successive numbers (DG x). 57

Chapter 2 Array Configuration and Management [WebBIOS] Logical drive icon The logical drive icon appears in the following color according to the logical drive status. table: Color of each logical drive status Color Display Status Description Green Optimal Online Normal Yellow Partially Degraded Partially critical Operating with redundancy. But a hard disk has been failed. Blue Degraded Critical Operation without redundancy Red Offline Offline Unavailable 2 Click the logical drive which you want to check details or operate. The status and settings of the selected logical drive are displayed. For MegaRAID SAS 8880EM2 / RAID 5/6 SAS based on LSI MegaRAID Properties area Policies area Operations area Expand Capacity setting area 58

MegaRAID SAS User's Guide For MegaRAID SAS 300-8E ROMB / MegaRAID SAS 8300XLP / MegaRAID SAS 8408E / MegaRAID SAS 8344ELP Properties area Policies area Operations area Expand Capacity setting area Properties Item RAID Level State Size Stripe Size table: Properties information Displays RAID level of the logical drive. Description Displays logical drive status. " Logical drive icon" (pg.58) Displays the total capacity of the logical drive. Displays the configurable stripe sizes of the logical drive. 59

Chapter 2 Array Configuration and Management [WebBIOS] Policies Item Access Read Disk Cache Write Disable BGI I/O table: Policies information Description Sets the I/O access. Do not change this. Sets the read policy. Do not change this. Sets the write cache of the hard disk. Do not change this. Displays and sets Write Policy. For RAID 5/6 SAS based on LSI MegaRAID, WTru (Write Through) or WBack (Write Back) is displayed. And below that is "Use wrthru for failure or missing battery" where you can change or display the settings of Write Policy. With a check at "Use wrthru for failure or missing battery", the Write Policy will be set to WTru (Write Through) when a battery is not installed or failed. Without a check, the Write Policy will be BadBBU that will operate a controller with WBack (Write Back) even when a battery is not installed or failed. It displays WThru (Write Through), WBack (Write Back), or BadBBU for a controller other than RAID 5/6 SAS based on LSI MegaRAID. Click [Change] after changing the settings to reflect the new settings. "1.3.1 Write Policy" (pg.28). Sets BGI. Do not change this. Sets DMA transmission mode. Do not change this. If a background task is in operation, the [Policies] items are only displayed, and cannot be modified. Operations The following operations can be executed on the logical drives. To execute, select the target function, and click [Go]. Del Locate Item Fast or FastInit Slow or SlowInit CC table: Information about the operations area Description Deletes the logical drive. "2.5.5 Deleting Logical Drive" (pg.95). Makes the failure LED of the hard disk installed in the logical drive blink for ten seconds to indicate the location. Performs fast initialization for a logical drive. "2.5.2 Fast Initialization of Logical Drive" (pg.90). Performs foreground initialization for a logical drive. "2.5.1 Initializing Logical Drive" (pg.89). Performs Make Data Consistent for a logical drive. "2.5.3 Make Data Consistent of Logical Drive" (pg.91). Expand Capacity setting The capacity of a logical drive can be expanded. For details, refer to "2.5.4 Expand Capacity of Logical Drive" (pg.92). 3 When the operation is finished, click [HOME] or to return to the main window. 60

MegaRAID SAS User's Guide 2.2.3 Checking Progress of Background Task To display the status of a task currently running in the background, follow the procedure below. You can check the progress of various kinds of initialization, Make Data Consistent, rebuild, and Expand Capacity for logical drives. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Click the button for the task in progress displayed at the bottom of the main window. The buttons to be displayed and background tasks displayed by pressing buttons are as follows: [Rebuild Progress] Rebuild [Initialization Progress] Foreground initialization [Background Initialization Progress] Background initialization [Check Consistency Progress] Make Data Consistent [Reconstruction Progress] Expand Capacity [VD Progress Info] All of Foreground initialization, Background initialization, Make Data Consistency, and Expand Capacity [PD Progress Info] Rebuild If the button for a task which is supposed to be running in the background is not shown, click [Scan Devices] in the main menu. 61

Chapter 2 Array Configuration and Management [WebBIOS] The names and status of running tasks are displayed in the left area of the window. If you select [Abort] for a target background task and click [Go] in the lower right of the window, that background task will be aborted. Normally, do not abort. Although the options [Fast Initialize], [Check Consistency], [Properties], and [Set Boot Drive] are displayed in the right area of the window, these functions are not supported in this window. Do not use them. 3 Check the information, then click [HOME] or to return to the main window. 62

MegaRAID SAS User's Guide 2.2.4 Confirming Array Controller Information This window displays array controller information. The settings can be changed. Before creating an array configuration, make sure to check the array controller settings. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Click [Adapter Properties] from the main menu. The [Adapter Information] window appears. Item Firmware Version WebBIOS Version SubVendorID SubDeviceID HostInterface PortCount NVRAM Size Memory Size Firmware Time Serial Number Min Stripe Size Max Stripe Size Virtual Disk Count Physical Disk Count FW Package Version table: Array controller information Description Displays the firmware version of the array controller. Displays the WebBIOS version. Displays the Sub Vendor ID. Displays the Sub Device ID. Displays the interface type between the server and the array controller. Displays the number of SAS ports. Displays the NVRAM size. Displays the cache memory size. Displays the time stamp of the firmware. Displays array controller serial number. However, the value displayed here is invalid for the onboard array controllers. Displays the minimum configurable stripe size of logical drives. Displays the maximum configurable stripe size of logical drives. Displays the number of created logical drives. Displays the number of hard disk units connected. Displays the total version that includes the array controller's firmware and BIOS, etc. 63

Chapter 2 Array Configuration and Management [WebBIOS] 3 Click [Next]. The [Adapter Properties] window appears. To change properties To change the property settings of the array controller, edit or select the item to change, and then click [Submit] to confirm the change. Before changing the properties, check the changeable and fixed items in " Array controller settings" (pg.64). 4 Check the description, and click [HOME] or to return to the main window. Array controller settings The initial value for each property item of the array controller is as follows. table: Array controller property initial values Item Value Description Battery Backup Present or None Displays whether a battery is installed or not. "Present" (detected) or "None" (not detected) will appear depending on whether a battery is detected or not. Set Factory Defaults No Selecting "Yes" and saving the settings allows you to reset each setting of the array controller to the initial value. Cluster Mode Disabled (Do not change) This is an optional setting for supporting multi-initiator environments. This item is set to "Disabled" and must not be changed. Rebuild Rate [Note 1] Sets the priority when performing a rebuild. The higher this value, the higher the priority of rebuild over I/Os from the server. BGI Rate [Note 1] Sets the priority when running background initialization. The higher this value, the higher the priority of background initialization over I/Os from the server. CC Rate [Note 1] Sets the priority when performing Make Data Consistent. The higher this value, the higher the priority of Make Data Consistent over I/Os from the server. Reconstruction Rate [Note 1] Sets the priority when expanding capacity. The higher this value, the higher the priority of Expand Capacity over I/Os from the server. 64

MegaRAID SAS User's Guide Adapter BIOS Coercion Mode Enabled (Do not change) None (Do not change) PDF Interval 300 (Do not change) Alarm Control Disabled (Do not change) Enables or disables the array controller BIOS. This item is set to "Enabled" and must not be changed. Sets a capacity rounding function to make varying hard disk capacities equal to each other. This item is set to "None" and must not be changed. Sets the PFA (=S.M.A.R.T.) tracking interval for the hard disk. This item is set to "300" and must not be changed. If a buzzer for notifying errors is installed in the array controller, this enables or disables the buzzer. This item is set to "Disabled" and must not be changed. Patrol Read Rate 20 Sets the priority when running Patrol Read. The higher this value, the higher priority the Patrol Read has over I/Os from the server. This setting is not supported. Cache Flush Interval 4 (Do not change) Spinup Drive Count 2 (Do not change) Spinup Delay 6 (Do not change) StopOnError Disabled (Do not change) table: Array controller property initial values Item Value Description Sets the data write interval from the cache memory to the hard disk when running in the Write Back mode. This item is set to "4" and must not be changed. Sets the number of hard disk units that start rotating at the same time. This item is set to "2" and must not be changed. Sets the interval to the next start after starting the hard disk rotation for the first time. This item is set to "6" and must not be changed. During POST, sets whether to stop the POST as soon as a specific message appears. This item is set to "Disabled" and must not be changed. [Note 1]: 80 for: MegaRAID SAS 300-8E ROMB, MegaRAID SAS 8300XLP, MegaRAID SAS 8408E, and MegaRAID SAS 8344ELP 30 for: MegaRAID SAS 8880EM2, RAID 5/6 SAS based on LSI MegaRAID 65

Chapter 2 Array Configuration and Management [WebBIOS] 2.3 Creating/Deleting Disk Array Configuration This section explains how to configure a new disk array. 2.3.1 Creating Disk Array Configuration Overview of array configuration procedure Follow the procedure below to configure disk arrays. Before starting the disk array configuration, refer to "2.2.4 Confirming Array Controller Information" (pg.63) and check the array controller properties. Select whether to create a new disk array configuration or to add a logical drive to the current disk array configuration. Create a disk group. Only for MegaRAID SAS 8880EM2 and RAID 5/6 SAS based on LSI MegaRAID, create a disk group (span) to be included in a logical drive. Create the logical drive. Select a RAID level Set Write Policy Set capacity of the logical drive Initialize the logical drive. COMPLETE 66

MegaRAID SAS User's Guide How to create disk array configuration The procedure to create a disk array configuration is different when you use MegaRAID SAS 8880EM2 or RAID 5/6 SAS based on LSI MegaRAID and when you use other types of this array controller. Refer to the procedure of the array controller you are using. When using MegaRAID SAS 8880EM2 or RAID 5/6 SAS based on LSI MegaRAID " For MegaRAID SAS 8880EM2 / RAID 5/6 SAS based on LSI MegaRAID" (pg.67) When using MegaRAID SAS 300-8E ROMB, MegaRAID SAS 8300XLP, MegaRAID SAS 8408E, or MegaRAID SAS 8344ELP " For MegaRAID SAS 300-8E ROMB / MegaRAID SAS 8300XLP / MegaRAID SAS 8408E / MegaRAID SAS 8344ELP" (pg.78) For MegaRAID SAS 8880EM2 / RAID 5/6 SAS based on LSI MegaRAID When the disk array is configured with RAID 10, note that two or more logical drives cannot be defined in a RAID 10 disk group. Follow the procedure below to configure a disk array. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Click [Configuration Wizard] on the main menu. The [Configuration Wizard] window appears. 3 Select [New Configuration] to delete all the current disk array configuration and create a new disk array configuration, or [Add Configuration] to add a logical drive to the current disk array configuration. Then click [Next]. 67

Chapter 2 Array Configuration and Management [WebBIOS] When you select [New Configuration] when a disk array configuration already exists, a warning message appears telling that the existing configuration will be all deleted. If you want to delete the current configuration, click [Yes] to proceed. To add a new logical drive while keeping the existing logical drive, click [No] to select [Add Configuration]. The following window appears. [Auto Configuration: With Redundancy (Recommended)] and [Auto Configuration: Without Redundancy] are not supported. Do not select them. 4 Select [Custom Configuration], and click [Next]. The [DG Definition] window appears. 5 Create a disk group. The procedure to create a disk group is different depending on the RAID level. Do not configure the disk group of RAID 10 and other RAID levels at the same time when RAID 10 exists with the other RAID levels (RAID 0 / RAID 1 / RAID 5). After configuring a disk group, add another disk group by selecting [Configuration Wizard] [Add Configuration] (pg.67) again. 68

MegaRAID SAS User's Guide Configuring RAID 0 / RAID 1 / RAID 5 / RAID 6 Follow the procedure below. 1. While pressing the [Ctrl] key, select all the hard disks to be added to a disk group from the [Physical Drives] area. Selectable hard disks are indicated as "UNCONF GOOD" in blue. table: Number of hard disk units needed for each RAID level RAID level Necessary number of hard disk units RAID 0 1 or more RAID 1 2 RAID 5 3 or more RAID 6 3 or more 2. Click [AddtoArray]. The hard disks to be included in a disk group are confirmed and added to the [Disk Groups] area. 69

Chapter 2 Array Configuration and Management [WebBIOS] 3. Click [Accept DG]. The disk group added to [Disk Groups] is confirmed, and the disk capacity in the disk group is displayed. 4. Repeat the steps above to set as many disk groups as needed. Add as many hard disks as needed to configure an array. Hard disks in a single disk group should be of the same model (with the same capacity and rotation speed). Hard disks that are already included in a disk group are indicated as "ONLINE" in green. If an incorrect disk group is set by mistake, click [Back], and start again from Step 3 (pg.67). 5. Click [Next]. The [Span Definition] window appears. 70

MegaRAID SAS User's Guide 6. Check that a disk group has been selected in the [Array With Free Space] area, and click [Add to SPAN]. Now the disk group is confirmed and added to the [Span] area. 7. Click [Next]. The [VD Definition] window appears. Configuring RAID 10 Follow the procedure below. 1. While pressing the [Ctrl] key, select two hard disks to be set to a disk group from the [Physical Drives] area. The selected two hard disks will be mirrored. Selectable hard disks are indicated as "UNCONF GOOD" in blue. 71

Chapter 2 Array Configuration and Management [WebBIOS] 2. Click [AddtoArray]. The hard disks to be included in a disk group are confirmed and added to the [Disk Groups] area. 3. Click [Accept DG]. The disk group added to [Disk Groups] is confirmed, and the disk capacity in the disk group is displayed. 72

MegaRAID SAS User's Guide 4. Repeat the steps above until you arrange all the hard disks to configure RAID 10 into multiple disk groups. Add as many hard disks as needed to configure an array. To configure RAID 10, you need to create multiple disk groups here. Here is an example four hard disks create two disk groups. Hard disks in a single disk group should be of the same model (with the same capacity and rotation speed). Hard disks that are already included in a disk group indicated as "ONLINE" in green. If an incorrect disk group is set by mistake, click [Back], and start again from Step 3 (pg.67). 5. Click [Next]. The [Span Definition] window appears. 73

Chapter 2 Array Configuration and Management [WebBIOS] 6. Check that a disk group has been selected in the [Array With Free Space] area, and click [Add to SPAN]. Now the disk group is confirmed and added to the [Span] area. Repeat the steps, confirming a disk group and adding it to the [Span] area, until all the disk groups in the [Array With Free Space] area are moved to the [Span] area. To configure RAID 10, you need to create disk groups in advance. 7. Click [Next]. The [VD Definition] window appears. 74

MegaRAID SAS User's Guide 6 Create a logical drive. Enter settings for a logical drive. The setting items and their default values for a logical drive are as follows. table: Logical drive settings Item Default value Remarks RAID Level Depends on the number of hard disk units in a disk group Sets a RAID level. Stripe Size 64 KB Do not change. Access Policy RW Do not change. Read Policy Normal Do not change. Write Policy WThru Set Write Policy. Select WThru (Write Through) or WBack (Write Back). For the OS installation, make sure to set WThru. You can change the Write Policy after the OS installation. Wrthru for BAD BBU Checked With a check, when there is no battery or at a time of a failure, Write Policy is changed from Write Back to Write Through. Without a check, it remains BadBBU which works as Write Back even when there is no battery or at a time of a failure. IO Policy Direct Do not change. Disk Cache Disable Do not change. Policy Disable BGI No Do not change. Select Size The maximum capacity of the logical drive Sets the capacity of the logical drive in MB. Only one RAID level can be configured in a disk group. Before changing the Write Policy setting, make sure to refer to "1.3.1 Write Policy" (pg.28). 75

Chapter 2 Array Configuration and Management [WebBIOS] Configuring RAID 0 / RAID 1 / RAID 5 / RAID 6 The default [Select Size] value is the maximum capacity of a RAID level that was selected when moved to the [VD Definition] window. When you want to manually change a RAID level or create it by dividing logical drives in a disk group, select a RAID level and enter the appropriate number for its capacity. You can refer to [Next LD, Possible RAID Levels] on the right side of the window for "Possible RAID level: maximum capacity". For example, as in the figure below, when the number shows "R0:277888 R5:208416 R6:138944", the maximum configurable RAID capacity is 277,888 MB, 208,416 MB, and 138,944 MB for RAID 0, RAID 5, RAID 6 respectively. Entering a number smaller than the maximum value for a logical drive capacity will allow you to create multiple logical drives. Configuring RAID 10 Set [RAID Level] to "RAID 10". Only one logical drive with the maximum capacity of the disk group can be created. Do not change the value of [Select Size]. 7 Click [Accept]. The logical drive is confirmed and added to the area in the right side of the window. When there is available logical drive capacity left in a disk group, click [Back] to show the [Span Definition] window. Then create as many logical drives as needed until using up all the available areas. 8 Click [Next]. The [Preview] window appears. The [Preview] window example (for RAID 0 / RAID 1 / RAID 5 / RAID 6) 76

MegaRAID SAS User's Guide The [Preview] window example (for RAID 10) If you create an incorrect logical drive by mistake, click [Back], and start again from the [DG Definition] window. When RAID 10 already exists, the same logical drive number appears straddling disk groups in the logical drive. 9 Click [Accept]. The message "Save this Configuration?" appears. 10 Click [Yes]. All the settings up to here are saved, and the [Confirm Page] window appears. 11 To perform fast initialization of the logical drive here, click [Yes]. To perform other initialization or fast initialization later, click [No]. Clicking [Yes] at this point will start fast initialization. After the fast initialization which finishes soon, background initialization will automatically start. To perform foreground initialization, click [No] to return to the main window, and then perform foreground initialization by following "2.5.1 Initializing Logical Drive" (pg.89). 77

Chapter 2 Array Configuration and Management [WebBIOS] For MegaRAID SAS 300-8E ROMB / MegaRAID SAS 8300XLP / MegaRAID SAS 8408E / MegaRAID SAS 8344ELP When RAID 10 configures a disk array, note that two or more logical drives cannot be defined in the RAID 10 disk group. Follow the procedure below to configure a disk array. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Click [Configuration Wizard] on the main menu. The [Configuration Wizard] window appears. 3 Select [New Configuration] to delete all the current disk array configurations and create a new disk array configuration, or [Add Configuration] to add a logical drive to the current disk array configuration. Then click [Next]. When you select [New Configuration] when a disk array configuration already exists, a warning message appears telling that the existing configuration will be all deleted. If you want to delete the current configuration, click [Yes] to proceed. To add a new logical drive while keeping the existing logical drive, click [No] to select [Add Configuration]. 78

MegaRAID SAS User's Guide The following window appears. [Auto Configuration: With Redundancy (Recommended)] and [Auto Configuration: Without Redundancy] are not supported. Do not select them. 4 Select [Custom Configuration], and click [Next]. The [DG Definition] window appears. 79

Chapter 2 Array Configuration and Management [WebBIOS] 5 Create a disk group. Do not configure the disk group of RAID 10 and other RAID levels at the same time when RAID 10 exists with the other RAID levels (RAID 0 / RAID 1 / RAID 5). After configuring a disk group, add another disk group by selecting [Configuration Wizard] [Add Configuration] (pg.67) again. Configuring RAID 0 / RAID 1 / RAID 5 Follow the procedure below. 1. While pressing the [Ctrl] key, select all the hard disks to be added to a disk group from the [Physical Drives] area. Selectable hard disks are indicated as "UNCONF GOOD" in blue. table: Number of hard disk units needed for each RAID level RAID level Necessary number of hard disk units RAID 0 1 or more RAID 1 2 RAID 5 3 or more 2. Click [Accept DG]. The disk group is confirmed and displayed in the [Disk Groups] area. 3. Repeat the steps above to set as many disk groups as needed. Add as many hard disks as needed to configure an array. Hard disks in a single disk group should be of the same model (with the same capacity and rotation speed). Hard disks that are already included in a disk group are indicated as "ONLINE" in green. If an incorrect disk group is set by mistake, click [Back], and start again from Step 3 (pg.78). Configuring RAID 10 Follow the procedure below. 1. While pressing the [Ctrl] key, select two hard disks to be set to a disk group from the [Physical Drives] area. The selected two hard disks will configure a mirrored array. Selectable hard disks are indicated as "UNCONF GOOD" in blue. 2. Click [Accept DG]. The disk group is confirmed and displayed in the [Disk Groups] area. 3. Repeat the steps above to set all the hard disks which configure RAID 10 to the disk groups. Hard disks in a single disk group should be of the same model (with the same capacity and rotation speed). 80

MegaRAID SAS User's Guide Hard disks that are already included in a disk group indicated as "ONLINE" in green. If an incorrect disk group is set by mistake, click [Back], and start again from Step 3 (pg.78). 6 Click [Next]. The [VD Definition] window appears. 7 Create a logical drive. 1. Select a disk group to create a logical drive from the right area. To create a logical drive, start from the disk group with the smallest number. For example, if there are disk groups named "DG 0" and "DG 1", start creating a logical drive from the disk group "DG 0". Do not create a logical drive for a new disk group while free space remains in the old disk group. Wait until no free space remains in the old disk group to create a logical drive for a new disk group. 81

Chapter 2 Array Configuration and Management [WebBIOS] To configure RAID 10, select all the disk groups created in Step 5 while pressing the [Ctrl] key. 2. Enter settings for a logical drive. The setting items and their default values for a logical drive are as follows. table: Logical drive settings Item Default value Remarks RAID Level RAID 0 Sets a RAID level. Stripe Size 64KB Do not change. Access Policy RW Do not change. Read Policy Normal Do not change. Write Policy WThru Sets WThru (Write Through), WBack (Write Back), or BadBBU. For the OS installation, make sure to set WThru. You can change the Write Policy after the OS installation. IO Policy Direct Do not change. Disk Cache Policy Disable Do not change. Disable BGI No Do not change. Select Size (Blank) Sets the capacity of a logical drive in MB. 82

MegaRAID SAS User's Guide Configuring RAID 0 / RAID 1 / RAID 5 Only one RAID level can be configured in a disk group. Before changing the default Write Policy setting, make sure to refer to "1.3.1 Write Policy" (pg.28). Logical drive capacity is displayed in the "RAID level = maximum value" format on a disk group in the right area. Do not set a larger value than the maximum value. For example, as in the figure below, when "DG 0:R0=138944MB, R1=69472MB" is displayed, the maximum configurable RAID capacity for disk group 0 is 138,944 MB for RAID 0, and 69,472 MB for RAID 1. Entering a number smaller than the maximum value for a logical drive capacity will allow you to create multiple logical drives. Configuring RAID 10 Set [RAID Level] to "RAID 1". Only one logical drive with the maximum capacity of disk groups can be created. In [Select Size], enter the total capacity of the selected disk groups. For instance, when "DG 0:R0=138944MB, R1=69472MB" is displayed on a disk group in the right area, capacity of one disk group is 69,472 MB. When RAID 10 is configured with three disk groups (six hard disks), the capacity to be entered in [Select Size] is 208,416 MB. 69,472 MB 3 = 208,416 MB Before changing the Write Policy setting, make sure to refer to "1.3.1 Write Policy" (pg.28). 8 Click [Accept]. The logical drive is added under the disk group in the [Configuration] area. 83

Chapter 2 Array Configuration and Management [WebBIOS] If there is free space left to set other logical drives, or if there is a disk group where no logical drive has been set, the [VD Definition] window appears again. Return to Step 7, and create logical drives until no free space remains in the disk group. When there is no more free space in the disk group, the [Preview] window appears. When the disk group settings are finished, the [Preview] window appears. The [Preview] window example (for combination of RAID 1 and RAID 5) The [Preview] window example (for RAID 10) If you create an incorrect logical drive by mistake, click [Back], and start again from the [DG Definition] window. If RAID 10 is configured, the first RAID 1 disk group is displayed as "RAID 10", and the other disk groups are displayed as "(Contd)". 84

MegaRAID SAS User's Guide 9 After checking that contents of the [Preview] window are correct, click [Accept]. The message "Save this Configuration?" appears. 10 Click [Yes]. All the settings up to here are saved, and the [Confirm Page] window appears. 11 To perform fast initialization of the logical drive here, click [Yes]. To perform other initialization or fast initialization later, click [No]. Clicking [Yes] at this point will start fast initialization. After a short while, background initialization will automatically be started. To perform foreground initialization, click [No] to return to the main window, and then perform foreground initialization by referring to "2.5.1 Initializing Logical Drive" (pg.89). When neither fast initialization nor foreground initialization is performed, background initialization will be performed. 12 Click [HOME] or to return to the main window. 2.3.2 Deleting Disk Array Configuration To delete the entire current disk array configuration, use the [Clear Configuration] function in [Configuration Wizard]. When the disk array configuration is deleted, the data on the hard disk can no longer be accessed. The configuration information for the disk array is completely deleted, and all hard disks are restored to the unused status. Normally, this function should not be used. If this function is used, the current settings on the array controller are deleted and all the data on the hard disks connected to the array controller is also deleted. If using this function, make a plan beforehand and proceed carefully. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) When multiple array controllers are installed, select a controller you want to delete disk array in the controller selection window. 85

Chapter 2 Array Configuration and Management [WebBIOS] 2 Select [Configuration Wizard] from the main menu. The [Configuration Wizard] window appears. 3 Select [Clear Configuration], and click [Next]. When the following warning message appears, click [Yes]. 4 If the [Configuration Preview] window appears, click [Accept]. When the message "Save this Configuration?" appears, click [Yes]. Allocation of all logical drives is deleted, and the status of all hard disks connected to the array controller becomes "Unused (UNCONF GOOD)". If a hard disk is in the "UNCONF BAD" state after clicking [Yes], it means that the hard disk is failed. That hard disk cannot be used. Refer to "Chapter 6 Replacing Hard Disk" (pg.249) to replace the failed hard disk. 5 Click [HOME] or to return to the main window. 86

MegaRAID SAS User's Guide 2.4 Setting/Releasing Spare Disk This section explains how to set a hard disk in the unused status as a spare disk and how to release an existing spare disk. 2.4.1 Setting Spare Disk To set a hard disk in the unused status (UNCONF GOOD) as a spare disk, follow the procedure below. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Select a hard disk in "Unused (UNCONF GOOD)" state to set as a spare disk from the [Physical Drives] view in the main window. 3 Select [Make Global HSP] at the bottom of the window, and click [Go]. 1. 2. Click here to select Click Make sure that the hard disk has been set as a spare disk and that [Physical Drive State] is indicated as "HOTSPARE" or "GL HOTSPARE". 4 Check the settings, and click [HOME] or to return to the main window. 87

Chapter 2 Array Configuration and Management [WebBIOS] 2.4.2 Releasing Spare Disk To restore a hard disk that has been set as a spare disk to the unused status (UNCONF GOOD), follow the procedure below. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Select the spare disk to restore to the unused status (UNCONF GOOD) from the [Physical Drives] view of the main window. 3 Select [Remove HOTSPARE] at the bottom of the window, and click [Go]. 1. Click here to select 2. Click Make sure that the spare disk has been released and that [Physical Drive State] is indicated as "UNCONF GOOD". 4 Check the settings and click [HOME] or to return to the main window. 88

MegaRAID SAS User's Guide 2.5 Operations on Logical Drive This section explains the following operations on logical drives: Initializing Logical Drive pg.89 Fast Initialization of Logical Drive pg.90 Make Data Consistent of Logical Drive pg.91 Expand Capacity of Logical Drive pg.92 2.5.1 Initializing Logical Drive To initialize a logical drive in the foreground, follow the procedure below. Note that initialization of a logical drive will delete all the data on the target logical drive. For an overview of the initialization functions, refer to "1.3.2 Logical Drive Initialization" (pg.30). This array controller supports background initialization. If you do not want to initialize the logical drive in the background, perform this function beforehand to initialize the logical drive. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Click the logical drive to initialize in the [Virtual Drives] view in the main window. 3 Select "Slow" or "SlowInit" in [Operations], and click [Go]. A confirmation window appears. 4 Click [Yes]. The initialization of the logical drive starts. To return to the previous window, click [No]. During initialization, you can check the progress by clicking [Initialization Progress] or [VD Progress Info] in the lower part of the window. For details, refer to "2.2.3 Checking Progress of Background Task" (pg.61). 5 When the initialization is complete, click [HOME] or to return to the main window. 89

Chapter 2 Array Configuration and Management [WebBIOS] 2.5.2 Fast Initialization of Logical Drive To fast initialize logical drives, follow the procedure below. For an overview of the fast initialization function, refer to " Fast initialization" (pg.32). When background initialization is in progress, fast initialization cannot be performed. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Click the logical drive to fast initialize in the [Virtual Drives] view in the main window. 3 Select "Fast" or "FastInit" in [Operations], and click [Go]. A confirmation window appears. 4 Click [Yes]. The fast initialization of the logical drive starts. To return to the previous window, click [No]. 5 When the initialization is complete, click [HOME] or to return to the main window. 90

MegaRAID SAS User's Guide 2.5.3 Make Data Consistent of Logical Drive To make data consistency of a logical drive, follow the procedure below. For an overview of the Make Data Consistent function, refer to "1.3.4 Make Data Consistent" (pg.35). 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Make sure that the [Virtual Drives] view appears. If the [Configured Drives] view appears, select [Logical View] in the main menu to switch to the [Virtual Drives] view. 3 Click the logical drive whose consistency is to be checked in the [Virtual Drives] view. 4 Select "CC" in [Operations], and click [Go]. Make Data Consistent for the logical drive starts. During Make Data Consistent, you to check the progress by clicking [Check Consistency Progress] or [VD Progress Info] in the lower part of the window. For details, refer to "2.2.3 Checking Progress of Background Task" (pg.61). 5 When the Make Data Consistent is complete, click [HOME] or to return to the main window. 91

Chapter 2 Array Configuration and Management [WebBIOS] 2.5.4 Expand Capacity of Logical Drive To expand the capacity and change the RAID level for an existing logical drive, follow the procedure below. There are certain conditions to note and precautions to take for. Read "1.3.5 Expand Capacity" (pg.36) carefully before expanding the capacity. Operations before Expand Capacity Before performing Expand Capacity, follow the procedure below to prepare. 1 Back up the data to prepare for unexpected problems. 2 Perform Make Data Consistent (pg.91) on the logical drive whose capacity is to be expanded, and check if it completes successfully. If Make Data Consistent is not complete successfully, do not expand the capacity of the logical drive. 3 Install the hard disk to be added to the server. Install it as a hard disk under this array controller. Procedure of Expand Capacity 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Make sure that the [Virtual Drives] view appears. If the [Configured Drives] view appears, select [Logical View] in the main menu to switch to the [Virtual Drives] view. 92

MegaRAID SAS User's Guide 3 Click the logical drive whose capacity is to be expanded in the [Virtual Drives] view. A information window of the logical drive appears. 4 Configure the settings for Expand Capacity. 1. Select an option for Expand Capacity. 2. Select a RAID level. 3. Select a hard disk to be added. 4. Review the settings above, and click here. 1. Select an option for Expand Capacity. To change the RAID level only, select [Migration only]. To expand the capacity by changing the RAID level and adding a hard disk, select [Migration with addition]. Do not select [Remove physical drive]. 93

Chapter 2 Array Configuration and Management [WebBIOS] 2. Select a RAID level to be set after Expand Capacity. 3. Select a hard disk to be added for Expand Capacity. If [Migration only] is selected for Expand Capacity option, skip this step. Multiple hard disks can be selected at a time by holding down the [Ctrl] key. Depending on the number of additional hard disk units and the RAID level after the Expand Capacity, only the RAID level may be changed without expanding the capacity. For example, when converting a RAID 0 logical drive into a RAID 5 logical drive by adding a hard disk unit, the capacity will not be expanded, because the added capacity of one hard disk unit is used for the RAID 5 parity data capacity. 5 Click [Go]. A confirmation window appears. 6 Click [Yes]. [Reconstruction Progress] or [VD Progress Info] appears, and the Expand Capacity process starts. Wait for a while until the process is finished. During Expand Capacity, you can check the progress by clicking [Reconstruction Progress] or [VD Progress Info] in the lower part of the window. For details, refer to "2.2.3 Checking Progress of Background Task" (pg.61). If the message "Unacceptable Reconstruction parameter" appears, the RAID level after Expand Capacity may not be appropriate. If the message "Failed to start operation on Logical Drive" appears, a background task may be under execution. Wait until the background task operations are finished before performing Expand Capacity. If the RAID level of the logical drive after Expand Capacity has data redundancy (that is RAID 1, RAID 5 or RAID 6), background initialization will be performed after Expand Capacity is complete. While Expand Capacity is in progress, do not turn off, reset or restart the server. Doing so may cause loss of data on the target logical drive. If the server is turned off during the Expand Capacity process, access to the hard disk is resumed automatically after the server restarts. In this case, wait until the hard disk access LED goes off, ensure that no access is made to the hard disk, create an array configuration again, and then restore the data backed up before the task. 7 When Expand Capacity is complete, click [HOME] or to return to the main window. 94

MegaRAID SAS User's Guide 2.5.5 Deleting Logical Drive When all the logical drives in a disk group are deleted, the hard disks in that disk group are restored to unused status. Delete logical drives in descending order of the logical drive IDs. 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Make sure that the [Virtual Drives] view appears. If the [Configured Drives] view appears, select [Logical View] in the main menu to switch to the [Virtual Drives] view. 3 Click the last logical drive (at the bottom) in the [Virtual Drives] view. 95

Chapter 2 Array Configuration and Management [WebBIOS] 4 Select "Del" in [Operations], and click [Go]. 1. Click here to select 2. Click A confirmation window appears. 5 Click [Yes]. The logical drive is deleted. 6 Click [HOME] or to return to the main window. When all the logical drives in a disk group are deleted, the hard disks in that disk group are restored to the "UNCONF GOOD" status. 96

MegaRAID SAS User's Guide 2.6 Rebuilding Hard Disk To perform a manual rebuild, follow the procedure below. Usually, a rebuild is performed automatically after replacing a hard disk. For details on the hard disk replacement and the automatic rebuild, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 1 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 2 Select the hard disk to rebuild (in "FAILED" or "OFFLINE" status) in the [Physical Drives] view of the main window. The detailed information window for the selected hard disk appears. Regarding the detailed information window for hard disks, refer to "2.2.1 Checking Hard Disk Status" (pg.53). 3 Select [Rebuild Drive] in the function area at the bottom of the window, and click [Go]. The hard disk status changes to "REBUILD" and the rebuild starts. Wait for a while until the rebuild is complete. The rebuild is complete when the progress bar for the rebuild reaches its end and the hard disk status changes to "ONLINE". If the server restarts or shuts down during rebuild, the rebuild is resumed from where it was stopped the next. 4 When the rebuild is complete, click [HOME] or to return to the main window. 97

Chapter 2 Array Configuration and Management [WebBIOS] 2.7 Reusing Hard Disk This section explains how to reuse a hard disk that has been connected to this array controller in another system. 2.7.1 Cautions when Reusing Hard Disk Before you can reuse a hard disk that has been connected to this array controller in another system, the following information must be deleted from the hard disk. Disk array configuration information OS partition table If a hard disk with this information remaining is used in another system, it may cause unexpected behavior in the system and problems such as loss of existing data. Delete the configuration information, etc. completely from the hard disk before reusing the hard disk. 2.7.2 Clearing Information from Hard Disk To reuse a hard disk, follow the procedure below to delete all the information from the hard disk. This procedure should be performed in the old system before connecting the hard disk to the new system. This procedure deletes the entire disk array configuration in the actual array controller. 1 If there is a disk array configuration on the hard disk, delete the disk array configuration by referring to "2.3.2 Deleting Disk Array Configuration" (pg.85). After the disk array configuration is deleted, all the hard disks are displayed as being in the "UNCONF GOOD" status. If there is a hard disk in the "UNCONF BAD" status, it means that that hard disk is failed and cannot be used. 2 Initialize each hard disk by creating RAID 0. The procedures are different between MegaRAID SAS 8880EM2 or RAID 5/6 SAS based on LSI MegaRAID and the other controllers. For MegaRAID SAS 8880EM2 / RAID 5/6 SAS based on LSI MegaRAID 1. Click [Configuration Wizard] on the main menu. The [Configuration Wizard] window appears. 2. Select [New Configuration], and click [Next]. 3. Select [Custom Configuration], and click [Next]. The [DG Definition] window appears. 98

MegaRAID SAS User's Guide 4. From the [Physical Drives] area, select a single hard disk and click [AddtoArray], and [Accept DG]. Repeat this for each hard disk in the "UNCONF GOOD" status. 5. When the hard disk becomes "ONLINE", click [Next]. The [Span Definition] window appears. 99

Chapter 2 Array Configuration and Management [WebBIOS] 6. From the disk groups displayed in the [Array With Free Space] area, select one, click [Add to SPAN], and then [Next]. The [VD Definition] window appears. 7. With the default logical drive setting, click [Accept]. Logical drives will be created. 8. Click [Back] to return to the [Span Definition] window. Repeat Step 6 and 7 to every disk group created. 9. After creating a logical drive to every disk group created, click [Next] on the [VD Definition] window. The message "Save this Configuration?" appears. 10. Click [Yes]. The message "All data on the new Virtual Disks will be lost. Want to Initialize?" appears. 11. Click [No]. Be sure to click [No] to delete all the data here. Clicking [Yes] will start fast initialization. 12. Click [HOME] to return to the main window. For MegaRAID SAS 300-8E ROMB / MegaRAID SAS 8300XLP / MegaRAID SAS 8408E / MegaRAID SAS 8344ELP 1. Click [Configuration Wizard] on the main menu. The [Configuration Wizard] window appears. 2. Select [New Configuration], and click [Next]. 3. Select [Custom Configuration], and click [Next]. The [DG Definition] window appears. 100

MegaRAID SAS User's Guide 4. From the [Physical Drives] area, select a single hard disk, and click [Accept DG]. Repeat this for each hard disk in the "UNCONF GOOD" status. 5. When the hard disk becomes "ONLINE", click [Next]. The [VD Definition] window appears. 101

Chapter 2 Array Configuration and Management [WebBIOS] 6. Select a disk group from the [Configuration] area, and create a logical drive with the following settings. table: Logical drive settings when deleting a hard disk Item Default value Remarks RAID Level RAID 0 Same as the default value Stripe Size 64KB Same as the default value Access Policy RW Same as the default value Read Policy Normal Same as the default value Write Policy WThru Same as the default value IO Policy Direct Same as the default value Disk Cache Policy Disable Same as the default value Disable BGI No Same as the default value Select Size Maximum Specifies the maximum capacity of the logical drive. 7. Perform Step 6 for each disk group. When the settings for all the logical drives are complete, the [Preview] window appears. 8. Click [Accept]. The message "Save this Configuration?" appears. 9. Click [Yes]. The message "All data on the new Virtual Disks will be lost. Want to Initialize?" appears. 10. Click [No]. Be sure to click [No] to delete all the data here. Clicking [Yes] will start fast initialization. 11. Click [HOME] or to return to the main window. 102

MegaRAID SAS User's Guide 3 Initialize all the logical drives created. 1. Select [Virtual Disks] in the main window. The [Virtual Disks] window appears. 2. While pressing the [Ctrl] key, select all the displayed logical drives. 3. Select [Slow Initialize], and click [Go]. 4. When the following confirmation window appears, click [Yes]. The initialization of the logical drive starts. Wait for a while. 5. When the initialization is complete, click [HOME] or to return to the main window. 4 Delete the disk array configuration created in Step 2. "2.3.2 Deleting Disk Array Configuration" (pg.85) This completes deleting all the information from the hard disk. The hard disk can now be reused in other systems. 103

Chapter 2 Array Configuration and Management [WebBIOS] 104

Chapter 3 Preparations This chapter explains how to update the device drivers and apply hotfix. 3.1 Updating Device Driver.......................... 106 3.2 Applying Hotfix................................ 112 105

Chapter 3 Preparations 3.1 Updating Device Driver This section explains how to update device drivers installed in the server. The procedure to update device drivers varies depending on the OS you use. Refer to the following section to update device drivers. For Windows Server 2008 "3.1.1 Updating Driver (Windows Server 2008)" (pg.106) For Windows Server 2003 and Windows 2000 Server "3.1.2 Creating Driver Disk (Windows Server 2003 / Windows 2000 Server)" (pg.108) "3.1.3 Updating Driver (Windows Server 2003)" (pg.108) "3.1.4 Updating Driver (Windows 2000 Server)" (pg.110) Even when the OS is newly installed in the server, the drivers need to be updated when those included in Array Controller Document & Tool CD are the latest. For the latest information on the software supplied with the array controller such as Array Controller Document & Tool CD, refer to the Fujitsu PRIMERGY website (http://primergy.fujitsu.com). 3.1.1 Updating Driver (Windows Server 2008) The procedure to update device drivers is different when you use the full installation and when you use Server Core installation. Full installation 1 Log on to Windows with Administrator privileges. 2 Insert Array Controller Document & Tool CD into the CD/DVD drive. 3 From [Server Manager], select [Diagnostics] and [Device Manager]. 4 Double-click [Storage controllers]. A list of array controllers is displayed. 5 Double-click one of the following array controller names. The name starting with "LSI Logic MegaRAID SAS" The name starting with "LSI MegaRAID SAS" RAID 5/6 SAS based on LSI MegaRAID 6 Select the [Driver] tab, and click [Update Driver]. The [Update Driver Software] window appears. 106

MegaRAID SAS User's Guide 7 Click [Browse my computer for driver software]. 8 Click [Let me pick from a list of device drivers on my computer]. 9 Click [Have Disk]. 10 Click [Browse], and select the following file. For Windows Server 2008 (32-bit) [CD/DVD drive]:\drivers\mega-sas\w2k8\oemsetup.inf For Windows Server 2008 (64-bit) [CD/DVD drive]:\drivers\mega-sas\w2k8x64\oemsetup.inf 11 Click [OK], check that the driver is displayed in the list, and then click [Next]. 12 After updating the device driver is complete, click [Close], and close the [Update Driver Software] window. 13 Click [OK], and close the [Properties] window. 14 If more than one array controllers are installed, update all the device drivers. Repeat Step 5 to 13 until all the device drivers are updated. 15 After updating all the device drivers is complete, eject Array Controller Document & Tool CD from the CD/DVD drive. 16 Restart the system. Server Core installation 1 Log on to Windows with Administrator privileges. 2 Insert Array Controller Document & Tool CD into the CD/DVD drive. 3 Execute the following command at the command prompt. For Windows Server 2008 (32-bit) pnputil -i -a [CD/DVD drive]:\drivers\mega-sas\w2k8\oemsetup.inf For Windows Server 2008 (64-bit) pnputil -i -a [CD/DVD drive]:\drivers\mega-sas\w2k8x64\oemsetup.inf 4 After updating all the device drivers is complete, eject Array Controller Document & Tool CD from the CD/DVD drive. 5 Restart the system. 107

Chapter 3 Preparations 3.1.2 Creating Driver Disk (Windows Server 2003 / Windows 2000 Server) Before updating device drivers, create driver disks from Array Controller Document & Tool CD by following the procedure below. 1 Prepare formatted floppy disks. 2 Insert Array Controller Document & Tool CD into the CD/DVD drive. 3 Copy the necessary drivers for the OS from the following folder in the CD-ROM to the floppy disks. To label the floppy disks with the floppy disk names, also refer to the following table. table: Driver disks OS Folder name Floppy disk name Windows 2000 Server \Drivers\Mega-SAS\W2K MegaRAID SAS Windows 2000 Server Drivers Disk Windows Server 2003 \Drivers\Mega-SAS\W2K3 MegaRAID SAS Windows Server 2003 Drivers Disk Windows Server 2003 x64 \Drivers\Mega-SAS\W2K3x64 MegaRAID SAS Windows Server 2003 for x64 Edition Drivers Disk For the versions of the device drivers, refer to ReadmeEN.html in Array Controller Document & Tool CD. 3.1.3 Updating Driver (Windows Server 2003) 1 Log on to Windows with Administrator privileges. 2 Exit all the software before the update. 3 Select [System] from [Control Panel]. 4 Select the [Hardware] tab, and click [Device Manager]. 5 Double-click [SCSI and RAID Controller]. A SCSI adapter list is displayed. 108

MegaRAID SAS User's Guide 6 Double-click your array controller from the following list: LSI Logic MegaRAID SAS PCI Express ROMB LSI Logic MegaRAID SAS 8300XLP RAID Controller LSI Logic MegaRAID SAS 8408E RAID Controller LSI Logic MegaRAID SAS 8344ELP RAID Controller LSI Logic MegaRAID SAS 8880EM2 RAID Controller RAID 5/6 SAS based on LSI MegaRAID The [Properties] window appears. 7 Select the [Driver] tab, and click [Update Driver]. The [Hardware Update Wizard] window appears. 8 Select [No, not this time], and click [Next]. 9 Select [Install from a list or specific location], and click [Next]. 10 Select [Don't search. I will choose the driver to install.], and click [Next]. 11 Insert the following floppy disk in the floppy disk drive, and click [Have Disk]. For Windows Server 2003 "MegaRAID SAS Windows Server 2003 Drivers Disk" For Windows Server 2003 x64 "MegaRAID SAS Windows Server 2003 for x64 Edition Drivers Disk" 12 Type "A:\" in [Copy manufacturer's file from], and click [OK]. 13 Select the same model name as what you specified in Step 6 from the model field, and click [Next]. The files are copied. 14 When file copying is complete, click [Done], and close the [Hardware Update Wizard] window. 15 Click [Close] to close the [Properties] window. If the [Change System Settings] window appears, click [No]. 16 If more than one array controllers are installed, update all the device drivers. Repeat Step 6 to 15 until all the device drivers are updated. 17 When all the device drivers are updated, restart the system. 109

Chapter 3 Preparations To make sure all the device drivers are correctly read, check that a controller from the following list is displayed in [SCSI and RAID controller] of [Device Manager]. (One of the following controller names is displayed depending on your array controller type.) LSI Logic MegaRAID SAS PCI Express ROMB LSI Logic MegaRAID SAS 8300XLP RAID Controller LSI Logic MegaRAID SAS 8408E RAID Controller LSI Logic MegaRAID SAS 8344ELP RAID Controller LSI Logic MegaRAID SAS 8880EM2 RAID Controller RAID 5/6 SAS based on LSI MegaRAID Also, to check the version of the device driver currently being read, double-click a controller name above, select the [Driver] tab and refer to the version information. 3.1.4 Updating Driver (Windows 2000 Server) 1 Log on to Windows with Administrator privileges. 2 Exit all the software before the update. 3 Select [System] from [Control Panel]. 4 Select [Hardware] tab, and click [Device Manager]. 5 Double-click [SCSI and RAID Controller]. A SCSI adapter list is displayed. 6 Double-click your array controller from the following list: LSI Logic MegaRAID SAS PCI Express ROMB LSI Logic MegaRAID SAS 8300XLP RAID Controller LSI Logic MegaRAID SAS 8408E RAID Controller The [Properties] window appears. 7 Select the [Driver] tab, and click [Update Driver]. The [Device Driver Upgrade Wizard] window appears. 8 Click [Next]. 9 Select [Display known drivers for this device and select a driver from the list.], and click [Next]. 10 Click [Have Disk], and insert the "MegaRAID SAS Windows 2000 Server Drivers Disk" in the floppy disk drive. 11 Type "A:\" in [Copy manufacturer's file from], and click [OK]. 110

MegaRAID SAS User's Guide 12 Select the same model name as what you specified in Step 6 from the model field, and click [Next]. The device driver installation starts. 13 When the installation is complete, click [Done], and close the [Device Driver Update Wizard] window. 14 Click [Close] to close the [Properties] window. If the [Change System Settings] window appears, click [No]. 15 If more than one array controllers are installed, update all the device drivers. Repeat Step 6 to 14 until all the device drivers are updated. 16 When all device drivers are updated, restart the system. To make sure all the device drivers are correctly read, check that a controller from the following list is displayed in [SCSI and RAID controller] of [Device Manager]. (One of the following controller names is displayed depending on your array controller type.) LSI Logic MegaRAID SAS PCI Express ROMB LSI Logic MegaRAID SAS 8300XLP RAID Controller LSI Logic MegaRAID SAS 8408E RAID Controller Also, to check the version of the device driver currently being read, double-click a controller name above, select the [Driver] tab and refer to the version information. 111

Chapter 3 Preparations 3.2 Applying Hotfix To use this array controller with a newly installed OS in the Windows environment, make sure to apply hotfix. For hotfix, use PRIMERGY Startup Disc ( or ServerStart Disc1 CD-ROM depending on the server) supplied with the server. 1 Log on to Windows with Administrator privileges. 2 Exit all the software. 3 Insert PRIMERGY Startup Disc (or ServerStart Disc1 CD-ROM depending on the server) into the CD/DVD drive. 4 Run the following program to apply hotfix. For Windows 2000 Server [CD/DVD drive]:\hotfix\w2k\enu\windows2000-kb904374-x86-enu.exe For Windows Server 2003 Service Pack 1 [CD/DVD drive]:\hotfix\w2k3\windowsserver2003-kb916048-x86-enu.exe For Windows Server 2003 x64 Service Pack 1 [CD/DVD drive]:\hotfix\w2k3x64\windowsserver2003.windowsxp-kb916048-x64- ENU.exe 112

Chapter 4 Array Configuration and Management [ServerView RAID] This chapter explains an overview of and product requirements for ServerView RAID, and describes how to install and use the program. 4.1 Overview of and Requirements for ServerView RAID.... 114 4.2 Installing ServerView RAID [Windows]............... 121 4.3 Installing HDD Check Scheduler [Windows]........... 124 4.4 Installing Battery Recalibration Scheduler [Windows].... 129 4.5 Starting and Exiting ServerView RAID Manager........ 133 4.6 ServerView RAID Manager Window Layout........... 137 4.7 Settings..................................... 147 4.8 Operating RAID Configuration..................... 156 4.9 Checking Each Status [ServerView RAID Manager]..... 171 113

Chapter 4 Array Configuration and Management [ServerView RAID] 4.1 Overview of and Requirements for ServerView RAID This section explains ServerView RAID. ServerView RAID is used to monitor, manage, maintain, and configure the array controllers and the hard disks and logical drives that are connected to the array controllers. 4.1.1 Overview of ServerView RAID This application software runs on the OS to monitor and manage the array controller. ServerView RAID is an application software that allows you to manage a disk array system connected to the array controllers (RAID controllers). Using ServerView RAID, which is a client-server application, you can manage the array controllers via a network as well as in a standalone environment. ServerView RAID includes the ServerView RAID service and ServerView RAID Manager. The ServerView RAID service and ServerView RAID Manager use HTTPS to communicate with each other. ServerView RAID Service Installed on the server, this program operates as an OS service and monitors the operation of the array controllers. ServerView RAID Manager A graphical user interface that uses a Web browser and Java. You manage and configure the array controllers with ServerView RAID Manager. ServerView RAID ManagerGUI ServerView RAID service (service) Array controller 114

MegaRAID SAS User's Guide To ensure stable operation of PRIMERGY, install ServerView RAID (or GAM) when using the array controller. If ServerView RAID (or GAM) is not installed, failures will not be detected. You will also be unable to perform maintenance correctly. Make sure to install ServerView RAID (or GAM). Install either ServerView RAID or GAM, but do not install both. For which to use, refer to ReadmeEN.html in the Array Controller Document & Tool CD. If, by mistake, you have installed GAM rather than ServerView RAID on the server where ServerView RAID is only acceptable, uninstall GAM and HDD Check Scheduler, and then install ServerView RAID and HDD Check Scheduler. Do not operate the server with both ServerView RAID and GAM installed. Only start ServerView RAID Manager when management or maintenance of arrays is necessary. From a security point of view, it is not recommended to run ServerView RAID Manager continuously. If the accessed server shuts down while you are logged in to ServerView RAID Manager, ServerView RAID Manager cannot communicate with ServerView RAID service and cannot respond, making it impossible to operate ServerView RAID Manager. In this case, close the Web browser in which ServerView RAID Manager is running. ServerView RAID uses SSL to encrypt communication. Since server certification is not supported, you are responsible for ensuring the reliability of the network server. 4.1.2 Requirements for ServerView RAID An appropriate server environment is required for using ServerView RAID properly. Make sure that the following requirements are met. ServerView RAID (server requirements) You must have the following environments to use ServerView RAID. Category Hard disk table: Requirements for ServerView RAID For Windows 150 MB or more free space in the installation partition of the OS Description For Linux 150 MB or more free space under /opt Application TCP/IP, SNMP service, and ServerView must be installed. OS OS supported by servers with this array controller installed Web browser Internet Explorer 6 or later Mozilla Firefox 1.0.4 or later SeaMonkey 1.0.3 or later Java Java 2 Runtime Environment Standard Edition V1.5.0_06 or later 115

Chapter 4 Array Configuration and Management [ServerView RAID] Make sure to install the device drivers and ServerView RAID specified by Fujitsu. Make sure to configure the network settings properly. If there is a problem with the network configuration, you may be unable to monitor the status of arrays by ServerView or events may not be notified. To monitor the array controllers, refer to the OS event log notified by ServerView (Source: Fujitsu ServerView Services) or by ServerView RAID, or /var/log/messages (Source: ServerView RAID). Refer to the events by ServerView instead of ServerView RAID when events are notified from both. For the list of logs notified by ServerView, refer to "Appendix A List of ServerView RAID Event Log" (pg.276). In the following environments, Web browsers and Java are not necessary for the servers where ServerView RAID will be installed. However, to manage the array system, the environments where these software run properly on the server or the client PC over the network is necessary. The environment in which Windows Server 2008 is installed by the Server Core installation option RHEL5(Intel64), RHEL-AS4(EM64T), and RHEL-ES4(EM64T) ServerView RAID Manager (when managed from a client PC) When ServerView RAID Manager is used on a client PC different from the server, the following environments are required for the client PC. Category Network Input device Processor Memory Monitor table: Requirements for ServerView RAID Manager Description Network connection with TCP/IP available A mouse or other pointing device Pentium 500MHz or higher (1GHz or higher recommended) 512 MB or more (1 GB or more recommended) 800 600 or better resolution (1024 768 or more recommended), 256 or more colors OS Windows Server 2008 Windows Server 2003 SP1 or later Windows Vista Windows XP Windows 2000 Service Pack 4 or later Windows 2000 Professional Service Pack 4 or later Web browser Internet Explorer 6 or later Java Java 2 Runtime Environment Standard Edition V1.5.0_06 or later 116

MegaRAID SAS User's Guide 4.1.3 Access Privileges to ServerView RAID To use the ServerView RAID functions, you must log in to ServerView RAID Manager. When you log in, user authentication is performed against your user account registered with the OS. The available functions depend on the user account. There are two levels of access privileges as described below. User privileges User privileges are mainly used to check statuses of array controllers, hard disks, and logical drives. To use User privileges, log in to ServerView RAID Manager with the user names and passwords which belong to the "raid-usr" group. Create "raid-usr" groups as necessary. With User privileges, you can check detailed information, settings, and statuses of the RAID subsystems such as array controllers, hard disks, and logical drives. However, you cannot change the disk array configuration, rebuild hard disks, or change parameters related to array controllers and ServerView RAID. Arrays cannot be configured with User privileges. It is recommended that you log in with User privileges unless you need to operate arrays or change settings. Administrator privileges Administrator privileges are used for management, maintenance, and configuration of array controllers, hard disks, and logical drives. To use the Administrator privileges, log in to ServerView RAID Manager as a user belonging to the "raid-adm" group, the "Administrators" group (for Windows), or "root" group (for Linux). In addition to the functions available with User privileges, it is possible to use all other functions including creating/changing disk array configurations, rebuilding hard disks, making logical drive data consistent, and changing the hard disk status. When using ServerView RAID with Administrator privileges, certain operations may cause loss of data in the array controller. Read this chapter and use ServerView RAID carefully. You must create the "raid-adm" group. Make sure to create the group with the name "raid-adm". 117

Chapter 4 Array Configuration and Management [ServerView RAID] 4.1.4 Using ServerView RAID in Linux Environment To use ServerView RAID in a Linux environment, you need to install device drivers for Linux and ServerView RAID. For Linux, refer to the PRIMERGY page on the Fujitsu website (http://primergy.fujitsu.com/) and the information about Linux. For the RHEL5(Intel64), RHEL-AS4(EM64T), or RHEL-ES4(EM64T) environment The Java plug-ins do not work on the servers with the RHEL5(Intel64), RHEL-AS4(EM64T), or RHEL- ES4(EM64T) system. You cannot manage the array controllers with ServerView RAID Manager running directly on these servers. If you want to manage the array controllers on servers with the RHEL5(Intel64), RHEL-AS4(EM64T), or RHEL-ES4(EM64T) system, run ServerView RAID Manager on a Windows server or a client PC, and manage the array controllers remotely. For the remote management configuration, refer to "4.1.5 Operations via Network" (pg.118). 4.1.5 Operations via Network In a network environment, arrays on the servers can be monitored and managed from a server or a Windows client PC connected to the network. When you manage the disk array on the server with ServerView RAID installed from ServerView RAID Manager on another server or a client PC, the configuration is as follows. Windows server or Client PC Monitoring and setting the disk array Server ServerView RAID Manager Notifying ServerView RAID events Array controller ServerView RAID service Each ServerView RAID Manager program can manage only one server. When you manage disk arrays on multiple servers from a single client PC or server, start one ServerView RAID Manager program for each server. If there is an intervening firewall in the environment, you need to configure the network settings so that the port used by ServerView RAID is not blocked. ServerView RAID uses the TCP port 3173. Depending on the OS, for example Windows Server 2008, the fire wall is enabled by the default setting. To use ServerView RAID Manager from the other servers or client PC which are connected to the network, release the blockage for the port to be used for ServerView RAID. 118

MegaRAID SAS User's Guide Interaction with ServerView AlarmService ServerView RAID can log the events of the array controller in the OS event log on the server using ServerView AlarmService. When you monitor the disk array remotely, you can also log the events of the array controller in the OS event log on the client PC by installing ServerView Console or ServerView AlarmService on the client PC. When you install ServerView AlarmService on the client PC as well, the configuration is as follows. Windows server or Client PC Monitoring and setting the disk array Windows server or Linux server ServerView RAID Manager ServerView AlarmService OS event log Notifying ServerView RAID events (SNMP Trap) SNMP Service Notifying ServerView RAID events (SNMP Trap) ServerView AlarmService Array controller ServerView RAID service OS event log Make sure to install ServerView on the server. 119

Chapter 4 Array Configuration and Management [ServerView RAID] 4.1.6 When Using Both ServerView RAID and GAM When using both ServerView RAID and GAM on a network, you can use ServerView RAID Manager and GAM Client simultaneously on the management client PC or Windows server. You can use the following configuration. Management client PC or Windows server Notifying events Monitoring and setting the disk array Server Array controller ServerView RAID service GAM Client Monitoring and setting the disk array Server ServerView RAID Manager GAM Server Notifying GAM events Array controller If ServerView is installed on the management client PC or Windows server, you can start ServerView RAID Manager or the GAM Client program for a server managed by ServerView by making use of ServerView's interaction with RAID Manager. For details about RAID Manager interaction, refer to "RAID Manager Linking" of "ServerView User's Guide". When you install GAM Client on a Windows server where ServerView RAID is already installed, install only GAM Client. Do not install GAM Server. One GAM Client can manage up to a maximum of 100 GAM Servers. When managing more than 100 servers at the same time, one Windows server or one client PC to be used as GAM Client is necessary per 100 servers. If multiple versions of GAM Server coexist, use the version of GAM Client that corresponds to the latest version of GAM Server, or a later version. 120

MegaRAID SAS User's Guide 4.2 Installing ServerView RAID [Windows] This section explains how to install ServerView RAID on a Windows server. You may or may not overwrite installation on ServerView RAID depending on the version. With the version v2.0.x or v2.1.x, overwriting installation cannot be done. Uninstall ServerView RAID and install the version of ServerView RAID you want to install. With the version v2.3.x or later, you can overwrite installation. Just go on installing the later version. With all the versions, you can not overwrite installation with the older version. If you need to do that for some reasons, uninstall ServerView RAID first and then install the version of ServerView RAID you want to install. To log events that occur in the array controllers in the OS event log, make sure to install ServerView and configure the event log settings. For details, refer to "ServerView User's Guide". 4.2.1 Installing ServerView RAID Follow the procedure below to install ServerView RAID. 1 Log on to Windows with Administrator privileges. 2 Before installing ServerView RAID, complete the following preparations: Check that ServerView is installed and working properly. Insert Array Controller Document & Tool CD supplied with the array controller into the CD/ DVD drive. Exit all application software. Exit all application software before starting the installation. In particular, if you install the software while [Event Viewer], [Services], or [Computer Management] is running, the installation may fail. 3 Click [Start] [Run...]. Enter the following path, and click [OK]. [CD/DVD drive]:\raidtool\svraid\windows\install.bat The message "Installation of ServerView RAID has started." is displayed at the command prompt, and the installation of ServerView RAID starts. In the Server Core installation environment for Windows Server 2008, the [Start] menu is not available. Execute the following command. cd /d [CD/DVD drive]:\raidtool\svraid\windows install.bat 121

Chapter 4 Array Configuration and Management [ServerView RAID] 4 When the message "Installation of ServerView RAID has finished." is displayed at the command prompt, press the [Enter] key. The command prompt window closes. 5 Restart the system. 6 Once the system is up and running, create and configure a Windows user account. User account for Administrator privileges 1. Create a user account for the Administrator privileges of ServerView RAID. 2. Create the "raid-adm" group. 3. Configure the user account for the Administrator privileges so that it belongs to the "raid-adm" group or the "Administrators" group. User account for User privileges 1. Create a user account for the User privileges of ServerView RAID. 2. Create the "raid-usr" group. 3. Configure the user account for the User privileges so that it belongs to the "raid-usr" group. Do not include the user account for the User privileges to the "raid-adm" group. When creating the account for ServerView RAID, uncheck [User must change password at next logon]. Normally, [Password never expires] should be checked. Unless the account is configured as above, you may be unable to log in to ServerView RAID Manager properly, because the program will consider the user account invalid if the password of the account has expired or has not been set, without displaying any warning message. Create each user account as an OS user account. You must create the "raid-adm" group. Make sure to create the group with the name "raidadm". 122

MegaRAID SAS User's Guide 4.2.2 Uninstalling ServerView RAID Follow the procedure below to uninstall ServerView RAID. In general, do not uninstall ServerView RAID except for certain cases such as when you need to update ServerView RAID. 1 Log on to Windows with Administrator privileges. Exit all programs before starting the uninstallation. If you uninstall the software while [Event Viewer], [Services], or [Computer Management] is running, the uninstallation will fail. Make sure to exit all programs. 2 Click [Start] [Control Panel]. 3 Double-click [Add or Remove Applications] (or [Add or Remove Programs] depending on the OS). 4 Select [ServerView RAID] from the application list, and click [Delete] (or [Change/Remove] depending on the environment). The message "Are you sure you want to remove ServerView RAID from the computer?" appears. 5 Click [Yes]. The uninstallation process starts. 6 When the following message appears after the uninstallation, click [Yes]. You must restart your system for the configuration changes made to ServerView RAID to take effect... The system restarts. In the Server Core installation environment for Windows Server 2008, follow the procedure below to uninstall ServerView RAID. 1. Log on to Windows with Administrator privileges. 2. Execute the following command at the command prompt. wmic product where name="serverview RAID Manager" call uninstall 3. Restart the system. 123

Chapter 4 Array Configuration and Management [ServerView RAID] 4.3 Installing HDD Check Scheduler [Windows] HDD Check Scheduler is a tool to detect or recover media errors on hard disks, by periodically performing Make Data Consistent of the media. This section explains how to install HDD Check Scheduler on the Windows server. Make sure to install HDD Check Scheduler. Also, make sure to install ServerView RAID before installing HDD Check Scheduler. During the HDD check, the I/O performance for the logical drives decreases to some extent (up to about 30%). After the installation, set a start date and time that minimizes the effect on the business by referring to "4.3.2 Changing HDD Check Start Time" (pg.125). When you install HDD Check Scheduler, the check is set to start every day at 12:00 by default. For an overview of the HDD Check Scheduler function, refer to "1.3.7 HDD Check Scheduler" (pg.38). HDD Check Scheduler performs Make Data Consistent periodically using the standard OS scheduling function (AT command for Windows, or Cron for Linux). It is recommended that you perform the HDD check, at least once a week. If the server restarts or shuts down during HDD check, the HDD check is resumed from where it was stopped the next time. 4.3.1 Installing HDD Check Scheduler Follow the procedure below to install HDD Check Scheduler. 1 Log on to Windows with Administrator privileges. 2 Before installing HDD Check Scheduler, complete the following preparations: Check that ServerView RAID is installed. Make sure that ServerView is installed and working properly. Insert Array Controller Document & Tool CD supplied with the array controller into the CD/ DVD drive. Exit all application software. Exit all application software before starting the installation. In particular, if you install the software while [Event Viewer], [Services], or [Computer Management] is running, the installation may fail. 124

MegaRAID SAS User's Guide 3 Click [Start] [Run...]. Enter the following file name, and click [OK]. [CD/DVD drive]:\raidtool\hddcs\windows\install.bat The message "Installing HDD Check Scheduler." appears at the command prompt. In the Server Core installation environment for Windows Server 2008, the [Start] menu is not available. Execute the following command. cd /d [CD/DVD drive]:\raidtool\hddcs\windows install.bat 4 When the message "Installation of HDD Check Scheduler has finished." appears, press the [Enter] key. The command prompt window closes. When you install HDD Check Scheduler for the first time, the HDD check is set to start every day at 12:00 by default. If you reinstall and overwrite HDD Check Scheduler, the previously set start time setting will be kept. 4.3.2 Changing HDD Check Start Time The start time for the HDD check can be changed by using the AT command of the OS. To change the start time, follow the procedure below. It is recommended that you perform the HDD check regularly, at least once a week. 1 Log on to Windows with Administrator privileges. 2 Click [Start] [All Programs] [Accessories] [Command Prompt] to open the command prompt. In the Server Core installation environment for Windows Server 2008, the command prompt is already open. This step is not necessary. 3 Type "AT", and press the [Enter] key. 125

Chapter 4 Array Configuration and Management [ServerView RAID] 4 Write down the [ID] value for the row with the following value in the [Command Line] column. For Windows Server 2008 (64-bit), and Windows Server 2003 x64 C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat For Windows Server 2008 (32-bit), Windows Server 2003, and Windows 2000 Server C:\Program Files\Fujitsu Siemens\RAID\FJutils\HDDchk.bat Example: If the AT command output is as follows, the ID is 13. C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 1 Each M F 10:00 AM "C:\Program Files (x86)\fujitsu \F5FBFE01\ServerView Services\WebServer\ClearMyLogs.exe" 13 Each M T 12:00 PM "C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat" 5 Delete the currently registered HDD check schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 13 /delete 6 Register a new schedule. By using the AT command, register the following command line together with the time of execution. For Windows Server 2008 (64-bit) and Windows Server 2003 x64 C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat For Windows Server 2008 (32-bit), Windows Server 2003, and Windows 2000 Server C:\Program Files\Fujitsu Siemens\RAID\FJutils\HDDchk.bat To specify the scheduled execution, enter the following AT command. AT <Start Time> /every:<day of Week> "Command Line" To perform the HDD check every Friday at 18:00, for Windows Server 2003 and Windows 2000 Server AT 18:00 /every:f "C:\Program Files\Fujitsu Siemens\RAID\ FJutils\HDDchk.bat" To perform the HDD check every day at 21:00, for Windows Server 2003 x64 AT 21:00 /every:m,t,w,th,f,s,su "C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat" 126

MegaRAID SAS User's Guide To specify the day of the week, use the following abbreviations. Monday Tuesday Wednesday Thursday Friday Saturday Sunday M T W Th F S Su For details on how to use the AT command, refer to Windows Help. To see whether the HDD check is properly performed at the specified time, check that the log entry below is recorded in the OS application log at the specified time. If no log entries below are recorded at the specified time, review the settings. Source : Fujitsu ServerView Services Type : Information Event ID : 1 Description: Adapter LSI MegaRAID SAS xx (x): MDC started on Logical Drive x (Server xxx) [Note]: Each "x" in the descriptions corresponds to a string of characters or digits, depending on your operating environment. 4.3.3 Uninstalling HDD Check Scheduler Follow the procedure below to uninstall HDD Check Scheduler. 1 Log on to Windows with Administrator privileges. 2 Click [Start] [All Programs] (or [Programs]) [Accessories] [Command Prompt]. The command prompt window opens. In the Server Core installation environment for Windows Server 2008, the command prompt is already open. This step is not necessary. 3 Type "AT", and press the [Enter] key. 127

Chapter 4 Array Configuration and Management [ServerView RAID] 4 Write down the [ID] value for the row with the following value in the [Command Line] column. For Windows Server 2008 (64-bit) and Windows Server 2003 x64 C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat For Windows Server 2008 (32-bit), Windows Server 2003 and Windows 2000 Server C:\Program Files\Fujitsu Siemens\RAID\FJutils\HDDchk.bat Example: If the AT command output is as follows, the ID is 13. C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 1 Each M F 10:00 AM "C:\Program Files (x86)\fujitsu \F5FBFE01\ServerView Services\WebServer\ClearMyLogs.exe" 13 Each M T 12:00 PM "C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat" 5 Delete the currently registered HDD check schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 13 /delete 128

MegaRAID SAS User's Guide 4.4 Installing Battery Recalibration Scheduler [Windows] Battery Recalibration Scheduler is a tool to periodically perform battery recalibration for the array controllers below. This section explains how to install Battery Recalibration Scheduler on a Windows server. MegaRAID SAS 300-8E ROMB with a battery MegaRAID SAS 8408E MegaRAID SAS 8344ELP with a battery MegaRAID SAS 8880EM2 with a battery RAUD 5/6 SAS based on LSI MegaRAID with a battery Before using this tool, also refer to "1.3.8 Battery Backup Unit (BBU)" (pg.39). Recalibration requires 12 hours with the server power on. During recalibration, the logical drive in Write Back mode shifts automatically to Write Through mode for safety. Therefore, refer to "4.4.2 Changing Battery Recalibration Start Date and Time" (pg.130) to set a start date and time that minimizes the effect on the business. MegaRAID SAS 8300XLP does not require the installation of this tool or periodical recalibration. Battery Recalibration Scheduler performs recalibration periodically using the standard OS scheduling function (AT command for Windows, or Cron for Linux). It is recommended that you perform battery recalibration about once a month or every two months. If the server restarts or shuts down during battery recalibration, the battery recalibration is resumed from where it was stopped the next time. 4.4.1 Installing Battery Recalibration Scheduler Follow the procedure below to install Battery Recalibration Scheduler. 1 Log on to Windows with Administrator privileges. 2 Before installing Battery Recalibration Scheduler, complete the following preparations: Check that ServerView RAID is installed. Make sure that ServerView is installed and working properly. Insert Array Controller Document & Tool CD supplied with the array controller into the CD/ DVD drive. Exit all application software. 129

Chapter 4 Array Configuration and Management [ServerView RAID] Exit all application software before starting the installation. In particular, if you install the software while [Event Viewer], [Services], or [Computer Management] is running, the installation may fail. 3 Click [Start] [Run...]. Enter the following file name and click [OK]. [CD/DVD drive]:\raidtool\batcs\windows\install.bat The message "Installing Battery Recalibration Scheduler." appears at the command prompt. In the Server Core installation environment for Windows Server 2008, the [Start] menu is not available. Execute the following command. cd /d [CD/DVD drive]:\raidtool\batrcs\windows install.bat 4 When the message "Installation of Battery Recalibration has finished." appears, press the [Enter] key. The command prompt window closes. When you install Battery Recalibration Scheduler for the first time, the recalibration is set to start the first day of every month at 11:00 by default. If you reinstall and overwrite Battery Recalibration Scheduler, the previously set start date and time settings will be kept. 4.4.2 Changing Battery Recalibration Start Date and Time The start date and time for the battery recalibration can be modified using the AT command of the operating system. To modify the start date and time, follow the procedure below. It is recommended that you perform battery recalibration about once a month or every two months. 1 Log on to Windows with Administrator privileges. 2 Click [Start] [All Programs] [Accessories] [Command Prompt] to open the command prompt. In the Server Core installation environment for Windows Server 2008, the command prompt is already open. This step is not necessary. 3 Type "AT", and press the [Enter] key. 130

MegaRAID SAS User's Guide 4 Write down the [ID] value for the row with the following value in the [Command Line] column. For Windows Server 2008 (64-bit) and Windows Server 2003 x64 C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\Batrec.bat For Windows Server 2008 (32-bit), Windows Server 2003, and Windows 2000 Server C:\Program Files\Fujitsu Siemens\RAID\FJutils\Batrec.bat Example: If the AT command output is as follows, the ID is 7. C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 1 Each M F 10:00 AM "C:\Program Files (x86)\fujitsu \F5FBFE01\ServerView Services\WebServer\ClearMyLogs.exe" 6 Each M T 12:00 PM "C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat" 7 Each 1 11:00 PM "C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\Batrec.bat" 5 Delete the currently registered battery recalibration schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 7 /delete 6 Register a new schedule. Using the AT command, register the following command line together with the date and time of execution. For Windows Server 2008 (64-bit) and Windows Server 2003 x64 C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\Batrec.bat For Windows Server 2008 (32-bit), Windows Server 2003, and Windows 2000 Server C:\Program Files\Fujitsu Siemens\RAID\FJutils\Batrec.bat To specify the scheduled execution, enter the following AT command. AT <Start Time> /every:<date> "Command Line" To perform the Battery Recalibration every 15th at 8:00, for Windows Server 2003 and Windows 2000 Server AT 8:00 /every:15 "C:\Program Files\Fujitsu Siemens\RAID\FJutils\Batrec.bat" For details on how to use the AT command, refer to Windows Help. To confirm whether battery recalibration is configured correctly, type "AT" at the command prompt and check the displayed result. Check that the date, time, and the command line set in Step 6 are correct. 131

Chapter 4 Array Configuration and Management [ServerView RAID] 4.4.3 Uninstalling Battery Recalibration Scheduler Follow the procedure below to uninstall Battery Recalibration Scheduler. 1 Log on to Windows with Administrator privileges. 2 Click [Start] [All Programs] (or [Programs]) [Accessories] [Command Prompt]. The command prompt window opens. In the Server Core installation environment for Windows Server 2008, the command prompt is already open. This step is not necessary. 3 Type "AT", and press the [Enter] key. 4 Write down the [ID] value for the row with the following value in the [Command Line] column. For Windows Server 2008 (64-bit) and Windows Server 2003 x64 C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\Batrec.bat For Windows Server 2008 (32-bit), Windows Server 2003, and Windows 2000 Server C:\Program Files\Fujitsu Siemens\RAID\FJutils\Batrec.bat Example: If the AT command output is as follows, the ID is 7. C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 1 Each M F 10:00 AM "C:\Program Files (x86)\fujitsu \F5FBFE01\ServerView Services\WebServer\ClearMyLogs.exe" 6 Each M T 12:00 PM "C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\HDDchk.bat" 7 Each 1 11:00 PM "C:\Program Files (x86)\fujitsu Siemens\RAID\FJutils\Batrec.bat" 5 Delete the currently registered battery recalibration schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 7 /delete 132

MegaRAID SAS User's Guide 4.5 Starting and Exiting ServerView RAID Manager This section explains how to start and exit ServerView RAID Manager. 4.5.1 Preparations and Precautions for Using ServerView RAID Manager You need to configure the Web browser settings before using ServerView RAID Manager. Also, note the following precautions. General precautions ServerView RAID Manager use a Java plug-in. You must install Java on the server or client PC on which you use ServerView RAID Manager. If Java is not already installed, refer to "Installing Java 2 Runtime Environment Standard Edition" under "Chapter 2 Installation" of "ServerView User's Guide" in PRIMERGY Startup Disc, and install Java. On servers with the RHEL-AS4(EM64T) or RHEL-ES4(EM64T) system, Java plug-ins do not work. Therefore, when you manage the array controllers in the RHEL-AS4(EM64T) or RHEL- ES4(EM64T) environment, you need a separate Windows server or client PC to run ServerView RAID Manager. When you use ServerView RAID Manager on the server, do not use proxy servers for the IP address and localhost. Do not use the [Back], [Forward], and [Refresh] buttons in the Web browser. In the Server Core installation environment for Windows Server 2008, the Web browser does not run. To manage the array controller, prepare the environment described in " ServerView RAID Manager (when managed from a client PC)" (pg.116), and start up ServerView RAID Manager. When using Internet Explorer as your Web browser Enable SSL. Select [Tools] [Internet Options] [Advanced], and enable SSL 3.0 and TSL 1.0 under the [Security]. Start the Web browser and add the zone setting of Web site as follows. 1. In Internet Explorer, select [Tools] [Internet Options]. 2. Click the [Security] tab, and select [Intranet] (or [Local intranet] in Internet Explorer 7) or [Trusted sites]. 3. Click [Sites] (or [Advanced] in Windows 2000), and add the following URL of the server where you installed ServerView RAID: https://<the name or IP address of the server>. 4. Click [Close]. 5. Click [Custom Level]. 6. Set [Enable] for [Run ActiveX controls and plug-ins]. 133

Chapter 4 Array Configuration and Management [ServerView RAID] When using Mozilla Firefox / SeaMonkey as your Web browser Select [Edit] [Configuration], and enable the following items in the configuration window. [SSL 2.0] and [SSL 3.0] [Enable Java] 4.5.2 Starting and Logging in ServerView RAID Manager You can start ServerView RAID Manager in several ways as follows: Starting from the [Start] button Start ServerView RAID Manager from the [Start] button. This method can only be used on the Windows server on which ServerView RAID is installed. Starting from ServerView Start ServerView RAID Manager by utilizing the RAID Manager linking function of ServerView. For information about how to start ServerView RAID Manager from ServerView, refer to "RAID Manager Linking" of "ServerView User's Guide" in PRIMERGY Startup Disc. Starting by specifying a server name or IP address directly Start ServerView RAID Manager by specifying the host name or IP address of the server directly. Using this method, you can manage the array controllers on the server from a remote client PC. 1 Start up ServerView RAID Manager. Starting from the [Start] button 1. Click [Start] [All Programs] (or [Program]) [Fujitsu Siemens] [ServerView RAID Manager Start]. Starting by specifying the server name or IP address directly 1. Start up the Web browser. 2. Enter the following URL, and press the [Enter] key. https://<the name or the IP address of the server>:3173/ If a "Security Alert" message appears, click [Yes] to continue. If the following message appears on the startup of Java, click [Yes] to continue. "The web site's certificate is invalid. Do you want to continue?" "The web site's certificate cannot be verified. Do you want to continue?" If the message "The name of the site does not match the name on the certificate. Do you want to run the application?" appears on the startup of Java, click [Run] to continue. 134

MegaRAID SAS User's Guide When ServerView RAID Manager starts, the login window appears. If you leave the Java startup pop-up window open for a long time while starting ServerView RAID Manager, the following window may appear and ServerView RAID Manager cannot start. In this case, close the Web browser, and then start ServerView RAID Manager again. 2 Enter the user name in [Username:]. When logging in with Administrator privileges Enter a user name that belongs to the "raid-adm", "Administrators" (for Windows) or "root" (for Linux) group. When logging in with User privileges Enter a user name that belongs to the "raid-usr" group. 3 Enter a password in [Password]. 4 Click [Login]. The main window of ServerView RAID Manager appears. 135

Chapter 4 Array Configuration and Management [ServerView RAID] If you enter the wrong password for logging in, you may be unable to enter the password again. In this case, click [Username:], and then enter the user account and the correct password. In ServerView RAID Manager, the available functions are limited by the access privileges. For details about access privileges, refer to "4.1.3 Access Privileges to ServerView RAID" (pg.117). Do not change the status of the language selection button in the login window. Leave it as. The following event logged after login may indicate that the operation to finish ServerView RAID Manager last time was not appropriate. This, however, does not interrupt any of your operation at the next login. ID: 10172 Event: User xxxxx (amrui) logged out 4.5.3 Exiting ServerView RAID Manager This section explains how to exit ServerView RAID Manager. 1 In the ServerView RAID Manager menu bar, click [File] [Exit]. The login window appears. 2 Exit the Web browser. 136

MegaRAID SAS User's Guide 4.6 ServerView RAID Manager Window Layout This section explains the windows, buttons, and menu items displayed when using ServerView RAID Manager. 4.6.1 Startup Window Layout and Functions When ServerView RAID Manager starts, the following main window appears. Object window Menu bar Toolbar Tree view Event window Object window The window displays information about the object (device) selected in the tree view. Use the tabs at the top of the object window to switch the information displayed. [General] tab This tab displays information about the object, such as the IP address, the OS, and the capacities of the hard disks and logical drives. [Settings] tab This tab displays the settings of the object. It also allows you to modify the settings of the object. [Layout] tab This tab allows you to see the configuration of the object. [Background activities] tab This tab allows you to see the statuses of the background tasks currently running for the object. 137

Chapter 4 Array Configuration and Management [ServerView RAID] The information displayed in each tab of the object window varies with the selected object. The [Background activities] tab is not displayed when no background tasks are running. Menu bar ServerView RAID Manager provides menus for performing various functions. For details on the ServerView RAID Manager menus, refer to "4.6.2 Menu Layout and Functions" (pg.139). Toolbar The toolbar contains the following buttons (Back/Forward) These buttons allow you to go back/forward in the object selection history for the tree view. (Change Access Mode) This button allows you to switch the access mode of the object. Access modes represent the operation privileges for ServerView RAID Manager. The following access modes are available: Read only access You can only see the information. Write access You can make any modifications. When you want to operate the disk array or modify the settings for the controller and logical drives, you must set "Write access" mode. When you log in to ServerView RAID Manager with Administrator privileges, you can use "Write access mode". When you log in with User privileges, you can only use "Read only access" mode. "4.5.2 Starting and Logging in ServerView RAID Manager" (pg.134) When you log in with Administrator privileges in "Read only access" mode and perform any modification such as modifying the settings, the following pop-up window appears. Click [Yes] to automatically switch the access mode to "Write access" mode. Tree view The tree view displays all the objects related to the array controller as icons in a tree. For details, refer to "4.6.3 Layout of Tree View" (pg.142). 138

MegaRAID SAS User's Guide Event window ServerView RAID monitors the operation of the array controller and hard disks connected to the array controller. When any behavior that is considered as an event (for example, a failure of a hard disk or an allocation of a spare disk) occurs, ServerView RAID Manager is notified of it and displays it in the event window. The following information is displayed. Event Severity table: Event window Description Priority level of the event. The following icons are used: Information Warning Error Date Source ID Event Date and time when the event occurred. Object (device) where the event occurred. ID of the event Event description ServerView RAID Manager displays only the last 100 events in the event window. You can not refer to the old events because events before the last 100 are overwritten with the newer events. To refer to the old events that you can not view on the ServerView RAID Manager event window, these can be obtained in Windows event viewer or in /var/log/messages of Linux. 4.6.2 Menu Layout and Functions This section explains the layout and functions of ServerView RAID Manager menu items. File menu Exit Menu table: File menu Function Select this item to close the current session and return to the login window. Actions menu The items in the [Actions] menu vary with the object selected in the tree view and the status of the object. Note that grayed out menu items are disabled. The [Actions] menu is the same as the right-click menu for the object selected in the tree view. 139

Chapter 4 Array Configuration and Management [ServerView RAID] When array controller is selected Menu Scan configuration Create logical drive Delete all logical drives Delete last logical drive Alarm on / Alarm off Start Patrol Read Cancel Patrol Read Clear configuration Manage foreign configuration table: Actions menu (when array controller is selected) Function Select this item to redetect all devices connected to the array controller. Select this item to set a disk array configuration. This function enables you to define disk groups and logical drives. For details, refer to "4.8.1 Creating Disk Array Configuration (Create logical drive)" (pg.156). Select this item to delete all the logical drives connected to the array controller. Select this item to delete the last created logical drive (i.e. the logical drive with the highest drive number). Not supported. Do not use them. Select this item to clear all the disk array configurations of the array controller. For details, refer to "4.8.6 Deleting Disk Array Configuration Information" (pg.167). Not supported. Do not use it. When hard disk is selected Menu Locate device Stop location Make online Make offline Make ready Replace missing Disk Start rebuild Cancel rebuild table: Actions menu (when hard disk is selected) Function Turns on the hard disk failure LED to indicate the disk location. Turns off the hard disk failure LED that is turned on with [Locate device]. Forces the status of the hard disk to online (operational). Note: Do not use this function unless you are instructed to do so, for example during maintenance. This operation may cause loss of data. Forces the status of the hard disk to offline (failed). Note: Do not use this function unless you are instructed to do so, for example during preventive replacement of a hard disk or during maintenance. This operation may cause loss of data. Changes the status of the failed hard disk to available. Note: Do not use this function unless you are instructed to do so, for example during maintenance. Reassigns a hard disk to the disk group. Note: Do not use this function unless you are instructed to do so, for example during maintenance. Starts a rebuild for the selected hard disk. Stops the rebuild running for the selected hard disk. Note: Do not use this function unless you are instructed to do so, for example during maintenance. 140

MegaRAID SAS User's Guide Menu Manage hot spare table: Actions menu (when hard disk is selected) Function Sets or manages spare disks. For how to operate, refer to "4.8.4 Setting and Releasing Spare Disk" (pg.164). When logical drive is selected Menu Migrate logical drive Delete logical drive Locate logical drive Stop location Start MDC Cancel MDC Cancel BGI Start rebuild Cancel rebuild Start initialization Cancel initialization table: Actions menu (when logical drive is selected) Function Increases the capacity of the logical drive. Not supported. Do not use it. Turns on the hard disk failure LEDs of the hard disks that make up the selected logical drive to indicate their locations. Turns off the hard disk failure LEDs that are turned on with [Locate logical drive]. Starts Make Data Consistent for the selected logical drive. Stops Make Data Consistent running for the selected logical drive. Stops the initialization process running in the background. Even if the background initialization process is stopped, it will automatically be restarted after a certain period of time. Starts a rebuild for the selected logical drive. Stops the rebuild running for the selected logical drive. Note: Do not use this function unless you are instructed to do so, for example during maintenance. Not supported. Do not use it. Note: If you use this function, the data on the logical drive will be deleted. Please be careful. Not supported. Do not use it. When battery backup unit is selected Menu Recalibration table: Actions menu (when battery backup unit is selected) Function Starts the battery recalibration. Note: Note that a logical drive set to Write Back mode operates in Write Through mode during battery recalibration. When ServerView RAID Manager is selected table: Actions menu (when ServerView RAID Manager is selected) Menu Function Write configuration Not supported. Do not use this. 141

Chapter 4 Array Configuration and Management [ServerView RAID] When File log is selected Clear log Menu table: Actions menu (when File log is selected) Function Deletes the event log file. Note: Do not use it. Event log files are used when investigating failures. If you delete event log files, it may become difficult to investigate when failures occur. Help menu Menu Info about ServerView RAID Contents and index table: Help menu Function Displays the version information of ServerView RAID. Not supported. 4.6.3 Layout of Tree View This section explains the icons in the tree view displayed at the left of the main ServerView RAID Manager window. In the tree view, you can see the statuses of the hard disks and logical drives. Server Array controller Battery backup unit Hard disk Port Logical drive File log System log Plug-in 142

MegaRAID SAS User's Guide The screen will be as shown below when a SES (SCSI Enclosure Service) device and/or a SAS Expander are connected to the array controller. Array controller Port SAS Expander Hard disk SES device Fan unit under SES device Power unit under SES device Temperature sensor under SES device Server At the top of the tree, the name of the connected server is displayed. Array controller The array controller installed on the server is displayed. The controller number is displayed in ( ). Battery backup unit The battery backup unit installed on the array controller is displayed. This only applies to the array controller with battery. The status of battery unit is not supported in ServerView RAID Manager. Use the event logs to monitor the battery. The battery backup unit may be displayed with the ( ) icon during battery recalibration or battery charging, but this does not mean a failure. 143

Chapter 4 Array Configuration and Management [ServerView RAID] Port The port of the array controller is displayed. The port number is displayed in ( ). A port name does not indicate hard disk types connected, but port types. It always displays "SAS Port". When a number of ports are widely linked, the number of ports are consecutively displayed such as "SAS Port (0-1-2-3)." Hard disk The vendor name and product name of the hard disk connected to the array controller are displayed. The numbers displayed in ( ) are: a hard disk slot number such as (3), or an enclosure number and a slot number such as (1:4). You can also check the present status of the hard disk with icons. table: Hard disk status icon Icon Displayed in Status Description Gray Online (Operational) A part of the array and functioning properly White Unused (Available) Unused or available White with a "+" mark hotspare (Global hot spare) A spare disk Gray with an "x" mark Failed/Offline (Failed/Offline) A part of the array, but has failed or is unreadable/unwritable White with an "x" mark Failed (Failed) Not a part of the array, and has failed White (dotted frame) with an "x" mark Unrecognizable (Failed (missing)) Not recognized, or failed Gray Rebuild Rebuild in progress Gray with an "!" mark Failure Predicted (S.M.A.R.T. error) Failure expected Gray with yellow background Locating [Locate device] function running 144

MegaRAID SAS User's Guide White with a blue "!" mark table: Hard disk status icon Icon Displayed in Status Description Incorrect configuration information detected (Foreign configuration) Information about an old array or an externalarray configuration exists on the hard disk. This icon is displayed in the following cases: When a hard disk that has been used before is installed. When a hard disk that has failed once is detected again. If the hard disk is in the failed status at the same time as it is in this status, regard it as a failed hard disk. When you select the icon of each hard disk, more detailed information is displayed in the object window. Note that you cannot see some information for unrecognizable hard disks. For details, refer to "4.9.3 Checking Hard Disk Status" (pg.174). S.M.A.R.T. error and Foreign configuration may be displayed in combination with another status. Fan unit, power unit, and temperature sensor under SES device Monitoring these devices is not supported in ServerView RAID. Logical drive The logical drive created under the array controller is displayed. The logical drive number is displayed in ( ). You can also see the status of the logical drive. table: Logical drive status icons Icon Status Description Online Normal (Operational) Partially critical (Partially Degraded) Critical (Degraded) Offline (Failed) Operating with redundancy. But a hard disk has been failed. Operating without redundancy Not available Locating [Locate logical drive] function running Incorrect configuration information detected (Foreign configuration) This icon may be displayed when the hard disk in the "Foreign configuration" status exists. This displays the information in the hard disk and is not actual logical drive (cannot be seen from the OS). This icon may be displayed in the following cases: When a failed hard disk exists. When a hard disk that has been used before is connected. 145

Chapter 4 Array Configuration and Management [ServerView RAID] When you select the icon of the logical drive, more detailed information is displayed in the object window. For details, refer to "4.9.4 Checking Logical Drive Status" (pg.176). File log, system log, and plug-in You can configure the log files and the OS event log for ServerView RAID. For details, refer to "4.7.4 Setting ServerView RAID" (pg.153). 146

MegaRAID SAS User's Guide 4.7 Settings This section explains how to change each parameter for the array controllers, the logical drives, and the event logging. 4.7.1 Changing Array Controller Setting You can check and change the array controller settings. The array controller settings must be set within the ranges in the following table. You may only change items specified as default in the table. If the settings are outside of the specified ranges, correct operation may not be ensured. Apart from the array controller settings, you may be required to make various settings on the server itself. For how to set up the server, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select the array controller ( ). 3 Click the [Settings] tab in the object window. The current settings are displayed in the tab. 147

Chapter 4 Array Configuration and Management [ServerView RAID] Tasks 4 Click [Edit] to set each item. The setting window appears. For the items that can be set, refer to the following. When you set Tasks " Tasks" (pg.148) When you set Others " Others" (pg.149) 5 Click [OK]. A confirmation window appears. 6 Click [Apply]. You will see the change reflected on the [Settings] tab in the object window. You can primarily set the options associated with background tasks. table: The array controller Tasks Option Setting Description Init mode Normal initialization Cannot be changed Not supported. BGI rate [Note 1] Default Sets the background initialization priority for normal I/O. Migration rate [Note 1] Default Sets the Expand Capacity priority for normal I/O. MDC rate [Note 1] Default Sets the Make Data Consistent priority for normal I/O. Patrol Read mode Disabled Cannot be changed Not supported. Auto rebuild Enabled Cannot be changed Enables automatic rebuilds after hard disk replacements. Rebuild rate [Note 1] Default Sets the rebuild priority for normal I/O. [Note 1]: 80 for: MegaRAID SAS 300-8E ROMB, MegaRAID SAS 8300XLP, MegaRAID SAS 8408E, and MegaRAID SAS 8344ELP 30 for: MegaRAID SAS 8880EM2, RAID 5/6 SAS based on LSI MegaRAID Increasing values increases each background task performance although it may also decrease the rate of normal I/O. 148

MegaRAID SAS User's Guide Others table: The array controller Others Option Setting Description Alarm control Disabled Cannot be changed S.M.A.R.T. poll interval Spinup drive count 5 Cannot be changed 2 Cannot be changed Spinup delay 6 Cannot be changed Auto flush interval 4 Cannot be changed Coercion mode None Cannot be changed Cluster enable No Fixed Not supported. If a buzzer for notifying errors is installed in the array controller, this enables or disables the buzzer. Specifies a polling interval for S.M.A.R.T. detection. Sets the number of the hard disk units that start rotating at the same time. Specifies the rotation start interval between hard disks. Sets the data write interval from the cache memory to the hard disk. Sets a capacity rounding function to make varying hard disk capacities equal to each other. 149

Chapter 4 Array Configuration and Management [ServerView RAID] 4.7.2 Configuring Battery Backup Unit You can check and change the operation mode of the battery backup unit. This setting is automatically changed to the appropriate setting after the installation of a RAID management tool. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select the battery backup unit ( ). 3 Click the [Settings] tab in the object window. The current setting is displayed in the tab. 4 Click [Edit], and set "Disabled". table: Battery backup unit Misc.Properties Option Setting Description Auto learn mode Disabled Setting value during the operation This item sets the battery recalibration mode. Make sure to set it to "Disabled". 5 Click [OK]. A confirmation window appears. 6 Click [Apply]. You will see the change reflected on the [Settings] tab in the object window. 150

MegaRAID SAS User's Guide 4.7.3 Setting Logical Drive You can check and change the logical drives. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select a logical drive ( ). 3 Click the [Settings] tab in the object window. The current settings are displayed in the tab. 4 Click [Edit] to set each item. The setting window appears. For the items that can be set, refer to the following. " Cache" (pg.152) Setting a name for a logical drive in [Logical drive] is not supported. Do not use this item. 5 Click [OK]. A confirmation window appears. 6 Click [Apply]. You will see the change reflected on the [Settings] tab in the object window. 151

Chapter 4 Array Configuration and Management [ServerView RAID] Cache You can set options for the access method of logical drives. table: Logical drive Cache Option Setting Description Default write mode Write-through Default Sets Write Policy of the logical drive. The following modes can be selected: Write-through, Write-back, Always write-back (Independent of BBU state) (Bad BBU). For details about Write Policy, refer to "1.3.1 Write Policy" (pg.28). Default read mode No read-ahead Cannot be changed Not supported. Do not change them. Default cache mode Direct Cannot be changed Disk cache mode Disabled Cannot be changed Before you change [Default write mode], make sure that you understand Write Policy. Using the Writeback or Bad BBU without preventing power off due to blackout may cause data loss. For details, refer to "1.3.1 Write Policy" (pg.28). 152

MegaRAID SAS User's Guide 4.7.4 Setting ServerView RAID You can configure the log files and the OS event log for ServerView RAID. 1 Start ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select the icon you want. 3 Click the [Settings] tab in the object window. The current settings are displayed in the tab. When you select File log in the tree view When you select System log in the tree view 4 Click [Edit] to set each item. A setting window appears. For the items that can be set, refer to the following. When you set File log " File log" (pg.154) When you set System log " System log" (pg.155) 153

Chapter 4 Array Configuration and Management [ServerView RAID] ServerView RAID, Multiplexer and LSIStoreLib-Plugin are not supported, so do not use these items. 5 Click [OK]. A confirmation window appears. 6 Click [Apply]. You will see the change reflected on the [Settings] tab in the object window. File log table: File log Option Setting Description Logging Enabled Cannot be changed Log level Information Cannot be changed File name RAIDLog.xml Cannot be changed File size 10 Cannot be changed Enables event logging to the log file. Always use the program with this option "Enabled". Sets a priority level of events for which logging will be performed. Events that have a priority level equal to or greater than the one set here will be logged in the log file. Do not change the setting from "Information". Log file name. Sets the log file size limit. Log method Rotation Cannot be Sets the method of log rotation. changed Max. save files 5 Default Maximum number of log files. It is recommended that you set this value equal to or greater than the default value. 154

MegaRAID SAS User's Guide The log files for ServerView RAID are stored as RAIDLog.xml (or RAIDLog<number>.xml) in the following folders. OS Windows Server 2008 (32-bit), Windows Server 2003, Windows 2000 Server Windows Server 2008 (64-bit), Windows Server 2003 x64 Linux table: Destination for log files Destination for log files C:\Program Files\Fujitsu Siemens\RAID\web\public\ C:\Program Files (x86)\fujitsu Siemens\RAID\web\public\ /opt/fsc/raid/web/public/ System log Do not delete or edit the log files since they may be used when investigation is necessary. The maximum file size to be used ranges from 20 MB to 110 MB, depending on the settings. Browsing or monitoring the log files is not supported. table: System log Option Setting Description Logging Enabled Default Enables event logging to the OS event log. Log level Information Default Sets a priority level of events that are logged in the OS event log. Events that have a priority level equal to or greater than the one set here will be logged. "Debug" is not supported. ServerView RAID logs events that occur on the array controller in the OS application log as events of the "Source: ServerView RAID". However, when ServerView is installed, events that occur on the array controller are logged also as events of the "Source: Fujitsu ServerView Services". The event log for which the settings can be changed under [System Logging] is the one that contains the "Source: ServerView RAID". 155

Chapter 4 Array Configuration and Management [ServerView RAID] 4.8 Operating RAID Configuration This section explains the procedure for creating a new disk array configuration, expanding the capacity, and setting a hot spare. Disk array operations require logging in to ServerView RAID Manager with Administrator privileges. They also require that the access mode is set to "Write access". For details about access modes, refer to " Toolbar" (pg.138). 4.8.1 Creating Disk Array Configuration (Create logical drive) To create a new array configuration, or to add an array configuration in free space of a hard disk, follow the procedure below. During rebuild, the array configuration cannot be created or deleted. Perform these procedures after the rebuild is complete. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select the array controller ( ), right-click, and then click [Create logical drive] from the displayed menu. The [Parameters] tab of the [Create logical drive] window appears. 156

MegaRAID SAS User's Guide 3 Configure the logical drives. 1. Select a RAID level for the logical drive. Click of [RAID level], and select a RAID level for the logical drive from the displayed list. The configurable RAID levels are limited according to the number of the unused hard disk units and the free space in an existing disk group. When there is free space in a predefined disk group, you can only set the same RAID level as that of the logical drive that is already set in the disk group. (Logical drives with different RAID levels cannot be created within the same group.) Do not change the default stripe size (64 [K]). 2. Specify the capacity of the logical drive. When creating a single logical drive, a logical drive with RAID 10, or the last logical drive in the disk group In [Capacity of the logical drive], select [Use maximum capacity]. The maximum capacity of the logical drive is set as a capacity. When multiple logical drives are created in the disk group In [Capacity of the logical drive], select [Use only] and specify a suitable capacity. The default unit for entering the capacity of the logical drive is GB. If you set in MB, select [MB] from the pull-down menu (KB unit is not supported). The capacity of the logical drive may be automatically adjusted to match the block size. If multiple logical drives are created in a single disk group, you cannot expand the capacity of the logical drive with the Expand Capacity function later. If you create a RAID 10 logical drive, the capacity of the logical drive cannot be specified. 3. Set Write Policy of the logical drive. Click of [Write mode] in [General parameters], and select Write Policy from the displayed list. Before using Write Policy, refer to "1.3.1 Write Policy" (pg.28), and make sure you understand what it does. If there is no battery backup module, the logical drive is created in "Write-through" mode even if "Write-back" is enabled. To enable Write Policy if there is no battery backup module, set [Write mode] to "Always write-back (independent of BBU state)". 4. Set other parameters. Change the settings as follows if different. table: Parameter settings Parameter Setting Read mode No read-ahead Cache mode Direct Disk cache mode Disabled Init mode No initialization Name (Blank) 157

Chapter 4 Array Configuration and Management [ServerView RAID] 4 Click the [Layout] tab to allocate a hard disk to a logical drive. When you set the logical drive to Write Back or Bad BBU and click the [Layout] tab, the following confirmation message appears. If the settings are acceptable, click [Yes]. The following window appears. 1. In the left tree of the window, select an unused hard disk ( ) or a hard disk with free space ( ) that belongs to the existing disk group. 2. Click [Add]. The hard disk you select or add is displayed in [Current layout]. When selecting a hard disk with free space that is included in an existing disk group, all the hard disks in that group are additionally displayed in [Current layout]. 158

MegaRAID SAS User's Guide 3. Repeat the steps above to allocate as many hard disks as you need for the RAID level you set. table: RAID levels and the necessary number of the hard disk units RAID level The number of the hard disk units RAID 0 1 or more RAID 1 2 RAID 5 3 or more RAID 6 3 or more RAID 10 Even number equal to or larger than four Number of used blocks on the hard disks Number of start blocks on the hard disks Indicates that the selected hard disk is configuring the existing logical drive x. (In this figure, logical drive 0) Used capacity If "Available" is displayed in the [Used by] column after selecting the hard disk that belongs to the disk group in the tree, there is some free space in the hard disk. Use hard disks of the same capacity and type within a logical drive. If a mixture of different hard disks are connected, check them in advance by referring to "4.9.3 Checking Hard Disk Status" (pg.174). If you allocate a new hard disk to a logical drive, a disk group is automatically configured together with the logical drive. If there is a hard disk with free space in the existing disk group, you cannot allocate an unused hard disk to a logical drive. Use all the free space in the hard disks of the disk group before allocating an unused hard disk to the logical drive. If you allocate a hard disk with free space in the existing disk group to a logical drive, add all of the hard disks within that disk group. 159

Chapter 4 Array Configuration and Management [ServerView RAID] When removing hard disks from a logical drive In the [Current layout] area on the right side of the window, select a hard disk, and click [Remove]. 5 Click [Create]. If you cannot click [Create] because it is grayed out, the following causes are possible. Review the allocation of hard disks in Step 4. The number of hard disks is not enough for the selected RAID level. All of the hard disks within the existing disk group have not been added. A confirmation window appears. 6 Type "yes", and click [OK]. A new disk array configuration is created. 4.8.2 Expanding Capacity of Logical Drive (Migrate logical drive) You can expand the capacity and change the RAID level for the existing logical drives. There are certain conditions to note and precautions to take for Expand Capacity. Read "1.3.5 Expand Capacity" (pg.36) carefully before expanding the capacity. The capacity of RAID 10 logical drives cannot be expanded. 1 Back up the data to prepare for unexpected problems. 2 Start ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 3 Perform Make Data Consistent (pg.168) on a logical drive whose capacity is to be expanded, and check if it completes successfully. If Make Data Consistent is not complete successfully, do not expand the capacity of the logical drive. 160

MegaRAID SAS User's Guide 4 In the tree view, select the logical drive ( ), right-click, and then click [Migrate logical drive] from the displayed menu. The following window appears. 5 Select a RAID level to be set after Expand Capacity. Click of [RAID level], and select a RAID level for the logical drive from the displayed list. After Expand Capacity, the RAID level of the logical drive will be converted to the RAID level selected here. If you convert a RAID 1, 5 or 6 logical drive to RAID 0, note that the logical drive will lose its redundancy. 6 Click the [Layout] tab. 161

Chapter 4 Array Configuration and Management [ServerView RAID] 7 Add hard disks to the disk group. 1. In the left tree of the window, select the unused hard disk ( ). 2. Click [Add]. The hard disk you select or add is displayed in [Current layout]. 3. Repeat the steps above to add further hard disks. You can add hard disks until the total number of the hard disk units in the disk group displayed in [Current layout] reaches 32. Make sure that the number of the hard disk units in the disk group does not exceed 32. Use additional hard disks of the same capacity and type as the existing hard disks in the disk group. If a mixture of different hard disks are connected, check them in advance by referring to "4.9.3 Checking Hard Disk Status" (pg.174). When removing hard disks from a logical drive In the [Current layout] area on the right side of the window, select a hard disk, and click [Remove]. 8 Click [Apply]. A confirmation window appears. 9 Type "yes" and click [OK]. The Expand Capacity process starts. Wait until the process is complete. Do not turn off, reset or restart the server while Expand Capacity is in progress. Doing so may cause loss of data in the target logical drive. If the server is turned off during Expand Capacity, access to the hard disk is automatically resumed after the server restarts. In this case, wait until the hard disk access LED on the hard disk turns off to ensure that no access is occurred to the hard disk, redo the array configuration, and then restore the data backed up before the expansion. If the RAID level of a logical drive after Expand Capacity has data redundancy (RAID 1, RAID 5 or RAID 6), background initialization will be performed after the Expand Capacity is complete. 162

MegaRAID SAS User's Guide 4.8.3 Deleting Existing Logical Drive (Delete last logical drive) This section explains how to delete the existing logical drive. When you delete the logical drive, start deleting them from the last created one (the drive with the highest number). Do not delete them in any order than starting from the last created logical drive. It is recommended that you back up data on all of the logical drives as a safeguard against accidental data loss due to operational errors before you delete logical drives. For how to delete all logical drives (i.e. to clear an array configuration), refer to "4.8.6 Deleting Disk Array Configuration Information" (pg.167). If a background task is running, wait until the background task is complete before you start deleting logical drives. 1 Back up the data to prepare for unexpected problems. 2 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 3 In the tree view, select the array controller ( ) where the logical drive to be deleted is created, right-click, and then click [Delete last logical drive] from the displayed menu. Do not click [Delete all logical drive]. If you click [Delete all logical drive], all the logical drives are deleted at once. If you select a wrong logical drive, click [Cancel] in a confirmation window and start again from the beginning. A confirmation window appears. 4 Type "yes", and click [OK]. The last created logical drive is deleted. If you accidentally delete wrong a logical drive, recreate the array configuration, and then restore the data backed up before the deletion. 163

Chapter 4 Array Configuration and Management [ServerView RAID] If all the logical drives in a disk group are deleted, the disk group will also automatically be deleted, and the hard disks that composed the disk group will be marked as unused. 4.8.4 Setting and Releasing Spare Disk There are some conditions and notes to configure a spare disk. Make sure to read " Hot spare rebuild" (pg.33) before setting a hot spare. Setting spare disks This section explains how to set spare disks. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select an unused hard disk ( ), right-click, and then click [Manage hot spare] from the displayed menu. 3 Click [Create]. A confirmation window appears. 4 Click [Yes]. The icon of the unused hard disk changes to a spare disk icon ( ). 164

MegaRAID SAS User's Guide If a disk group exists that consists of hard disks with a capacity larger than that of the hard disk to be set as a spare, the following message appears when you click [Manage hot spare]. If you want to set the spare disk, click [Yes]. Use a spare disk of the same capacity and type as the hard disks in the disk group. When there are two or more disk groups that use different hard disk capacities, set a spare disk for each type of hard disk. Releasing spare disks This section explains how to release spare disks. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select the hard disk ( ) where the spare disk is to be released, right-click, and then click [Manage hot spare] from the displayed menu. 3 Click [Delete]. The confirmation widow appears. 165

Chapter 4 Array Configuration and Management [ServerView RAID] 4 Click [Yes]. The spare disk icon changes to an unused hard disk icon ( ). 4.8.5 Changing Write Policy For this array controller, you can set Write Policy for each logical drive. To change Write Policy of a logical drive, follow the procedure below. Before using Write Policy, refer to "1.3.1 Write Policy" (pg.28) and make sure you understand what it does. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select a logical drive ( ) whose Write Policy you want to change. 3 Click the [Settings] tab, and click [Edit] in [Cache]. The [Edit:LogicalDrive_n] ("n" stands for a number) window appears. The current Write Policy is displayed in [Default write mode]. The setting values for other than [Default write mode] are not allowed to be changed. Do not change them. 4 Click of [Default write mode], and select Write Policy you want to set from the displayed list. 166

MegaRAID SAS User's Guide 5 Click [OK]. The [Apply changed settings] window appears. 6 Click [Apply]. The Write Policy changes to the new mode. If you change [Default write mode] to "Write-back" or "Write cache bad BBU", the following warning message appears. If you want to enable the write cache, click [Yes]. 4.8.6 Deleting Disk Array Configuration Information Disk array configuration information refers to the information created when the disk array is configured, such as the disk group and logical drive capacity, the RAID level, and which hard disk belongs to which disk group. You can delete all the existing logical drives defined in the array controller and make all the connected hard disk unused by deleting the disk array configuration information. You should only delete disk array configuration information when instructed to do so during maintenance. All data on the logical drives will be lost by this operation. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 167

Chapter 4 Array Configuration and Management [ServerView RAID] 2 In the tree view, select the array controller ( ) whose disk array configuration information you want to delete, right-click, and then click [Clear configuration] from the displayed menu. A confirmation window appears. 3 Type "yes", and click [OK]. The existing disk array configuration information is deleted, and all logical drives are deleted. 4.8.7 Make Data Consistent Make Data Consistent, also called redundant data regeneration, is a function that regenerates the parity or mirrored data on redundant logical drives and makes the array configuration consistent. When a server failure or a blackout results in a dirty OS shutdown, the array may become inconsistent. When a dirty shutdown occurs, perform this function as soon as possible. Make Data Consistent can be performed on logical drives with redundancy such as RAID 1, RAID 5, RAID 6, and RAID 10 logical drives in the online status (operational), and a RAID 6 logical drive in the partially degraded status. Make Data Consistent cannot be performed on logical drives without redundancy such as RAID 0 logical drives or logical drives in the critical (degraded) or the offline (failed) status. In addition to keeping the data consistent in the array, Make Data Consistent also automatically corrects any media errors on the hard disks. Make Data Consistent can be performed simultaneously for several logical drives. To perform Make Data Consistent, follow the procedure below. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 168

MegaRAID SAS User's Guide 2 In the tree view, select a logical drive ( ) where Make Data Consistent is to be performed, right-click, and then click [Start MDC] from the displayed menu. A confirmation window appears. 3 Click [Yes]. When Make Data Consistent starts, [MDC running] in [Activity] of the object window and the progress of the Make Data Consistent are displayed. When the progress bar reaches 100% and the window closes, the check is complete. 4.8.8 Rebuild Rebuild is the operation to recover a logical drive in the critical status (degraded) to the online status (operational). For details about the rebuild, refer to "1.3.3 Rebuild" (pg.33). To perform a manual rebuild, follow the procedure below. Usually, a rebuild is automatically performed after replacing the hard disk. For how to replace the hard disk and how to perform a rebuild, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select a logical drive in the degraded status ( ), right-click, and then click [Start rebuild] from the displayed menu. A confirmation window appears. 3 Click [Yes]. When the rebuild starts, [Rebuilding] in [Activity] of the object window and the progress of the rebuild are displayed. When the progress bar reaches 100% and the window closes, the rebuild is complete. 169

Chapter 4 Array Configuration and Management [ServerView RAID] 4.8.9 Recalibration Recalibration is a process to recalibrate the battery capacity. To perform recalibration manually, follow the procedure below. Perform battery recalibration when: You purchased an array controller with a battery backup unit installed in. You additionally purchased a battery backup unit and install it into the array controller. You replaced parts including a battery. Battery recalibration requires 12 hours with the server power on. If the server is turned off and on during battery recalibration, the recalibration restarts from the beginning. During recalibration, the logical drive in Write Back mode shifts automatically to Write Through mode for safety. When the battery recalibration is complete, Write Policy is automatically back to Write Back. 1 Start up ServerView RAID Manager, and log in with Administrator privileges. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, select a battery backup unit ( ), right-click, and then click [Recalibration] from the displayed menu. The recalibration starts. 170

MegaRAID SAS User's Guide 4.9 Checking Each Status [ServerView RAID Manager] The following information can be checked by using ServerView RAID Manager. Information about events or errors that have occurred: "Appendix A List of ServerView RAID Event Log" (pg.276) Server information: "4.9.1 Checking the Server Condition" (pg.171) Array configuration or the array controller information: "4.9.2 Checking Array Controller Status" (pg.172) Hard disk information: "4.9.3 Checking Hard Disk Status" (pg.174) Logical drive information: "4.9.4 Checking Logical Drive Status" (pg.176) Information about tasks running in the background: "4.9.5 Checking Progress of Background Task" (pg.178) 4.9.1 Checking the Server Condition Selecting a server name displayed on the top of the tree view displays information about the server on which ServerView RAID is installed. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 Click the name of the server you want to check in the tree view. The detailed information about the selected server is displayed. table: Detailed information about servers Item Category Description System Name Server name. IP Address Displays the server IP address. The loopback address (127.0.0.1) may be displayed depending on the server network settings. Fully Qualified Domain Name Server FQDN. 171

Chapter 4 Array Configuration and Management [ServerView RAID] table: Detailed information about servers Item Category Description Operation System Vendor OS vendor name. Version OS type and version. Edition OS edition. Service pack Service pack applied to OS. Processor architecture CPU type of OS. [Note]: All of the items may not be displayed depending on the OS type or settings. 4.9.2 Checking Array Controller Status Selecting the array controller in the tree view displays detailed information about the array controller in the object window. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 Click the array controller ( ) you want to check in the tree view. The detailed information about the selected array controller is displayed in the [General] tab. 172

MegaRAID SAS User's Guide table: Detailed information about the array controller Item Category Description Adapter Name The model name of the array controller. ID The array controller number. Vendor The vendor name of the array controller. Product The product name of the array controller. Serial number The serial number of the array controller. Memory size The cache memory size of the array controller. FlashROM size The size of the array controller's FlashROM. NVRAM size The size of the array controller's NVRAM. Firmware package version The version included the array controller's firmware and BIOS, etc. Firmware version The version of the array controller's firmware. Firmware build time The build date and time of the array controller's firmware. BIOS version The version of the array controller's BIOS. Driver version The version of the array controller's driver. PCI Location Bus The bus number for the array controller. Function The function number for the array controller. Device The device number for the array controller. Ports Number The number of ports on the array controller. Protocol The protocol supported by the array controller. Devices(max) Physical Logical maximum number of physical devices. Logical Logical maximum number of logical devices. Properties S.M.A.R.T. support Support status of S.M.A.R.T. Alarm present Indicates whether the alarm is present or not. Inconsistency handling Operation method for data consistency check. For this array controller, MDC (Make Data Consistent) is used. Correctable errors These counters are not supported. Uncorrectable errors These counters indicate the number of recovery attempts made by the array controller for temporary or minor errors. They can be ignored unless the hard disk is in the failed status. Cluster active This is not supported. Status The current status of the array controller is displayed. Additional Status Displays the additional information on the current state of the array controller. Regarding the information displayed in the [Settings] tab, refer to "4.7.1 Changing Array Controller Setting" (pg.147). The [Background activities] tab appears if a background task is running. Regarding the information displayed in the [Background activities] tab, refer to " Checking with [Background activities] of array controller" (pg.179). 173

Chapter 4 Array Configuration and Management [ServerView RAID] 4.9.3 Checking Hard Disk Status Detailed information about hard disks connected to the array controller is displayed in the object window. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 Click the hard disk ( ) you want to check in the tree view. The detailed information about the selected hard disk is displayed. When selecting the [General] tab 174

MegaRAID SAS User's Guide table: Detailed information about hard disks Item Category Description Disk Name Hard disk model name. The number in ( ) indicates a hard disk slot number, or an enclosure number and a slot number. Device number The device number for the hard disk. Enclosure number The enclosure number of the hard disk that it is installed in. It is only displayed to the hard disks connected to a SAS Expander. Slot The slot number where the hard disk is installed. Vendor The hard disk vendor name. Product The hard disk product name. Type The protocol type of the hard disk. SAS address 00 The SAS address of the hard disks. Serial number The serial number of the hard disk. Firmware version The version of the hard disk's firmware. Transfer width The data transfer width of the hard disk is displayed. Physical size The physical capacity of the hard disk. Config. size The hard disk's available capacity when connected to the array controller. Foreign configuration Indicates whether configuration information significantly different from the current information is present or not. Usually, "No" is displayed. When there is a failed hard disk, "Yes" may be displayed. Misc errors Not supported. This counter indicates the number of recovery attempts made by the array controller for temporary or minor errors. There is no problem unless the hard disk is in the failed status. S.M.A.R.T. errors The counter for S.M.A.R.T. failure predictions for the hard disk. Media errors Not supported. This counter indicates the number of recovery attempts made by the array controller for temporary or minor errors. There is no problem unless the hard disk is in the failed status. Activity The running tasks for the hard disk are displayed. Idle: No tasks are running. Rebuilding: A rebuild is in progress Estimated time remaining Status This category is displayed only while a background task is running. For details, refer to "4.9.5 Checking Progress of Background Task" (pg.178). The current status of the hard disk is displayed. 175

Chapter 4 Array Configuration and Management [ServerView RAID] When selecting the [Layout] tab Item Config. size Usage table: Layout information about hard disks Description Available hard disk capacity when connecting to the array controller. Logical drive information that the hard disk is allocated to. 4.9.4 Checking Logical Drive Status Detailed information about logical drives is displayed in the object window. You can change the displayed information by switching the tabs at the top of the object window. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, click the logical drive you want to browse ( ). Detailed information about the selected logical drive is displayed in the object window. 3 Click the tab you want to check in the object window. When you select the [General] tab 176

MegaRAID SAS User's Guide table: Detailed information about logical drives Item Category Description Logical drive Name The name of the logical drive. Logical drive number Logical drive number. RAID level The RAID level set for the logical drive. Stripe size The size of the striping used by the logical drive. Logical size The logical size of the logical drive. Physical size The physical size of the logical drive. Operating system device name The hard disk location of the logical drive in the OS. Initialization status Activity Estimated time remaining The status of the initialized logical drive: Initialized: Already initialized Uninitialized: Not initialized Since the array controller supports background initialization, logical drives are automatically initialized even if the status is "Uninitialized". In this case, there is no problem. The running tasks for the logical drive. When background tasks are running, the running tasks and their progress are displayed on the progress bar. For details, refer to "4.9.5 Checking Progress of Background Task" (pg.178). This is displayed only while a background task is running. For details, refer to "4.9.5 Checking Progress of Background Task" (pg.178). The current status of the logical drive. Status Cache Write mode The current Write Policy setting of the logical drive. Read mode Cache mode These are not supported. Disk cache mode When you select the [Layout] tab Item Disk Used capacity Partitions table: Logical drive layout information Description The hard disks that compose the target logical drive are displayed. The capacity of hard disks used by the target logical drive. The partition information of the target logical drive. 177

Chapter 4 Array Configuration and Management [ServerView RAID] Regarding the information displayed in the [Settings] tab, refer to "4.7.3 Setting Logical Drive" (pg.151). 4.9.5 Checking Progress of Background Task There are four types of background tasks: initialization, rebuild, Make Data Consistent, and Expand Capacity. ServerView RAID Manager enables you to view the progress of these tasks with progress bars. The progress bar displays the approximate time for the task to complete (only when using ServerView RAID v2.1.x or later). The background tasks whose progress can be checked are as follows. table: Background task Background task Activities How to check background tasks Background initialization BGI Checking with detailed information about logical drives Make Data Consistent MDC running Checking with [Background activities] of the array controller Rebuild Rebuilding Checking with detailed information about hard disks Expand Capacity Migrating (rebuild only) Checking with detailed information about logical drives With this procedure, you can check the background tasks being performed on the selected logical drive. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, click the logical drive ( ). 3 Click the [General] tab. You can check the type and progress of the background tasks being performed in [Activity]. 178

MegaRAID SAS User's Guide If rebuilds are simultaneously being performed for the multiple hard disks in a RAID 10 configuration, the progress displayed in [Activity] of the logical drive is the average value for each rebuild progress of the hard disks. When using ServerView RAID v2.1.x or later, the approximate time to complete the background task is displayed in [Estimated time remaining]. However, use this time only as a guideline since the actual time may be different depending on the system load. Checking with [Background activities] of array controller With this procedure, you can check all the background tasks being performed on the selected array controller. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, click the array controller ( ). 3 Click the [Background activities] tab. The logical drives where background tasks are currently being performed are displayed in [Logical drive], and you can check the types and progress of the background tasks being performed in [Activity]. When you click the button ( ) to the right of the information for each logical drive, the target logical drive is selected in the tree view, so that you can view the detailed information about that logical drive. While a rebuild is in progress, both the progress for the logical drive and the progress for the hard disk are displayed. When using ServerView RAID v2.1.x or later, the approximate time to complete the background task is displayed in [Activity]. However, use this time only as a guideline since the actual time may be different depending on the system load. 179

Chapter 4 Array Configuration and Management [ServerView RAID] Checking with detailed information about hard disks (rebuild only) With this procedure, you can check the rebuild being performed on the selected hard disk. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, click a hard disk being rebuilt ( ). 3 Click the [General] tab. You can check the progress of the rebuild being performed in [Activity]. When using ServerView RAID v2.1.x or later, the approximate time to complete the background task is displayed in [Estimated time remaining]. However, use this time only as a guideline since the actual time may be different depending on the system load. 180

Chapter 5 Array Configuration and Management [GAM] This chapter explains an overview of and requirements for, and how to install and use Global Array Manager (GAM). 5.1 Overview of and Requirements for GAM............. 182 5.2 Installing GAM [Windows]...................... 189 5.3 Installing HDD Check Scheduler [Windows]......... 194 5.4 Installing Battery Recalibration Scheduler [Windows].... 198 5.5 Starting and Exiting GAM......................... 202 5.6 GAM Window Layout........................... 204 5.7 Settings..................................... 213 5.8 Operating RAID Configuration..................... 218 5.9 Viewing Information [GAM]....................... 234 5.10 Maintenance Functions.......................... 245 181

Chapter 5 Array Configuration and Management [GAM] 5.1 Overview of and Requirements for GAM This section explains Global Array Manager (GAM). GAM is used to monitor, manage, maintain, and configure an array controller, and hard disks and logical drives that are connected to the array controller. 5.1.1 Overview of GAM GAM is an application software that monitors and manages the array controller on the OS. GAM also allows you to manage a disk array system connected to an array controller (RAID controller). The GAM functions are provided by the interaction between GAM Server and GAM Client. GAM Server (monitoring function) GAM Server monitors, collects, and notifies information about the status and the resource usage of the disk array. GAM Client (management function) GAM Client provides fault management, highly reliable messaging, and excellent OS support. You can manage the maintenance of disk arrays and hard disks from the server or from a client PC connected to the network. To ensure stable operation of PRIMERGY, install GAM (or ServerView RAID) when using the array controller. If GAM (or ServerView RAID) is not installed, failures will not be detected. You will also be unable to perform maintenance correctly. Install either GAM or ServerView RAID, but do not install both. For which to use, refer to ReadmeEN.html in Array Controller Document & Tool CD. If, by mistake, you have installed ServerView RAID rather than GAM on the server where GAM is only acceptable, uninstall ServerView RAID and HDD Check Scheduler, and then install GAM and HDD Check Scheduler. Do not operate the server with both ServerView RAID and GAM installed. Start up GAM Client only when management or maintenance of arrays is necessary. From a security point of view, it is not recommended to run GAM Client continuously. While signing on from GAM Client, if the accessed server shuts down, GAM Client cannot communicate with GAM Server and cannot respond. This prevents the user from operating it. In that case, wait for the GAM Server that is to be accessed to start up again, or forcibly exit GAM Client. 182

MegaRAID SAS User's Guide 5.1.2 Requirements for GAM An appropriate server environment is required for GAM Server and GAM Client. Use hardware and software that meet the following conditions. GAM Server You must have the following environment to use GAM Server. Item Hard disk Application OS table: Requirements for GAM Server Contents For Windows 64 MB or more free space in installation partition of the OS For Linux 64 MB or more free space under /usr and /var TCP/IP, SNMP service, and ServerView must be installed. OS supported by servers with this array controller installed Apply the latest Service Pack for the OS. Make sure to install the device drivers and GAM specified by Fujitsu. Make sure to configure the network settings properly. If there is a problem with the network configuration, you may not be able to monitor the status of arrays by ServerView or events may not be notified. When you change a host name or an IP address of the server with a GAM Server installed on it, uninstall the GAM Server after the change and install it again. GAM Client (when managed from a client PC) When GAM Client is installed on a client PC different from the server, the following environment is required for the client PC. Item Network Input device Processor Memory Hard disk Monitor table: Requirements for GAM Client Contents Network connection with available TCP/IP A mouse or other pointing device Pentium or later 256 MB or more 32 MB or more free space 800 600 or better resolution (1024 768 or more recommended) OS Windows Server 2003 Windows XP Windows 2000 Service Pack 4 or later Windows 2000 Professional Service Pack 4 or later 183

Chapter 5 Array Configuration and Management [GAM] 5.1.3 Access Privileges to GAM To use the GAM functions, you must sign on to GAM. User authentication is based on the user accounts registered in the OS. Note that the available functions vary depending on the user account used to sign on. There are three levels of access privileges as described below. Guest privileges When using GAM with Guest privileges, it is not necessary to sign on. With Guest privileges, only the RAID status and occurring events can be checked. It is not possible to set or change parameters. User privileges User privileges are mainly used to monitor statuses of controllers, hard disks, and logical drives. To use User privileges, sign on with a user name and passwords registered on the OS. With User privileges, in addition to the functions available with Guest privileges, several parameters can be changed. You can also view the detailed status of the selected controller and RAID subsystem. Note that you cannot manage operations such as changing disk array configurations, rebuilding drives, and changing parameters related to controllers and drivers. RAID cannot be configured with User privileges. It is recommended that you sign on with User privileges when only monitoring RAID or only checking its status. Administrator privileges Administrator privileges are used for management, maintenance, and configuration of controllers, hard disks, and logical drives. To use Administrator privileges, sign on as "gamroot". In addition to the monitoring functions available with Guest and User privileges, it is possible to use all other functions including creating/changing a RAID configuration, rebuilding drives, making logical drive data consistent, and changing a drive status. When using GAM with Administrator privileges, certain operations may cause loss of data in the array controller. Read this chapter and use GAM carefully. If GAM information cannot be monitored from ServerView, the network settings may be incorrect. In this case, check the network settings again. 184

MegaRAID SAS User's Guide 5.1.4 Using GAM in Linux Environment To use GAM in a Linux environment, you need to install device drivers and GAM. For Linux, refer to the PRIMERGY page on the Fujitsu website (http://primergy.fujitsu.com/) GAM Client can only be installed on servers or PCs running Windows. Even when monitoring the array controller on Linux servers using GAM Client, GAM Client cannot be installed on Linux servers. Prepare a Windows server or client PC and install GAM Client on it. The following figure shows a system configuration that GAM Client on a Windows server or on a client PC manages a Linux server. Windows server or Client PC Monitoring and setting GAM Linux server GAM Server Notifying GAM event GAM Client Array controller On Linux servers, you need to install GAM Server and edit the configuration file to set the GAM event recipient and enable storing event logs after the installation. For details, refer to the "Installation Guide" for Linux at the above URL. If there is an intervening firewall in the environment, you need to configure the network settings so that the port used by the GAM protocol is not blocked. GAM uses the TCP port 157 and 158. 185

Chapter 5 Array Configuration and Management [GAM] 5.1.5 Using GAM in Network Environment In a network environment, arrays on multiple servers can be monitored and managed from a Windows client connected to the network. The following figure shows a system configuration in which GAM Client on the other Windows client manages GAM Server. Windows server Monitoring and setting GAM GAM Server Array controller Windows server or Client PC GAM Client Notifying GAM event Linux server Monitoring and setting GAM Notifying GAM event GAM Server Array controller Monitoring and setting GAM Notifying GAM event Windows server GAM Server Array controller You need to configure the server receiving GAM events during GAM installation. For details, refer to Step 12 in "5.2.1 Installing GAM" (pg.189). If there is an intervening firewall in the environment, you need to configure the network settings so that the port used by the GAM protocol is not blocked. GAM uses the TCP port 157 and 158. One GAM Client can manage up to a maximum of 100 GAM Servers. When managing more than 100 servers at the same time, one Windows server or one client PC to be used as GAM Client is necessary per 100 servers. If multiple versions of GAM Server coexist, use the version of GAM Client that corresponds to the latest version of GAM Server, or a later version. 186

MegaRAID SAS User's Guide Interaction with ServerView AlarmService The following figure shows the interaction between ServerView and AlarmService when GAM Client on the other Windows server (client) manages GAM Server. OS event logs can be stored both on the GAM Server and on the Windows server (client). When storing OS event logs on the GAM Client Server/PC Windows server or Client PC Monitoring and setting GAM Windows server or Linux server GAM Server GAM Client ServerView AlarmService OS event log Notifying GAM event (SNMP Trap) SNMP Service Notifying GAM event (SNMP Trap) Array controller ServerView AlarmService OS event log ServerView must be installed on the GAM Client server/pc as well. Make sure to install ServerView on the server. 187

Chapter 5 Array Configuration and Management [GAM] 5.1.6 When Using Both ServerView RAID and GAM When using both ServerView RAID and GAM on a network, you can use ServerView RAID Manager and GAM Client simultaneously on the management client PC or Windows server. You can use the following configuration. Management client PC or Windows server Notifying events Monitoring and setting the disk array Server Array controller ServerView RAID service GAM Client Monitoring and setting the disk array Server ServerView RAID Manager GAM Server Notifying GAM events Array controller If ServerView is installed on the management client PC or Windows server, you can start ServerView RAID Manager or the GAM Client program for a server managed by ServerView by making use of ServerView's interaction with RAID Manager. For details about the RAID Manager linking, refer to "RAID Manager Linking" of "ServerView User's Guide". When you install GAM Client on a Windows server where ServerView RAID is already installed, install only GAM Client. Do not install GAM Server. 188

MegaRAID SAS User's Guide 5.2 Installing GAM [Windows] This section explains how to install GAM on a Windows server. You cannot reinstall GAM over itself. Make sure to uninstall any existing version of GAM before reinstalling GAM. Restart the OS after installing or uninstalling GAM. If Service Pack 1 of Windows Server 2003 has been applied, the following message may pop up when restarting the system just after installing or uninstalling GAM. A problem has been detected and Windows has shut down to prevent damage to your computer. Name: SNMP Service There will be no problem with operations. Click [Close] to close the message. To log events in OS event logs, make sure to install ServerView and configure the event-logging settings. For details, refer to "ServerView User's Guide". 5.2.1 Installing GAM Follow the procedure below to install GAM. 1 Log on to Windows with Administrator privileges. 2 Before installing GAM, complete the following preparations: Check that TCP/IP is installed and working properly. Check that ServerView is installed and working properly. Insert Array Controller Document & Tool CD supplied with the array controller into the CD/ DVD drive. Exit all application software. Exit all application software before starting the installation. In particular, if you install the software while [Event Viewer] or [Computer Management] is running, the installation may fail. 3 Click [Start] [Run...]. Enter the following path, and click [OK]. [CD/DVD drive]:\raidtool\gam\windows\install.bat The [Global Array Manager Setup] wizard starts up. 4 On the [Welcome] screen, click [Next]. The [Software License Agreement] screen appears. 189

Chapter 5 Array Configuration and Management [GAM] 5 Click [Yes]. The [Select Components] screen appears. Make sure the boxes next to [Global Array Manager Server] and [Global Array Manager Client] are checked. 6 Select [Global Array Manager Server], and click [Change]. The [Select Sub-components] screen appears. Make sure [Program Files] and [SNMP] are checked. 7 Confirm the settings, and click [Continue]. The [Select Components] screen appears again. 8 Click [Next]. The [Choose Destination Location] screen appears. 9 Click [Next]. The installation location for GAM appears. If GAM is already installed, a warning message will appear to confirm overwriting. After clicking [Cancel] to close the warning message box, click [Cancel] and then [Exit Setup] to quit the GAM setup. If the command prompt appears, click the [X] button to close the command prompt. After uninstalling GAM Client and GAM Server, perform the installation again. 190

MegaRAID SAS User's Guide 10 Confirm the installation location, and click [Next]. The files are copied. 11 Specify the client receiving events from GAM Server. In the text box, enter the name of the computer where GAM Client is being installed, and click [Next]. The [Server Event Logging] screen appears. If GAM Client is installed on the same computer as GAM Server, enter the name of the server. To specify multiple clients receiving events, enter the servers' computer names or IP addresses separated by spaces. You can specify up to 25 clients receiving events at a time. If the IP address or computer name of the client is changed after GAM Server has been installed, events cannot be correctly notified. In this case, GAM Server needs to be first uninstalled and then reinstalled. If the IP address is automatically obtained from the DHCP server, the IP address may be changed depending on the timing when the system is turned on/off or restarted. If DHCP is being used, it is recommended that you specify a computer name. 12 Make sure [Enable event logging on the server machine] is checked, and click [Next]. This option must be enabled. 13 When the full path name of the configuration file appears, click [OK]. The [Setup Complete] screen appears. 14 Click [Finish] to exit. Follow the message displayed at the command prompt, press the [Enter] key and carry on processing until the command prompt closes. 15 Restart the system. 191

Chapter 5 Array Configuration and Management [GAM] 16 After the restart, create the user account "gamroot" with GAM Administrator privileges and a user account with GAM User privileges (e.g. gamuser) as Windows user accounts. Assign the user account "gamroot" to the Administrators group. When creating the account with GAM Administrator privileges, uncheck the [User must change password at next logon] checkbox. Also check the [Password never expires] checkbox. If you do not make the above settings, you may be unable to sign on to GAM. Create each user account as an OS user account. 5.2.2 Uninstalling GAM Follow the procedure below to uninstall GAM. Uninstall GAM only when reinstalling or updating it. Do not operate the server without GAM. Uninstalling GAM Client 1 Log on to Windows with Administrator privileges. Exit all programs before starting the uninstallation. If you uninstall the software while [Event Viewer] or [Computer Management] is running, the uninstallation will fail. Make sure to exit all programs. 2 Click [Start] [Control Panel]. 3 Double-click [Add or Remove Applications] (or [Add or Remove Programs] depending on the OS). 4 Select [Mylex Global Array Manager Client v.x.xx-xx] from the application list, and click [Change/Remove]. The message "Are you sure you want to completely remove 'Mylex Global Array Manager Client vx.xx-xx' and all of its components?" appears. 5 Click [Yes]. The uninstallation process starts. 6 When the uninstallation is complete, click [OK]. 192

MegaRAID SAS User's Guide Uninstalling GAM Server 1 Log on to Windows with Administrator privileges. Exit all programs before starting the uninstallation. If you uninstall the software while [Event Viewer] or [Computer Management] is running, the uninstallation will fail. Make sure to exit all programs. 2 Click [Start] [Control Panel]. 3 Double-click [Add or Remove Applications] (or [Add or Remove Programs] depending on the OS). 4 Select [Mylex Global Array Manager Server v.x.xx-xx] from the application list and click [Change/Remove]. The message "Are you sure you want to completely remove 'Mylex Global Array Manager Server vx.xx-xx' and all of its components?" appears. 5 Click [Yes]. The uninstallation process starts. 6 When the uninstallation is complete, click [OK]. 7 Select [LSI 1030 Storage SNMP Agent] from the application list, and click [Remove]. The message "Are you sure you want to remove LSI 1030 Storage SNMP Agent vx.xx-xx from your computer?" appears. 8 Click [Yes]. The uninstallation process starts. 9 Select [LSI SWR IDE Storage SNMP Agent] from the application list, and click [Remove]. The message "Are you sure you want to remove LSI SWR IDE Storage SNMP Agent vx.xx-xx from your computer?" appears. 10 Click [Yes]. The uninstallation process starts. 11 Restart the system. 193

Chapter 5 Array Configuration and Management [GAM] 5.3 Installing HDD Check Scheduler [Windows] HDD Check Scheduler is a tool to detect or recover media errors on hard disks by periodically performing Make Data Consistent of the media. This section explains how to install HDD Check Scheduler on a Windows server. Make sure to install HDD Check Scheduler. Make sure to install GAM before installing HDD Check Scheduler. When updating GAM (or reinstalling GAM after uninstalling it), make sure to reinstall HDD Check Scheduler after updating GAM. During the HDD check, the I/O performance for the logical drives decreases to some extent (up to about 30%). After the installation, set a start time that minimizes the effect on the business by referring to "5.3.2 Changing HDD Check Start Time" (pg.195). When you install HDD Check Scheduler, the check is set to start every day at 12:00 by default. For an overview of the HDD Scheduler functions, refer to "1.3.7 HDD Check Scheduler" (pg.38). HDD Check Scheduler performs Make Data Consistent periodically using the standard OS scheduling function ("AT" command for Windows, or Cron for Linux). It is recommended that you perform the HDD check, at least once a week. If the server restarts or shuts down during HDD check, the HDD check is resumed from where it was stopped the next time. 5.3.1 Installing HDD Check Scheduler Follow the procedure below to install HDD Check Scheduler. 1 Log on to Windows with Administrator privileges. 2 Before installing HDD Check Scheduler, complete the following preparations: Check that GAM is installed. Check that ServerView is installed and working properly. Insert Array Controller Document & Tool CD supplied with the array controller into the CD/ DVD drive. Close all applications. Close all applications before starting the installation. In particular, if you install the software while [Event Viewer] or [Computer Management] is running, the installation may fail. 194

MegaRAID SAS User's Guide 3 Click [Start] [Run...]. Enter the following file name, and click [OK]. [CD/DVD drive]:\raidtool\hddcs\windows\install.bat The message "Installing HDD Check Scheduler." appears at the command prompt. 4 When the message "Installation of HDD Check Scheduler has finished." appears, press the [Enter] key to close the command prompt. When you install HDD Check Scheduler for the first time, the HDD check is set to start every day at 12:00 by default. If you reinstall HDD Check Scheduler over itself, the previously start time setting will be kept. 5.3.2 Changing HDD Check Start Time The start time for the HDD check can be changed by using the AT command of the OS. To change the start time, follow the procedure below. It is recommended that you perform the HDD check periodically, at least once a week. 1 Log on to Windows with Administrator privileges. 2 Click [Start] [All Programs] [Accessories] [Command Prompt] to open the command prompt. 3 Type "AT", and press the [Enter] key. 4 Write down the [ID] value for the row with the following value in the [Command Line] column. For Windows Server 2003 x64 C:\WINDOWS\SysWOW64\GAMSERV\HDDchk.bat For Windows Server 2003 C:\WINDOWS\system32\GAMSERV\HDDchk.bat For Windows 2000 Server C:\WINNT\system32\GAMSERV\HDDchk.bat Example: If the AT command output is as follows, the ID is 11. C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 1 Each M F 10:00 AM "C:\ProgramFiles\ Fujitsu\F5FBFE01\ServerView Services\WebServer\ClearMyLogs.exe" 11 Each M T W Th F 12:00 PM C:\WINDOWS\system32\ GAMSERV\HDDchk.bat 195

Chapter 5 Array Configuration and Management [GAM] 5 Delete the currently registered HDD check schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 11 /delete 6 Register a new schedule. By using the AT command, register the following command line together with the time of execution. For Windows Server 2003 x64 C:\WINDOWS\SysWOW64\GAMSERV\HDDchk.bat For Windows Server 2003 C:\WINDOWS\system32\GAMSERV\HDDchk.bat For Windows 2000 Server C:\WINNT\system32\GAMSERV\HDDchk.bat To specify the scheduled execution, enter the following AT command. AT <Start Time> /every:<day of Week> "Command Line" To perform the HDD check every Friday at 18:00, on Windows Server 2003 (except for Windows Server 2003 x64) at 18:00 /every:f "C:\Windows\system32\GAMSERV\HDDchk.bat" To perform the HDD check every day at 21:00, on Windows 2000 Server at 21:00 /every:m,t,w,th,f,s,su "C:\WINNT\system32\GAMSERV\ HDDchk.bat" To specify the day of the week, use the following abbreviations. Monday Tuesday Wednesday Thursday Friday Saturday Sunday M T W Th F S Su For details on how to use the AT command, refer to Windows Help. To see whether the HDD check is properly performed at the specified time, check that the log entry below is recorded in the OS application log at the specified time. If no log entries below are recorded at the specified time, review the settings. Source : Fujitsu ServerView Services Type : Information EventID : 1 Description: Consistency check / Make Data Consistent is started. 196

MegaRAID SAS User's Guide 5.3.3 Uninstalling HDD Check Scheduler Follow the procedure below to uninstall HDD Check Scheduler. 1 Log on to Windows with Administrator privileges. Exit all programs before starting the uninstallation. If you uninstall the software while [Event Viewer] or [Computer Management] is running, the uninstallation will fail. Make sure to exit all programs. 2 Click [Start] [All Programs] [Accessories] [Command Prompt]. The command prompt window opens. 3 Type "AT", and press the [Enter] key. 4 Write down the [ID] value for the row with the following value in the [Command Line] column. For Windows Server 2003 x64 C:\WINDOWS\SysWOW64\GAMSERV\HDDchk.bat For Windows Server 2003 C:\WINDOWS\system32\GAMSERV\HDDchk.bat For Windows 2000 Server C:\WINNT\system32\GAMSERV\HDDchk.bat Example: If the AT command output is as follows, the ID is 11. C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 1 Each M F 10:00 AM "C:\Program Files\ Fujitsu\F5FBFE01\ServerView Services\WebServer\ClearMyLogs.exe" 11 Each M T W Th F 12:00 PM C:\WINDOWS\system32\ GAMSERV\HDDchk.bat 5 Delete the currently registered HDD check schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 11 /delete 197

Chapter 5 Array Configuration and Management [GAM] 5.4 Installing Battery Recalibration Scheduler [Windows] Battery Recalibration Scheduler is a tool to periodically perform battery recalibration for MegaRAID SAS 300-8E ROMB equipped with a battery. This section explains how to install Battery Recalibration Scheduler on a Windows server. When updating GAM (or reinstalling GAM after uninstalling it), make sure to reinstall Battery Recalibration Scheduler after updating GAM. Before using this tool, also refer to "1.3.8 Battery Backup Unit (BBU)" (pg.39). Recalibration requires 12 hours with the server power on. During recalibration, the logical drive in Write Back mode shifts automatically to Write Through mode for safety. Therefore, refer to "5.4.2 Changing Battery Recalibration Start Date and Time" (pg.199) to set a start date and time that minimizes the effect on the business. MegaRAID SAS 8300XLP does not require the installation of this tool or periodical recalibration. Battery Recalibration Scheduler performs recalibration periodically using the standard OS scheduling function (AT command for Windows, or Cron for Linux). It is recommended that you perform battery recalibration about once a month or every two months. If the server restarts or shuts down during battery recalibration, the battery recalibration is resumed from where it was stopped the next time. 5.4.1 Installing Battery Recalibration Scheduler Follow the procedure below to install Battery Recalibration Scheduler. 1 Log on to Windows with Administrator privileges. 2 Before installing Battery Recalibration Scheduler, complete the following preparations: Check that GAM is installed. Make sure that ServerView is installed and working properly. Insert Array Controller Document & Tool CD supplied with the array controller into the CD/ DVD drive. Close all applications. Close all applications before starting the installation. In particular, if you install the software while [Event Viewer], [Services], or [Computer Management] is running, the installation may fail. 198

MegaRAID SAS User's Guide 3 Click [Start] [Run...]. Enter the following file name, and click [OK]. [CD/DVD drive]:\raidtool\batrcs\windows\install.bat The message "Installing Battery Recalibration Scheduler." appears at the command prompt. 4 When the message "Installation of Battery Recalibration Scheduler has finished." appears, press the [Enter] key. The command prompt window closes. When you install Battery Recalibration Scheduler for the first time, the recalibration is set to start the first day of every month at 11:00 by default. If you reinstall Battery Recalibration Scheduler over itself, the previously start date and time settings will be kept. 5.4.2 Changing Battery Recalibration Start Date and Time The start date and time for the battery recalibration can be modified using the AT command of the OS. To modify the start date and time, follow the procedure below. It is recommended that you perform battery recalibration about once a month or every two months. 1 Log on to Windows with Administrator privileges. 2 Click [Start] [All Programs] (or [Programs]) [Accessories] [Command Prompt] to open the command prompt. The command prompt window opens. 3 Type "AT", and press the [Enter] key. 4 Write down the [ID] value for the row with the following value in the [Command Line] column. For Windows Server 2003 x64 C:\WINDOWS\SysWOW64\GAMSERV\Batrec.bat For Windows Server 2003 C:\WINDOWS\system32\GAMSERV\Batrec.bat For Windows 2000 Server C:\WINNT\system32\GAMSERV\Batrec.bat Example: If the AT command output is as follows, the ID is 4. C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 3 Each M T 12:00 PM "C:\WINDOWS\system32\GAMSERV\HDDchk.bat" 4 Each 1 11:00 PM "C:\WINDOWS\system32\GAMSERV\Batrec.bat" 199

Chapter 5 Array Configuration and Management [GAM] 5 Delete the currently registered battery recalibration schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 4 /delete 6 Register a new schedule. Using the AT command, register the following command line together with the date and time of execution. For Windows Server 2003 x64 C:\Windows\SysWOW64\GAMSERV\Batrec.bat For Windows Server 2003 C:\windows\system32\GAMSERV\Batrec.bat For Windows 2000 Server C:\WINNT\system32\GAMSERV\Batrec.bat To specify the scheduled performance, enter the following AT command. AT <Start Time> /every:<date> "Command Line" To perform the battery recalibration every 15th at 8:00, (For Windows Server 2003) AT 8:00 /every:15 "C:\Windows\system32\GAMSERV\Batrec.bat" For details on how to use the AT command, refer to Windows Help. To confirm whether the battery recalibration is configured correctly, type "AT" at the command prompt and check the displayed result. Check that the date, time, and the command line set in Step 6 are correct. 200

MegaRAID SAS User's Guide 5.4.3 Uninstalling Battery Recalibration Scheduler Follow the procedure below to uninstall Battery Recalibration Scheduler. 1 Log on to Windows with Administrator privileges. Exit all programs before starting the uninstallation. If uninstalling the software while [Event Viewer] or [Computer Management] is running, the uninstallation will fail. Make sure to exit all programs. 2 Click [Start] [All Programs] (or [Programs]) [Accessories] [Command Prompt]. The command prompt window opens. 3 Type "AT", and press the [Enter] key. 4 Write down the [ID] value for the row with the following value in the [Command Line] column: For Windows Server 2003 x64 C:\Windows\SysWOW64\GAMSERV\Batrec.bat For Windows Server 2003 C:\windows\system32\GAMSERV\Batrec.bat For Windows 2000 Server C:\WINNT\system32\GAMSERV\Batrec.bat Example: If the AT command output is as follows, the ID is 4. (For Windows Server 2003) C:\Documents and Settings\Administrator>at Status ID Day Time Command Line ---------------------------------------------------------------------- 3 Each M T 12:00 PM "C:\WINDOWS\system32\GAMSERV\HDDchk.bat" 4 Each 1 11:00 PM "C:\WINDOWS\system32\GAMSERV\Batrec.bat" 5 Delete the currently registered battery recalibration schedule. Execute the following command. AT <ID in Step 4> /delete Example: AT 4 /delete 201

Chapter 5 Array Configuration and Management [GAM] 5.5 Starting and Exiting GAM This section explains how to start and exit GAM. 5.5.1 Starting GAM and Signing On GAM requires user authentication to limit the availability of functions according to the usage. You have to sign on to GAM to obtain User access privileges or higher. When GAM starts for the first time after the installation, the [Define Server Groups] window appears. Refer to "5.7.1 Server group and server settings" (pg.213) to configure the settings. Only start GAM Client when management or maintenance of arrays is necessary. From a security point of view, it is not recommended to run GAM Client continuously. While signing on from GAM Client, if the accessed server shuts down, GAM Client cannot communicate with GAM Server and cannot respond. This prevents the user from operating it. In that case, wait for the GAM Server that is to be accessed to start up again, or forcibly exit GAM Client. 1 To start up GAM, click [Start] [Programs] (or [All Programs] in Windows Server 2003) [Mylex Global Array Manager Client]. If a server group or server is already defined, the [Global Status View] window appears. GAM starts with Guest privileges. To use User or Administrator privileges, you have to sign on. 2 When you double-click the server icon in the [Global Status View] window, or perform operations that require the sign on. You can also open the [Sign On] window by selecting [Sign on] from the [Administration] menu. If GAM Client and GAM Server are installed on different servers (for a Linux system, etc.), enter the password for the GAM Server. 202

MegaRAID SAS User's Guide 3 Sign on to GAM. 1. Enter your user name. When signing on with User privileges Enter a user name in [Username:]. When signing on with Administrator privileges Enter "gamroot" in [Username:]. 2. Enter a password in [Password:]. If [Remember password for this session] is checked, uncheck it. Note that if this option is checked, you can automatically sign on to different servers with the same password. To avoid automatically accessing servers, it is recommended to keep this option unchecked. Even if this option is checked, you need to sign on again once you exit GAM Client. 3. Click [Sign-on]. In GAM, the available functions are limited according to access privileges. For detail about the access privileges, refer to "5.1.3 Access Privileges to GAM" (pg.184). 5.5.2 Exiting GAM To exit GAM, click [Exit] from [File] in the GAM menu bar. 203

Chapter 5 Array Configuration and Management [GAM] 5.6 GAM Window Layout The following provides a description of the windows, buttons, and menu items displayed when using GAM. 5.6.1 Startup Window Layout and Functions When GAM starts a window consisting of [Global Status View] and [Log Information Viewer] appear. Server Group Selection box Controller Selection box Menu bar Toolbar Global Status View Controller icon Server icon Log Information Viewer Menu bar You can perform GAM functions such as [Controller View] and [RAID Assist] with this menu bar. For details on the GAM menus, refer to "5.6.2 Menu Layout and Functions" (pg.207). Toolbar The tool bar contains the buttons for frequently used GAM functions. For details on the toolbar, refer to "5.6.3 Toolbar Icon" (pg.209). Server Group Selection box The server group names which are registered in GAM Client are displayed. Clicking switches the server groups to be managed. Controller Selection box You can select the array controller to be operated from Controller Selection box. Clicking displays the onboard array controller connected to the currently selected server, or the controller ID and type (e.g. MegaRAID SAS 300-8E ROMB) of the array card. 204

MegaRAID SAS User's Guide Global Status View The servers in the currently selected server group are displayed. Server icon The information of the server is displayed. Server name (e.g. PRIMERGY) or IP address (e.g. 192.168.1.5) OS (e.g. W2K3: Windows Server 2003, Linux: Linux) Status of the server For detail about the status, refer to the following table. table: Server status Icon Color Server status Green Normal. Yellow Waiting for server connection. Red The server is down or disconnected. The following are possible causes. Network problem occurs. The server is not powered on. The server IP or host name has been changed. GAM Server is not installed or not running on the server. 205

Chapter 5 Array Configuration and Management [GAM] Controller icon Information of the array controller on the servers are displayed. Controller Icons are displayed to the right of the server icons. The numbers in the parentheses indicate the numbers of the connected array controllers. The array controllers have the following statuses. table: Array controller status Icon Color Array Controller Status Green The array controller or the logical drives created under the controller are operating normally. Yellow The array controller or the logical drives created under the controller are in Critical status, or there is trouble with the connected hard disks. Red The array controller or the logical drive created under the controller is not operating properly. Log Information Viewer Displays events on the array controller. Events Event ID Severity Source Source Time Device Address Description Sequence (Seq) Local Time table: Log Information Viewer Details The level of the event to be reported is indicated with an icon that signifies Information, Caution, Warning, or Others. The ID assigned to the event being reported is also displayed. Priority level of the event. IP address or name of the server that sent the event. Time when the event occurred. Other data regarding the addresses of related devices, operations in question, and the reason why the event was sent. Event description Event sequence number Time when the event occurrence was signaled to GAM Client. Log Information Viewer only displays events that occur while the GAM Client is running. To see all the events that have occurred on the array controller, refer to the event log for the operating system. 206

MegaRAID SAS User's Guide 5.6.2 Menu Layout and Functions This section describes the function of the GAM menu items. File menu Menu Open Configuration Save Configuration Clear Configuration table: File menu Function Not supported. Do not use it. Saves the current RAID configuration in a file. For details, refer to " Saving disk array configuration information" (pg.232). Clears all the RAID configurations of the currently selected array controller. For details, refer to " Deleting disk array configuration information" (pg.233). The above menus are only available when you sign on with Administrator privileges to open [Controller View]. Executing [Open Configuration] or [Clear Configuration] deletes the existing array configurations and all the data on the logical drives. Do not perform these operations unless directed by your maintenance engineer. View menu Menu Global Status View Controller View Log Information Viewer Foreground Initialize Status Background Initialize Status Rebuild Status Make Data Consistent Status Expand Capacity Status table: View menu Function Displays the [Global Status View] window. By default settings, [Global Status View] opens automatically when GAM starts up. Displays the [Controller View] window. Displays information for each device and the status of hard disks or logical drives connected to the controller. Displays the [Log Information Viewer] window. This window displays events or errors that occurred in the array controller. [Log Information Viewer] opens automatically when GAM Client starts up. Shows the progress of foreground initialization on a logical drive. This can only be selected while foreground initialization is in progress. This function is not supported. Shows the progress of background initialization on a logical drive. This can only be selected while background initialization is in progress. Shows the rebuild progress. This can only be selected while rebuild is in progress. Shows the Make Data Consistent progress. This can only be selected while Make Data Consistent is in progress. Shows the Expand Capacity process progress. This can only be selected while the Expand Capacity process is in progress. Note: Do not restart or shut down the system while the Expand Capacity process is in progress. Doing so may cause loss of data. 207

Chapter 5 Array Configuration and Management [GAM] Menu Patrol Read Status Error Table table: View menu Function Not supported. Do not use it. Not supported. Do not use it. Administration menu Menu Sign On Define Server Groups Select Current Server Group Select Current Controller RAID Assist Initialize Logical Drives Controller Information Enclosure Information Controller Options Physical Device Options Intelligent BBU Scan Devices Advanced Functions Settings Alarm Sound Consistency Check with Restoration table: Administration menu Function Allows you to sign on when using GAM's monitoring and setting functions. Signing on with a User account registered on the server enables you to use the monitoring function (available with User privileges). Signing on with "gamroot" enables you to use GAM's setting and management functions (available with Administrator privileges). Sets a server group and the names or IP addresses of servers in the group. Selects a server group. Functions in the same manner as when the [Server Selection] box is operated directly. Note: Make sure to select a server group registered with [Define Server Groups]. Selects a controller to be managed. It performs the same function as when the [Controller Selection] box is operated directly. Sets a disk array configuration. This function allows you to define disk groups and logical drives. For details, refer to "5.8.1 Starting RAID Assist Function" (pg.218). Not supported. Do not use it. Note: This function may cause data loss on logical drives. Displays the main information for the currently selected array controller. Displays SES and SAF-TE Enclosure Management information. Note: The Enclosure Information function is not supported. Hard disk cabinets cannot be monitored with this function. Sets parameters for the currently selected array controller. For details, refer to "5.7.2 Setting and Changing Controller Options" (pg.214). Not supported. Do not use it. When a battery backup unit is installed, its status appears. Redetects the hard disks connected to the array controller. Other functions. Selecting this displays a submenu. Flash Utility Updates the firmware on the array controller. Note: Do not update the firmware except the versions supported by Fujitsu. Not supported. Do not use it. Not supported. Do not use it. Not supported. Do not use it. 208

MegaRAID SAS User's Guide 5.6.3 Toolbar Icon The toolbar icons at the top of the [GAM] window allow you to start up frequently used functions. Icon table: Toolbar icons Function Displays the [RAID Assist] window and starts creating a RAID configuration. It perform the same function as when [RAID Assist] is selected from the [Administration] menu. Rescans the devices. It perform the same function as when [Scan Devices] is executed from the [Administration] menu. Displays array controller information. It perform the same function as when [Controller Information] is selected from the [Administration] menu. Not supported. Do not use it. Opens the [Sign On] window. It perform the same function as when [Sign On] is selected from the [Administration] menu. Not supported. Displays Help. 209

Chapter 5 Array Configuration and Management [GAM] 5.6.4 Starting Controller View and the Window Layout The [Controller View] window allows you to monitor the status of hard disks or logical drives. To open the [Controller View] window, select [Controller View] from the GAM [View] menu (pg.207). If the [Sign On] window opens, sign on by referring to "5.5.1 Starting GAM and Signing On" (pg.202). The following window appears. The [Controller View] window displays the information below regarding the controller currently selected in the [Controller Selection] box. The number of enclosures connected to controllers Each enclosure appears on the left. 210

MegaRAID SAS User's Guide Hard disk Displays information about each hard disk. The hard disk status icon appears as follows. table: Hard disk status icon Icon Color Status Status Green Online The hard disk is running normally. (On Line) Red Yellow Green + white Yellow Not applied Target ID ("Enclosure ID" and "Hard Disk ID" are not supported for this array controller. Indefinite values are displayed.) Capacity of hard disk. Type and status of hard disk. Failure / Offline (Dead / Unconfigured Bad) Rebuild (Rebuilding) Spare disk (Hot Spare) Failure expected (Critical) Unused (Unconfigured) The hard disk has a failure, cannot be recognized, or cannot be read and written. A rebuild of the hard disk is in progress. This hard disk is set as a spare disk. This hard disk is expected to fail. This hard disk is unused or available. Double-click the icon of each hard disk to see more detailed information. For details, refer to "5.9.3 Viewing Hard Disk Information" (pg.237). If a hard disk is in an unrecognizable state, detailed information may not be displayed even when double-clicking its hard disk icon. 211

Chapter 5 Array Configuration and Management [GAM] Logical drive Displays information about each logical drive. Logical Drive ID RAID Level of logical drive. Capacity of logical drive. Status of logical drive. The logical drive status icon appears as follows. table: Logical drive status icon Icon Color Status Description Green On Line The logical drive is opening normally. Yellow Critical The logical drive is operating without redundancy. Red Off Line The logical drive is not available. Green Make Data Consistent Make Data Consistent is performing on the logical drive. Double-click the icon of each logical drive to see more detailed information. For details, refer to "5.9.4 Viewing Logical Drive Information" (pg.240). 212

MegaRAID SAS User's Guide 5.7 Settings This section explains how to set parameters. The following two items are to be set. Server group and server settings Controller option settings 5.7.1 Server group and server settings The setting window for the server groups automatically opens when the GAM Client is started for the first time. Add a server group and servers according to the following procedures. 1 Click [Add] below the [Server Groups] area. 2 Enter a name for the server group to be added using the [Adding Item] window. 3 Click [OK]. The added server group name appears in the [Server Groups] area. 4 Select the added server group, and click [Add] below the [Servers] area. The [Adding Item] window appears. 5 Enter the name or the IP address of the server computer that you want to monitor in the [Adding Item] window. When you enter the computer name or IP address of the server to be monitored, do not specify "localhost" or the loopback address "127.0.0.1". 6 Click [OK]. The entered server name is added in the [Server] area. 213

Chapter 5 Array Configuration and Management [GAM] 7 Click [OK] to close the [Define Server Groups] window. Check that the registered server appears in [Global Status View]. You can also set a server group by selecting [Define Server Groups] from the [Administration] menu (pg.208). You can set up to a maximum of 100 servers which to be monitored. 5.7.2 Setting and Changing Controller Options You can check and change array controller options from [Controller Options] in the [Administration] menu. The controller options must be set within the ranges displayed in the following table. You may change only options specified as default in the table. If the settings are out of the specified ranges, correct operation may not be ensured. [Controller Options] tab 214

MegaRAID SAS User's Guide table: Array controller properties (set with Controller Options tab) Option Parameter Setting Description Global Parameters Startup Parameters Clustering Parameters Cache Line Size (MB) Enable Automatic Rebuild Management Enable Cannot be changed Enables automatic rebuild after hard disk replacement. Rate [Note 1] 80 Default Sets the rebuild priority for normal I/O. Enable Background Initialization Enable Fixed Enables the background initialization function. Rate [Note 1] 80 Default Sets the background initialization priority for normal I/O. Check Consistency / Make Data Consistent Rate [Note 1] 80 Default Sets the Make Data Consistent priority for normal I/O. MORE Rate [Note 1] 80 Default Sets the Expand Capacity priority for normal I/O. Enable Auto Drive Sizing Disable Fixed Enables rounding of the hard disk capacity. Disk Spin-up On Command Fixed Not supported Devices per Spins 2 Cannot be changed Initial Delay 0 Fixed Not supported Delay Between Spins 6 Cannot be changed Enable Clustering Disable Fixed Not supported Controller Host ID 0 Fixed Not supported User Selected 16 Fixed Not supported Active 16 Fixed Not supported Sets the number of hard disks that start rotating at the same time. Specifies the rotation start interval between hard disks. [Note 1]: While increasing the values for them increases background task performance, it may also decrease the rate of normal I/O. 215

Chapter 5 Array Configuration and Management [GAM] [Advanced] tab Selecting the [Advanced] tab enables you to set [Advanced] options. table: Array controller properties (set with the [Advanced] tab) Options Parameter Setting Description Temporarily Offline RAID Array Enable Disable Fixed Not supported Device Health Monitoring (S.M.A.R.T/ I.E.C) Patrol Read Enable Polling Enable Fixed Enables the hard disk failure prediction function (S.M.A.R.T.). Polling Interval 300 Cannot be changed Start Automatically on Start Up Disable Cannot be changed Specifies a polling interval for S.M.A.R.T detection. Not supported Delay between Iterations 168 Default Not supported Apart from the controller options, you may be required to configure various settings for the server. For how to set up the server, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 216

MegaRAID SAS User's Guide 5.7.3 Configuring Battery Backup Unit You can check and change the operation mode of the battery backup unit. The settings will be automatically corrected after RAID management tool installation. 1 Start up GAM Client, and sign on with Administrator privileges. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Intelligent BBU] from the [Administration] menu. The following window appears. 3 Select "Disable" from [Auto Learn Mode]. table: Intelligent battery backup unit properties Options Setting Description Auto Learn Mode Disabled Setting value during the operation This item sets the battery recalibration mode. Make sure to set it to "Disable". 4 Click [Cancel]. The settings of [Auto Learn Mode] are reflected and the window closes. 217

Chapter 5 Array Configuration and Management [GAM] 5.8 Operating RAID Configuration To create a new disk group or logical drive with GAM, use the RAID Assist function. RAID Assist enables you to create or add an array configuration, expand the capacity and set a hot spare. Disk array operations require signing on with Administrator privileges. 5.8.1 Starting RAID Assist Function You can start RAID Assist function by selecting [RAID Assist] from the [Administration] menu. The following window appears. You cannot use the RAID Assist function while rebuilding, expanding the capacity, or running Make Data Consistent. Operating RAID Assist may cause loss of the disk array configuration or data files. Read the manual carefully before use. Automatic Configuration and Assisted Configuration... are not supported. Do not use them. Use Manual Configuration... to create a RAID configuration. RAID Assist cannot configure the RAID 10 logical drive. To configure the RAID 10 logical drive, use WebBIOS. "2.3.1 Creating Disk Array Configuration" (pg.66) 218

MegaRAID SAS User's Guide Manual Configuration To start creating the array configuration, click [Manual Configuration...]. The following window appears. Edit Configuration Used for setting/releasing a hot spare or for deleting a logical drive. Do not use Edit Configuration for any other purposes than to delete a logical drive or to set/release a hot spare. New Configuration Deletes the old array configuration and data (if any) and applies a new array configuration to the array controller. This function is used when no array configuration exists on the array controller, or when creating a new array configuration after deleting the existing array configuration completely. Executing New Configuration deletes the existing configuration information. Use this function with caution since it will delete all the existing logical drives and the data inside. Delete Logical Drives Not supported. Do not use it. Use [Edit Configuration] to delete logical drives. Add Logical Drive Used for setting up an additional logical drive while keeping the existing array configuration. One or more logical drives must be set for the selected array controller, and there must be free space on an unused (Unconfigured) hard disk or in a disk group. 219

Chapter 5 Array Configuration and Management [GAM] Expand Arrays To expand the logical drive's capacity, an unused (Unconfigured) hard disk is added to an existing disk group and data on the drive are reconfigured. The data on the logical drive to be expanded will not be destroyed. 5.8.2 Creating New Disk Array Configuration (New Configuration) To create a new array configuration, or to create a new array configuration after completely deleting the existing array configuration, follow the procedure below. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [RAID Assist] from the [Administration] menu. The [RAID Assist] window appears. 3 Click [Manual Configuration]. The [Manual Configuration] window appears. 4 Click [New Configuration]. The following window appears. At this point, create a disk group. The disk group appears in the [Disk Arrays] area at the top left of the window. Move the required hard disks using a drag-and-drop operation. A list of unused hard disks is displayed in the [Unused Disk Drives] area at the bottom of the window. Select an unused hard disk from the list and drag it to the [Disk Array A0] area. The dragged-and-dropped hard disk will become part of the disk group A0. A maximum of 32 hard disks can be added to one disk group. To select multiple hard disks, hold the [Ctrl] key and select the hard disks. For RAID 1, only two hard disks can be used in one disk group. 220

MegaRAID SAS User's Guide Do not use the [Add Array] at this point. To create multiple disk groups, use [Add Logical Drive] after completing this procedure. Use hard disks of the same capacity and type in one disk group. If a mixture of different hard disks are connected, check them in advance by referring to "5.9.3 Viewing Hard Disk Information" (pg.237). Removing hard disks from the disk group You can remove hard disks from the disk group by dragging them from the [Disk Arrays] area and dropping them into the [Unused Disk Drives] area. Recreating the configuration Click [Clear All] if you want to recreate a disk group from the beginning. All the settings are cleared. Setting a spare disk Select an unused hard disk and click [Make Spare]. The icon of the unused hard disk changes to a spare disk icon. To release a spare disk, select a spare disk and click [Remove Spare]. Do not drag a hard disk assigned as a spare disk into the [Disk Arrays] area. Use a spare disk of the same capacity and type as the hard disks in the disk group. When there are two or more disk groups that use different hard disks, set a spare disk for each type of hard disk. 5 When you complete creating a disk group, click the [Logical Drives] tab to configure the logical drives. 1. Select a RAID level for the logical drive. Click under [RAID Level] and select the RAID level for the logical drive from the displayed list. GAM cannot configure the RAID 10 logical drive. To configure the RAID 10 logical drive, use WebBIOS. " How to create disk array configuration" (pg.67) 221

Chapter 5 Array Configuration and Management [GAM] 2. Enter the capacity of the logical drive to be created in [Capacity-Logical], or enter the physical capacity to be used in [Capacity-Physical]. When creating only a single logical drive in the disk array, or when creating the last logical drive in the disk group, do not change the default size. When creating multiple logical drives, enter a smaller value than the entire capacity to leave the required capacity for the others. Enter the size in either of the two capacity fields. When a value is entered in one field, the other value is automatically calculated and displayed. The capacity of the logical drive may be automatically adjusted to match the block size. Do not add multiple logical drives at a time. To create multiple logical drives in a single disk group, use the [Add Logical Drive] function (pg.223) after completing this procedure. If multiple logical drives are created in a single disk group, you cannot expand the capacity with the [Expand Capacity] function later. 3. Check the [Write Cache] checkbox to use the logical drive in Write Back mode. Before using Write Back mode, refer to "1.3.1 Write Policy" (pg.28). If there is no battery backup unit, the logical drive is created in Write Through mode even if [Write Cache] is enabled. To use a logical drive in Write Back mode, refer to "5.8.7 Changing Write Policy" (pg.231) and change the write mode settings after creating the logical drive. 4. Check that the other parameters are set to the following values. If any difference is found, change the settings as follows. table: Parameter settings Parameter Setting Init Drive? Disabled Stripe Size 64 KB Read Control Normal Direct I/O Enabled This array controller does not support the foreground initialization function on the OS. Logical drives with redundancy are automatically initialized with background initialization. 6 Click [Add Drive] to register the new logical drive. 222

MegaRAID SAS User's Guide 7 Click [Apply] to enable the disk array configuration. When Write Cache is enabled When Write Cache is enabled, the following confirmation window appears. Click [YES]. The final confirmation window appears. 8 Type "YES", and click [OK]. A new disk array configuration is created. 9 A message that the logical drive is available appears. Click [OK] to exit. 5.8.3 Adding Logical Drive to Existing Disk Array Configuration (Add Logical Drive) This section explains the procedure for adding a new logical drive to the existing disk array configuration. To add a new logical drive, it is necessary to have either an unused drive or free space in the existing disk group. Follow the procedure below to execute Add Logical Drive. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [RAID Assist] from the [Administration] menu. The [RAID Assist] window appears. 3 Click [Manual Configuration...]. The [Manual Configuration] window appears. 223

Chapter 5 Array Configuration and Management [GAM] 4 Click [Add Logical Drive]. The [Disk Arrays] tab window appears. If there is free space in the defined disk group, the [Logical Drives] tab appears. In this case, refer to "5.8.2 Creating New Disk Array Configuration (New Configuration)" (pg.220) and follow the procedure from Step 5 to assign logical drives. 5 Click [Add Array] to configure a new disk group. An empty disk group is added in the [Disk Arrays] area. 6 Select an unused hard disk from the [Unused Disk Drives] area and drag it to the [Disk Arrays] area. The dragged-and-dropped hard disk will become part of the disk group. A maximum of 32 hard disks can be added to one disk array. To select multiple hard disks, hold the [Ctrl] key and select the hard disks. For RAID 1, two hard disks can be used in one disk group. To remove hard disks from the disk group, drag them from the [Disk Arrays] area and drop them into the [Unused Disk Drives] area. Do not add multiple disk groups at a time. To create an additional disk group, complete this procedure, apply the configuration, and then perform Add Logical Drive again. Use hard disks of the same capacity and type in one disk group. If a mixture of different hard disks are connected, check them in advance by referring to "5.9.3 Viewing Hard Disk Information" (pg.237). Setting a spare disk Select an unused hard disk and click [Make Spare]. The icon of the unused hard disk changes to a spare disk icon. To release a spare disk, select the spare disk to be released and then click [Remove Spare]. 224

MegaRAID SAS User's Guide Do not drag a hard disk assigned as a spare disk into the [Disk Arrays] area. Use a spare disk of the same capacity and type as the hard disks in the disk group. When there are two or more disk groups that use different hard disks, set a spare disk for each type of hard disk. 7 When the configuration of the disk group and the spare disks are complete, click the [Logical Drives] tab to start configuring logical drives. The procedure for configuring logical drives is the same as that for creating a new disk array. Refer to "5.8.2 Creating New Disk Array Configuration (New Configuration)" (pg.220) and follow the procedure from Step 5. 5.8.4 Expanding Capacity of Logical Drive (Expand Array) You can expand the capacity and change the RAID level for existing logical drives. There are certain conditions to note and precautions to perform Expand Capacity. Read "1.3.5 Expand Capacity" (pg.36) carefully before expanding the capacity. The capacity of the RAID10 logical drive cannot be expanded. 1 To prepare for unexpected problems, back up the data. 2 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 3 Perform Make Data Consistent (pg.245) on the logical drive whose capacity is to be expanded, and check if it completes successfully. If the Make Data Consistent is not complete successfully, do not perform Expand Capacity. 4 Select [RAID Assist] from the [Administration] menu. The [RAID Assist] window appears. 5 Click [Manual Configuration]. The [Manual Configuration] window appears. 225

Chapter 5 Array Configuration and Management [GAM] 6 Click [Expand Array]. The following window appears. 7 Select a logical drive to be expanded and click [OK]. The [Disk Arrays] tab window appears. 8 Select an unused hard disk from the [Unused Disk Drives] area, and drag it to the disk group displayed at the top left of the window. You can add hard disks until the total number of drives in the disk group reaches 32. Make sure that the number of hard disks in the disk group does not exceed 32. Use additional hard disks of the same capacity and type as the existing hard disks in the disk group. If a mixture of different hard disks are connected, check them in advance by referring to "5.9.3 Viewing Hard Disk Information" (pg.237). 226

MegaRAID SAS User's Guide 9 When you complete adding hard disks to the disk group, click the [Logical Drives] tab to configure the logical drives. 10 Select a RAID level for the logical drive. Click under [RAID Level] and select the RAID level for the logical drive from the displayed list. After Expand Capacity is performed, the RAID level of the logical drive will be converted to the RAID level selected here. If you convert a RAID 1 or 5 logical drive to RAID 0, note that the logical drive will lose its redundancy. Do not change the Stripe Size. 11 Click [Apply]. When the following confirmation window appears, click [YES]. The final confirmation window appears. 12 Type "YES" and click [OK]. A window appears displaying the progress of Expand Capacity, and the Expand Capacity process starts. Wait until the process is complete. 227

Chapter 5 Array Configuration and Management [GAM] Do not turn off, reset or restart the server while Expand Capacity is in progress. Doing so may cause loss of data in the target logical drive. If the server is turned off during Expand Capacity, access to the hard disk is automatically resumed after the server restarts. In this case, wait until the hard disk access LED goes off to ensure that no access is made to the hard disk, recreate the array configuration, and then restore the data backed up before expanding the capacity. If the RAID level of the logical drive after Expand Capacity has data redundancy (that is RAID 1 or RAID 5), background initialization will be performed after the Expand Capacity is complete. 5.8.5 Deleting Existing Logical Drive (Edit Configuration) Use [Edit Configuration] to delete logical drives. The [Delete Logical Drive] function in [Manual Configuration] is not supported. Do not use it. Do not use [Edit Configuration] for operations other than deleting a logical drive or setting/releasing spare disks. To create a disk group or logical drives, use [New Configuration] or [Add Logical Drive]. Make sure to back up data before you delete a logical drive. You can only delete the last created logical drive (the drive with the highest number). For how to delete all logical drives (i.e. to clear an array configuration), refer to "5.8.8 Saving and Deleting Disk Array Configuration Information" (pg.232). 1 To prepare for unexpected problems, back up the data. 2 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 3 Select [RAID Assist] from the [Administration] menu. The [RAID Assist] window appears. 4 Click [Manual Configuration...]. The [Manual Configuration] window appears. 5 Click [Edit Configuration]. When the [Disk Arrays] tab appears, click the [Logical Drives] tab to display a list of the currently defined logical drives. Do not change the disk group configuration on the [Disk Arrays] tab. If you change the disk group configuration, all the existing logical drives will be deleted. If by mistake the disk group configuration be accidentally changed, click [Cancel] and restart the procedure. 228

MegaRAID SAS User's Guide 6 Click [Delete Drive]. One logical drive will be deleted at a time in descending order from the last in the list. Do not click [Clear All]. If you click [Clear All], all the logical drives are deleted at once. If you select a wrong logical drive, click [Cancel] in a confirmation window and start again from the beginning. 7 When completing the deletion of logical drive from the list, reconfirm that you have not deleted any wrong logical drives. When the deletion is done properly, click [Apply]. The following warning message appears. 8 Click [Yes]. A confirmation window appears. 9 Type "YES", and click [OK]. If you accidentally delete a wrong logical drive, recreate the array configuration, and then restore the data backed up before the deletion. If all the logical drives in a disk group are deleted, the disk group will also automatically be deleted, and the hard disks that composed the disk group will be marked as unused. 229

Chapter 5 Array Configuration and Management [GAM] 5.8.6 Setting and Releasing Spare Disk (Edit Configuration) This section explains how to set and release spare disks. Do not use [Edit Configuration] for any other operations than deleting a logical drive or setting/ releasing spare disks. To create a disk group or logical drives, use [New Configuration] or [Add Logical Drive]. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [RAID Assist] from the [Administration] menu. The [RAID Assist] window appears. 3 Click [Manual Configuration...]. The [Manual Configuration] window appears. 4 Click [Edit Configuration]. The [Disk Arrays] tab window appears. 5 Select an unused hard disk in the [Unused Disk Drives] area, and click [Make Spare]. The icon of the unused hard disk changes to a spare disk icon. To release a spare disk, select a spare disk icon, and click [Remove Spare]. If a disk group exists that consists of hard disks with a capacity larger than that of the hard disk to be set as a spare disk, the following message appears when you click [Make Spare]. Click [OK]. Do not drag a spare disk or an unused hard disk into the [Disk Arrays] area. Use a spare disk of the same capacity and type as the hard disks in the disk group. When there are two or more disk groups that use different hard disks, set a spare disk for each type of hard disk. 6 When the configuration of the spare disk is complete, make sure the existing logical drives displayed on the right side have not been changed. When the configuration is done properly, click the [Logical Drives] tab. 230

MegaRAID SAS User's Guide 7 Click [Apply]. A confirmation window appears. 8 Type "YES", and click [OK]. 5.8.7 Changing Write Policy For this array controller, you can set Write Policy for each logical drive. To change Write Policy of a logical drive, follow the procedure below. Before using Write Back, refer to "1.3.1 Write Policy" (pg.28) and make sure that you understand the idea. If Write Policy is set to "Enabled" (Write Back) when the battery is not installed, the "BadBBU" mode is set on WebBIOS. For details about "BadBBU", refer to " Notes on Write Back (or Bad BBU)" (pg.29). If Write Policy is changed on GAM when the battery is installed, it is switched between the "WriteThrough" mode and the "WriteBack" mode by default. However, if the "BadBBU" mode is set on WebBIOS, it is switched between the "WriteThrough" mode and the "BadBBU" mode when seeing from GAM. To set the "WriteBack" mode again, change the Write Policy on WebBIOS. To use the "BadBBU" mode when the battery is installed, change the Write Policy on WebBIOS. When the battery fails or Write Policy is set to "WriteBack" during battery reconditioning, "Disabled" appears in Write Cache and not changed. It is automatically changed to "Enabled" when the battery is reconditioned. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Controller View] from the [View] menu. "5.6.4 Starting Controller View and the Window Layout" (pg.210) 3 In the [Controller View] window, double-click the icon of the logical drive whose Write Policy you want to change. The [Logical Drive Information] window appears. 231

Chapter 5 Array Configuration and Management [GAM] 4 The current Write Policy appears in [Write Cache] of the [Logical Drive Information] window. Disabled: Operating in the Write Through mode Enabled: Operating in the Write Back or the BadBBU mode 5 Click the [Enable Write Cache] or [Disable Write Cache] button at the bottom of the window. The Write Policy changes to the new mode. 5.8.8 Saving and Deleting Disk Array Configuration Information Disk array configuration information refers to the information created when the disk array is configured, such as the disk group or the logical drive capacity, RAID level, and which hard disk belongs to which disk group. Saving disk array configuration information You can save the current disk array configuration information on a floppy disk or a hard disk. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Save Configuration] from the [File] menu. 3 Enter a file name, and click [Save As]. The current disk array configuration information is saved in a file. 232

MegaRAID SAS User's Guide Deleting disk array configuration information You can delete the disk array configuration information. Do not delete the disk array configuration information unless instructed by your maintenance engineers. All data on the logical drives will be lost by this operation. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Clear Configuration] from the [File] menu. When the following confirmation window appears, click [YES]. If you want to quit deleting the disk array configuration information, click [No]. The final confirmation window appears. 3 Type "YES", and click [OK]. The disk array configuration information is deleted. 233

Chapter 5 Array Configuration and Management [GAM] 5.9 Viewing Information [GAM] The following information can be viewed using GAM. Information about events or errors that have occurred: "Appendix B List of GAM Event Log" (pg.313) Array configuration or controller information: "5.9.2 Viewing Array Controller Information" (pg.236) Hard disk information: "5.9.3 Viewing Hard Disk Information" (pg.237) Logical drive information: "5.9.4 Viewing Logical Drive Information" (pg.240) Information about tasks running in the background: "5.9.5 Checking Progress of Background Tasks" (pg.242) 5.9.1 Events GAM monitors the operations of the array controllers and hard disks connected to the array controllers. If a behavior that should be treated as an event (a serious event such as a hard disk failure or an event related to spare disk allocation) is found, GAM is notified of that event. Events such as disk array system errors, information, or management tasks are displayed in [Log Information Viewer]. To monitor the array controllers, use the OS event log (the application log; Source: Fujitsu ServerView Services). However, if the GAM Client has not been started, or if the network has a failure, [Log Information Viewer] cannot monitor the log for events that occur on the array controller. Although the log notified by GAM (source: gamevlog) is logged, ignore it since it is not supported. Also, if there are logs for the array controller which are notified by ServerView around the event log, refer to them. For the list of logs notified by ServerView, refer to "Appendix B List of GAM Event Log" (pg.313). To enter events or errors into the OS event log, it is necessary to install ServerView. To install and configure ServerView, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. The GAM Server records detailed information about occurred events in the file "GAMEVLOG.LOG". This file may be used when analysis is necessary. (Viewing or monitoring of this file is not supported.) The path where "GAMEVLOG.LOG" is stored is as follows, depending on the OS. For Windows 2000 Server For Windows Server 2003 For Windows Server 2003 x64 For Linux C:\WINNT\system32\GAMSERV\GAMEVLOG.LOG C:\Windows\system32\GAMSERV\GAMEVLOG.LOG C:\Windows\SysWOW64\GAMSERV\GAMEVLOG.LOG /var/log/gamevlog.log While GAM Client is running, if the IP address of the server to be monitored is changed or the LAN connection is disconnected, the message "Lost connection to server, or server is down" will be logged every 10 minutes. In such case, confirm the network status and that GAM Server is working properly, and then restart the GAM Client. 234

MegaRAID SAS User's Guide Log Information Viewer Log Information Viewer is automatically displayed when GAM Client starts up and an array controller is detected. The event histories displayed in the Log Information Viewer are stored in the file GAM2CL.LOG. This file may be used when investigation is necessary. (Viewing or monitoring of this log file is not supported). GAM2CL.LOG is stored in the following locations. For Windows Server 2003, Windows 2000 Server For Windows Server 2003 x64 C:\Program Files\Mylex\Global Array Manager Client\gam2cl.log C:\Program Files (x86)\mylex\global Array Manager Client\gam2cl.log To manually open the Log Information Viewer, select [Log Information Viewer] from the [View] menu. For the meaning of each item displayed in the Log Information Viewer, refer to "5.6.1 Startup Window Layout and Functions" (pg.204). Displaying detailed information about each event When detailed information about an event displayed in Log Information Viewer is needed, open the [Event Information] window. To open the [Event Information] window, double-click the event in the Log Information Viewer. The detailed information about the selected event appears. Click [OK] to close the window. 235

Chapter 5 Array Configuration and Management [GAM] 5.9.2 Viewing Array Controller Information Using Controller View By using Controller View, you can view the status of the array controller and the hard disks or logical drives connected to the array controller. For how to start Controller View, and for details about its icons, refer to "5.6.4 Starting Controller View and the Window Layout" (pg.210). Displaying detailed information about array controllers 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Controller Information] from the [Administration] menu. The [Controller Information] window appears. You can configure the controller options by clicking [Controller Options...]. For details about the controller options, refer to "5.7.2 Setting and Changing Controller Options" (pg.214). Click [Close] to close the window. 236

MegaRAID SAS User's Guide Detailed information about array controllers The following information appears. Item Model Firmware Version Intelligent BBU Package Version BIOS Version Cache Size FlashROM Size Bus Type Channels Physical Devices Max. Physical Devices Logical Drives Max. Logical Drives Bus # Device # IRQ table: Detailed Information about array controllers Description The model name of the array controller. The version of the array controller's firmware. Indication of whether there is a battery backup unit for the array controller. The version of the array controller's package. The version of the array controller's BIOS. The cache size of the array controller. The size of the array controller's FlashROM. The type of the host-side bus. Not supported. The number of hard disks connected to the array controller. Not supported. The number of logical drives. Not supported. The bus number for the array controller. The device number for the array controller. The IRQ number. 5.9.3 Viewing Hard Disk Information By using Controller View, you can view the detailed information about the hard disks connected to the controller. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Controller View] from the [View] menu. "5.6.4 Starting Controller View and the Window Layout" (pg.210) Each drive column indicates the hard disks connected to each enclosure or to each channel of the controller. 237

Chapter 5 Array Configuration and Management [GAM] 3 Double-click the icon of a hard disk you want to see the information. The detailed information about the selected hard disk appears. Channel, Target, and LUN information displayed in the title bar is not supported. Click [Close] to close the window. Detailed information about hard disks The following information appears. Item Vendor Product Revision Bus Width Sync / Linked / Soft Reset / CmdQue / ANSI Version Serial Enclosure Slot Nego. Transfer Speed Nego. Bus Width Sector Size Physical Capacity Config. Size Status table: Detailed information about hard disks Description Information about the hard disk vendor. Product model name of the hard disk. The version of the hard disk's firmware. The bus width. The value for this array controller is "Serial". Not supported. Serial number of the hard disk. Enclosure ID that the hard disk is connected to. The slot number where the hard disk is installed. Not supported. The current bus width. The value for this array controller is "Serial". The sector size. The physical capacity of the hard disk. The hard disk's available capacity when connected to this array controller. The current status of the hard disk. For details about the status, refer to " Hard disk" (pg.211). 238

MegaRAID SAS User's Guide Soft Errors / Parity Errors / Hard Errors / Misc Errors PFA Count Item table: Detailed information about hard disks Description Not supported. The number of recovery attempts made by the array controller for temporary or minor errors. This can be ignored unless the hard disk gets a failure status. The counter for S.M.A.R.T. failure predictions for the hard disk. Function buttons You can perform the following operations by using the buttons. [Rebuild] button This button is available only when the status of the hard disk is Failure. Click this button to perform a rebuild of the hard disk."5.10.2 Rebuild" (pg.246) [Make Ready] button Not supported. Do not use the [Make Ready] button unless instructed by your maintenance engineers. [Make Online] / [Make Offline] button This button forcibly changes the status of the hard disk to "Online" or "Offline". Do not use the [Make Online] or [Make Offline] button unless instructed by your maintenance engineer. In particular, always perform a rebuild to restore the hard disk status from "Offline" status to "Online" status. This operation may cause loss of data. [Locate...] button This button blinks the failure LED of the hard disk and indicate the location of the hard disk. [ReplaceMissing] button This button reassigns a hard disk to the disk group. Do not use the [ReplaceMissing] button unless you are instructed to do so by your maintenance engineer. [Close] button This button closes the detailed hard disk information window. 239

Chapter 5 Array Configuration and Management [GAM] 5.9.4 Viewing Logical Drive Information By using Controller View, you can view the detailed information about the defined logical drives. The logical drives are displayed on the right in the [Controller View] window. Each icon represents one logical drive. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Controller View] from the [View] menu. "5.6.4 Starting Controller View and the Window Layout" (pg.210) 3 Double-click the icon of a logical drive you want to see the information. The detailed information about the selected logical drive appears. The logical drive number appears on the title bar. Click [Close] to close the window. 240

MegaRAID SAS User's Guide Detailed information about logical drives The following information appears. Item RAID Level Fault Tolerant Optimized for Speed Optimized for Capacity Logical Size Physical Size Stripe Size Status Write Cache Read Control I/O Control Used Array Capacity table: Detailed information about logical drives Description The RAID level set for the logical drive. Indicates whether the logical drive has a redundancy feature or not. The setting whether the priority of logical drive's RAID level is placed on speed or not. The setting whether the priority of logical drive's RAID level is placed on capacity or not. The logical size of the logical drive. The physical size of the logical drive. The striping size used by the logical drive. The current status of the logical drive. For details about the status, refer to " Logical drive" (pg.212). The current Write Policy set for the logical drive. Disabled: Write Through mode Enabled: Write Back or Bad BBU mode Not supported. Not supported. Displays the ratio of the logical drive's capacity to the disk group's total capacity. The RAID 10 logical drive is displayed as RAID 1 on GAM. In the case of the RAID 10 logical drive, two or more disk groups are displayed on the graph of Used Array Capacity. Function buttons You can perform the following operations by using the buttons. [Locate] button This button blinks the failure LED of all the hard disks that compose the logical drive to indicate their locations. [Make Data Consistent] button This button starts Make Data Consistent. For details, refer to "5.10.1 Make Data Consistent" (pg.245). [Enable Write Cache] / [Disable Write Cache] button This button changes Write Policy of the logical drive. The available button depends on the current Write Policy. For details, refer to "5.8.7 Changing Write Policy" (pg.231). [Show Bad Data Blocks] button Not supported. [Close] button This button closes the detailed logical drive information window. 241

Chapter 5 Array Configuration and Management [GAM] 5.9.5 Checking Progress of Background Tasks There are four kinds of background tasks: initialization, rebuild, Make Data Consistent and Expand Capacity. GAM allows you to check the progress of these tasks with progress bars. From the pace of the progress bar, you can figure out the approximate time for the task to complete. Background Initialization Status When the background initialization of a logical drive is in progress, you can check its progress by selecting [Background Initialize Status] from the [View] menu. You can cancel the initialization of logical drives by checking the [Cancel] checkbox for each logical drive and clicking [Cancel]. Click [Close] to close the [Background Initialize Status] window. The color of the background task progress bar is different for each logical drive. Even if you cancel the background initialization, it is automatically restarted after a certain interval. Rebuild Status When a rebuild is in progress, you can check its progress by selecting [Rebuild Status] from the [View] menu. Click [Close] to close the [Rebuild Status] window. 242

MegaRAID SAS User's Guide When there are multiple logical drives assigned to a disk group, the rebuild begins from the logical drive with the lowest number in ascending order. For example, when two logical drives 0 and 1 are assigned to one disk group, after the rebuild of the logical drive 0 is completed 100%, the rebuild of the logical drive 1 starts from 0%. Do not cancel the rebuild. Make Data Consistent Status When Make Data Consistent is in progress, you can check its progress by selecting [Make Data Consistent Status] from the [View] menu. You can cancel the Make Data Consistent of logical drives by checking the [Cancel] checkbox for each logical drive and clicking [Cancel]. Click [Close] to close the [Consistency Check Status] window. Expand Capacity Status When Expand Capacity is in progress, you can check its progress by selecting [Expand Capacity Status] from the [View] menu. Click [Close] to close the [Expand Capacity Status] window. Do not restart or shut down the system while the Expand Capacity process is in progress. Doing so may cause data loss. Do not perform any other GAM operations while the Expand Capacity process is in progress. 243

Chapter 5 Array Configuration and Management [GAM] Calculating approximate time needed for a background task For background initialization, rebuild, Make Data Consistent, and Expand Capacity, you can figure out approximate time that the task takes from start to finish from the pace of the progress bar. 1 Measure the period of time required for the progress bar to advance 1%. 2 Calculate the approximate time that the task takes from start to finish, using the following formula. For background initialization, Make Data Consistency, and Expand Capacity (Period measured in Step 1) 100 For Rebuild (Period in Step 1) 100 (total LD capacity in DG / capacity of LD in rebuild) * DG: Disk Group, LD: Logical Drive Use the calculated time only as a guideline. The actual time may be different from the calculated time depending on the system load and other factors. 244

MegaRAID SAS User's Guide 5.10Maintenance Functions GAM provides the following maintenance functions. Making data consistency for logical drive (Make Data Consistent)pg.245 Rebuilding offline hard disks (Rebuild)pg.246 5.10.1 Make Data Consistent Make Data Consistent, also called redundant data regeneration, is a function that regenerates the parity or mirrored data on redundant logical drives and makes the array configuration consistent. When the OS does not shut down properly because of a server failure or a blackout, the array may become inconsistent. In this case, perform this function as soon as possible. Make Data Consistent can be performed on logical drives with redundancy such as RAID 1, RAID 5, or RAID10 logical drives in the online status). Make Data Consistent cannot be performed on logical drives without redundancy such as RAID 0 logical drives or logical drives in the critical/offline status. In addition to keeping the data consistent in the array, Make Data Consistent also automatically corrects any media errors on the hard disks. Make Data Consistent can be performed simultaneously for several logical drives. To make data consistency, follow the procedure below. 1 Start up GAM, and sign on with Administrator privileges. "5.5 Starting and Exiting GAM" (pg.202) 2 From the [View] menu, select [Controller View]. "5.6.4 Starting Controller View and the Window Layout" (pg.210) 3 On the [Controller View] window, double-click an icon of a logical drive where Make Data Consistent is to be performed. The detailed information about the logical drive appears in the [Logical Drive Information] window. 4 Click [Make Data Consistent] to start Make Data Consistent. When Make Data Consistent starts, the [Make Data Consistent Status] window appears and displays the progress of the Make Data Consistent. When the progress bar reaches 100% and the window closes, the check is complete. 245

Chapter 5 Array Configuration and Management [GAM] 5.10.2 Rebuild To execute a rebuild manually, follow the procedure below. Usually, a rebuild is automatically executed after replacing the hard disk. For how to replace the hard disk and how to perform a rebuild, refer to "Chapter 6 Replacing Hard Disk" (pg.249). 1 Start up GAM, and sign on with Administrator privileges. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Controller View] from the [View] menu. "5.6.4 Starting Controller View and the Window Layout" (pg.210) 3 Double-click the icon of a hard disk with "Failure" state ( ) in the [Controller View] window. The detailed information about the hard disk appears in the [Disk Device Information] window. 4 Click [Rebuild]. The [Rebuild Status] window appears, and the rebuild starts. When the rebuild is complete, the following window appears, and the hard disk and the related logical drives are restored to Online status. 5 Click [OK] to exit. 246

MegaRAID SAS User's Guide 5.10.3 Recalibration Recalibration is a process to recalibrate the battery capacity. To perform recalibration manually, follow the procedure below. Perform battery recalibration when: You purchased an array controller with a battery backup unit installed in. You additionally purchased a battery backup unit and install it into the array controller. You replaced parts including a battery. Battery recalibration requires 12 hours with the server power on. If the server is turned off and on during battery recalibration, the recalibration restarts from the beginning. During recalibration, the logical drive in Write Back mode shifts automatically to Write Through mode for safety. When the battery recalibration is complete, Write Policy is automatically back to Write Back. 1 Start up GAM, and sign on with Administrator privileges. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Intelligent BBU] from the [Administration] menu. The following window appears. 3 Click [Recalibrate]. The battery recalibration starts. 4 Click [Cancel]. The window closes. 247

Chapter 5 Array Configuration and Management [GAM] 248

Chapter 6 Replacing Hard Disk This chapter explains maintenance related issues, such as hard disk replacement. Since the procedure varies depending on the management tool used, read the description concerning your management tool. 6.1 How to Replace Hard Disk [ServerView RAID]......... 250 6.2 How to Replace Hard Disk [GAM].................. 265 249

Chapter 6 Replacing Hard Disk 6.1 How to Replace Hard Disk [ServerView RAID] This section explains maintenance related issues, such as replacing hard disks using ServerView RAID. 6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID] Check the slot number of the target hard disk before replacing it. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 In the tree view, check the hard disk icon. A failed hard disk is indicated with the / icon. A hard disk that has been predicted to fail is indicated with the The slot number can be checked on the hard disk icon as follows. icon. Check here for the slot number. In this example, it is "2". Check the number shown below to specify the location of the hard disk under SAS expander. Check here for the hard disk enclosure number and slot number. In this example, the enclosure is "1", and the slot is "3". If there are any hard disks being rebuilt (indicated with ), wait until the rebuild is complete. After the rebuild is complete, check the hard disk status again. 3 For the details, refer to the [General] tab of the object window. Detailed information about the selected hard disk appears. If "S.M.A.R.T. errors" appears in [Status], a failure prediction sign (S.M.A.R.T.) has been informed. The detailed information may not appear depending on the failure condition of the hard disk. 250

MegaRAID SAS User's Guide 4 If there is a failed hard disk or a hard disk that is predicted to fail, replace it according to your situation. When there is a failed hard disk When your server supports hot swap, replace the hard disk by referring to "6.1.2 Replacing Failed Hard Disk [ServerView RAID] (With Hot Swap Support)" (pg.251). When your server does not support hot swap, replace the hard disk by referring to "6.1.3 Replacing Failed Hard Disk [ServerView RAID] (No Hot Swap Support)" (pg.253). When there is a hard disk that is predicted to fail Refer to "6.1.4 Preventive Replacement of Hard Disk [ServerView RAID]" (pg.255) to replace the hard disk with the failure prediction sign. When a hard disk failed and another is predicted to fail First, replace the failed hard disk, and perform a rebuild. After that, check that the failure indication of the replaced hard disk has disappeared, i.e. that the logical drive status is "Operational", and then replace the hard disk that is predicted to fail, as a preventive measure. If the hard disk that is predicted to fail is replaced before the failed hard disk, a rebuild cannot be performed and data will be lost. Failed hard disks can also be checked by using WebBIOS. Start up WebBIOS, and check the [Physical Drives] view. For details, refer to "2.2.1 Checking Hard Disk Status" (pg.53). 6.1.2 Replacing Failed Hard Disk [ServerView RAID] (With Hot Swap Support) When a hard disk fails, replace it with a new one as soon as possible. This section explains how to replace hard disks when your server supports hot swap (changing disks with the server running). For how to remove and install hard disks, and for hot swap support, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Replace the failed hard disk with a new one of the same model (with the same capacity and rotation speed) as a rule. Never remove any other hard disks than a failed one while the server is turned on. 1 Check the disk number of the failed hard disk, and locate the disk. The bay number is the number of the slot number plus one. The slot number can be identified with Step 1 and 2 of "6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID]" (pg.250). Example: If the slot number is 2, the location of the disk is bay 3. 251

Chapter 6 Replacing Hard Disk 2 On the server, check that a hard disk failure LED is lit for the bay corresponding to the failed hard disk. For bay locations and the LED location, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 3 Pull the failed hard disk about an inch (1 to 3 cm) to disconnect it from the connector. For how to remove hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Do not completely pull out the hard disk from the server here. 4 Wait at least one minute until the hard disk motor stops rotating. 5 Pull out the failed hard disk completely from the hard disk unit bay. 6 Install a new hard disk at the same location as the failed hard disk once was. 7 Check the following. The checking procedure differs depending on whether a spare disk was set before the hard disk failure. If a spare disk was set Shortly after installing a new hard disk, it automatically becomes a spare disk, and the corresponding hard disk failure LED goes off. In the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Global Hot Spare" in the object window. If no spare disk was set Shortly after installing a new hard disk, a rebuild automatically starts for the newly installed hard disk. When the rebuild starts, the hard disk failure LED that was lit starts blinking, and then it goes off when the rebuild is complete. After the rebuild is complete, in the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Operational" in the object window. When the following event is logged in the OS event log, or in the event window of ServerView RAID Manager, the rebuild is complete. ("X" indicates the number of the hard disk where the rebuild was performed.) Source: ServerView RAID event, or ServerView RAID Manager event window ID: 10267 Event: <Controller type and its number>: Rebuild complete on Disk X Events via ServerView Source : Fujitsu ServerView Services Type : Information Event ID : 1 Description: <Controller type and its number>: Rebuild complete on Disk X 252

MegaRAID SAS User's Guide For the approximate time to complete a rebuild, refer to " Estimated time for rebuild" (pg.34) or "4.9.5 Checking Progress of Background Task" (pg.178). If the server restarts or shuts down during a rebuild, the rebuild is resumed from where it was stopped the next time. 6.1.3 Replacing Failed Hard Disk [ServerView RAID] (No Hot Swap Support) When a hard disk fails, replace it with a new one as soon as possible. This section explains how to replace hard disks when your server does not support hot swap (changing disks with the server running). For how to remove and install hard disks, and for hot swap support, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Replace the failed hard disk with a new one of the same model (with the same capacity and rotation speed) as a rule. Never remove any other hard disks than a failed one while the server is turned on. 1 Check the disk number of the failed hard disk, and locate the disk. The bay number is the number of the slot number plus one. The slot number can be identified with Step 1 and 2 of "6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID]" (pg.250). Example: If the slot number is 2, the location of the disk is bay 3. For bay locations, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 2 Shut down the server, and disconnect the AC cable. 3 Disconnect all the cables that are connected to the failed hard disk, and completely pull out the failed hard disk. For how to remove hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 4 Install a new hard disk at the same location as the failed hard disk once was. Connect all the cables that were connected to the failed one to the new one. 5 Turn on the server, and start the OS. 253

Chapter 6 Replacing Hard Disk 6 Start up ServerView RAID Manager to check the following. The checking procedure differs depending on whether a spare disk was set before the hard disk failure. If a spare disk was set The newly installed hard disk automatically becomes a new spare disk. After the rebuild is complete, in the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Global Hot Spare" in the object window. The replacement procedure is complete here. If no spare disk was set The newly installed hard disk becomes a hard disk without an array configuration. In the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Available" in the object window. Proceed to the following steps. 7 Install the new hard disk into the array configuration. 1. Log in ServerView RAID Manager with Administrative privileges. 2. Check the logical drive status. In the tree view of ServerView RAID Manager, select an logical drive icon in critical or partially critical status ( ). Check if its [Status] is "Degraded" or "Partially Degraded" in the object window. 3. In the ServerView RAID Manager tree view, right-click the hard disk icon of the newly installed ( ) to select [Replace missing disk] from the displayed menu. When the [Write access needed!] window appears, click [Yes] to proceed. 4. Select an empty slot ( ), and click [Replace]. The [Confirmation needed] window appears. 5. Click [Yes]. The procedure above will result in the status of the new hard disk in "Offline" ( ). It is not a hard disk failure only after this operation. 8 Rebuild the new hard disk. 1. In the ServerView RAID Manager tree view, right-click the icon of the new hard disk ( ), and select [Start rebuild] from the displayed menu. The [Confirmation needed] window appears. 254

MegaRAID SAS User's Guide 2. Click [Yes]. A rebuild starts. After the rebuild, select the icon of the new hard disk in the ServerView RAID Manager tree view to see if [Status] is "Operational" in the object window. When the following events are logged in the OS event log, or in the event window of the ServerView RAID Manager, the rebuild is complete. ("X" indicates the number of the hard disk where the rebuild was performed.) Source: ServerView RAID events, or ServerView RAID Manager event window ID: 10267 Event: <Controller type and its number>: Rebuild complete on Disk X For events via ServerView Source : Fujitsu ServerView Services Type : Information Event ID : 1 Description: <Controller type and its number>: Rebuild complete on Disk X For the approximate time to complete a rebuild, refer to " Estimated time for rebuild" (pg.34) or "4.9.5 Checking Progress of Background Task" (pg.178). If the server restarts or shuts down during a rebuild, the rebuild is resumed from where it was stopped the next time. 6.1.4 Preventive Replacement of Hard Disk [ServerView RAID] When the hard disk failure prediction function (PFA/S.M.A.R.T.) reports a hard disk as "S.M.A.R.T. errors", it means that the disk may fail in the near future. If the hard disk status becomes "S.M.A.R.T. errors", replace that disk as a preventive measure. The procedure for preventive replacement of a hard disk differs depending on the RAID level (redundancy availability) on the logical drive whose hard disk is to be replaced, and,with a RAID level with redundancy, whether the hard disk supports hot swap. When the logical drive does not have redundancy (RAID 0) " Preventive replacement of hard disk with RAID 0 configuration" (pg.269) When the logical drive has redundancy (RAID 1, RAID 5, RAID 6, or RAID 10), and supports hot swap " Preventive replacement of hard disk with RAID 1, RAID 5, RAID 6, or RAID 10 configuration (with hot swap support)" (pg.258) When the logical drive has redundancy (RAID 1, RAID 5, RAID 6, or RAID 10), but does not support hot swap " Preventive replacement of a hard disk with RAID 1, RAID 5, RAID 6, or RAID 10 configuration (no hot swap support)" (pg.261) 255

Chapter 6 Replacing Hard Disk Follow the procedure in " Checking availability of redundancy" (pg.268) if you do not know whether the logical drive has redundancy or not. For how to remove and install hard disks, and for hot swap support, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Replace the hard disk that is predicted to fail with a new one of the same model (with the same capacity and rotation speed) as a rule. It is recommended that you back up the data before replacing a hard disk as a preventive measure. If there is a failed hard disk at this point, replace the disk first by referring to "6.1.2 Replacing Failed Hard Disk [ServerView RAID] (With Hot Swap Support)" (pg.251) or "6.1.3 Replacing Failed Hard Disk [ServerView RAID] (No Hot Swap Support)" (pg.253). If any hard disks are being rebuilt, wait until the rebuild process is complete. Checking availability of redundancy Following the procedure below if you do not know whether the logical drive has redundancy or not. 1 Start up ServerView RAID Manager, and log in. "4.5 Starting and Exiting ServerView RAID Manager" (pg.133) 2 Check the disk number of the hard disk that is predicted to fail. The bay number is the number of the slot number plus one. The slot number can be identified with Step 1 and 2 of "6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID]" (pg.250). Example: If the slot number is 2, the location of the disk is bay 3. 3 In the tree view, select a logical drive ( ), and click the [Layout] tab in the object window. A list of the hard disks that compose the selected logical drive appears. 4 Check whether the hard disk, which was predicted to have a failure in Step 2, is included in the hard disks that configure the logical drive. If there is a hard disk that is predicted to fail Click the [General] tab and check the RAID level. Depending on the RAID level, replace the hard disk as a preventive measure. If there is no hard disk that is predicted to fail Repeat the steps from Step 3 to check the other logical drives. 256

MegaRAID SAS User's Guide Preventive replacement of hard disk with RAID 0 configuration When a hard disk to be replaced configures RAID 0 logical drive, the disk array must be reconfigured and the data must be restored from the backup. Follow the procedure below to replace a hard disk as a preventive measure. Replace the hard disk that is predicted to fail with a new one of the same model (with the same capacity and rotation speed) as a rule. 1 Back up all the data on the hard disks connected to this array controller. 2 By using ServerView RAID Manager, check the disk number of the hard disk that has a failure prediction sign ( ), and locate that disk. The bay number is the number of the slot number plus one. The slot number can be identified with Step 1 and 2 of "6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID]" (pg.250). Example: If the slot number is 2, the location of the disk is bay 3. If there is a failed hard disk at this point, replace the disk first by referring to "6.1.2 Replacing Failed Hard Disk [ServerView RAID] (With Hot Swap Support)" (pg.251) or "6.1.3 Replacing Failed Hard Disk [ServerView RAID] (No Hot Swap Support)" (pg.253). If any hard disks are being rebuilt, wait until the rebuild process is complete. 3 Restart the server. 4 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 5 Select [Configuration Wizard] from the main menu. The [Configuration Wizard] window appears. 257

Chapter 6 Replacing Hard Disk 6 Select [Clear Configuration], and click [Next]. When the following warning message appears, click [Yes]. By selecting [Clear Configuration], all the data stored on the hard disks connected to this array controller is deleted. 7 Exit WebBIOS, and turn off the server. "2.1.2 Exiting WebBIOS" (pg.52) 8 Replace all the hard disks that are predicted to fail with new ones. For how to replace hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 9 When the hard disk replacement is complete, turn on the server. 10 Create a disk array configuration by using WebBIOS. "2.3.1 Creating Disk Array Configuration" (pg.66) 11 Restore the backup data. Preventive replacement of hard disk with RAID 1, RAID 5, RAID 6, or RAID 10 configuration (with hot swap support) When a hard disk to be replaced as a preventive measure configures RAID 1, RAID 5, RAID 6, or RAID 10 logical drive, and the server supports hot swap, follow the procedure below for the preventive replacement. If a hard disk that is predicted to fail configures RAID 0 array, do not perform this procedure for the preventive replacement. For the replacement procedure, refer to " Preventive replacement of hard disk with RAID 0 configuration" (pg.269). Replace the hard disk that is predicted to fail with a new one of the same model (with the same capacity and rotation speed). When two or more hard disks are predicted to fail, replace one at a time. Back up the data before the preventive replacement of a hard disk. Do not follow this procedure if your server does not support hot swap. Even with RAID 1 / RAID 5 / RAID 6 / RAID 10, when your server does not support hot swap, refer to " Preventive replacement of a hard disk with RAID 1, RAID 5, RAID 6, or RAID 10 configuration (no hot swap support)" (pg.261). 258

MegaRAID SAS User's Guide 1 By using ServerView RAID Manager, check the disk number of the hard disk that has a failure prediction sign ( ), and locate that disk. The bay number is the number of the slot number plus one. The slot number can be identified with Step 1 and 2 of "6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID]" (pg.250). Example: If the slot number is 2, the location of the disk is bay 3. If there is a failed hard disk at this point, replace the disk first by referring to "6.1.2 Replacing Failed Hard Disk [ServerView RAID] (With Hot Swap Support)" (pg.251) or "6.1.3 Replacing Failed Hard Disk [ServerView RAID] (No Hot Swap Support)" (pg.253). If any hard disks are being rebuilt, wait until the rebuild process is complete. 2 Make sure that there are no media errors by performing Make Data Consistent. "4.8.7 Make Data Consistent" (pg.168) 3 In the tree view, select the hard disk ( ) with a failure prediction sign. Detailed information about the selected hard disk appears in the object window. If [Status] is "S.M.A.R.T. errors", a failure prediction sign (S.M.A.R.T.) has been informed. 4 In the tree view, select the target hard disk, right-click, and then click [Locate device] from the displayed menu to check the location of the target hard disk on the server. A hard disk failure LED corresponding to the hard disk starts blinking. For bay locations and the LED location, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 259

Chapter 6 Replacing Hard Disk The hard disk confirmed here has received a failure prediction sign (the target hard disk for preventive replacement). Out a mark on this disk to identify it later. 5 After checking the location, in the tree view, select the target hard disk, rightclick, and then click [Stop location] from the displayed menu. The failure LED goes off. 6 In the tree view, select the target hard disk, right-click, and then click [Make Offline] from the displayed menu. The following message appears. Are you sure you want to set this physical disk to offline? 7 Type "yes", and click [OK]. If a hot spare disk has been set, a rebuild automatically starts. The hard disk failure LED for the disk to be replaced is now lit. If the hard disk failure LED is not lit, right-click the disk to be replaced, and then click [Locate device] from the displayed menu again to check the location of the disk to be replaced. Should you change a hard disk other than the one to be replaced to [Make Offline], continue to perform the preventive replacement procedure for the hard disk as [Make Offline] and complete the rebuild. After that, perform the preventive replacement procedure for the correct disk to be replaced. 8 In the object window, check that [Status] of the hard disk to be replaced has changed to "Failed", "Offline" or "Available". 9 Pull the hard disk identified in Step 4 about an inch (1 to 3 cm) to disconnect it from the connector. For how to remove hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Make sure to remove a right hard disk. Removing a wrong hard disk may cause a loss of data. Do not pull out the hard disk completely from the server at this point. 10 Wait at least one minute until the hard disk motor stops rotating. 11 Pull out the hard disk that is predicted to fail completely from the hard disk unit bay. 12 Install a new hard disk at the same location as the removed hard disk once was. 260

MegaRAID SAS User's Guide 13 Check the following. If a spare disk was set Shortly after installing a new hard disk, it automatically becomes a spare disk. In the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Global Hot Spare" in the object window. If no spare disk was set Shortly after installing a new hard disk, a rebuild automatically starts for the newly installed hard disk. When the rebuild starts, the hard disk failure LED that was lit starts blinking, and then it goes off when the rebuild is complete. After the rebuild is complete, in the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Operational" in the object window. When the following event is logged in the OS event log, or in the event window of the ServerView RAID Manager, the rebuild is complete. ("X" indicates the number of the hard disk where the rebuild was performed.) Source: ServerView RAID event, or ServerView RAID Manager event window ID: 10267 Event: <Controller type and its number>: Rebuild complete on Disk X Event via ServerView Source : Fujitsu ServerView Services Type : Information Event ID : 1 Description: <Controller type and its number>: Rebuild complete on Disk X If the server restarts or shuts down during a rebuild, the rebuild is resumed from where it was stopped the next time. For the approximate time to complete the rebuild, refer to " Estimated time for rebuild" (pg.34) or "4.9.5 Checking Progress of Background Task" (pg.178). If you replace a hard disk while the Locate Device function is in progress, the disk may be displayed with the ( ) icon after the replacement, indicating that the Locate Device function is in progress. If this happens, right-click on the hard disk and select [Stop location] from the displayed menu. Preventive replacement of a hard disk with RAID 1, RAID 5, RAID 6, or RAID 10 configuration (no hot swap support) When a hard disk to be replaced as a preventive measure configures RAID 1, RAID 5, RAID 6, or RAID 10 logical drive, and the server does not support hot swap, follow the procedure below for the preventive replacement. If a hard disk that is predicted to fail configures RAID 0 array, do not perform this procedure for the preventive replacement. Refer to " Preventive replacement of hard disk with RAID 0 configuration" (pg.269) for the replacement procedure. Replace the hard disk that is predicted to fail with a new one of the same model (with the same capacity and rotation speed). When two or more hard disks are predicted to fail, replace one at a time. Back up the data before the preventive replacement of a hard disk. 261

Chapter 6 Replacing Hard Disk 1 By using ServerView RAID Manager, check the disk number of the hard disk that has a failure prediction sign ( ), and locate the disk. The slot number identified in Step 1 to 2 in "6.1.1 Checking Hard Disk to Be Replaced [ServerView RAID]" (pg.250) plus one is the bay number. Example: If the slot number is 2, the location of the disk is bay 3. To specify a bay location, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. If there is a failed hard disk at this point, replace the disk first by referring to "6.1.2 Replacing Failed Hard Disk [ServerView RAID] (With Hot Swap Support)" (pg.251) or "6.1.3 Replacing Failed Hard Disk [ServerView RAID] (No Hot Swap Support)" (pg.253). If any hard disks are being rebuilt, wait until the rebuild process is complete. 2 Make sure that there are no errors on the media with Make Data Consistent. "4.8.7 Make Data Consistent" (pg.168) 3 In the tree view, right-click the hard disk icon ( ) with a failure prediction sign to select [Make Offline]. The [Confirmation needed] window appears. Type "yes", and click [OK]. A rebuild automatically starts if hot spare is set. 4 In the object window, check that [Status] of the hard disk to be replaced has been changed to "Failed", "Offline", or "Available". 5 Shut down the server, and disconnect the AC cable. 6 Disconnect all the cables that are connected to the failed hard disk, and completely pull out the failed hard disk. For how to remove hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 7 Install a new hard disk at the same location as the failed hard disk once was. Connect all the cables that were connected to the failed one to the new one. 8 Turn on the server, and start the OS. 262

MegaRAID SAS User's Guide 9 Start up ServerView RAID Manager to check the following. The checking procedure differs depending on whether a spare disk was set before the hard disk failure. If a spare disk was set The newly installed hard disk automatically becomes a spare disk. After the rebuild is complete, in the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Global Hot Spare" in the object window. The replacement procedure is complete here. If no spare disk was set The newly installed hard disk becomes a hard disk without an array configuration. In the tree view of ServerView RAID Manager, select the icon of the replaced hard disk to see if [Status] is "Available" in the object window. Proceed to the following steps. 10 Install the new hard disk into the array configuration. 1. Log in ServerView RAID Manager with Administrative privileges. 2. Check the logical drive status. In the tree view of ServerView RAID Manager, select an logical drive icon in critical or partially critical ( ). Check if its [Status] is "Degraded" or "Partially Degraded" in the object window. 3. In the ServerView RAID Manager tree view, right-click the hard disk icon of the newly installed ( ) to select [Replace missing disk] from the displayed menu. When the [Write access needed!] window appears, click [Yes] to proceed. 4. Select an empty slot ( ), and click [Replace]. The [Confirmation needed] window appears. 5. Click [Yes]. The procedure above will result in the status of the new hard disk in "Offline" ( ). It is not a hard disk failure only after this operation. 263

Chapter 6 Replacing Hard Disk 11 Rebuild the new hard disk. 1. In the ServerView RAID Manager tree view, right-click the icon of the new hard disk ( ), and select [Start rebuild] from the displayed menu. The [Confirmation needed] window appears. 2. Click [Yes]. A rebuild starts. After the rebuild, select the icon of the new hard disk in the ServerView RAID Manager tree view to see if the Status is "Operational". When the following events are logged in the OS event log, or in the event window of Server- View RAID Manager, the rebuild is complete. ("X" indicates the number of the hard disk where the rebuild was performed.) Source: ServerView RAID event, or ServerView RAID Manager event window ID: 10267 Event: <Controller type and its number>: Rebuild complete on Disk X Events via ServerView Source : Fujitsu ServerView Services Type : Information Event ID : 1 Description: <Controller type and its number>: Rebuild complete on Disk X For the approximate time to complete a rebuild, refer to " Estimated time for rebuild" (pg.34) or "4.9.5 Checking Progress of Background Task" (pg.178). If the server restarts or shuts down during a rebuild, the rebuild is resumed from where it was stopped the next time. 264

MegaRAID SAS User's Guide 6.2 How to Replace Hard Disk [GAM] This section explains maintenance related issues in GAM such as hard disk replacement. 6.2.1 Checking Hard Disk to Be Replaced [GAM] Check the slot number of the target hard disk before replacing it. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Controller View] from the [View] menu. "5.6.4 Starting Controller View and the Window Layout" (pg.210) 3 Check the displayed icon of the hard disk. A failed hard disk is indicated with the icon. A hard disk that has been predicted to fail is indicated with the The slot number can be checked on the hard disk icon as follows. icon. Slot number In this case, the slot number is "4". If there are any hard disks being rebuilt (indicated with ), wait until the rebuild is complete. After the rebuild is complete, check the hard disk status again. 4 Double-click the hard disk icon to check detailed information. Detailed information about the selected hard disk appears. You can check whether a failure prediction sign (S.M.A.R.T.) has been informed by referring to [PFA Count] in the [Device Errors] field. The detailed information may not appear depending on the failure condition of the hard disk. 265

Chapter 6 Replacing Hard Disk 5 If there is a failed hard disk or a hard disk that is predicted to fail, replace it according to your situation. If there is a failed hard disk Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk. If there is a hard disk that is predicted to fail Refer to "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268) to replace the hard disk with the failure prediction sign. When a hard disk failed and another is predicted to fail First, replace the failed hard disk, and perform a rebuild. After that, check that the failure indication of the replaced hard disk has disappeared, i.e. that the logical drive status is online, and then replace the hard disk that is predicted to fail, as a preventive measure. If the hard disk that is predicted to fail is replaced before the failed hard disk, a rebuild cannot be performed and data will be lost. Failed hard disks can also be checked by using WebBIOS. Start up WebBIOS, and check the [Physical Drives] view. For details, refer to "2.2.1 Checking Hard Disk Status" (pg.53). 6.2.2 Replacing Failed Hard Disk [GAM] When a hard disk fails, replace it with a new one as soon as possible. For how to remove and install hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Replace the failed hard disk with a new one of the same model (with the same capacity and rotation speed) as a rule. When connecting a hard disk that has previously been used in a disk array configuration on a general host adapter, perform a low level format of the hard disk on the host adapter in advance. Never remove any hard disks while the server is turned on, except to replace a failed disk. 1 Check the slot number of the failed hard disk, and locate the disk. The bay number is the number of the slot number plus one. The slot number can be identified with Step 1 to 3 in "6.2.1 Checking Hard Disk to Be Replaced [GAM]" (pg.265). Example: If the slot number is 4, the location of the disk is bay 5. 2 On the server, check that a hard disk failure LED is lit for the bay corresponding to the failed hard disk. For bay locations and the LED location, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 266

MegaRAID SAS User's Guide 3 Pull the failed hard disk about an inch (1 to 3 cm) to disconnect it from the connector. For how to remove hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Do not pull out the hard disk completely from the server at this point. 4 Wait at least one minute until the hard disk motor stops rotating. 5 Pull out the failed hard disk completely from the hard disk unit bay. 6 Install a new hard disk at the same location as the failed hard disk once was. 7 Check the following. The checking procedure differs depending on whether a spare disk was set before the hard disk failure. If a spare disk was set Shortly after installing a new hard disk, it automatically becomes a spare disk, and the corresponding hard disk failure LED goes off. Make sure that the status of the replaced hard disk has changed to "Hot Spare" in the [Disk Device Information] window (pg.237) of GAM. If no spare disk was set Shortly after installing a new hard disk, a rebuild automatically starts for the newly installed hard disk. When the rebuild starts, the hard disk failure LED that was lit starts blinking, and then it goes off when the rebuild is complete. When the rebuild is complete, make sure that the status of the replaced hard disk has changed to "On Line" in the [Disk Device Information] window (pg.237) of GAM. When the following event is logged in the OS event log, or in the Log Information Viewer of the GAM Client, a rebuild is complete. ([ctl] indicates the controller number and [chn] indicates the slot number of the hard disk.) For Log Information Viewer I-7 ctl:x chn:y tgt:z Rebuild is over. For OS event log Source :Fujitsu ServerView Services Type : Information Event ID : 1 Description: [ctl:x chn:y tgt:x] Rebuild is over. To confirm the completion of the rebuild in GAM Client, do not close GAM Client until the rebuild is complete. For the approximate time to complete a rebuild, refer to " Estimated time for rebuild" (pg.34) or " Calculating approximate time needed for a background task" (pg.244). If the server restarts or shuts down during a rebuild, the rebuild is resumed from where it was stopped the next time. 267

Chapter 6 Replacing Hard Disk 6.2.3 Preventive Replacement of Hard Disk [GAM] When the hard disk failure prediction function (PFA/S.M.A.R.T.) reports a hard disk as "Critical", it means that the disk may fail in the near future. If the hard disk status becomes "Critical", replace that disk as a preventive measure. For how to remove and install hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Replace the hard disk that is predicted to fail with a new one of the same model (with the same capacity and rotation speed) as a rule. It is recommended that you back up the data before replacing a hard disk as a preventive measure. If there is a failed hard disk at this point, replace the disk first by referring to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266). If any hard disks are being rebuilt, wait until the rebuild process is complete. Checking availability of redundancy The procedure for preventive replacement of a hard disk is different depending on the RAID level (availability of redundancy) on the logical drive that the hard disk to be replaced is installed in. When the logical drive does not have redundancy (RAID 0) " Preventive replacement of hard disk with RAID 0 configuration" (pg.269) When the logical drive has redundancy (RAID 1, RAID 5, or RAID 10) " Preventive replacement of hard disk with RAID 1, RAID 5, or RAID 10 configuration" (pg.271) Follow the procedure below if you do not know whether the logical drive has redundancy or not. 1 Start up GAM, and sign on. "5.5 Starting and Exiting GAM" (pg.202) 2 Select [Controller View] from the [View] menu. "5.6.4 Starting Controller View and the Window Layout" (pg.210) 268

MegaRAID SAS User's Guide 3 Right-click to select a logical drive icon displayed in the [Logical Drives] field in the [Controller View] window of GAM. Check the [Physical Devices] field. The icons for the hard disks configuring the selected logical drive appear in white. RAID level of a logical drive Right-click a logical drive Hard disk drives composing the logical drive Right-clicking the logical drive icon again reverts the display to its original status. 4 Check all the hard disks included in the disk array configuration for each logical drive. 5 Check which logical drive uses the failure-predicted hard disk ( ). 6 Check the RAID level for the logical drive. Preventive replacement of hard disk with RAID 0 configuration When a hard disk to be replaced configures RAID 0 logical drive, the disk array must be reconfigured and the data must be restored from the backup. Follow the procedure below to replace a hard disk as a preventive measure. Replace the hard disk that is predicted to fail with a new one of the same model (with the same capacity and rotation speed) as a rule. 1 Back up all the data on the hard disks connected to the array controller. 269

Chapter 6 Replacing Hard Disk 2 By using GAM, check the slot number of the hard disk with a failure prediction sign ( ) to locate the drive. The bay number is the number of the slot number plus one. The slot number can be identified in Step 1 to 3 in "6.2.1 Checking Hard Disk to Be Replaced [GAM]" (pg.265). Example: If the slot number is 4, the location of the disk is bay 5. If there is a failed hard disk at this point, replace the disk first by referring to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266). If any hard disks are being rebuilt, wait until the rebuild process is complete. 3 Restart the server. 4 Start up WebBIOS. "2.1.1 Starting Up WebBIOS" (pg.48) 5 Select [Configuration Wizard] from the main menu. The [Configuration Wizard] window appears. 6 Select [Clear Configuration], and click [Next]. When the following warning message appears, click [Yes]. 270

MegaRAID SAS User's Guide By performing [Clear Configuration], all the data stored on the hard disks connected to this array controller is deleted. 7 Exit WebBIOS, and turn off the server. "2.1.2 Exiting WebBIOS" (pg.52) 8 Replace all the hard disks that are predicted to fail with new ones. For how to replace hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. 9 When the hard disk replacement is complete, turn on the server. 10 Create a disk array configuration by using WebBIOS. "2.3.1 Creating Disk Array Configuration" (pg.66) 11 Restore the backup data. Preventive replacement of hard disk with RAID 1, RAID 5, or RAID 10 configuration When a hard disk to be replaced as a preventive measure configures RAID 1, RAID 5, or RAID 10 logical drive, follow the procedure below for the preventive replacement. If a hard disk that is predicted to fail configures a RAID 0 array, do not perform this procedure for the preventive replacement. Refer to " Preventive replacement of hard disk with RAID 0 configuration" (pg.269) for the replacement procedure. Replace the hard disk that is predicted to fail with a new one of the same model (with the same capacity and rotation speed) as a rule. When two or more hard disks are predicted to fail, replace one disk at a time. It is recommended that you back up the data before replacing a hard disk as a preventive measure. 1 By using GAM, check the slot number of the hard disk with a failure prediction sign ( ) to locate the disk. The bay number is the number of the slot number plus one. The slot number can be identified with Step 1 to 3 of "6.2.1 Checking Hard Disk to Be Replaced [GAM]" (pg.265). Example: If the slot number is 4, the location of the disk is bay 5. If there is a failed hard disk at this point, replace the disk first by referring to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266). If any hard disks are being rebuilt, wait until the rebuild process is complete. 2 Make sure that there are no media errors by performing Make Data Consistent. "5.10.1 Make Data Consistent" (pg.245) 271

Chapter 6 Replacing Hard Disk 3 Double-click the icon of the hard disk with the failure prediction sign. Detailed information about the selected hard disk appears. Make sure that its "Status" is indicated as "Critical". 4 Click [Locate] to check the location of the target hard disk on the server. A hard disk failure LED corresponding to the hard disk starts blinking. For bay locations and the LED location, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. The hard disk confirmed here has received a failure prediction sign (the target hard disk for preventive replacement). Put a mark on this disk to identify it later. 5 Click [OK] when the location is confirmed. The failure LED goes off. 6 Click the [Make Offline] button. When the [WARNING] window appears, type "YES", and click [OK]. If a hot spare is set, a rebuild automatically starts. The hard disk failure LED for the disk to be replaced is now lit. If the hard disk failure LED is not lit, click [Locate] again to check the location of the disk for the preventive replacement. If you select a wrong hard disk for [Make Offline], continue the preventive replacement procedure with that hard disk with [Make Offline] and complete a rebuild. After that, perform the preventive replacement procedure for the proper disk. 272

MegaRAID SAS User's Guide 7 Check that the following log entry appears in the GAM Log Information Viewer. Event ID : E-50 Description: Physical disk status changed to offline 8 Pull the hard disk identified in Step 4 about an inch (1 to 3 cm) to disconnect it from the connector. For how to remove hard disks, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. Make sure to remove a right hard disk. Removing a wrong hard disk may cause a loss of data. Do not pull out the hard disk completely from the server at this point. 9 Wait at least one minute until the hard disk motor stops rotating. 10 Pull out the hard disk that is predicted to fail completely from the hard disk unit bay. 11 Install a new hard disk at the same location as the removed hard disk once was. 12 Check the following. If a spare disk was set Shortly after installing a new hard disk, it automatically becomes a spare disk. Make sure that the status of the replaced hard disk has changed to "Hot Spare" in the [Disk Device Information] window (pg.237) of GAM. If no spare disk was set Shortly after installing a new hard disk, a rebuild automatically starts for the newly installed hard disk. When the rebuild starts, the hard disk failure LED that was lit starts blinking, and then it goes off when the rebuild is complete. When the rebuild is complete, make sure that the status of the replaced hard disk has changed to "On Line" in the [Disk Device Information] window (pg.237) of GAM. If the server restarts or shuts down during a rebuild, the rebuild is resumed from where it was stopped the next time. If the [Controller View] display is not updated, perform [Scan Devices] from the [Administration] menu. 273

Chapter 6 Replacing Hard Disk 274

Appendix This section explains event codes for ServerView RAID and GAM, and how to replace batteries. A List of ServerView RAID Event Log................. 276 B List of GAM Event Log........................... 313 C Replacing Battery.............................. 336 275

Appendix A List of ServerView RAID Event Log With ServerView, SNMP traps issued by ServerView RAID can be saved in the OS logs. ServerView RAID also has an own function to save logs. For Windows Events coming through ServerView are recorded as the source "Fujitsu ServerView Services" in the event viewer application log. Events from ServerView RAID are recorded as the source "ServerView RAID" in the event viewer application log. For Linux Events coming through ServerView are recorded as the source "Fujitsu ServerView Services" in the system log. Events from ServerView RAID are recorded as the source "ServerView RAID" in the system log. Refer to the events by ServerView instead of ServerView RAID when events are notified from both. The log also records the location of the device. Types of locations are as follows. Character string [Note 1] Server %s Adapter %s Disk %s Logical Drive %s table: Meaning of event log strings Meaning Target server name. Events coming through ServerView only save them. Displays the name and number of the array controller. As for this array controller, the name is "LSI MegaRAID SAS..." or "RAID 5/6 SAS based on LSI MegaRAID". Hard disk number. "x" or "y:x" appears in %s representing the hard disk number. Logical drive number. [Note 1]: The mark, %s, is replaced with a number or a character string. Unless ServerView is installed, logging into the OS event logs from the source "Fujitsu ServerView Services" will not be possible. Refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server to install and configure ServerView. 276

MegaRAID SAS User's Guide The relationship between event types and each event log function is as follows. Severity Description table: Event log types and descriptions Severity with ServerView RAID Manager OS event log type CRITICAL Severe error Error Error MAJOR Error Error Error MINOR Warning Warning Warning INFORMA Information (No action required) Informational Information List of ServerView RAID event log The ID numbers described below shows event IDs for the source "ServerView RAID", and Specific Numbers for SNMP traps. table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action 1 INFORMA 2 INFORMA 3 INFORMA Undefined Event Connection to ServerView RAID Manager established Disconnected from ServerView RAID Manager 4 CRITICAL Unexpected disconnect from ServerView RAID Manager 10000 INFORMA Unknown event 10002 MINOR Write access to ServerView RAID Manager revoked by user %s (%s) (Server %s) An unknown event has occurred. Connection to ServerView RAID has been confirmed. Connection to ServerView RAID has been cut. Connection to ServerView RAID has been unexpectedly cut. An unknown event has occurred. Write Access mode has been canceled. Another client has obtained Write Access mode. If there are errors before or after this event, perform the proper recovery action for that error. Check if you can log in to ServerView RAID Manager. If you can not, restart ServerView RAID service. If there are errors before or after this event, perform the proper recovery action for that error. 277

Appendix 10003 MAJOR Adapter %s: BBU temperature out of range 10004 MAJOR Adapter %s: BBU temperature unstable 10005 MAJOR Adapter %s: BBU temperature above threshold (Server %s) 10006 MAJOR Adapter %s: BBU voltage out of range 10007 MAJOR Adapter %s: BBU voltage unstable 10008 MAJOR Adapter %s: BBU voltage below threshold (Server %s) 10009 MAJOR Adapter %s: BBU communication error 10010 INFORMA 10011 INFORMA 10012 INFORMA Adapter %s: BBU reconditioning started Adapter %s: BBU status becomes normal Adapter %s: BBU good 10013 CRITICAL Adapter %s: BBU failed 10014 MAJOR Adapter %s: BBU fast charging failed table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Abnormal battery temperature is detected. Abnormal battery temperature is detected. Abnormal battery temperature is detected. Abnormal battery voltage is detected. Battery voltage is unstable. Battery voltage is below the threshold. An error occurred during the communication with the battery backup unit. Battery recalibration has started. The battery is now in normal state. The battery is normal. The battery backup unit has failed. Rapid battery charging has failed. Check the fan status in ServerView. If the fan has a failure, follow the corrective procedure of the server. If the fan has no failure, check the surrounding temperature. If the problem recurs after checking the surrounding temperature, replace the battery backup unit. Check the fan status in ServerView. If the fan has a failure, follow the corrective procedure of the server. If the fan has no failure, check the surrounding temperature. If the problem recurs after checking the surrounding temperature, replace the battery backup unit. Check the fan status in ServerView. If the fan has a failure, follow the corrective procedure of the server. If the fan has no failure, check the surrounding temperature. If the problem recurs after checking the surrounding temperature, replace the battery backup unit. Replace the battery backup unit. Replace the battery backup unit. Wait until the battery charging is complete. Replace the battery backup unit. Replace the battery backup unit. Replace the battery backup unit. 278

MegaRAID SAS User's Guide table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action 10015 MINOR Adapter %s: BBU charge count exceeded (Server %s) 10016 MINOR Adapter %s: BBU needs reconditioning 10023 MAJOR Adapter %s: Disk (%s) timed out 10024 INFORMA Adapter %s: Global hot spare created on disk (%s) (Server %s) 10025 MINOR Adapter %s: Global hot spare deleted on disk (%s) (Server %s) 10026 INFORMA Adapter %s: Dedicated hot spare created on disk (%s) 10027 MINOR Adapter %s: Dedicated hot spare deleted on disk (%s) 10028 INFORMA 10029 INFORMA Adapter %s: Disk (%s) marked available (Server %s) Adapter %s: Rebuild on disk (%s) started 10030 MAJOR Adapter %s: Rebuild on disk (%s) failed 10032 INFORMA Adapter %s: New disk (%s) detected 10033 MINOR Adapter %s: Disk (%s) removed 10038 MAJOR Adapter %s: Error on disk (%s) detected The number of battery recharging times has reached the maximum limit. Battery recalibration is necessary. The hard disk has been timed out. A spare disk has been created. A spare disk has been released. A dedicated spare disk has been created. A dedicated spare disk has been released. The hard disk is now unused. Rebuilding the hard disk has started. Rebuilding the hard disk has failed. A new hard disk has been detected. A hard disk was removed. An error has been detected on a hard disk. Replace the battery backup unit. Perform recalibration. Check that the hard disk is connected properly. If the hard disk has failed, replace and rebuild the new one. A dedicated spare disk is not supported. Reconfigure the spare disk. Check the current logical drive status. For the critical status: Replace the failed hard disk, and perform a rebuild again. For the offline status: Contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. 279

Appendix 10039 INFORMA Adapter %s: Channel %s was reset 10040 MAJOR Adapter %s: Retry I/ O on disk (%s) 10050 MAJOR Adapter %s: Temperature sensor %s in enclosure %s above threshold 10055 INFORMA 10056 INFORMA Adapter %s: Rebuild started on logical drive %s Adapter %s: Rebuild finished on logical drive %s 10057 MAJOR Adapter %s: Rebuild failed on logical drive %s 10058 MINOR Adapter %s: Rebuild aborted on logical drive %s 10068 INFORMA 10069 INFORMA Adapter %s: Migration started on logical drive %s Adapter %s: Migration finished on logical drive %s 10070 MAJOR Adapter %s: Migration failed on logical drive %s 10071 MINOR Adapter %s: Migration aborted on logical drive %s 10072 INFORMA Adapter %s: Patrol Read started (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action A channel has been reset. I/O retry for the hard disk has been performed. Temperature sensor in enclosure is above the threshold. Rebuilding the logical drive has started. Rebuilding the logical drive is complete. Rebuilding the logical drive has failed. Rebuilding the logical drive has been aborted. The Expand Capacity process of the logical drive has started. The Expand Capacity process of the logical drive is complete. The Expand Capacity process of the logical drive has failed. The Expand Capacity process of the logical drive has been aborted. Patrol Read has started. There is no problem as long as there is no hard disk failure, because the firmware has performed a recovery. Check the fan status in ServerView. If the fan has a failure, follow the corrective procedure of the server. If the fan has no failure, check the surrounding temperature. Check the current logical drive status. For the critical status: Replace the failed hard disk, and perform a rebuild again. For the offline status: Contact an office listed in the "Contact Information" of "Start Guide". Perform a rebuild again. Recreate the array configuration, and restore the data from the backup. Recreate the array configuration, and restore the data from the backup. Patrol Read is not supported. Check the array controller configuration. 280

MegaRAID SAS User's Guide 10073 INFORMA Adapter %s: Patrol Read finished 10074 MINOR Adapter %s: Patrol Read aborted (Server %s) 10075 INFORMA 10076 INFORMA 10077 INFORMA Adapter %s: Patrol Read stopped Adapter %s: Patrol Read paused (Server %s) Adapter %s: Patrol Read resumed 10078 MAJOR Adapter %s: Logical drive %s degraded 10079 CRITICAL Adapter %s: Logical drive %s failed 10080 INFORMA Adapter %s: Logical drive %s created 10081 MINOR Adapter %s: Logical drive %s deleted 10082 INFORMA 10085 INFORMA 10086 INFORMA 10087 INFORMA 10088 INFORMA Adapter %s: Logical drive %s operational Adapter %s: Initialization started on logical drive %s Adapter %s: Initialization finished on logical drive %s Adapter %s: BGI started on logical drive %s Adapter %s: BGI finished on logical drive %s 10089 MINOR Adapter %s: BGI canceled on logical drive %s 10090 MINOR Adapter %s: Initialization canceled on logical drive %s table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Patrol Read is complete. Patrol Read has been aborted. Patrol Read has stopped. Patrol Read has paused. Patrol Read has resumed. The logical drive is now critical. The logical drive is now offline. The logical drive has been created. The logical drive has been deleted. The logical drive is now online. Foreground initialization of the logical drive has started. Foreground initialization of the logical drive is complete. Background initialization of the logical drive has started. Background initialization of the logical drive is complete. Background initialization of the logical drive has been canceled. Foreground initialization of the logical drive has been canceled. Patrol Read is not supported. Check the array controller configuration. Replace the failed hard disk, and perform a rebuild. Contact an office listed in the "Contact Information" of "Start Guide". Background initialization resumes automatically after a certain period of time. 281

Appendix 10095 INFORMA Adapter %s: Alarm enabled 10096 MINOR Adapter %s: Alarm disabled 10101 INFORMA 10102 INFORMA 10103 INFORMA 10104 INFORMA 10105 INFORMA 10106 INFORMA 10107 INFORMA 10108 INFORMA 10109 INFORMA 10113 INFORMA 10114 INFORMA 10115 INFORMA 10116 INFORMA 10117 INFORMA 10118 INFORMA Adapter %s: Patrol Read disabled Adapter %s: Patrol Read set to automatic (Server %s) Adapter %s: Patrol Read set to manual Adapter %s: Patrol Read delay changed Adapter %s: Initialization set to normal Adapter %s: Initialization set to fast Adapter %s: Initiator ID changed (Server %s) Adapter %s: Automatic rebuild enabled Adapter %s: Automatic rebuild disabled Adapter %s: MDC rate changed to %s%% Adapter %s: BIOS enabled Adapter %s: BIOS disabled Adapter %s: BIOS stop on error enabled Adapter %s: BIOS stop on error disabled Adapter %s: Write cache on all disks enabled table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action An alarm has been enabled. An alarm has been disabled. Patrol Read has been disabled. Patrol Read has been set to Automatic mode. Patrol Read has been set to Manual mode. Patrol Read Delay has been changed. Initialization mode has been set to Normal. Initialization mode has been set to Fast. The Initiator ID has been changed. Auto Rebuild has been enabled. Auto Rebuild has been disabled. MDC Rate has been changed. BIOS has been enabled. BIOS has been disabled. BIOS Stop on Error has been enabled. BIOS Stop on Error has been disabled. Write cache has been enabled on all hard disks. Patrol Read is not supported. Check the array controller configuration. Patrol Read is not supported. Check the array controller configuration. Patrol Read is not supported. Check the array controller configuration. Do not change the Initiator ID. Check the configuration of the array controller, and set the correct value. Enabling write cache is not supported. Check the logical drive settings. 282

MegaRAID SAS User's Guide 10119 INFORMA 10120 INFORMA 10121 INFORMA 10124 INFORMA 10125 INFORMA 10131 INFORMA 10132 INFORMA 10133 INFORMA Adapter %s: Write cache on all disks disabled Adapter %s: Readahead on all disks enabled Adapter %s: Readahead on all disks disabled Adapter %s: Spinup drive count changed (after next reboot) Adapter %s: Spinup delay changed (after next reboot) (Server %s) Adapter %s: S.M.A.R.T. poll interval changed Adapter %s: Configuration rescanned (Server %s) Adapter %s: Configuration cleared 10151 MAJOR Adapter %s: BBU voltage problem detected 10152 MAJOR Adapter %s: BBU temperature problem detected 10153 INFORMA Adapter %s: BBU charging 10154 CRITICAL Adapter %s: BBU failed 10155 INFORMA 10156 INFORMA 10168 INFORMA Adapter %s: BBU normal Adapter %s: BBU discharging (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Adapter %s: Logical drive %s: Name changed Write cache has been disabled on all hard disks. Read Ahead has been enabled on all hard disks. Read Ahead has been disabled on all hard disks. Spinup Drive Count has been changed. Spinup Delay has been changed. S.M.A.R.T. poll interval has been changed. The array configuration has been rescanned. The array configuration has been deleted. Abnormal battery voltage is detected. Abnormal battery temperature is detected. The battery is charging. A battery has failed. The battery is normal. The battery is discharging. The name of the logical drive has been changed. Enabling Read Ahead is not supported. Check the logical drive settings. Replace the battery backup unit. Check the fan status in ServerView. If the fan has a failure, follow the corrective procedure of the server. If the problem recurs after checking the surrounding temperature, replace the battery backup unit. Replace the battery backup unit. 283

Appendix 10171 INFORMA 10172 INFORMA 10180 INFORMA User %s (%s) logged in User %s (%s) logged out Adapter %s: BBU charging started 10195 MAJOR Adapter %s: Unable to recover cache data from TBBU (Server %s) 10196 INFORMA 10199 INFORMA 10200 INFORMA 10202 INFORMA Adapter %s: Cache data recovered from TBBU successfully Adapter %s: BGI rate changed to %s%% Adapter %s: Adapter cache discarded due to memory/bbu problems (Server %s) Adapter %s: Cache data recovered successfully (Server %s) 10203 MINOR Adapter %s: Adapter cache discarded due to firmware version Incompatibility 10204 CRITICAL Adapter %s: Fatal firmware error: %s 10205 INFORMA Adapter %s: Factory defaults restored 10206 MAJOR Adapter %s: Flash downloaded image corrupt 10207 MAJOR Adapter %s: Flash erase error (Server %s) 10208 MAJOR Adapter %s: Flash timeout during erase 10209 MAJOR Adapter %s: Flash error table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The user has logged in. The user has logged out. Battery charging has started. Cache data could not be recovered from the battery backup unit. Cache data has been written from the battery backup unit. BGI Rate has been changed. Cache data has been deleted due to a problem with the cache memory or the battery backup unit. Cache data has been recovered successfully. Cache data has been discarded due to the difference in firmware versions. A fatal error has occurred in the firmware. Initial values have been restored. The downloaded firmware image is corrupted. The Flash erasure has failed. A timeout has occurred during the Flash erasure. The Flash has failed. Shut down the system properly, and then restart the system. Check the log entries around this entry, and take the necessary measure. Contact an office listed in the "Contact Information" of "Start Guide". Check the controller's settings, and change them to the correct values. Update the firmware again using a correct image. Update the firmware again. Update the firmware again. Update the firmware again. 284

MegaRAID SAS User's Guide 10210 INFORMA 10211 INFORMA Adapter %s: Flashing image: %s Adapter %s: Flash of new firmware image(s) complete 10212 MAJOR Adapter %s: Flash programming error 10213 MAJOR Adapter %s: Flash timeout during programming 10214 MINOR Adapter %s: Flash chip type unknown 10215 MAJOR Adapter %s: Flash command set unknown (Server %s) 10216 MAJOR Adapter %s: Flash verification failure 10217 INFORMA 10218 INFORMA 10219 INFORMA 10220 INFORMA Adapter %s: Flush rate changed to %s seconds Adapter %s: Hibernate command received from host Adapter %s: Event log cleared (Server %s) Adapter %s: Event log wrapped (Server %s) 10221 MAJOR Adapter %s: Multibit ECC error: ECAR=%s ELOG=%s (%s) 10222 MAJOR Adapter %s: Singlebit ECC error: ECAR=%s ELOG=%s (%s) 10223 MAJOR Adapter %s: Not enough adapter memory table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action A Flash of the image has been performed. A Flash of the new firmware image is complete. An error has occurred during the Flash programming. A timeout has occurred during the Flash programming. The chip type of the Flash is unknown. The Flash command is unknown. The Flash verification has failed. The Flush rate has been changed. The hibernation command was received from the server. The event log has been cleared. The event log has reached the maximum capacity and old log entries have been deleted. A multi-bit error has been detected. A single-bit error has been detected. There is not a sufficient amount of controller memory available. Update the firmware again. Update the firmware again. Update the firmware again using the correct image. Check if the target controller that needs to be updated is correct. Update the firmware again using a proper tool. Update the firmware again. Replace the cache memory or the battery backup unit. If the problem recurs, contact an office listed in the "Contact Information" of "Start Guide". Replace the cache memory or the battery backup unit. If the problem recurs, contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". 285

Appendix 10224 INFORMA 10225 INFORMA 10226 INFORMA 10227 INFORMA 10228 INFORMA 10229 INFORMA 10230 INFORMA Adapter %s: Patrol Read rate changed to %s%% Adapter %s: Migration rate changed to %s%% Adapter %s: Shutdown command received from host Adapter %s: Test event: '%s' (Server %s) Adapter %s: Time established as %s; (%s seconds since power on) (Server %s) Adapter %s: User entered firmware debugger (Server %s) Adapter %s: BGI corrected medium error (logical drive %s at LBA %s on disk (%s) at LBA %s) 10231 MAJOR Adapter %s: BGI completed with uncorrectable errors on logical drive %s 10232 MAJOR Adapter %s: BGI detected uncorrectable multiple medium errors (disk (%s) at LBA %s on logical drive %s) (Server %s) 10233 MAJOR Adapter %s: BGI failed on logical drive %s 10235 INFORMA table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Adapter %s: Logical drive %s: %s changed Patrol Read Rate has been changed. Rate of Expand Capacity has been changed. The shutdown command was received from the server. A test event has been issued. The current time has been set. The firmware has entered the debug mode. A medium error has been corrected during the background initialization. The background initialization is complete, but unrecoverable errors have been detected. Multiple unrecoverable media errors have been detected during the background initialization. Background initialization of the logical drive has failed. A property of the logical drive has been changed. If an unreadable file is found during operation, restore the file from the backup. If an unreadable file is found during operation, restore the file from the backup. Check the log entries around this entry, and take the necessary measure. 286

MegaRAID SAS User's Guide 10236 INFORMA Adapter %s: MDC corrected medium error (logical drive %s at LBA %s on disk (%s) at LBA %s) 10240 MAJOR Adapter %s: Initialization failed on logical drive %s 10242 INFORMA 10243 INFORMA 10244 INFORMA Adapter %s: Fast initialization started on logical drive %s Adapter %s: Full initialization started on logical drive %s Adapter %s: Logical drive %s: Property %s updated (Server %s) 10245 MAJOR Adapter %s: Migration detected uncorrectable multiple medium errors (logical drive %s at LBA %s on disk (%s) at LBA %s) 10247 INFORMA Adapter %s: Migration resumed on logical drive %s 10248 MAJOR Adapter %s: Resume migration of logical drive %s failed due to Configuration Mismatch (Server %s) 10249 INFORMA Adapter %s: State change on logical drive %s from operational to operational (Server %s) 10250 MINOR Adapter %s: Clear aborted on disk (%s) 10251 MAJOR Adapter %s: Clear failed on disk (%s) (error %s) (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action A medium error has been corrected during Make Data Consistent. Foreground initialization has failed. Fast initialization has started. Foreground initialization has started. A property of the logical drive has been changed. Multiple unrecoverable media errors have been detected during Expand Capacity. The Expand Capacity process has resumed. Resumption of Expand Capacity has failed due to a mismatched configuration. The logical drive is now online. PD Clear has been aborted. PD Clear has failed. Check the log entries around this entry, and take the necessary measure. If an unreadable file is found during operation, restore the file from the backup. Recreate the array configuration, and restore the data from the backup. 287

Appendix 10252 INFORMA 10253 INFORMA 10254 INFORMA Adapter %s: Clear progress on disk (%s) is %s (Server %s) Adapter %s: Clear started on disk (%s) Adapter %s: Clear finished on disk (%s) 10255 MAJOR Adapter %s: Error on disk (%s) (error %s) 10258 MAJOR Adapter %s: Hot spare S.M.A.R.T. polling failed on disk (%s) (error %s) 10259 MAJOR Adapter %s: Disk (%s) is not supported 10260 INFORMA Adapter %s: Patrol Read corrected medium error on disk (%s) at LBA %s 10262 MAJOR Adapter %s: Patrol Read found an uncorrectable medium error on disk (%s) at LBA %s 10263 MINOR Adapter %s: Predictive failure: Disk (%s) (Server %s) 10264 MAJOR Adapter %s: Puncturing bad block on disk (%s) at LBA %s table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action PD Clear is in progress. PD Clear has started. PD Clear is complete. An error has occurred on the hard disk. Error codes in (error x) mean as follows: 240: general failure 243: timeout 244: no drive 250: no path 252: no cable 02: check condition 08: busy These error codes indicate the reason why the array controller has disconnected the hard disk. They do not indicate the cause of a hard disk failure. S.M.A.R.T. polling of a spare disk has failed. An unsupported hard disk has been detected. A medium error has been corrected during Patrol Read. Multiple unrecoverable media errors have been detected during Patrol Read. A failure has been predicted for a hard disk. A medium error has been detected in the source disk drive during rebuild. Replace the failed hard disk, and perform a rebuild. Replace the hard disk. After changing the disk, configure it as a spare disk. Use a supported hard disk. If an unreadable file is found during operation, restore the file from the backup. Replace the hard disk as a preventive measure. If an unreadable file is found during operation, restore the file from the backup. 288

MegaRAID SAS User's Guide 10265 MINOR Adapter %s: Rebuild aborted by user on disk (%s) (Server %s) 10266 INFORMA 10267 INFORMA 10269 INFORMA 10270 INFORMA Adapter %s: Rebuild complete on logical drive %s Adapter %s: Rebuild complete on disk (%s) Adapter %s: Rebuild resumed on disk (%s) Adapter %s: Rebuild automatically started on disk (%s) (Server %s) 10273 MAJOR Adapter %s: Unrecoverable medium error during rebuild on disk (%s) at LBA %s (Server %s) 10274 INFORMA Adapter %s: Corrected medium error during recovery on disk (%s) at LBA %s 10275 MAJOR Adapter %s: Unrecoverable medium error during recovery on disk (%s) at LBA %s 10276 INFORMA 10277 INFORMA 10278 INFORMA Adapter %s: Unexpected sense: Disk (%s), CDB:%s, Sense:%s (Server %s) Adapter %s: State change on disk (%s) from available to available (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Adapter %s: State change by user on disk (%s) from available to available Rebuilding has been canceled. Rebuilding the logical drive is complete. Rebuilding the hard disk is complete. Rebuilding has resumed. Rebuilding the hard disk has started automatically. An unrecoverable medium error has been detected during rebuild. A medium error has been corrected. An unrecoverable media error has been detected. Sense information of the hard disk has been reported. The hard disk is now unused. The hard disk is now unused. Perform a rebuild again. If an unreadable file is found during operation, restore the file from the backup. If an unreadable file is found during operation, restore the file from the backup. There is no problem as long as the target hard disk is online because the array controller has performed a recovery. 289

Appendix 10281 MINOR Adapter %s: Dedicated hot spare (%s) no longer useful due to deleted array 10282 MINOR Adapter %s: SAS topology error: Loop detected 10283 MINOR Adapter %s: SAS topology error: Unaddressable device 10284 MINOR Adapter %s: SAS topology error: Multiple ports to the same SAS address 10285 MINOR Adapter %s: SAS topology error: Expander error 10286 MINOR Adapter %s: SAS topology error: SMP timeout 10287 MINOR Adapter %s: SAS topology error: Out of route entries 10288 MINOR Adapter %s: SAS topology error: Index not found 10289 MINOR Adapter %s: SAS topology error: SMP function failed 10290 MINOR Adapter %s: SAS topology error: SMP CRC error (Server %s) 10291 MINOR Adapter %s: SAS topology error: Multiple subtractive table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The dedicated spare disk is unavailable because the logical drive was deleted. A loop has been detected in the SAS topology. A device is unaddressable with the SAS topology. Multiple ports are connected to the same SAS address in the SAS topology. An error has been detected in the Expander. An SMP timeout has been detected. Route entries cannot be found. Index was not found. An error has been detected in an SMP function. A CRC error has been detected in SMP. An error has been detected in the SAS topology. The dedicated spare disk is not supported. Reconfigure the spare disk. Check the connections of the hard disks and cables. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Check the connections of the hard disks and cables. If the system connections are correct, yet a hard disk has failed, replace the hard disk, and perform a rebuild. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Check the connections of the hard disks and cables. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". 290

MegaRAID SAS User's Guide 10292 MINOR Adapter %s: SAS topology error: Table to table 10293 MINOR Adapter %s: SAS topology error: Multiple paths 10294 MAJOR Adapter %s: Unable to access disk (%s) 10295 MINOR Adapter %s: Dedicated hot spare not useful for all arrays 10296 MINOR Adapter %s: Global hot spare does not cover all arrays 10297 MINOR Adapter %s: Marking logical drive %s inconsistent due to active writes at shutdown (Server %s) 10298 INFORMA 10299 INFORMA 10300 INFORMA 10301 INFORMA 10302 INFORMA Adapter %s: BBU present Adapter %s: BBU not present (Server %s) Adapter %s: New BBU detected Adapter %s: BBU has been replaced Adapter %s: BBU temperature is normal table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action An error has been detected in the SAS topology. Multiple paths exist. The hard disk cannot be accessed. No array is available for the dedicated spare disk. No array is available for a spare disk. The consistency of the logical drive has been lost due to shutdown during the Write operation. A battery backup unit was detected. There is no battery backup unit. A new battery backup unit has been detected. The battery backup unit has been replaced. The temperature of the battery backup unit is now normal. Contact an office listed in the "Contact Information" of "Start Guide". Check the connections of the hard disk and cables. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. The dedicated spare disk is not supported. Reconfigure the spare disk. Check the array configuration to make the spare disk available. Check the capacity of the spare disk. Perform Make Data Consistent. If the system does not have any battery backup unit, no action is required. If the system does have a battery backup unit, replace the battery backup unit. 291

Appendix 10303 MINOR Adapter %s: BBU needs to be replaced - SOH bad (Server %s) 10304 INFORMA 10305 INFORMA 10306 INFORMA Adapter %s: BBU relearn started Adapter %s: BBU relearn in progress Adapter %s: BBU relearn finished 10307 MINOR Adapter %s: BBU relearn timed out 10308 INFORMA 10309 INFORMA 10310 INFORMA 10311 INFORMA 10312 INFORMA 10313 INFORMA Adapter %s: BBU relearn pending: BBU is under charge Adapter %s: BBU relearn postponed Adapter %s: BBU relearn will start in 4 days Adapter %s: BBU relearn will start in 2 day Adapter %s: BBU relearn will start in 1 day Adapter %s: BBU Relearn will start in 5 hours 10314 MINOR Adapter %s: BBU removed table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The battery backup unit needs to be replaced. Battery recalibration has started. Battery recalibration is in progress. Battery recalibration is complete. A timeout has occurred during the battery recalibration. The battery is charging and the battery recalibration is in wait mode. BBU recondition has been postponed. Battery recalibration will start within 4 days. Battery recalibration will start within 2 days. Battery recalibration will start within 1 day. Battery recalibration will start within 5 hours. The battery backup unit was removed. The event occurs right after the battery backup unit replacement. Due to the insufficiently charged battery after the replacement, this event may occur. Follow "Appendix C Replacing Battery" (pg.336). The event occurs during regular operation. Replace a battery backup unit. Perform recalibration again. Automatic Learn is not supported. Check the battery backup unit settings. Automatic Learn is not supported. Check the battery backup unit settings. Automatic Learn is not supported. Check the battery backup unit settings. Automatic Learn is not supported. Check the battery backup unit settings. Automatic Learn is not supported. Check the battery backup unit settings. Extremely low voltage of a battery after battery replacement may result in this event. If ID: 10298 "BBU present" is logged, there is no problem. Follow "Appendix C Replacing Battery" (pg.336). If ID: 10298 "BBU present" is not logged, check if a battery backup unit is correctly installed. 292

MegaRAID SAS User's Guide 10315 INFORMA Adapter %s: Enclosure (SES) discovered on %s 10317 MINOR Adapter %s: Enclosure %s communication lost 10318 INFORMA Adapter %s: Enclosure %s communication restored 10325 MINOR Adapter %s: Enclosure %s not supported; too many enclosures connected to port 10331 MINOR Adapter %s: Enclosure %s not responding (Server %s) 10332 MINOR Adapter %s: SAS/ SATA mixing not supported in enclosure; %s disabled 10333 INFORMA Adapter %s: Enclosure (SES) hot plug on %s was detected, but is not supported (Server %s) 10336 MINOR Adapter %s: Disk (%s) too small to be used for auto rebuild 10337 INFORMA Adapter %s: BBU enabled; changing Write-through logical drives to Write-back (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action SES enclosure has been detected. Communication with enclosure has been disconnected. Communication with enclosure has been restored. Enclosure is not working because the number of enclosures detected are more than those supported. Enclosure is not responding. SAS/SATA mix is not supported in enclosure. SES enclosure has been hot plugged. The hard disk capacity is insufficient for rebuild. The battery backup unit has been enabled. The logical drive operation mode has switched from the Write Through mode to the Write Back mode. Check the cable connections. If this problem recurs, contact an office listed in the "Contact Information" of "Start Guide". Check the cable connections. If this problem recurs, contact an office listed in the "Contact Information" of "Start Guide". Check the cable connections. If this problem recurs, contact an office listed in the "Contact Information" of "Start Guide". Replace it with the supported hard disk. Replace the hard disk with one of sufficient capacity. 293

Appendix 10338 MINOR Adapter %s: BBU disabled; changing Write-back logical drives to Writethrough 10339 INFORMA Adapter %s: Bad block table on disk (%s) is 80%% full 10340 MINOR Adapter %s: Bad block table on disk (%s) is full; unable to log Block %s 10341 MINOR Adapter %s: MDC aborted due to ownership loss on logical drive %s 10342 MINOR Adapter %s: BGI aborted due to ownership loss on logical drive %s 10343 MINOR Adapter %s: BBU/ charger problems detected; SOH bad 10344 MINOR Adapter %s: Singlebit ECC error: ECAR=%s ELOG=%s (%s); warning threshold exceeded (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The battery backup unit has been disabled. The logical drive operation mode has switched from the Write Back mode to the Write Through mode. The Bad Block Table usage rate has exceeded 80%. The Bad Block Table is full. Make Data Consistent has been aborted due to an ownership change. Background initialization has been aborted due to an ownership change. A battery charger problem has been detected. A single-bit error has been detected. The event occurs at the system start right after the battery backup unit replacement. Due to the insufficiently charged battery after the replacement, this event may occur. Follow "Appendix C Replacing Battery" (pg.336). The event occurs at the regular system start. Turn on the server for 12 hours or more after the event occurs, and charge the battery. Replace the battery backup unit if ID:10337 does not occur after 12hours. The event occurs during regular operation. There is no problem if it occurs during battery recalibration (including the scheduled one). If not, replace the battery backup unit. A large number of uncorrectable bad block occurred. If an unreadable file is found during operation, restore the file from the backup. A large number of uncorrectable bad block occurred. If an unreadable file is found during operation, restore the file from the backup. If necessary, perform Make Data Consistent again. Replace the battery backup unit. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Replace the cache memory or the battery backup unit. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". 294

MegaRAID SAS User's Guide table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action 10345 MINOR Adapter %s: Singlebit ECC error: ECAR=%s, ELOG=%s, (%s); critical threshold exceeded (Server %s) 10346 INFORMA 10353 INFORMA 10354 INFORMA Adapter %s: Singlebit ECC error: ECAR=%s ELOG=%s (%s); further reporting disabled Adapter %s: Retention test started on previous reboot Adapter %s: NVRAM Retention test passed (Server %s) 10355 MINOR Adapter %s: NVRAM retention test failed (Server %s) 10356 INFORMA Adapter %s: %s test finished %s passes successfully (Server %s) 10357 MINOR Adapter %s: %s test FAILED on %s pass. fail data: erroroffset=%s gooddata=%s baddata=%s (Server %s) 10358 INFORMA 10359 INFORMA 10360 INFORMA 10361 INFORMA Adapter %s: Selfcheck diagnostics finished Adapter %s: Foreign configuration detected Adapter %s: Foreign configuration imported (Server %s) Adapter %s: Foreign configuration cleared 10362 MINOR Adapter %s: NVRAM is corrupt; reinitializing (Server %s) A single-bit error has been detected. A single-bit error has been detected. The NVRAM retention test has started. The NVRAM retention test is complete. The NVRAM retention test has failed. The test is complete. The test has failed. The self-test is complete. Foreign Configuration has been detected. Foreign Configuration has been imported. Foreign Configuration has been cleared. The NVRAM had a failure and re-initialization is in progress. Replace the cache memory or the battery backup unit. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Replace the cache memory or the battery backup unit. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". 295

Appendix 10363 MINOR Adapter %s: NVRAM mismatch occurred 10364 MINOR Adapter %s: SAS wide port %s lost link on PHY %s 10365 INFORMA Adapter %s: SAS wide port %s restored link on PHY %s 10366 MINOR Adapter %s: SAS port %s PHY %s has exceeded the allowed error rate 10367 MINOR Adapter %s: Bad block reassigned on disk (%s) from LBA %s to LBA %s 10369 MINOR Adapter %s: Enclosure %s temperature sensor %s differential detected 10371 INFORMA 10372 INFORMA 10373 INFORMA Adapter %s: Time duration provided by host is not sufficient for self-checking Adapter %s: Disk (%s) on array %s row %s marked missing Adapter %s: Disk (%s) replaced missing on array %s row %s 10377 MINOR Adapter %s: Disk (%s) is not certified 10378 MINOR Adapter %s: Dirty cache data discarded by user table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action An NVRAM mismatch has occurred. The SAS wide port has lost its link. The SAS wide port has restored its link. The errors in the SAS port have exceeded the permissible limit. A bad block of the hard disk has been relocated. An enclosure temperature difference was detected. The system has not provided enough time for self-checking. The hard disk has been marked as missing. Replace Missing of the hard disk has been performed. The hard disk is not a certified disk. Dirty cache data has been discarded by a user. Contact an office listed in the "Contact Information" of "Start Guide". Check the connections of the hard disk and cables. If the system connections are correct, yet a hard disk has failed, replace the hard disk, and perform a rebuild. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. Replace the hard disk with a certified disk. Perform Make Data Consistent. 296

MegaRAID SAS User's Guide 10382 MINOR Adapter %s: Previous configuration completely missing at boot 10383 INFORMA 10385 INFORMA 10386 INFORMA Adapter %s: BBU charge complete Adapter %s: Dedicated hot spare (%s) imported as global due to missing arrays Adapter %s: Disk (%s) rebuild not possible as SAS/ SATA mixing is not supported in an array 10388 MAJOR Adapter %s: Logical drive %s partially degraded (Server %s) 10389 MINOR Adapter %s: BBU requires reconditioning; please initiate a LEARN cycle 10390 INFORMA 10391 INFORMA 10392 INFORMA 10393 INFORMA 10394 INFORMA 10395 INFORMA Adapter %s: Coercion mode changed Adapter %s: BBU automatic learn mode changed Adapter %s: BBU atomatic learn period changed Adapter %s: BBU learn delay interval changed Adapter %s: BBU next learn time changed Adapter %s: Max. disk count for Patrol Read changed to %s table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The previous configuration was not found during startup. The battery charging is complete. The dedicated spare disk has been imported as a spare disk because the target array does not exist. Rebuilding the hard disk is not possible because SAS/ SATA is not supported. The logical drive is now critical. Recalibration is necessary. The coercion mode has been changed. The Autolearn mode of the battery has been changed. The Autolearn period of the battery has been changed. The delay interval for the battery has been changed. The time for next battery recalibration has been changed. The Max. Physical Drive Count of Patrol Read has been changed. Turn off the server, and check that hard disks, cables, the power supply, etc. are connected properly. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. Perform recalibration on a battery backup unit. 297

Appendix 10396 INFORMA 10397 INFORMA Adapter %s: Continuous patrolling enabled Adapter %s: Continuous patrolling disabled 10398 MAJOR Adapter %s: MDC finished with errors on logical drive %s 10399 INFORMA 10400 INFORMA Adapter %s: Logical drive %s disabled because RAID-5 is not supported by this RAID key (Server %s) Adapter %s: Logical drive %s disabled because RAID-6 is not supported by this RAID key (Server %s) 10401 MINOR Adapter %s: Logical drive %s disabled because SAS drives are not supported by this RAID key 10402 MINOR Adapter %s: Disks missing 10403 INFORMA 10405 INFORMA 10406 INFORMA 10407 INFORMA Adapter %s: Rebuild rate changed to %s%% Adapter %s: S.M.A.R.T. poll interval changed to %s min Adapter %s: MDC started on logical drive %s Adapter %s: MDC finished on logical drive %s 10408 MAJOR Adapter %s: MDC failed on logical drive %s 10409 MINOR Adapter %s: MDC aborted on logical drive %s table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Continuous patrolling has been enabled. Continuous patrolling has been disabled. Make Data Consistent is complete, but multiple media errors were detected during the Make Data Consistent. The logical drive has been disabled because RAID 5 is not supported by the RAID key. The logical drive has been disabled because RAID 6 is not supported by the RAID key. The logical drive has been disabled because the SAS hard disks are not supported by the RAID key. The hard disk does not exist. Rebuild Rate has been changed. S.M.A.R.T. poll interval has been changed. Make Data Consistent has started. Make Data Consistent is complete. Make Data Consistent has failed. Make Data Consistent has been aborted. If an unreadable file is found during operation, restore the file from the backup. Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. If necessary, perform Make Data Consistent again. 298

MegaRAID SAS User's Guide 10412 MAJOR Adapter %s: State change on logical drive %s from operational to degraded (Server %s) 10413 MAJOR Adapter %s: State change on logical drive %s from operational to partially degraded 10414 CRITICAL Adapter %s: State change on logical drive %s from operational to failed 10415 INFORMA Adapter %s: State change on logical drive %s from degraded to operational (Server %s) 10416 MAJOR Adapter %s: State change on logical drive %s from degraded to degraded (Server %s) 10417 MAJOR Adapter %s: State change on logical drive %s from degraded to partially degraded (Server %s) 10418 CRITICAL Adapter %s: State change on logical drive %s from degraded to failed 10419 INFORMA Adapter %s: State change on logical drive %s from partially degraded to operational (Server %s) 10420 MAJOR Adapter %s: State change on logical drive %s from partially degraded to degraded (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The logical drive has changed from online to critical. The logical drive has changed from online to partially critical. The logical drive has changed from online to offline. The logical drive has been restored from critical to online. The logical drive is now critical. The logical drive is now partially critical. The logical drive has changed from critical to offline. The logical drive has been restored from critical to online. The logical drive is now critical. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. Contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. Contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. 299

Appendix 10421 MAJOR Adapter %s: State change on logical drive %s from partially degraded to partially degraded 10422 CRITICAL Adapter %s: State change on logical drive %s from partially degraded to failed 10423 INFORMA Adapter %s: State change on logical drive %s from failed to operational 10424 MINOR Adapter %s: State change on logical drive %s from failed to degraded (Server %s) 10425 MAJOR Adapter %s: State change on logical drive %s from failed to partially degraded 10426 CRITICAL Adapter %s: State change on logical drive %s from failed to failed 10427 CRITICAL Adapter %s: State change by user on disk (%s) from available to failed 10428 INFORMA 10429 INFORMA 10430 INFORMA Adapter %s: State change by user on disk (%s) from available to hot spare Adapter %s: State change by user on disk (%s) from available to rebuilding (Server %s) Adapter %s: State change by user on disk (%s) from available to operational (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The logical drive is now critical. The logical drive has changed from partially critical to offline. The logical drive has changed from offline to online. The logical drive has changed from offline to critical. The logical drive has changed from offline to partially critical. The logical drive is now offline. The hard disk has changed from unused to failed. The hard disk has changed from the unused to a spare disk. The hard disk has changed from unused to rebuilding. The hard disk has changed from unused to online. Replace the failed hard disk, and perform a rebuild. Contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. Contact an office listed in the "Contact Information" of "Start Guide". Replace the failed hard disk. 300

MegaRAID SAS User's Guide 10431 INFORMA Adapter %s: State change by user on disk (%s) from failed to available (Server %s) 10432 CRITICAL Adapter %s: State change by user on disk (%s) from failed to failed 10433 INFORMA 10434 INFORMA 10435 INFORMA 10436 INFORMA Adapter %s: State change by user on disk (%s) from failed to hot spare (Server %s) Adapter %s: State change by user on disk (%s) from failed to rebuilding (Server %s) Adapter %s: State change by user on disk (%s) from failed to operational Adapter %s: State change by user on disk (%s) from hot spare to available 10437 CRITICAL Adapter %s: State change by user on disk (%s) from hot spare to failed 10438 INFORMA 10439 INFORMA 10440 INFORMA Adapter %s: State change by user on disk (%s) from hot spare to hot spare Adapter %s: State change by user on disk (%s) from hot spare to rebuilding Adapter %s: State change by user on disk (%s) from hot spare to operational table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The hard disk has changed from failed to unused. The hard disk is now failed. The hard disk has changed from the failed to a spare disk. The hard disk has changed from failed to rebuilding. The hard disk has changed from failed to online. The hard disk has changed from a spare disk to the unused. The hard disk has changed from a spare disk to the failed. The hard disk is now a spare disk. The hard disk has changed from a spare disk to the rebuilding. The hard disk has changed from a spare disk to the online. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and configure the new disk as a spare disk. 301

Appendix 10441 INFORMA Adapter %s: State change by user on disk (%s) from rebuilding to available (Server %s) 10442 CRITICAL Adapter %s: State change by user on disk (%s) from rebuilding to failed 10443 INFORMA 10444 INFORMA 10445 INFORMA 10446 INFORMA Adapter %s: State change by user on disk (%s) from rebuilding to hot spare Adapter %s: State change by user on disk (%s) from rebuilding to rebuilding (Server %s) Adapter %s: State change by user on disk (%s) from rebuilding to operational (Server %s) Adapter %s: State change by user on disk (%s) from operational to available (Server %s) 10447 CRITICAL Adapter %s: State change by user on disk (%s) from operational to failed 10448 INFORMA 10449 INFORMA Adapter %s: State change by user on disk (%s) from operational to hot spare Adapter %s: State change by user on disk (%s) from operational to rebuilding (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The hard disk has changed from rebuilding to unused. The hard disk has changed from rebuilding to failed. The hard disk has changed from the rebuilding to a spare disk. The hard disk is now rebuilding. The hard disk has changed from rebuilding to online. The hard disk has changed from online to unused. The hard disk has changed from online to failed. The hard disk has changed from the online to a spare disk. The hard disk has changed from online to rebuilding. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. 302

MegaRAID SAS User's Guide 10450 INFORMA Adapter %s: State change by user on disk (%s) from operational to operational (Server %s) 10451 CRITICAL Adapter %s: State change on disk (%s) from available to failed 10452 INFORMA 10453 INFORMA 10454 INFORMA 10455 INFORMA Adapter %s: State change on disk (%s) from available to hot spare Adapter %s: State change on disk (%s) from available to rebuilding (Server %s) Adapter %s: State change on disk (%s) from available to operational (Server %s) Adapter %s: State change on disk (%s) from failed to available (Server %s) 10456 CRITICAL Adapter %s: State change on disk (%s) from failed to failed 10457 INFORMA 10458 INFORMA 10459 INFORMA 10460 INFORMA Adapter %s: State change on disk (%s) from failed to hot spare Adapter %s: State change on disk (%s) from failed to rebuilding (Server %s) Adapter %s: State change on disk (%s) from failed to operational (Server %s) Adapter %s: State change on disk (%s) from hot spare to available (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The hard disk is now online. The hard disk has changed from unused to failed. The hard disk has changed from the unused to a spare disk. The hard disk has changed from unused to rebuilding. The hard disk has changed from unused to online. The hard disk has changed from failed to unused. The hard disk is now failed. The hard disk has changed from the failed to a spare disk. The hard disk has changed from failed to rebuilding. The hard disk has changed from failed to online. The hard disk has changed from a spare disk to the unused. Replace the failed hard disk. Replace the failed hard disk, and perform a rebuild. 303

Appendix 10461 CRITICAL Adapter %s: State change on disk (%s) from hot spare to failed 10462 INFORMA 10463 INFORMA 10464 INFORMA 10465 INFORMA Adapter %s: State change on disk (%s) from hot spare to hot spare Adapter %s: State change on disk (%s) from hot spare to rebuilding (Server %s) Adapter %s: State change on disk (%s) from hot spare to operational (Server %s) Adapter %s: State change on disk (%s) from rebuilding to available (Server %s) 10466 CRITICAL Adapter %s: State change on disk (%s) from rebuilding to failed 10467 INFORMA 10468 INFORMA 10469 INFORMA 10470 INFORMA Adapter %s: State change on disk (%s) from rebuilding to hot spare (Server %s) Adapter %s: State change on disk (%s) from rebuilding to rebuilding (Server %s) Adapter %s: State change on disk (%s) from rebuilding to operational (Server %s) Adapter %s: State change on disk (%s) from operational to available (Server %s) 10471 CRITICAL Adapter %s: State change on disk (%s) from operational to failed table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The hard disk has changed from a spare disk to the failed. The hard disk is now a spare disk. The hard disk has changed from a spare disk to the rebuilding. The hard disk has changed from a spare disk to the online. The hard disk has changed from rebuilding to unused. The hard disk has changed from rebuilding to failed. The hard disk has changed from the rebuilding to a spare disk. The hard disk is now rebuilding. The hard disk has changed from rebuilding to online. The hard disk has changed from online to unused. The hard disk has changed from online to failed. Replace the failed hard disk, and configure the new disk as a spare disk. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. 304

MegaRAID SAS User's Guide 10472 INFORMA 10473 INFORMA 10474 INFORMA Adapter %s: State change on disk (%s) from operational to hot spare (Server %s) Adapter %s: State change on disk (%s) from operational to rebuilding (Server %s) Adapter %s: State change on disk (%s) from operational to operational (Server %s) 10475 MAJOR Adapter %s: MDC detected uncorrectable multiple medium errors (disk (%s) at LBA %s on logical drive %s) (Server %s) 10476 MAJOR Adapter %s: Disk (%s) missing after reboot 10477 MAJOR Adapter %s: Logical drive (%s) missing after reboot (Server %s) 10478 INFORMA Adapter %s: Disk (%s) appeared new after reboot (Server %s) table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The hard disk has changed from the online to a spare disk. The hard disk has changed from online to rebuilding. The hard disk is now online. Multiple media errors have been detected during Make Data Consistent. A hard disk was not found after reboot. A logical drive was not found after reboot. A new hard disk has been found after the reboot. If an unreadable file is found during operation, restore the file from the backup. This may occur at the restart after adding or removing option cards. Check the array condition with ServerView RAID Manager. If it is OK, there is no problem. Also, there is no problem if the condition is the same as the one described at "1.5.4 Using ServerView RAID under the Windows 2000 Server Environment" (pg.44). If there is a failed hard disk, replace it, and rebuild the new one. This may occur at the restart after adding or removing option cards. Check the array condition with ServerView RAID Manager. If it is OK, there is no problem. Also, there is no problem if, after changing array configuration, the condition is the same as the one described at "1.5.4 Using ServerView RAID under the Windows 2000 Server Environment" (pg.44). If this error occurs during regular operation, contact an office listed in the "Contact Information" of "Start Guide". 305

Appendix 10479 INFORMA 10485 INFORMA 10486 INFORMA 10487 INFORMA Adapter %s: Logical drive %s appeared new after reboot Adapter %s: State change on disk (%s) from available to offline Adapter %s: State change by user on disk (%s) from available to offline Adapter %s: State change by user on disk (%s) from failed to offline (Server %s) 10488 MINOR Adapter %s: State change by user on disk (%s) from hot spare to offline 10489 INFORMA Adapter %s: State change by user on disk (%s) from offline to available 10490 MAJOR Adapter %s: State change by user on disk (%s) from offline to failed 10491 INFORMA 10492 INFORMA 10493 INFORMA 10494 INFORMA Adapter %s: State change by user on disk (%s) from offline to hot spare Adapter %s: State change by user on disk (%s) from offline to offline Adapter %s: State change by user on disk (%s) from offline to operational Adapter %s: State change by user on disk (%s) from offline to rebuilding table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action A new logical drive has been found after the reboot. The hard disk has changed from unused to offline. The hard disk has changed from unused to offline. The hard disk has changed from failed to offline. The hard disk has changed from a spare disk to the offline. The hard disk has changed from offline to unused. The hard disk has changed from offline to failed. The hard disk has changed from the offline to a spare disk. The hard disk has changed from offline to offline. The hard disk has changed from offline to online. The hard disk has changed from offline to rebuilding. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. 306

MegaRAID SAS User's Guide 10495 MINOR Adapter %s: State change by user on disk (%s) from operational to offline 10496 MINOR Adapter %s: State change by user on disk (%s) from rebuilding to offline 10497 INFORMA Adapter %s: State change on disk (%s) from failed to offline 10498 MINOR Adapter %s: State change on disk (%s) from hot spare to offline 10499 INFORMA Adapter %s: State change on disk (%s) from offline to available (Server %s) 10500 MAJOR Adapter %s: State change on disk (%s) from offline to failed 10501 INFORMA 10502 INFORMA 10503 INFORMA 10504 INFORMA Adapter %s: State change on disk (%s) from offline to hot spare Adapter %s: State change on disk (%s) from offline to offline Adapter %s: State change on disk (%s) from offline to operational (Server %s) Adapter %s: State change on disk (%s) from offline to rebuilding (Server %s) 10505 MINOR Adapter %s: State change on disk (%s) from operational to offline 10506 MINOR Adapter %s: State change on disk (%s) from rebuilding to offline table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The hard disk has changed from online to offline. The hard disk has changed from rebuilding to offline. The hard disk has changed from failed to offline. The hard disk has changed from a spare disk to the offline. The hard disk has changed from offline to unused. The hard disk has changed from offline to failed. The hard disk has changed from the offline to a spare disk. The hard disk has changed from offline to offline. The hard disk has changed from offline to online. The hard disk has changed from offline to rebuilding. The hard disk has changed from online to offline. The hard disk has changed from rebuilding to offline. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. Replace the failed hard disk, and perform a rebuild. 307

Appendix 10509 INFORMA 10511 INFORMA 10513 INFORMA Adapter %s: Rebuild on disk (%s) resumed Adapter %s: BGI restarted on logical drive %s Adapter %s: Rebuild on logical drive %s resumed 10518 MAJOR Adapter %s: SAS port %s lost link 10519 INFORMA Adapter %s: SAS port %s restored link 10526 CRITICAL Adapter %s: Adapter missing after reboot 10527 INFORMA Adapter %s: Adapter appeared new after reboot 10528 MINOR Adapter %s: Rebuild aborted on disk (%s) 10536 CRITICAL Adapter %s: Command timeout on disk (%s), CDB:%s (Server %s) 10537 MINOR Adapter %s: Disk (%s) reset (type %s) 10540 MAJOR Adapter %s: Uncorrectable medium error logged for logical drive %s at LBA %s (on disk (%s) at LBA %s) 10541 MINOR Adapter %s: Medium error corrected on logical drive %s at LBA %s 10542 MINOR Adapter %s: Bad block table on disk (%s) is 100%% full table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Rebuild resumes on the hard disk. Background initialization resumes on the logical drive. Rebuild resumes on the logical drive. SAS link has been lost. SAS link has been restored. The controller has been missing. A new controller has been detected. Rebuild has been aborted. Command timeout has been occurred. Reset has been issued to the hard disk. An uncorrectable error has been detected. Media errors have been restored. Bad Block Table of the hard disk is full. Check the cable connection. If this occurs again, contact an office listed in the "Contact Information" of "Start Guide". This is not a problem if this occurs after removal of the array controller. Otherwise, contact an office listed in the "Contact Information" of "Start Guide". If there are failed hard disks, replace and rebuild them. If there are failed hard disks, replace and rebuild them. If there are failed hard disks, replace and rebuild them. If an unreadable file is found during operation, restore the file from the backup. A large number of uncorrectable bad block occurred. If an unreadable file is found during operation, restore the file from the backup. 308

MegaRAID SAS User's Guide 10544 CRITICAL Adapter %s: Adapter needs replacement, faulty IOP detected 10559 INFORMA 10560 INFORMA 10561 INFORMA Adapter %s: Additional information for failed disk (%s) - firmware version: %s, serial number %s, first use: %s, total running time: %s days Adapter %s: If the logical size of logical drive %s has changed use appropriate operating system tool to reread disk information (Server %s) Adapter %s: Enclosure device exposure changed 10572 MAJOR Adapter %s: Cache discarded for logical drive (missing or offline) %s (Server %s) 10586 INFORMA 10587 INFORMA Adapter %s: Microcode update started on disk (%s) Adapter %s: Microcode update completed on disk (%s) 10588 MINOR Adapter %s: Microcode update timeout on disk (%s) 10589 MINOR Adapter %s: Microcode update failed on disk (%s) 10590 INFORMA Adapter %s: Adapter properties changed table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action Failures have been detected in the controller IO processor. Information on additional SATA hard disk appears: the firmware version of the failed SATA, the serial number, the date of the first use, and the total running time. Information on the OS needs to be updated as the logical drive capacity has been changed. The setting for enclosure display has been changed. The data stored on a cache has been discarded for the logical drive was inaccessible. The update of the firmware for the hard disk has started. The update of the firmware is complete. A timeout has occurred in the update of the firmware for the hard disk. The update of the firmware for the hard disk has failed. The settings for the array controller has changed. Contact an office listed in the "Contact Information" of "Start Guide". Shut down the system properly, and then restart the system. If there is a failed hard disk, replace it, and perform a rebuild. If there is a failed hard disk, replace it, and perform a rebuild. 309

Appendix 10591 INFORMA 10592 INFORMA 10593 INFORMA 10594 INFORMA 10598 INFORMA Adapter %s: Patrol Read properties changed Adapter %s: CC schedule properties changed Adapter %s: BBU properties changed Adapter %s: MDC resumed on logical drive %s Adapter %s: Spinup delay changed to %s seconds 10601 MINOR Adapter %s: Periodic BBU relearn is pending. Please initiate manual learn cycle as automatic learn is not enabled (Server %s) 10617 MAJOR Adapter %s: Adapter cache pinned for missing or offline logical drive %s 10618 MAJOR Adapter %s: Adapter cache pinned for missing or offline logical drives 10619 INFORMA Adapter %s: Adapter cache discarded by user for missing logical drives table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The setting of Patrol Read has changed. The schedule of Make Data Consistent has changed. The settings for the battery backup unit has changed. Make Data Consistent has resumed. The delay time of the hard disk startup has changed. The periodical battery recalibration is on standby. Perform the battery recalibration manually. The cache of the array controller has been retained for the deleted logical drive. The cache of the array controller has been retained for the deleted multiple logical drives. The cache for the array controller has discarded. (Patrol Read is not supported. Check the settings of the array controller.) Perform the battery recalibration. Check the current status of the logical drive. For the critical status Replace the failed hard disk, and rebuild the new one. For the offline status Contact an office listed in the "Contact Information" of "Start Guide". Check the current statuses of the logical drives. For the critical status Replace the failed hard disk, and rebuild the new one. For the offline status Contact an office listed in the "Contact Information" of "Start Guide". Check the current status of the logical drive. For the critical status Replace the failed hard disk, and rebuild the new one. For the offline status Contact an office listed in the "Contact Information" of "Start Guide". 310

MegaRAID SAS User's Guide 10620 INFORMA Adapter %s: Adapter cache destaged for logical drive %s 10621 MINOR Adapter %s: MDC started on an inconsistent logical drive %s 10624 INFORMA Complete RAID configuration written to file '%s' on the server 10625 MINOR Adapter %s: Abnormalshutdown sequence detected 10626 MAJOR Adapter %s: Internal hardware error 10627 MAJOR Adapter %s: Internal hardware error during POST (Server %s) 10628 INFORMA 10629 INFORMA Adapter %s: Bad block table of disk (%s) threshold exceeded; table is about to overflow Adapter %s: Bad block table of disk (%s) cleared (Server %s) 10630 CRITICAL Adapter %s: Disk (%s) failed due to bad block table overflow (Server %s) 10631 CRITICAL Adapter %s: Disk (%s) failure detected during POST (Server %s) 10632 CRITICAL Adapter %s: Failure detected during POST table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The cache for the array controller has discarded. Make Data Consistent has started for the uninitialized logical drive. All the array configurations has been stored in the file. Unexpected shutdown has been detected. A failure has been detected in the array controller. A failure has been detected in the array controller. The usage rate of Bad Block Table exceeds the threshold. Bad Block Table has been deleted. The hard disk has failed because Bad Block Table exceed the capacity. A failure has been detected in the hard disk during POST. A failure has been detected during POST. Check the current status of the logical drive. For the critical status Replace the failed hard disk, and rebuild the new one. For the offline status Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". If there is a failed hard disk, replace it, and rebuild the new one. If there is a failed hard disk, replace it, and rebuild the new one. Contact an office listed in the "Contact Information" of "Start Guide". 311

Appendix 10633 CRITICAL Adapter %s: Multiple failure on disk (%s) detected, internal info %s 10634 INFORMA Adapter %s: BGI started due to adapter exchange 10635 CRITICAL Adapter %s: Disk (%s) failure detected by Patrol Read table: List of ServerView RAID event logs ID Severity Log entry Description Recovery action The multiple hard disks have been failed. The background initialization has started because the array controller was replaced. A failure has been detected in the hard disk during Patrol Read. Contact an office listed in the "Contact Information" of "Start Guide". (Patrol Read is not supported. Check the settings of the array controller.) 312

MegaRAID SAS User's Guide B List of GAM Event Log With ServerView installed, occurred events are recorded in the OS event logs by ServerView. For Windows Events are recorded by the event viewer application log from the source "Fujitsu ServerView Services". For Linux Events are recorded in the system log from the source "Fujitsu ServerView Services". The device address is filled in at the beginning of the event log (the bracketed part). The device address indicates where the event occurred. Character string ctl: chn: tgt: logdrv: table: Meaning of event log strings Meaning Controller ID Slot number of hard disk Not used by this array controller. Logical drive number Unless ServerView is installed, event logging to the event viewer will not occur. Refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server to install and configure ServerView. The correspondence between severity for GAM events (SNMP TRAP), severity for GAM Client, and the event log type is shown in the table below. table: Event log types and descriptions Severity Description Severity in GAM Client OS event log type CRITICAL Severe error 1 Error MAJOR Error 2 Error MINOR Warning 3 Warning INFORMA Information (No action required) 4 Information 313

Appendix List of GAM event logs The number within the parentheses in the GAM ID appears in hexadecimal format. The GAM IDs in decimal format represent Specific Numbers on SNMP trap. table: List of GAM event logs GAM ID Severity Description Details Corrective action 1 (0x001) Info/1 A physical disk has been placed online. A hard disk has become online. 2 (0x002) 3 (0x003) 4 (0x004) 5 (0x005) 6 (0x006) 7 (0x007) 8 (0x008) 9 (0x009) Info/1 Error/3 Error/3 Info/1 Info/1 Physical disk added as hot spare. Physical disk error found. Physical disk PFA condition found; this disk may fail soon. An automatic rebuild has started. A rebuild has started. A hard disk has been set as a hot spare. A bad sector was found on the hard disk. A mechanical failure of the device. The host device detected an invalid sequence. The target device is missing. A failure has been predicted for the hard disk. Rebuild has started automatically. Rebuild has started by user. Check the state of the target hard disk. If it has a failure, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace it and perform a rebuild. If the error occurred only temporarily and does not recur, the hard disk is recovered and no action is required. However, if this error occurs frequently, it is recommended that you replace the disk as a preventive measure by referring to "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268). Refer to "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268) and replace the hard disk as a preventive measure. Info/1 Rebuild is over. Rebuild is complete. Info/1 Error/3 Rebuild is cancelled. Rebuild stopped with error. Rebuild has been canceled. Rebuild failed for some reasons. Perform a rebuild again. Refer to "5.9.4 Viewing Logical Drive Information" (pg.240) to check the current status of the logical drive. For critical status Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild again. For offline status Contact an office listed in the "Contact Information" of "Start Guide". 314

MegaRAID SAS User's Guide 10 (0x00A) 11 (0x00B) 12 (0x00C) 13 (0x00D) 14 (0x00E) 15 (0x00F) 16 (0x010) 17 (0x011) 18 (0x012) 19 (0x013) 20 (0x014) 21 (0x015) 23 (0x017) Error/3 Error/3 Error/3 Info/1 Info/1 Info/1 Info/1 Info/1 Error/3 Error/3 Error/3 Warning/2 Rebuild stopped with error. New device failed. Rebuild stopped because logical drive failed. Physical disk has failed. A new physical disk has been found. A physical disk has been removed. A previously configured disk is now available. Expand Capacity Started. Expand Capacity Completed. Expand Capacity Stopped with error. SCSI command timeout on hard device. SCSI command abort on hard disk. SCSI command retried on hard disk. table: List of GAM event logs GAM ID Severity Description Details Corrective action Rebuild failed due to a failure found on the target hard disk for the rebuild. Rebuild failed due to failures in multiple hard disks. A hard disk has failed. A new hard disk was detected. A hard disk was removed. A hard disk has become undetectable. A hard disk is now in Unconfigured state. The Expand Capacity process has started. The Expand Capacity process is complete. Multiple hard disks failed and the Expand Capacity process terminated abnormally. A command timeout was detected. A SCSI command was aborted. A SCSI command was retried. Warning/2 Soft error found. An error was detected on a hard disk, but it was resolved. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Contact an office listed in the "Contact Information" of "Start Guide". Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Recreate the array configuration, and restore the data from backup. Because the controller is performing a recovery, there is no problem as long as there are no failed hard disks. Because the controller is performing a recovery, there is no problem as long as there are no failed hard disks. Because the controller is performing a recovery, there is no problem as long as there are no failed hard disks. Because the controller is performing a recovery, no action is required. If this error occurs frequently, refer to "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268) to replace the hard disk as a precautionary measure. 315

Appendix 24 (0x018) 25 (0x019) 28 (0x01C) 29 (0x01D) 30 (0x01E) 31 (0x01F) 32 (0x020) 33-41 (0x021-0x029) 42 (0x02A) 43-49 (0x02B - 0x031) 50 (0x032) 51 (0x033) 52 (0x034) 53 (0x035) Warning/2 Misc error found. An error was detected on a hard disk, but it was resolved. Info/1 SCSI device reset. The firmware issued a device reset. Error/3 Info/1 Info/1 Error/3 Error/3 Error/3 Error/3 Error/3 Error/3 Info/1 Error/3 Warning/2 Request Sense Data available. Initialization started. Initialization completed. Initialization failed. Initialization canceled. A physical disk failed because A physical disk set to failed state by host. A physical disk failed because Physical disk status changed to offline. Physical disk status changed to hot spare. Physical disk status changed to rebuild. Physical device ID did not match. table: List of GAM event logs GAM ID Severity Description Details Corrective action A hard disk reported sense information. A hard disk formatting has started. The hard disk format is complete. The hard disk format failed. The hard disk format was canceled. A hard disk has failed. A Make Offline has been executed by the controller. A hard disk has failed. A hard disk has become offline. The hard disk has become "Hot Spare". The hard disk has become rebuild. The hard disk ID does not match. Because the controller is performing a recovery, no action is required. If this error occurs frequently, refer to "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268) to replace the hard disk as a precautionary measure. Because the controller is performing a recovery, no action is required as long as the corresponding disk is Online. Wait until the format is complete. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk. Format the hard disk again. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Check the logs surrounding the process and perform necessary actions. 316

MegaRAID SAS User's Guide 54 (0x036) 55 (0x037) 56 (0x038) 57 (0x039) 58 (0x03A) 61 (0x03D) 62 (0x03E) 67 (0x043) 68 (0x044) Error/3 Warning/2 Warning/2 Error/3 Error/3 Info/1 Info/1 Error/3 Info/1 Physical disk failed to start. Physical disk negotiated different offset than config. Physical disk negotiated different bus width than config. Disk missing on startup. Rebuild startup failed due to lower disk capacity. A standby rebuild has started. Hot spare replaced with a smaller capacity physical disk. Physical Disk found on only one disk channel. Physical disk type is not approved by vendor. table: List of GAM event logs GAM ID Severity Description Details Corrective action A hard disk failed to start. A hard disk has negotiated an offset different from the configuration. A hard disk has negotiated a bus width different from the configuration. No hard disk was detected during startup. Insufficient hard disk space to perform the rebuild. Standby rebuild started. A hot spare with a smaller capacity was set. A hard disk is connected to only one disk channel. An installed hard disk is not vendor approved. Check that the hard disk is connected properly. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Check that the hard disk is connected properly. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Check that the hard disk is connected properly. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Check that the hard disk is connected properly. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk with a disk of the same model (with the same capacity and speed) as the other disks, and then perform a rebuild. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk with a disk of the same model (with the same capacity and speed) as the other disks, and then perform a rebuild. Check that the hard disk is connected properly. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Use a vendor supported hard disk. 317

Appendix 69 (0x045) 70 (0x046) 71 (0x047) 72 (0x048) 73 (0x049) 74 (0x04A) 75 (0x04B) 76 (0x04C) 77 (0x04D) 78 (0x04E) 79 (0x04F) Error/3 Error/3 Error/3 Error/3 Info/1 Info/1 Error/3 Info/1 Physical disk has acquired an inappropriate loop ID. Enclosure disk-slot operations are disabled while this condition persists. Physical disk port has failed or cannot operate at the configured channel speed. Mirror Race recovery failed for logical drive. Controller parameters checksum verification failed - restored default. Online controller firmware upgrade has started. Online firmware upgrade has completed successfully. Online firmware upgrade has failed. A Configuration On Disk (COD) with unsupported features has been detected. A hard disk has acquired an inappropriate loop ID. A hard disk has failed. The hard disk is not compatible with the system. The enclosure slot hardware failed. Mirror Race recovery failed. A mistake was found in the checksum of the controller parameters. An online controller firmware upgrade has started. An online firmware upgrade is successfully complete. An online firmware upgrade has failed. The array configuration information of the hard disk contains features that are not supported. An array configuration information created by another system was detected. Info/1 New battery found. A new battery was detected. Error/3 Error/3 Battery backup unit charger error. Battery will not hold a charge. table: List of GAM event logs GAM ID Severity Description Details Corrective action The battery backup unit charger has failed. A battery has failed. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Perform Make Data Consistent from GAM. For how to perform Make Data Consistent, refer to "5.10.1 Make Data Consistent" (pg.245). Check and correct the parameters in the [Adapter Properties] (pg.63) of WebBIOS. If the message still appears, contact an office listed in the "Contact Information" of "Start Guide". Perform the online controller firmware upgrade again. If the message still appears, contact an office listed in the "Contact Information" of "Start Guide". Connect a compatible hard disk. Replace the battery backup unit. Replace the battery backup unit. 318

MegaRAID SAS User's Guide 80 (0x050) 85 (0x055) 86 (0x056) 87 (0x057) 88 (0x058) 89 (0x059) 90 (0x05A) 95 (0x05F) 97 (0x061) 98 (0x062) 99 (0x063) 100 (0x064) 101 (0x065) 102 (0x066) Error/3 Info/1 Firmware entered unexpected state at run-time. Unable to recover medium error during patrol read. The firmware entered unexpected state at runtime. Recovery of a medium error failed during the Patrol Read operation. Info/1 Rebuild resumed. Rebuild restarted. Info/1 Info/1 Info/1 Error/3 Info/1 Patrol Read operation moved to next available target. Patrol Read operation moved to next available channel. Physical disk transfer speed changed. Channel is suspended due to some faults. Configured physical disk replaced by user by a smaller capacity disk. Patrol Read operation started with the next available hard disk. Patrol Read operation started with the next available channel. The transfer speed of the hard disk has changed due to an unknown error. An abnormal state was found in the channel. A hard disk has been replaced with a smaller capacity drive than configured. Check the installation of the cache memory or battery backup unit. If the message still appears, replace the cache memory or the battery backup unit. If a corrupted file is found, restore it from the backup. Info/1 PD Clear aborted. PD Clear was aborted. Check the previous logs and, perform necessary steps. Contact an office listed in the "Contact Information" of "Start Guide". Reconnect a proper hard disk. Error/3 PD Clear failed. PD Clear has failed. Replace the hard disk. Info/1 PD Clear started. PD Clear was started. Info/1 PD Clear completed. PD Clear is complete. Error/3 Error. An unknown error was detected. Error/3 Hot Spare SMART polling failed. table: List of GAM event logs GAM ID Severity Description Details Corrective action SMART polling of the Hot Spare failed. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. If the message still appears, contact an office listed in the "Contact Information" of "Start Guide". Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Then configure the spare disk by referring to "2.4 Setting/Releasing Spare Disk" (pg.87). 319

Appendix 103 (0x067) 104 (0x068) 105 (0x069) 106 (0x06A) 107 (0x06B) 110 (0x06E) 111 (0x06F) 112 (0x070) 113 (0x071) 114 (0x072) 115 (0x073) 116 (0x074) 119 (0x077) 120 (0x078) Info/1 Error/3 Error/3 Info/1 Error/3 Info/1 Info/1 Warning/2 Info/1 Info/1 Warning/2 Info/1 Warning/2 Warning/2 Patrol Read corrected medium error. Reassign write operation failed. Unrecoverable medium error during rebuild. Corrected medium error during recovery. Unrecoverable medium error during recovery. Dedicated Hot Spare no longer useful due to deleted array. Dedicated Hot Spare created. Dedicated Hot Spare disabled. Dedicated Hot Spare no longer useful for all arrays. Global Hot Spare created. Global Hot Spare disabled. Global Hot Spare does not cover all arrays. PD too small to be used for autorebuild. Bad block table on PD is 80% full. table: List of GAM event logs GAM ID Severity Description Details Corrective action A medium error was corrected by Patrol Read. A Reassign operation failed. An unrecoverable medium error was detected during rebuild process. A medium error was corrected. An unrecoverable medium error was detected. No corresponding disk group exists for the dedicated spare disk. A dedicated spare disk was created. A dedicated hot spare has been released. No corresponding disk group exists for the dedicated spare disk. A spare disk has been created. A spare disk has been released. The capacity of a spare disk is insufficient, and there is no array targeted for Hot Spare rebuild. The rebuild could not be started because the capacity of the hard disk is smaller then the other hard disk. The Bad Block Table usage rate has exceeded 80%. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. If a corrupted file is found, restore it from the backup. If a corrupted file is found, restore it from the backup. The dedicated spare disk is not supported. Reset the spare disk. The dedicated spare disk is not supported. Reset the spare disk. The dedicated spare disk is not supported. Reset the spare disk. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk with a disk of the same model (with the same capacity and speed) as the other disks, and then perform a rebuild. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk with a disk of the same model (with the same capacity and speed) as the other disks, and then perform a rebuild. Refer to "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268), and replace the hard disk as a preventive measure. 320

MegaRAID SAS User's Guide 121 (0x079) 126 (0x07E) 128 (0x080) 129 (0x081) 130 (0x082) 132 (0x084) 134 (0x086) 135 (0x087) 136 (0x088) 137 (0x089) 138 (0x08A) 139 (0x08B) 140 (0x08C) 141 (0x08D) 142 (0x08E) Error/3 Info/1 Bad block table on PD is full; unable to log blocks. Firmware corrected the 'Read' error. Info/1 Consistency check / Make Data Consistent is started. Info/1 Error/3 Error/3 Error/3 Error/3 Info/1 Info/1 Info/1 Info/1 Error/3 Error/3 Error/3 Consistency check / Make Data Consistent is finished. Consistency check / Make Data Consistent is cancelled. Consistency check / Make Data Consistent on logical drive failed. Logical drive has been made offline. Logical drive is critical. Logical drive has been placed online. An automatic rebuild has started on logical drive. A manual rebuild has started on logical drive. Rebuild on logical drive is over. Rebuild on logical drive is cancelled. Rebuild stopped with error. Rebuild stopped with error. New physical disk failed. table: List of GAM event logs GAM ID Severity Description Details Corrective action The Bad Block Table is full. The media error was corrected. Make Data Consistent has started. Make Data Consistent is successfully complete. Make Data Consistent was canceled. Make Data Consistent is aborted because the logical drive is no longer online. The logical drive is now offline. The logical drive is now in the critical status due to a hard disk failure. The logical drive is now online. Rebuild started automatically. Rebuild started manually. Rebuild is complete. Rebuild was canceled. Rebuild terminated abnormally. Rebuild terminated abnormally due to a failure on the target hard disk. Refer to "6.2.3 Preventive Replacement of Hard Disk [GAM]" (pg.268) and replace the hard disk as a preventive measure. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Then, redo Make Data Consistent. The logical drive(s) cannot continue running in this status. Recreate the array configuration and restore the data from backup. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Perform the rebuild again. Check the logs surrounding the process and perform necessary actions. Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. 321

Appendix 143 (0x08F) 144 (0x090) 145 (0x091) 146 (0x092) 147 (0x093) 148 (0x094) 149 (0x095) 150 (0x096) 151 (0x097) 152 (0x098) 153 (0x099) 154 (0x09A) 155 (0x09B) 156 (0x09C) 157 (0x09D) Error/3 Info/1 Info/1 Error/3 Error/3 Info/1 Info/1 Info/1 Info/1 Error/3 Rebuild stopped because logical drive failed. Logical drive initialization started. Logical drive initialization done. Logical drive initialization cancelled. Logical drive initialization failed. A logical drive has been found. A logical drive has been deleted. Expand Capacity Started. Expand Capacity Completed. Expand Capacity stopped with error. The source disk of the rebuild failed. An initialization of a logical drive has started. The initialization of the logical drive is complete. The initialization of the logical drive was canceled. The initialization terminated abnormally. The logical drive is now offline. A new logical drive has been detected. A logical drive was deleted. The Expand Capacity process has started. The Expand Capacity process is complete. The Expand Capacity process terminated abnormally due to failures in multiple hard disks. Error/3 Bad Blocks found. A bad block was detected during Make Data Consistent, rebuild, or Expand Capacity process. Info/1 Info/1 Error/3 Info/1 System drive size changed. System drive type changed. Bad data blocks found. Possible data loss. Logical drive LUN mapping has been written to config. table: List of GAM event logs GAM ID Severity Description Details Corrective action The capacity of the logical drive has changed. A new configuration was added. The Expand Capacity process is complete. Bad blocks were found on multiple hard disks at the same location. Logical drive LUN mapping has been written to config. Contact an office listed in the "Contact Information" of "Start Guide". Perform the initialization process again. Back up all the data on the logical drive, and refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk. Recreate the array configuration. Then restore the data from backup. The logical drive(s) cannot continue running in this state. Recreate the array configuration and restore the data from backup. During Make Data Consistent/ Expand Capacity: The bad block will be repaired, so there is no problem. During rebuild: If a corrupted file is found, restore it from the backup. If a corrupted file is found, restore it from the backup. 322

MegaRAID SAS User's Guide 158 (0x09E) 159 (0x09F) 162 (0x0A2) 163 (0x0A3) 164 (0x0A4) 165 (0x0A5) 166 (0x0A6) 167 (0x0A7) 176 (0x0B0) 177 (0x0B1) 178 (0x0B2) 179 (0x0B3) 180 (0x0B4) Error/3 Error/3 Info/1 Error/3 Info/1 Error/3 Info/1 Error/3 Info/1 Info/1 Info/1 Info/1 Error/3 Attempt to read data from block that is marked in Bad Data Table. Data for Disk Block has been lost due to Logical Drive problem. A standby rebuild has started on a logical drive. Reconstruct detected uncorrectable double medium errors. Reconstruction resumed. Reconstruction resume failed due to configuration mismatch. LD Properties updated. Marking LD inconsistent due to active writes at shutdown. Logical drive background initialization started. Logical drive background initialization stopped. Logical drive background initialization paused. Logical drive background initialization restarted. Logical drive background initialization failed. table: List of GAM event logs GAM ID Severity Description Details Corrective action An attempt has been made to read data logged in the BDT table. Due to a problem with the logical drive, cache data could not be written to the hard disk. Standby rebuild started. Due to media errors detected in the same position on multiple hard disks, data cannot be recovered. Reconstruction was resumed. Reconstruction resume terminated abnormally due to configuration mismatch. A parameter has changed on the logical drive. The consistency of the logical drive was lost due to shutdown during the writing process. BGI started. BGI stopped by the user or the firmware. BGI was paused for operations with higher priority. BGI resumed after a pause. BGI terminated abnormally. If a corrupted file is found, restore it from the backup. Check the logs surrounding the process and perform necessary actions. If the hard disk has failed, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. If a corrupted file is found, restore it from the backup. Recreate the array and restore the backup data. Perform Make Data Consistent from GAM. For how to perform Make Data Consistent, refer to "5.10.1 Make Data Consistent" (pg.245). Check the logs surrounding the process and perform necessary actions. 323

Appendix 181 (0x0B5) 182 (0x0B6) 185 (0x0B9) 256 (0x100) 257 (0x101) 258 (0x102) 259 (0x103) 272 (0x110) 273 (0x111) 274 (0x112) 275 (0x113) 288 (0x120) 289 (0x121) 290 (0x122) 291 (0x123) Info/1 Error/3 Info/1 Logical drive background initialization completed. Low battery charge level. Logical drive may have lost data. Unable to recover medium error during background initialization. BGI is complete. Data may be lost because the power has been off for a long time with data maintained in the cache memory. Unrecoverable medium error detected during background initialization. Perform Make Data Consistent (pg.245) from GAM. If necessary, restore the data from the backup after reconfiguring the array. If a corrupted file is found, restore it from the backup. Error/3 Fan failure. A fan failure was detected. Check the fan with ServerView, and take the necessary measures. Info/1 Fan has been restored. The fan was restored. Error/3 Fan failure. A fan failure was detected. Check the fan with ServerView, and take the necessary measures. Info/1 Error/3 Info/1 Error/3 Info/1 Error/3 Warning/2 Info/1 Storage cabinet fan is not present. Power supply failure. Power supply has been restored. Power supply failure. Storage cabinet power supply is not present. Over temperature. Temperature is above 70 degrees Celsius. Temperature is above 50 degrees Celsius. Normal temperature has been restored. table: List of GAM event logs GAM ID Severity Description Details Corrective action No fan was detected. The power supply has failed. The PSU was restored. The power supply has failed. No power supply was detected. Abnormal temperature was detected. Abnormal temperature was detected. The temperature is now normal. Error/3 Over temperature. Abnormal temperature was detected. Check the power supply status with ServerView, and take the necessary measures. Check the power supply status with ServerView, and take the necessary measures. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. 324

MegaRAID SAS User's Guide 292 (0x124) 320 (0x140) 321 (0x141) 322 (0x142) 323 (0x143) 324 (0x144) 325 (0x145) 326 (0x146) 327 (0x147) 328 (0x148) 329 (0x149) 330 (0x14A) 331 (0x14B) 332 (0x14C) 334 (0x14E) 335 (0x14F) 336 (0x150) Info/1 Storage cabinet temperature sensor is not present. No temperature sensor was detected. Error/3 Fan failure. A fan failure was detected. Check the fan with ServerView, and take the necessary measures. Info/1 Fan has been restored. The fan was restored. Info/1 Fan is not present. No fan was detected. Error/3 Info/1 Info/1 Error/3 Error/3 Info/1 Info/1 Error/3 Info/1 Error/3 Info/1 Info/1 Info/1 Power supply failure. Power supply has been restored. Power supply is not present. Temperature is over safe limit. Failure imminent. Temperature is above working limit. Normal temperature has been restored. Temperature sensor is not present. Enclosure access critical. Enclosure access has been restored. Enclosure access is offline. Enclosure services ready. Access to temperature sensor has been lost. Access to power supply status information has been lost. table: List of GAM event logs GAM ID Severity Description Details Corrective action The power supply has failed. The PSU was restored. No power supply was detected. Abnormal temperature was detected. Abnormal temperature was detected. The temperature is now normal. No temperature sensor was detected. An error occurred in the access to the enclosure. Enclosure access was restored. An error occurred in the access to the enclosure. Enclosure is ready for service. Access to the temperature sensor was lost. Access to the power supply status information was lost. Check the power supply status with ServerView, and take the necessary measures. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". 325

Appendix 337 (0x151) 338 (0x152) 339 (0x153) 340 (0x154) 346 (0x15A) 347 (0x15B) 348 (0x15C) 349 (0x15D) 350 (0x15E) 351 (0x15F) 352 (0x160) 353 (0x161) 354 (0x162) Info/1 Error/3 Error/3 Error/3 Error/3 Error/3 Access to fan status information has been lost. Physical disks in enclosure being spundown. Temperature is below working limit. Temperature is under safe limit. Failure imminent. Enclosure not supported; too many enclosures connected to port. Enclosure firmware mismatch. Access to the fan status information was lost. Abnormal temperature was detected. Abnormal temperature was detected. Abnormal temperature was detected. Enclosure is not supported. Too many enclosures are connected to the port. The enclosure firmware does not match. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. Check the condition of the system connections. Contact an office listed in the "Contact Information" of "Start Guide". Error/3 Sensor bad. The sensor is abnormal. Contact an office listed in the "Contact Information" of "Start Guide". Error/3 Enclosure phy bad. The enclosure PHY is abnormal. Error/3 Warning/2 Info/1 Info/1 Info/1 SAS/SATA mixing not supported in enclosure; PD disabled. Enclosure (SES) hotplug on was detected, but is not supported. Enclosure power supply has been turned off. Enclosure power supply has been turned on. Enclosure power supply cable has been removed. table: List of GAM event logs GAM ID Severity Description Details Corrective action The hard disk cannot be used, because SAS and SATA devices are mixed. An enclosure hotplug was detected, but is not supported. The enclosure power supply was turned off. The enclosure power supply was turned on. The enclosure power supply cable was removed. Contact an office listed in the "Contact Information" of "Start Guide". Check if any unsupported hard disks are installed. If there is an unsupported hard disk installed, replace it with a supported one. Insert the power supply cable. 326

MegaRAID SAS User's Guide 355 (0x163) 356 (0x164) 357 (0x165) 384 (0x180) 385 (0x181) 386 (0x182) 388 (0x184) 389 (0x185) 390 (0x186) 391 (0x187) 392 (0x188) 393 (0x189) 394 (0x18A) 395 (0x18B) 396 (0x18C) Info/1 Info/1 Info/1 Info/1 Enclosure power supply cable has been inserted. Enclosure fan returned to normal. Enclosure temperature difference has been detected. Array management server software started successfully. The enclosure power supply cable was inserted. The enclosure fan returned to normal. An enclosure temperature difference was detected. GAM Server started successfully. Error/3 Write back error Failed to write cache. Check the logs surrounding the process and perform necessary actions. Warning/2 Error/3 Info/1 Info/1 Error/3 Internal log structures getting full, PLEASE SHUTDOWN AND RESET THE SYSTEM IN THE NEAR FUTURE. Controller is dead. System is disconnecting from this controller. Controller has been reset. Controller is found. Controller is gone. System is disconnecting from this controller. Due to many configuration changes, the configuration change table is full. The SCSI array controller failed. The controller received a reset command. A controller was detected. The power to the controller was cut off. The controller was removed from the system. Info/1 Battery present. A battery backup unit was detected. Error/3 Battery is bad. The battery backup unit has failed. Info/1 Battery power OK. The battery is fully charged. (The charge level is above the set threshold.) Error/3 Info/1 Controller is gone. System is disconnecting from this controller. Controller powered on. table: List of GAM event logs GAM ID Severity Description Details Corrective action The power to the controller was cut off. The controller was removed from the system. A new controller was installed. Shut down the system properly, power off the server and turn it back on. If the same log still appears, contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Because the firmware is performing a recovery, there is no problem as long as there are no failed hard disks. Contact an office listed in the "Contact Information" of "Start Guide". Replace the battery backup unit. Contact an office listed in the "Contact Information" of "Start Guide". 327

Appendix 397 (0x18D) 398 (0x18E) 399 (0x18F) 400 (0x190) 401 (0x191) 402 (0x192) 403 (0x193) 404 (0x194) 405 (0x195) 406 (0x196) 407 (0x197) 408 (0x198) Info/1 Error/3 Warning/2 Info/1 Info/1 Info/1 Error/3 Error/3 Error/3 Error/3 Info/1 Info/1 Controller is online. Controller is gone. System is disconnecting from this controller. Controller's partner is gone, controller is in failover mode now. Battery reconditioning is started. Battery reconditioning is finished. Battery reconditioning is canceled. Installation aborted. Controller firmware mismatch. The Battery Backup Unit is removed. WARM BOOT failed. Battery calibration cycle started. Battery calibration cycle finished. table: List of GAM event logs GAM ID Severity Description Details Corrective action A controller came online. The power to the controller was cut off. The controller was removed from the system. The controller went Offline. Recalibration of the battery was started. Recalibration of the battery is complete. Recalibration of the battery was canceled. The configuration changed while the system was offline. The controller firmware has been replaced with an old version. The battery backup unit was removed. A memory error was detected during the boot process. Recalibration of the battery was started. Recalibration of the battery is complete. Contact an office listed in the "Contact Information" of "Start Guide". Perform the battery recalibration again. Shut down the server and check the hard disk connections. Check that the appropriate hard disks are installed, and remove any inappropriate hard disks. (For example, a hard disk for another system may have been installed by mistake.) If this does not resolve the problem, reconfigure the array and restore the backup data. Contact an office listed in the "Contact Information" of "Start Guide". Extremely low voltage of a battery after battery replacement may result in this event. If ID: 392 (0x188) "Battery present." is recorded, there is no problem. Follow "Appendix C Replacing Battery" (pg.336). If ID: 392 (0x188) "Battery present." is not recorded, check if a BBU is correctly installed. Replace the cache memory or the battery backup unit. 328

MegaRAID SAS User's Guide 409 (0x199) 410 (0x19A) 411 (0x19B) 412 (0x19C) 413 (0x19D) 414 (0x19E) 415 (0x19F) 416 (0x1A0) 418 (0x1A2) 425 (0x1A9) 426 (0x1AA) 427 (0x1AB) 428 (0x1AC) 431 (0x1AF) Info/1 Error/3 Info/1 Info/1 Info/1 Error/3 Error/3 Info/1 Error/3 Error/3 Error/3 Error/3 Error/3 Error/3 Battery calibration cycle was cancelled. Battery is not present. Controller entered conservative cache mode. Controller entered normal cache mode. Controller device start complete. Soft ECC error corrected. Hard ECC error corrected. Battery recondition suggested. BBU out of service. Controller boot ROM image needs to be reloaded. Controller is using default non-unique world-wide name. Mirror Race recovery failed. Mirror Race on critical drive. Controller improperly shutdown! Data may have been lost. table: List of GAM event logs GAM ID Severity Description Details Corrective action Recalibration of the battery was canceled. No battery was detected. The SCSI array controller entered Conservative Cache Mode. The SCSI array controller entered Normal Cache Mode. The controller device started. An ECC error was detected in the memory. An ECC error was detected in the memory. Recalibration of the battery is recommended. The battery cannot supply power to the cache memory. An inappropriate firmware image was loaded. The controller's MAC address was lost, or not set. Mirror Race recovery failed. The hard disk has a failure. The consistency of the logical drive was lost due to an invalid shutdown process. Perform the battery recalibration again. If a battery backup unit is used, check its connection. If this error recurs, replace the battery backup unit. Replace the memory module or the battery backup unit. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Replace the memory module or the battery backup unit. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Perform the battery recalibration. Replace the battery backup unit. Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Perform Make Data Consistent from GAM. For how to perform Make Data Consistent, refer to "5.10.1 Make Data Consistent" (pg.245). Refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. Perform Make Data Consistent from GAM. For how to perform Make Data Consistent, refer to "5.10.1 Make Data Consistent" (pg.245). 329

Appendix 440 (0x1B8) 442 (0x1BA) 443 (0x1BB) 444 (0x1BC) 446 (0x1BE) 447 (0x1BF) 448 (0x1C0) 449 (0x1C1) 450 (0x1C2) 451 (0x1C3) 452 (0x1C4) 453 (0x1C5) 455 (0x1C7) 456 (0x1C8) 457 (0x1C9) 458 (0x1CA) Error/3 Info/1 Info/1 Info/1 Info/1 Error/3 Info/1 Info/1 Info/1 Info/1 Info/1 Info/1 Info/1 Info/1 Error/3 Info/1 Error in Mirror Race Table. Patrol Read iteration completed. Controller entered 'Write Through' cache mode. Controller entered 'Write Back' cache mode. Data in Cache flushed during power up. Data in Cache not flushed during power up. Consistency Check rate changed. Background initialization rate changed. Flush rate changed. Patrol Read Rate changed. Rebuild rate changed. Reconstruction rate changed. Alarm disabled by user. Alarm enabled by user. Unable to recover cache data due to configuration mismatch. Cache data recovered successfully. table: List of GAM event logs GAM ID Severity Description Details Corrective action An error occurred in the Mirror Race Table. Patrol Read is complete. The controller entered 'Write Through' cache mode. The controller entered 'Write Back' cache mode. Data in the cache memory was flushed at the time of system boot. Data in the cache memory failed to flush at the time of system boot due to an abnormal configuration. The rate of Make Data Consistent has been changed. The background initialization rate has been changed. The flush rate has been changed. The Patrol Read Rate has been changed. The rebuild rate has been changed. The reconstruction rate has been changed. The user disabled the alarm. The user enabled the alarm. The cache data cannot be recovered due to a configuration mismatch. The cache data was recovered successfully. Check the array configuration. If the array configuration is correct, perform Make Data Consistent. If the array configuration is invalid, reconfigure the array and restore the data from the backup. Check the array configuration. If the array configuration is correct, perform Make Data Consistent. If the array configuration is invalid, reconfigure the array and restore the data from the backup. If the array configuration is correct, perform Make Data Consistent (pg.245). If the array configuration is invalid, reconfigure the array and restore the data from the backup. 330

MegaRAID SAS User's Guide 460 (0x1CC) 461 (0x1CD) 462 (0x1CE) 463 (0x1CF) 464 (0x1D0) 465 (0x1D1) 466 (0x1D2) 467 (0x1D3) 468 (0x1D4) 469 (0x1D5) 470 (0x1D6) 471 (0x1D7) 472 (0x1D8) 473 (0x1D9) 474 (0x1DA) 475 (0x1DB) Info/1 Info/1 Factory defaults restored. Hibernate command received from host. A factory default was restored. A hibernate command was received from the host. Info/1 Event log cleared. The NVRAM log was cleared. Info/1 Error/3 Info/1 Info/1 Info/1 Info/1 Info/1 Error/3 Info/1 Info/1 Error/3 Error/3 Info/1 Event log wrapped. Not enough controller memory. Patrol Read paused. Patrol Read resumed. Patrol Read started. Shutdown command received from host. User entered firmware debugger. Battery temperature is high. Battery is charging. Battery is discharging. Battery needs to be replaced. Battery relearn timed out. Battery relearn pending: Battery is under charge. table: List of GAM event logs GAM ID Severity Description Details Corrective action The NVRAM log was wrapped. Insufficient controller memory was detected. Patrol Read operation was temporarily paused. Patrol Read operation was resumed. Patrol Read operation was started. A shutdown command was received from the host. The user entered firmware debug mode. Abnormal temperature was detected. The battery is charging. The battery is discharging. Replacement of the battery is required. The battery recalibration was timed out. The recalibration of the battery is pending because the battery is being charged. Reconfigure the controller if necessary. Replace the cache memory or the battery backup unit. Check the fan status with ServerView. In the event of a fan failure, take the necessary measures. If there is no fan failure, adjust the ambient temperature. If this error recurs, replace the battery backup unit. The event occurs right after the battery backup unit replacement. Due to the insufficiently charged battery after the replacement, this event may occur. Follow "Appendix C Replacing Battery" (pg.336) The event occurs during regular operation. Replace the battery backup unit. Perform the battery recalibration again. If this error recurs, replace the battery backup unit. 331

Appendix 478 (0x1DE) 479 (0x1DF) 480 (0x1E0) 481 (0x1E1) 482 (0x1E2) 483 (0x1E3) 512 (0x200) 514 (0x202) 515 (0x203) 516 (0x204) 517 (0x205) Info/1 Warning/2 Info/1 Info/1 Info/1 Info/1 BBU enabled; changing WT logical drives to WB. BBU disabled; changing WB logical drives to WT. Battery charge complete Battery Relearn request received. Controller Cache discarded by user. Previous configuration is completely missing at boot. The battery backup unit became enabled and the logical drive was changed from Write Through mode to Write Back mode. The battery backup unit became disabled and the logical drive was changed from Write Back mode to Write Through mode. The battery charge is complete. A request to recalibration the battery was received. The user discarded the Controller Cache. The array configuration was missing at boot time. Info/1 System started.- The server or GAM Server started. Info/1 User logged in.- The user logged in to the server. Info/1 User logged out.- The user logged out from the server. Info/1 Server alive. Reconnected to the server. The server rebooted. Error/3 Lost connection to server, or server is down. table: List of GAM event logs GAM ID Severity Description Details Corrective action The network connection to the server was lost. The server shut down. The event occurs at the system start right after the battery backup unit replacement. Due to the insufficiently charged battery after the replacement, this event may occur. Follow "Appendix C Replacing Battery" (pg.336). The event occurs at the system start during regular operation. Turn on the server for 12 hours or more after the event occurs to charge the battery. Replace the battery backup unit if ID:478(0x1DE) does not occur after 12 hours. The event occurs during regular operation. There is no problem if it occurs during battery recalibration (including the scheduled one). If not, replace the battery backup unit. Perform recalibration for a battery backup unit. Shut down the system and check the connection to the hard disk. Check that the cabinet is properly connected. Check the state of the power supply. Check the network. Check that GAM Server was properly executed. Check that the server is running. 332

MegaRAID SAS User's Guide 518 (0x206) 519 (0x207) 520 (0x208) 521 (0x209) 522 (0x20A) 700 (0x2BC) 701 (0x2BD) 702 (0x2BE) 703 (0x2BF) 800 (0x320) 801 (0x321) 802 (0x322) 803 (0x323) 804 (0x324) 805 (0x325) Info/1 Info/1 Info/1 Info/1 Automatic reboot count has changed. Battery test has started. Battery test complete. Battery test canceled. Error/3 Battery test failed - battery bad. Automatic reboot count has changed. A battery test started. The battery test is complete. The battery test was canceled. Errors were detected during the battery test. Info/1 Event log empty. The content of the event log has become blank. Info/1 Event log entries lost. Event Log entries were lost. Info/1 Request Sense. Sense Information was reported. Replace the battery backup unit. Info/1 Set real time clock. The clock was set. Info/1 Info/1 Warning/2 Warning/2 Warning/2 Warning/2 New Configuration Received. Configuration Cleared. Configuration Invalid. Configuration On Disk Access Error. Configuration on disk converted. Configuration On Disk Import Failed. table: List of GAM event logs GAM ID Severity Description Details Corrective action A new array configuration was issued. The array configuration was cleared. The array configuration information is invalid. The array configuration information could not be read from the hard disk. The array configuration information on the hard disk was converted. The array configuration information could not be imported. Because the firmware is performing a recovery, there is no problem as long as there are no failed hard disks. Check that the hard disk is connected properly. If this does not resolve the problem, recreate the array and recover the backup data. Check the array configuration. If there is a failed hard disk, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace it and perform a rebuild. If the array configuration is invalid, reconfigure the array and restore the data from the backup. Shut down the server and check the hard disk connections. Check that the appropriate hard disks are installed, and remove any inappropriate hard disks. (For example, a hard disk for another system may have been installed by mistake.) If this does not resolve the problem, reconfigure the array and restore the backup data. 333

Appendix 806 (0x326) 807 (0x327) 808 (0x328) 810 (0x32A) 811 (0x32B) 812 (0x32C) 960 (0x3C0) 961 (0x3C1) 962 (0x3C2) 963 (0x3C3) 964 (0x3C4) Info/1 Info/1 Info/1 Info/1 Error/3 Info/1 Error/3 Error/3 Error/3 Error/3 Error/3 A debug dump exists on this system. A debug dump exists on this system. No valid Configuration On Disk (COD) found. MegaRAID firmware initialization started. Unable to recover cache data from TBBU. Cache data recovered from TBBU successfully. SAS topology error: Loop detected. SAS topology error: Unaddressable device. SAS topology error: Multiple ports to the same SAS address. SAS topology error: Expander error. SAS topology error: SMP timeout. table: List of GAM event logs GAM ID Severity Description Details Corrective action A debug dump exists on this system. A debug dump exists on this system. No valid Configuration On Disk (COD) found. The initialization of the MegaRAID firmware started. Cache data cannot be recovered from the battery backup unit. Data was successfully written to the cache memory. Loop detected in the SAS topology. Device is unaddressable in the SAS topology. Multiple ports were connected to the same SAS address in the SAS topology. An error was detected in the Expander. SMP timeout was detected. Check if the hard disk connected was previously used in another system. If a hard disk that was previously used in another system is connected, that hard disk must be formatted completely before use. Shut down the system properly, and then restart the system. Check the condition of the system connections. If this error recurs, even though the connections are correct, contact an office listed in the "Contact Information" of "Start Guide". Check the condition of the system connections. If the system connections are correct but there is a failed hard disk, refer to "6.2.2 Replacing Failed Hard Disk [GAM]" (pg.266) to replace the hard disk and perform a rebuild. If this error recurs, contact an office listed in the "Contact Information" of "Start Guide". Check the condition of the system connections. If this error recurs, even though the connections are correct, contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". 334

MegaRAID SAS User's Guide 965 (0x3C5) 966 (0x3C6) 967 (0x3C7) 968 (0x3C8) 969 (0x3C9) 970 (0x3CA) 971 (0x3CB) Error/3 Error/3 Error/3 Error/3 Error/3 Error/3 Error/3 SAS topology error: Out of route entries. SAS topology error: Index not found. SAS topology error: SMP function failed. SAS topology error: SMP CRC error. SAS topology error: Multiple subtractive. SAS topology error: Table to table. SAS topology error: Multiple paths. table: List of GAM event logs GAM ID Severity Description Details Corrective action Route entries cannot be found. Index was not found. An error was detected in an SMP function. A CRC error was detected in SMP. An error was detected in the SAS topology. An error was detected in the SAS topology. Multiple paths exist. Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Contact an office listed in the "Contact Information" of "Start Guide". Check the condition of the system connections. If this error recurs, even though the connections are correct, contact an office listed in the "Contact Information" of "Start Guide". 335

Appendix C Replacing Battery The battery for the cache memory is mounted in models that come with a battery backup unit. The battery is a consumable part and its capacity decreases over time. Replace the battery periodically. The length of the data holding time will decrease gradually over the life of the battery. The speed of this decrease depends on the environment, especially on high temperatures. As an example, if the temperature surrounding the server is 25 degrees Celsius, change the battery after two to three years of the purchase. In a worst-case scenario, if you continue to use the battery at an extremely low capacity, the battery fluid may leak. Regarding battery replacement, there are cases when you need to buy a new server or replace a server, or when you need to contact an office listed in the "Contact Information" of "Start Guide". For the details, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server. For how to physically replace a battery, refer to "User's Guide" in PRIMERGY Startup Disc supplied with the server, or the manual come with a battery backup unit. Check if the event that detected the following battery is recorded. When using ServerView RAID Events via ServerView Source : Fujitsu ServerView Services Type : Information Event ID : 1 Description: <Type and number of controller>: BBU present Source: ServerView RAID events, or ServerView RAID Manager event window Source : ServerView RAID ID : 10298 Description: <Type and number of controller>: BBU present When using GAM Source : Fujitsu ServerView Services Type : Information Event ID : 1 Description: Battery present When the voltage of the replaced battery is extremely low, both messages shown above and below appear at a time resulting in no batteries detected. By charging the battery more than 60 minutes, a battery will be normally detected from the next restart. 336

MegaRAID SAS User's Guide When using ServerView RAID Events via ServerView Source : Fujitsu ServerView Services Type : error Event ID : 3 Description: <Type and number of controller>: BBU removed Source: ServerView RAID events, or ServerView RAID Manager event window. Source :ServerView RAID ID : 10314 Description: <Type and number of controller>: BBU removed When using GAM Source : Fujitsu ServerView Services Type : Information Event ID : 3 Description: The Battery Backup Unit is removed. The batteries are nickel hydride and lithium ion. The batteries are recyclable. At the ends of their useful lives, under various states and local laws, it may be illegal to dispose of these batteries into the municipal waste stream. Check with your local solid waste officials for details in your area for recycling options or proper disposal. 337

Appendix 338

MegaRAID SAS User's Guide Index A Actions menu...................... 139 Add Logical Drive................... 219 Administration menu................. 208 Administrator privileges............... 117 Array controller Notes.......................... 43 Controller icons.................. 206 Controller options................ 214 Detailed information (GAM)......... 236 Detailed information (ServerView RAID)172 Detailed information (WebBIOS)...... 63 Hotfix.......................... 112 Property........................ 64 Setting and changing.............. 147 Setting and changing options....... 214 Specifications.................... 17 Updating device driver............. 106 Work flow....................... 14 B Background initialization............... 31 Checking progress (GAM).......... 242 Checking progress (ServerView RAID) 178 Checking progress (WebBIOS)....... 61 Background task............. 61, 178, 242 Bad BBU........................... 29 Battery backup unit............... 39, 143 Battery recalibration scheduler Installation...................... 129 Uninstallation.................... 132 BBU.............................. 39 C Checking availability of redundancy. 256, 268 Controller icon.................. 143, 206 Controller options................... 214 Controller Selection Box.............. 204 Controller View..................... 210 Administration menu.............. 208 File menu...................... 207 View menu..................... 207 Viewing hard disk information....... 237 Viewing logical drive information..... 240 D Device drivers Creating driver disk............... 108 Updating drivers (Windows Server 2003).............................108 Updating drivers (Windows 2000 Server).............................110 Disk array...........................18 Add Logical Drive.................223 Creating configuration (GAM)........220 Creating configuration (ServerView RAID).............................156 Creating configuration (WebBIOS).....66 Deleting configuration (GAM)........233 Deleting configuration (ServerView RAID).............................167 Deleting configuration (WebBIOS).....85 Disk group........................24 RAID Assist......................218 Saving configuration...............232 Disk group..........................24 Configuration example..............25 Expand Capacity...................36 Logical drive......................24 E Edit Configuration............219, 228, 230 Event window.......................139 Events............................234 GAM...........................313 ServerView RAID.................276 Expand Array...................220, 225 Expand Capacity..........36, 92, 160, 225 Checking progress (GAM)...........243 Checking progress (ServerView RAID).178 Checking progress (WebBIOS)........61 Expand Capacity Status...............243 F Fast initialization......................32 Checking progress.................61 File log............................146 File menu......................139, 207 Foreground Initialization................30 G GAM..............................182 Access privileges.................184 Events..........................313 Exiting..........................203 Installation.......................189 339

Maintenance Functions............ 245 RAID Assist function............... 218 Requirements.................... 183 Setting/releasing spare disk......... 230 Sign on......................... 202 Starting......................... 202 Toolbar......................... 209 Uninstallation.................... 192 Window Layout................... 204 GAM Client........................ 182 Installation...................... 189 Server group and server settings..... 213 Uninstallation.................... 192 GAM Server........................ 182 Installation...................... 189 Uninstallation.................... 193 Global Array Manager................ 182 Global Status View................... 205 Guest............................. 184 H Hard disk Checking availability of redundancy256, 268 Checking status.................. 174 Detailed information (Controller View). 237 Detailed information (WebBIOS)...... 53 Disk group....................... 24 Formatting....................... 98 Logical drive...................... 24 Manual rebuild............... 169, 246 Notes........................... 42 Preventive replacement (GAM)...... 268 Preventive replacement (ServerView RAID)............................ 255 RAID levels...................... 18 Replacing............... 251, 253, 266 Status icon.................. 144, 211 Hard disk failure prediction function....... 38 HDD Check Scheduler................. 38 Installation.................. 124, 198 Uninstallation................ 127, 201 Help menu......................... 142 Hotfix............................. 112 I Installation Battery Recalibration Schedule...... 129 GAM........................... 189 GAM Client...................... 189 GAM Server..................... 189 HDD Check Scheduler..... 124, 194, 198 ServerView RAID................. 121 L Log Information Viewer........... 206, 235 Logical drive........................ 24 Add to disk array configuration....... 223 Background initialization........ 178, 242 Checking status.................. 176 Configuration example.............. 25 Controller View................... 240 Deleting................. 95, 163, 228 Expand Capacity.. 92, 160, 178, 225, 243 Initialization...................... 30 Make Data Consistent........... 35, 91 RAID Levels...................... 18 Rebuild.......................... 33 Status........................... 26 Status Icon.................. 145, 212 Logical drive initialization Fast initialization................... 89 M Make Data Consistent...... 35, 91, 168, 245 Checking progress (GAM).......... 243 Checking progress (ServerView RAID) 178 Checking progress (WebBIOS)....... 61 Make Data Consistent Status........... 243 MegaRAID SAS 300-8E ROMB.......... 17 MegaRAID SAS 8300XLP.............. 17 MegaRAID SAS 8344ELP.............. 17 MegaRAID SAS 8408E................ 17 MegaRAID SAS 8880EM2.............. 17 Migrate Logical Drive................. 160 Migrating.......................... 178 Mirroring........................... 20 Mirroring + striping.................... 22 N New Configuration............... 219, 220 P PFA............................... 38 PFA Count/S.M.A.R.T............. 255, 268 Plug-in............................ 146 Port.............................. 144 R RAID.............................. 18 RAID Assist........................ 218 Add Logical Drive................. 219 Create......................... 220 Edit Configuration................. 219 Expand Array.................... 220 New Configuration................ 219 340

MegaRAID SAS User's Guide RAID Levels........................ 18 Redundancy..................... 18 RAID 5/6 SAS based on LSI MegaRAID... 17 Rebuild.................... 33, 169, 246 Checking progress (GAM).......... 242 Checking progress (ServerView RAID) 178 Checking progress (WebBIOS)....... 61 Hot spare rebuild.................. 33 Manual rebuild.................... 33 Rebuild Status................... 242 Recalibration................... 170, 247 Redundancy........................ 18 Redundant data regeneration...... 168, 245 Replacing battery................... 336 S S.M.A.R.T.......................... 38 SAS.............................. 17 Server group and server settings....... 213 Server icon.................... 143, 205 Server Selection box................. 204 ServerView........................ 183 ServerView AlarmService......... 119, 187 ServerView RAID Access privileges................ 117 Events......................... 276 Installation...................... 121 Log in......................... 134 Requirements................... 115 Setting and releasing spare disks.... 164 Uninstallation.................... 123 ServerView RAID Manager Change Access Mode............. 138 Exiting......................... 136 Starting........................ 134 Window Layout.................. 137 Spare disk.......................... 33 Releasing............... 88, 164, 230 Setting.................. 87, 164, 230 Starting Controller View.................. 210 GAM.......................... 202 RAID Assist..................... 218 ServerView RAID Manager......... 134 WebBIOS....................... 48 Status Hard disk........................ 26 Logical drive..................... 26 Status icon Hard disk................... 144, 211 Logical drive................ 145, 212 Striping............................ 20 Striping + Parity...................... 21 System log........................ 146 T Tree view..........................142 U Uninstallation Battery Recalibration Scheduler......132 GAM...........................192 GAM Client......................192 GAM Server.....................193 HDD Check Scheduler.....127, 197, 201 ServerView RAID.................123 User..............................184 User privileges......................117 V View menu.........................207 W WebBIOS...........................47 [Configured Drives] View.............51 [Physical Drives] view...............51 [Virtual Drives] view.................51 Exiting...........................52 Starting..........................48 Viewing array controller..............63 Viewing background task............61 Viewing hard disk..................53 Viewing logical drive................57 Window layout....................50 Write Back..........................29 Write Cache.........................28 Write Policy..........................28 Changing (GAM)..................231 Changing (ServerView RAID)........166 Write Through........................28 341

MegaRAID SAS User's Guide B7FY-2401-01ENZ0-00 Issued on July, 2008 Issued by FUJITSU LIMITED The contents of this manual may be revised without prior notice. Fujitsu assumes no liability for damages to third party copyrights or other rights arising from the use of any information in this manual. No part of this manual may be reproduced in any form without the prior written permission of Fujitsu.