Silicon Graphics CHALLENGE RAID Owner's Manual

CHALLENGE® RAID Owner’s Guide
Document Number 007-2532-004
CONTRIBUTORS
Written by Carolyn Curtis Illustrated by Cheri Brown, Dan Young, and Carolyn Curtis Production by Heather Hermstad Engineering contributions by Sammy Wilborn, Albert Lui, Dale Witt, Carl Strasen,
Brad Eacker, and Paul Tsien
Cover design and illustration by Rob Aguilar, Rikk Carey, Dean Hodgkinson,
Erik Lindholm, and Kay Maitz
© Copyright 1995, Silicon Graphics, Inc.— All Rights Reserved This document contains proprietary and confidential information of Silicon
Graphics, Inc. The contents of this document may not be disclosed to third parties, copied, or duplicated in any form, in whole or in part, without the prior written permission of Silicon Graphics, Inc.
RESTRICTED RIGHTS LEGEND Use, duplication, or disclosure of the technical data contained in this document by
the Government is subject to restrictions as set forth in subdivision (c) (1) (ii) of the Rights in Technical Data and Computer Software clause at DFARS 52.227-7013 and/or in similar or successor clauses in the FAR, or in the DOD or NASA FAR Supplement. Unpublished rights reserved under the Copyright Laws of the United States. Contractor/manufacturer is Silicon Graphics, Inc., 2011 N. Shoreline Blvd., Mountain View, CA 94043-1389.
Silicon Graphics, CHALLENGE, and IRIS are registered trademarks and IRIX, XFS, IRIS FailSafe, POWER CHALLENGE, and POWER Channel are trademarks of Silicon Graphics, Inc. Oracle Parallel Server and OPS are trademarks of Oracle Corporation.
CHALLENGE® RAID Owner’s Guide Document Number 007-2532-004
Contents
List of Figures vii List of Tables ix About This Guide xi
1. Features of the CHALLENGE RAID Storage System 1
Storage System Components 5
SCSI-2 Interface 5 CHALLENGE RAID Storage-Control Processor 6 Storage System Chassis 8 Disk Modules 10
Data Availability and Performance 11
Data Redundancy 11 Enhanced Performance: Disk Striping 12 Enhanced Performance: Storage System Caching 12 Data Reconstruction and Rebuilding After Disk Module Failure 13
RAID Levels 14
RAID-0 Group: Nonredundant Array 14 RAID-1: Mirrored Pair 15 RAID-1_0 Group: Mirrored RAID-0 Group 16
RAID-5: Individual Access Array 18 RAID Hot Spare 20 Using the CHALLENGE RAID Command Line Interface 21
2. Storage System Configurations 23
Basic Configuration 25 Dual-Interface/Dual-Processor Configuration 26 Split-Bus Configuration 28 Dual-Bus/Dual-Initiator Configuration 31
iii
Contents
3. Operating the Storage System 35
Checking CHALLENGE RAID Storage System Status 36
Using the raid5 Command 37 Getting Device Names With getagent 38 Getting General System Information 39 Getting Information About Disks 39 Getting Information About Other Components 42 Displaying the CHALLENGE RAID Unsolicited Event Log 43
Shutting Down the CHALLENGE RAID Storage System 44 Restarting the CHALLENGE RAID Storage System 45
4. Configuring Disks 49
Binding Disks Into RAID Units 49 Getting Disk Group (LUN) Information 53 Changing LUN Parameters 56 Dual Interfaces, Load Balancing, and Device Names 57
5. Maintaining Disk Modules 59
Identifying and Verifying a Failed Disk Module 59 Setting Up the Workplace for Replacing or Installing Disk Modules 62 Replacing a Disk Module 63
Ordering Replacement Disk Modules 63 Unbinding the Disk 64 Removing a Failed Disk Module 65 Installing a Replacement Disk Module 69 Updating the Disk Module Firmware 71
Installing an Add-On Disk Module Array 72
Ordering Add-On Disk Module Arrays 72 Installing Add-On Disk Modules 73 Creating Device Nodes and Binding the Disks 77
iv
6. Identifying Failed System Components 79
Power Supply 80 Fan Module 80 Battery Backup Unit 81 Storage-Control Processor 81
Diagnosing SP Failure 81
Using the Auto-Reassign Capability 82
7. Caching 85
Setting Cache Parameters 86 Viewing Cache Statistics 87 Upgrading CHALLENGE RAID to Support Caching 90 Changing Unit Caching Parameters 91
A. Technical Specifications 93
B. The raid5 Command Line Interface 97
bind 99 chglun 102 clearlog 104 clearstats 104 firmware 104 getagent 105 getcache 106 getcontrol 109 getcrus 109 getdisk 110 getlog 114 getlun 116 setcache 120 unbind 121
Contents
Index 123
v
List of Figures
Figure 1-1 CHALLENGE RAID Storage System, Deskside Version:
Front View 2
Figure 1-2 CHALLENGE RAID Rack 3 Figure 1-3 Computing SCSI Cable Length Example: Single-Host
Configuration Only 4
Figure 1-4 CHALLENGE RAID Server With One SP 6 Figure 1-5 SPs Connected to the Same CHALLENGE Chassis 7 Figure 1-6 SPs Connected to Different CHALLENGE Chassis 7 Figure 1-7 Disk Module Locations (Chassis Front View) 8 Figure 1-8 SCSI-2 Bus and Internal Buses (Front View) 9 Figure 1-9 Disk Modules and Status Lights 10 Figure 1-10 RAID-1 Mirrored Pair (Hardware Mirrored Pair) 15 Figure 1-11 Distribution of User Data in a RAID-1_0 Group 17 Figure 1-12 Distribution of User and Parity Data in a RAID-5 Group 19 Figure 1-13 Hot Spare Example 21 Figure 2-1 Dual-Interface/Dual-Processor Configuration Example 27 Figure 2-2 Split-Bus Configuration Example 30 Figure 2-3 Dual-Bus/Dual-Initiator Configuration Example 33 Figure 3-1 CHALLENGE RAID Indicator Lights 36 Figure 3-2 Disk Module Locations 40 Figure 3-3 Turning Off (On) Power (Back of CHALLENGE RAID
Chassis) 45
Figure 3-4 CHALLENGE RAID Indicator Lights 45 Figure 3-5 Unlocking the Fan Module 46 Figure 3-6 Enabling an SP’s Power 47 Figure 5-1 Disk Module Status Lights 60 Figure 5-2 Disk Module Locations 61
vii
List of Figures
Figure 5-3 Attaching the ESD Clip to the ESD Bracket on a Deskside
Storage System 66
Figure 5-4 Attaching the ESD Clip to the ESD Bracket on a Rack
Storage System 66
Figure 5-5 Pulling Out a Disk Module 67 Figure 5-6 Removing a Disk Module 68 Figure 5-7 Engaging the Disk Module Rail 69 Figure 5-8 Engaging the Disk Module Guide 70 Figure 5-9 Inserting the Replacement Disk Module 70 Figure 5-10 Marking the Label for Disk Module A0 74 Figure 5-11 Disk Drive Locations 75 Figure 5-12 Engaging the Disk Module Rail 75 Figure 5-13 Engaging the Disk Module Guide 76 Figure 5-14 Inserting a Disk Module 76 Figure 6-1 Unlocking the Fan Module 83 Figure 6-2 Opening the Fan Module 83 Figure 6-3 Disabling an SP’s Power 84 Figure B-1 Disk Module Locations 111
viii
List of Tables
Table 2-1 CHALLENGE RAID Configurations 24 Table 2-2 Error Recovery: Basic Configuration 25 Table 2-3 Error Recovery: Dual Interface/Dual-Processor
Configuration 26
Table 2-4 Error Recovery: Split-Bus Configuration 28 Table 2-5 Error Recovery: Dual-Bus/Dual-Initiator
Configuration 31
Table 3-1 Output of raid5 getagent 38 Table 3-2 Output of raid5 getdisk 41 Table 3-3 Output of raid5 getcrus 43 Table 4-1 Output of raid5 getlun 54 Table 5-1 Ordering Replacement Disk Modules 63 Table 5-2 Ordering Add-On Disk Module Sets 72 Table 6-1 Field-Replaceable Units 79 Table 7-1 Output of raid5 getcache 88 Table A-1 CHALLENGE RAID Deskside Chassis Specifications 93 Table A-2 CHALLENGE RAID Rack Specifications 94 Table B-1 raid5 Parameters 97 Table B-2 Output of raid5 getagent 105 Table B-3 Output of raid5 getcache 107 Table B-4 Output of raid5 getcrus 110 Table B-5 Output of raid5 getdisk 112 Table B-6 getlog Error Codes 114 Table B-7 Output of raid5 getlun 118
ix
About This Guide
The CHALLENGE® RAID storage system provides a compact, high-capacity, high-availability source of disk storage for the complete line of Silicon Graphics® CHALLENGE servers running IRIX™ 5.3 and 5.3 with XFS™: CHALLENGE S, CHALLENGE DM, CHALLENGE L, and CHALLENGE XL. It also works with IRIX 6.1 and 6.2 on the POWER CHALLENGE L and XL.
The CHALLENGE RAID storage system uses high-availability disk storage in as many as 20 disk modules. For even more storage, the CHALLENGE RAID rack storage system offers up to four RAID chassis assemblies, each with as many as 20 disk modules. The chassis assemblies in a CHALLENGE RAID rack can be connected to one or more SCSI buses on CHALLENGE servers separately or in combination.
RAID levels 0, 1, 1_0 (0+1), and 5 are supported, as well as disks configured as hot spares. In addition, a basic CHALLENGE RAID storage system provides storage-system caching.
Structure of This Guide
This guide contains the following chapters:
Chapter 1, “Features of the CHALLENGE RAID Storage System,”
introduces the main CHALLENGE RAID components and summarizes
RAID levels and data availability and performance features.
Chapter 2, “Storage System Configurations,” explains CHALLENGE
RAID configurations in detail: basic, dual-interface/dual-processor,
and split-bus.
Chapter 3, “Operating the Storage System,” describes how to check
status, identify failing components, and start and shut down the
storage system.
xi
About This Guide
Chapter 4, “Configuring Disks,” explains how to use the command line interface to group disks into RAID-5 groups and how to display or change information on groups of disks.
Chapter 5, “Maintaining Disk Modules,” explains how to replace a failed disk module and add a disk module array.
Chapter 6, “Identifying Failed System Components,” explains how to get status information on failed components other than disk modules.
Chapter 7, “Caching,” explains how to determine, set up, and change caching parameters.
Appendix A, “Technical Specifications,” summarizes technical information for the CHALLENGE RAID deskside storage system.
Appendix B, “The raid5 Command Line Interface,” lists and explains all parameters of the raid5 command.
An index completes this guide.
Conventions
xii
In command syntax descriptions and examples, square brackets ( [ ] ) surrounding an argument indicate an optional argument. Variable parameters are in italics. Replace these variables with the appropriate string or value.
In text descriptions, IRIX filenames are in italics. The names of keyboard keys are printed in boldface typewriter font and enclosed in angle brackets, such as <Enter> or <Esc>.
Messages and prompts that appear on-screen are shown in fixed-width type. Entries that are to be typed exactly as shown are in boldface fixed-width type.
Compliance Statements
This section lists various domestic and international hardware compliance statements that pertain to the system.
FCC Warning
This equipment has been tested and found compliant with the limits for a Class A digital device, pursuant to Part 15 of the FCC rules. These limits are designed to provide reasonable protection against harmful interference when the equipment is operated in a commercial environment. This equipment generates, uses, and can radiate radio frequency energy and if not installed and used in accordance with the instruction manual, may cause harmful interference to radio communications. Operation of this equipment in a residential area is likely to cause harmful interference, in which case the user will be required to correct the interference at personal expense.
VDE 0871/6.78
This equipment has been tested to and is in compliance with the Level A limits per VDE 0871.
European Union Statement
This device complies with the European Directives listed on the “Declaration of Conformity” which is included with each product. The CE mark insignia displayed on the device is an indication of conformity to the aforementioned European requirements.
International Special Committee on Radio Interference (CISPR)
This equipment has been tested to and is in compliance with the Class A limits per CISPR publication 22.
Canadian Department of Communications Statement
This digital apparatus does not exceed the Class A limits for radio noise emissions from digital apparatus as set out in the Radio Interference Regulations of the Canadian Department of Communications.
xiii
About This Guide
Attention
Le present appareil numerique n’emet pas de bruits radioelectriques depassant les limites applicables aux appareils numeriques de Classe A prescrites dans le Reglement sur le Brouillage Radioelectrique etabli par le Ministere des Communications du Canada.
Japanese Compliance Statement
xiv
Chapter 1
1. Features of the CHALLENGE RAID Storage
System
RAID (redundant array of inexpensive disks) technology provides redundant disk resources in disk-array configurations that make the storage system more highly available and improve reliability and performance. RAID was first defined by D. Patterson, G. Garth, and R. Katz of the University of California, Berkeley, in their 1987 paper, “A Case for Redundant Arrays of Inexpensive Disks (RAID)” (University of California, Berkeley, Report No. UCB/CSD/87/391). That paper defines various levels of RAID.
This chapter introduces the CHALLENGE RAID disk-array storage system. It explains:
CHALLENGE RAID storage system components
data availability and performance
RAID levels
the RAID hot spare
using the CHALLENGE RAID command line interface
Figure 1-1 is an external view of the deskside version of the CHALLENGE RAID storage system.
1
Chapter 1: Features of the CHALLENGE RAID Storage System
2
Figure 1-1 CHALLENGE RAID Storage System, Deskside Version: Front View
Note: In Figure 1-1, the front cover is removed for clarity.
Figure 1-2 is an external view of the CHALLENGE RAID rack, with the maximum of four chassis assemblies installed. Each chassis assembly in a CHALLENGE RAID rack corresponds to one deskside CHALLENGE RAID chassis.
Figure 1-2 CHALLENGE RAID Rack
3
Chapter 1: Features of the CHALLENGE RAID Storage System
The CHALLENGE RAID storage system provides a compact, high-capacity, high-availability source of disk storage for your CHALLENGE S, DM, L, or XL server system. The storage system offers a large capacity of high-availability disk storage in multiple disk modules that you can replace when the storage system is turned on.
The CHALLENGE RAID storage system connects by a small computer system interface (SCSI-2) differential bus to a SCSI-2 interface in a CHALLENGE server.
Only Silicon Graphics service personnel authorized to open the computer cabinet and replace parts should install or replace the SCSI-2 interface. You can replace the disk modules by following instructions in Chapter 5 in this guide.
A CHALLENGE server can support multiple CHALLENGE RAID storage systems. The various storage system configurations, along with their availability and performance features, are explained in Chapter 2 in this guide.
The number of CHALLENGE RAID deskside storage systems or rack chassis assemblies that can be connected on one SCSI bus is limited by the recommended SCSI bus length limit of 60 feet, as diagrammed in Figure 1-3.
SCSI differential
20 ft
3 ft
CHALLENGE
Total length in this example: 57 feet Recommended maximum length: 60 feet
8 ft
RAID
chassis
assembly
5 ft
5 ft
8 ft
RAID
chassis
assembly
terminator
8 ft
RAID
chassis
assembly
Figure 1-3 Computing SCSI Cable Length Example: Single-Host Configuration
Only
4
Storage System Components
Storage System Components
Caution: Although the SCSI bus absolute length limit is 80 feet, exceeding 60 feet on the SCSI bus is not recommended. When cable lengths exceed 60 feet, problems can occur on the SCSI mezzanine card, the IO4B board, or both.
The CHALLENGE RAID deskside storage system, or each chassis assembly in the CHALLENGE RAID rack, consists of these components:
one or more host SCSI-2 interfaces that are either
native to the POWER Channel™ 2 I/O controller (IO4 board) in the
CHALLENGE server
HIO add-on cards (mezzanine cards) on the POWER Channel 2 I/O
controller
one or two storage-control processors (SPs)
5 to 20 disk modules in groups of five
one fan module
two or three power supplies (VSCs, or voltage semi-regulated
converters)
one battery backup unit (BBU) for storage system caching (optional)
The CHALLENGE server holds the SCSI-2 interface(s); the CHALLENGE RAID storage system chassis holds the other components.
SCSI-2 Interface
The SCSI-2 interface transfers data between host memory and the SCSI-2 differential bus. A cable connects the SCSI-2 interface to the SP(s) in the storage-system cabinet, as diagrammed in Figure 1-4. The SCSI-2 interface has an operating system-specific device name and ID number.
5
Chapter 1: Features of the CHALLENGE RAID Storage System
CHALLENGE RAID Storage-Control Processor
The storage-control processor (SP) is a printed circuit board with memory modules that resides in the storage-system cabinet. It controls the disk modules in the storage system through a synchronous SCSI-2 bus. An SP has five internal fast/narrow SCSI buses, each supporting four disk modules, for a total of 20 disk modules.
Figure 1-4 diagrams a CHALLENGE RAID storage system with one SP.
CHALLENGE server
SCSI-2 interface
Figure 1-4 CHALLENGE RAID Server With One SP
SCSI-2 bus
SP A
CHALLENGE RAID
For higher performance, a CHALLENGE RAID storage system can support an additional SP. The second SP provides a second path to the storage system, so both SPs can connect to the same host or two different hosts, as diagrammed in Figure 1-5 and Figure 1-6. With two SPs, the storage system can support storage system caching, whereby each SP temporarily stores modified data in its memory and writes the data to disk at the most expedient time.
Note that the disks are owned by the SP, and not by the host, the SCSI-2 bus, or the storage system.
6
Storage System Components
SCSI-2 interface
SCSI-2 interface
CHALLENGE server
SCSI-2 bus
SCSI-2 bus
SP A
SP B
CHALLENGE RAID
Figure 1-5 SPs Connected to the Same CHALLENGE Chassis
SCSI-2 interface
Second CHALLENGE server
SCSI-2 bus
SCSI-2 bus
SP B
SCSI-2 interface
CHALLENGE RAID
SP A
First CHALLENGE server
Figure 1-6 SPs Connected to Different CHALLENGE Chassis
7
Chapter 1: Features of the CHALLENGE RAID Storage System
Storage System Chassis
The CHALLENGE RAID storage system chassis contains compartments for disk modules, SPs, fan module, power supplies, and battery backup unit. The disk modules face front, and the SP(s), power supplies, battery backup unit, and fan module are accessible from the back.
Each disk module has an ID (the module ID) based on its position in the storage system. The disk modules are inserted in the following order:
modules A0, B0, C0, D0, and E0 (array 0)
modules A1, B1, C1, D1, and E1 (array 1)
modules A2, B2, C2, D2, and E2 (array 2)
modules A3, B3, C3, D3, and E3 (array 3)
Figure 1-7 diagrams this placement. Individual disk modules have disk position labels attached.
Deskside Chassis assembly in rack
A1 B1 C1 D1 E1 A3 B3 C3 D3 E3
8
A0 B0 C0 D0 E0 A2 B2 C2 D2 E2
5 to 20 disk modules in groups of 5
Figure 1-7 Disk Module Locations (Chassis Front View)
A0 B0 C0 D0
A1 B1 C1 D1 E1
E0 A2 B2 C2 D2
A3 B3 C3 D3 E3
E2
Storage System Components
Through the SP, the SCSI-2 bus is split into five internal fast/narrow SCSI buses—A, B, C, D, and E—that connect the slots for the disk modules. For example, internal bus A connects the modules in slots A0, A1, A2, and A3, in that order. Figure 1-8 diagrams this configuration.
Deskside Chassis assembly in rack
Internal bus A
A1 B1 C1 D1 E1 A3 B3 C3 D3
E3
A0 B0 C0 D0 E0 A2
B2 C2 D2
E2
Internal bus A Internal bus B
Internal bus C Internal bus D
Internal bus E
Internal bus B
Internal bus C
Internal bus D
A0 B0 C0 D0 E0
A1 B1 C1 D1 E1
Internal bus E
A2 B2 C2 D2 E2
A3 B3 C3 D3 E3
Figure 1-8 SCSI-2 Bus and Internal Buses (Front View)
9
Chapter 1: Features of the CHALLENGE RAID Storage System
Disk Modules
A disk module, also called a disk drive module, consists of a disk drive, a power regulator board, internal cabling, and a plastic carrier. The carrier has a handle for inserting and removing the module. Figure 1-9 indicates disk modules in the CHALLENGE RAID chassis and their status lights.
Fault light (amber)
Deskside
Busy light (green)
Figure 1-9 Disk Modules and Status Lights
Ready light (green)
Ready light (green)
Busy light (green)
Fault light (amber)
Rack
Three status lights on the module indicate the following:
Ready light (green): lights while the disk module is powered up and ready for use
Busy light (green): lights while the drive is in use, for example, during formatting or user I/O operations
10
Fault light (amber): lights when the module is shut down by the SP because the module failed; also lights after you replace the drive, while the replacement drive spins up to speed
A label attached to the carrier’s side shows the disk module’s model number and capacity . You can also determine the capacity of a disk module and other features using the command line interface; see Chapter 3 in this guide.
Data Availability and Performance
The CHALLENGE RAID storage system hardware implements data availability and performance enhancements in these ways:
data redundancy
enhanced performance: disk striping
enhanced performance: storage system caching
data reconstruction and rebuilding after disk module failure
This section discusses these features.
Data Redundancy
RAID technology provides redundant disk resources in disk-array and disk-mirror configurations that make the storage system more highly available. Data redundancy varies for the different RAID levels supported by CHALLENGE RAID: RAID-0, RAID-1, RAID-1_0, and RAID-5.
Data Availability and Performance
Because the CHALLENGE RAID storage system has five internal SCSI-2 buses, RAID-5 provides redundancy for up to five groups of disk modules. A RAID-5 group maintains parity data that lets the disk group survivea disk module failure without losing data. In addition, the group can survive a single SCSI-2 internal bus failure if each disk module in the group was bound on an independent SCSI-2 internal bus.
A RAID-1 mirrored pair, or a RAID-1_0 group, which uses RAID-1 technology, duplicates data on two disk modules. If one disk module fails, the other module provides continuing access to stored information. Similarly, a RAID-1 mirrored pair or RAID-1_0 group can survive a single SCSI internal bus failure if you bind each disk module on an independent SCSI internal bus.
11
Chapter 1: Features of the CHALLENGE RAID Storage System
Enhanced Performance: Disk Striping
In disk striping, the SP lays out data records, usually lar ge data r ecor ds or a number of small records for the same application, across multiple disks. For most applications, these disks can be written to or read from simultaneously and independently. Because multiple sets of read/write heads work on the same task at once, disk striping can enhance performance.
The amount of information read from or written to each module makes up the stripe element size (for example, 128 sectors). The stripe size is the number of data disks in a group multiplied by the stripe element size. For example, assume a stripe element size of 128 sectors (the default). If the RAID-5 group has five disks (four data disks and one parity disk), multiply by 4 the stripe element size of 128 to yield a stripe size of 512 sectors.
Enhanced Performance: Storage System Caching
Caching is available for CHALLENGE RAID storage systems that have two SPs, each with at least 8 MB of memory, a battery backup unit, and disk modules in slots A0 through E0. W ith storage system caching enabled, each SP temporarily stores requested information in its memory.
12
Caching can save time in two ways:
For a read request, if data is sought after the request is already in the read cache, the storage system avoids accessing the disk group to retrieve the data.
For a write request, if the information in the write cache is modified by the request and thus must be written to disk, the SP can keep the modified data in the cache and write it back to disk at the most expedient time instead of immediately. Write caching, in particular, can enhance storage system performance by reducing write time response.
To ensure data integrity, each SP maintains a mirror image of the other SP’s caches. If one SP fails, the data in its caches is available from the other SP.
You enable storage system caching and specify basic cache parameters, and enable or disable read and write caches for individual disk units using the command line interface, as explained in Chapter 7.
Data Availability and Performance
Note: The optional battery backup unit must be present in the
CHALLENGE RAID chassis for systems using cache to ensure that data is committed to disk in the event of a power failure.
Data Reconstruction and Rebuilding After Disk Module Failure
All RAID levels except RAID-0 provide data redundancy: the storage system reads and writes data from and to more than one disk at a time. Also, the system software writes parity information that lets the array continue operating if a disk module fails. When a disk module in one of these RAID levels fails, the data is still available because the SP canreconstruct it from the surviving disk(s) in the array.
Data rebuilding occurs when
a hot spare is available
the failed disk module is replaced with a new disk module
If a disk module has been configured (bound) as a hot spare, it is available as a replacement for a failed disk module. (See “RAID Hot Spare” later in this chapter .) When a disk module in any RAID level except RAID-0 fails, the SP automatically writes to the hot spare and rebuilds the group using the information stored on the surviving disks. Performance is degraded while the SP rebuilds the data and parity on the new module. However , the storage system continues to function, giving users access to all data, including data stored on the failed module.
Similarly, when a new disk module is inserted to replace a failed one, the SP automatically writes to it and rebuilds the group using the information stored on the surviving disks. As for the hot spare, performance is degraded during rebuilding, but data is accessible.
The length of the rebuild period, during which the SP re-creates the second image after a failure, can be specified when RAID levels are set and disks ar e bound into RAID units. These processes are explained in “Binding Disks Into RAID Units” in Chapter 4.
13
Chapter 1: Features of the CHALLENGE RAID Storage System
RAID Levels
The CHALLENGE RAID system supports these levels of RAID:
RAID-0 group: nonredundant array
RAID-1: mirrored pair
RAID-1_0 group: mirrored RAID-0 group
RAID-5 group: individual access array
Caution: Use only CHALLENGE RAID disk modules to replace failed disk modules. CHALLENGE RAID disk modules contain proprietary firmware that the storage system requires for correct functioning. Using any other disks, including those from other Silicon Graphics systems, can cause failure of the storage system. Swapping disk modules within a CHALLENGE RAID storage system is also not recommended, particularly disk modules in slots A0, B0, C0, and A3, which contain the licensed internal code, and those in slots D0 and E0, which serve with A0, B0, and C0 as the storage system cache vault.
14
RAID-0 Group: Nonredundant Array
Three to sixteen disk modules can be bound as a RAID-0 group. A RAID-0 group uses striping; see “Enhanced Performance: Disk Striping,” earlier in this chapter. You might choose a RAID-0 group configuration when fast access is more important than high availability. On IRIX 5.3 with XFS you can software-mirror the RAID-0 group to provide high availability.
Caution: The hardware does not maintain parity information on any disk module for RAID-0 the way it does for other RAID levels. Failure of a disk module in this RAID level results in loss of data.
RAID Levels
RAID-1: Mirrored Pair
In the RAID-1 configuration, two disk modules can be bound as a mirrored pair. In this disk configuration, the SP duplicates (mirrors) the data records and stores them separately on each disk module in the pair. The disks in a RAID-1 pair cannot be split into its individual units (as can a software mirror composed of two individual disk units).
Features of this RAID level include
fault tolerance
automatic mirroring: no commands are required to initiate it
physical separation of images
faster write operation than RAID-5
With a RAID-1 mirrored pair, the storage system writes the same data to both disk modules in the mirror, as shown in Figure 1-10.
First module
= User data (primary image)
201 43
. . .
= User data (secondary image)
Figure 1-10 RAID-1 Mirrored Pair (Hardware Mirrored Pair)
Second module
201 43
. . .
T o achieve the maximum fault tolerance, configure the mirr or with each disk module on a different internal SCSI bus; for example, the primary image on A0, the secondary image on B0, and so on.
15
Chapter 1: Features of the CHALLENGE RAID Storage System
RAID-1_0 Group: Mirrored RAID-0 Group
A RAID-1_0 configuration mirrors a RAID-0 group, creating a primary RAID-0 image and a secondary RAID-0 image for user data. This arrangement consists of four , six, eight, ten, twelve, fourteen, or sixteen disk modules. These disk modules make up two mirror images, with each image including two to eight disk modules. A RAID-1_0 group uses striping and combines the speed advantage of RAID-0 with the redundancy advantage of mirroring.
Figure 1-11 illustrates the distribution of user data with the default stripe element size of 128 sectors (65,536 bytes) in a six-module RAID-1_0 group. Notice that the disk block addresses in the stripe proceed sequentially fr om the first mirrored disk modules to the second mirror ed modules, to the third mirrored image disk modules, then from the first mirrored disk modules, and so on.
A RAID-1_0 group can survive the failure of multiple disk modules, providing that one disk module in each image pair survives. Thus, for highest availability and performance, the disk modules in an image pair must be on a different SCSI bus from the disk modules in the other image pair. For example, the RAID-1_0 group shown in Figure 1-11 has three disk modules in each image of the pair.
16
RAID Levels
= User data (primary image)
= User data (secondary image)
Secondary image
Primary image
Stripe element size
Stripe size
Stripe
Blocks
0-127 384-511 768-895 1152-1279
128-255
0-127 384-511 768-895 1152-1279
128-255 1280-1407512-639 896-1023
First module of primary image
Second module of primary image
1280-1407896-1023512-639
Third module of primary image
First module of secondary image
Second module of secondary image
1536-1663
1664-1791
1792-1919256-383 1408-15351024-1151640-767
1536-1663
1664-1791
Third module of secondary image
256-383 1792-19191408-15351024-1151640-767
Figure 1-11 Distribution of User Data in a RAID-1_0 Group
When you bind disk modules into a RAID-1_0 group, you must select them in this order: p1, s1, p2, s2, p3, s3, and so on, with primary (p1, p2, p3) and secondary (s1, s2, s3) disk modules on a separate internal SCSI buses.
17
Chapter 1: Features of the CHALLENGE RAID Storage System
RAID-5: Individual Access Array
This configuration usually consists of five disk modules (but can have three to sixteen) bound as a RAID-5 group. Because there are five internal SCSI-2 buses in the CHALLENGE RAID system, an array of five disk modules (or fewer) provides the greatest level of data redundancy.
A RAID-5 group maintains parity data that lets the disk group survivea disk module failure without losing data. In addition, in CHALLENGE RAID storage systems, the group can survive a single SCSI-2 internal bus failure if each disk module in the group was bound on an independent SCSI-2 internal bus. For highest data availability for a RAID-5 group, the disk modules making up the group should be on different SCSI internal buses (A, B, C, and so on).
With RAID-5 technology, the hardware writes parity information to each module in the array. If a module fails, the SP can reconstruct all user data from the user data and parity information on the other disk modules. After you replace a failed disk module, the SP automatically rebuilds the disk array using the information stored on the remaining modules. The rebuilt disk array contains a replica of the information it would have contained had the disk module never failed.
18
A RAID-5 group uses disk striping; see “Enhanced Performance: Disk Striping,” earlier in this chapter for an explanation of this feature. Figure 1-12 illustrates user and parity data with the default stripe element size of 128 sectors (65,536 bytes) in a five-module RAID-5 group. The stripe size comprises all stripe elements. Notice that the disk block addresses in the stripe proceed sequentially from the first module to the second, third, and fourth, fifth, then back to the first, and so on.
= User data = Parity data
Stripe element size
Stripe size
Stripe
Blocks
0-127
128-255 640-767 1152-1279
256-383 768-895
384-511
Parity
First module
1024-1151512-639 1536-1663 Parity
Second module
Third module
Parity
Fourth module
1280-1407
Parity
1792-1919
RAID Levels
...
2048-2175
...
2176-23031664-1791
...
2304-2431
...
Fifth module
Parity
896-1023
1408-1535
1920-2047
2432-2559
...
Figure 1-12 Distribution of User and Parity Data in a RAID-5 Group
For each write operation to a RAID-5 group, the CHALLENGE RAID storage system must perform the following steps:
1. Read data from the sectors being written and parity data for those
sectors.
2. Recalculate the parity data.
3. Write the new user and parity data.
19
Chapter 1: Features of the CHALLENGE RAID Storage System
RAID Hot Spare
A hot spare is a dedicated replacement disk unit on which users cannot stor e information. The capacity of a disk module that you bind as a hot spare must be at least as great as the capacity of the largest disk module it might r eplace.
Note: The hot spare is not available for RAID-0, because this RAID level
does not provide data redundancy.
If any disk in a RAID-5 group, RAID-1 mirrored pair, or RAID-1_0 group fails, the SP automatically begins rebuilding the failed disk module’s structure on the hot spare. When the SP finishes rebuilding, the disk group functions as usual, using the hot spare instead of the failed disk. When you replace the failed disk, the SP starts copying the data from the former hot spare onto the replacement disk. When the copy is done, the disk group consists of disk modules in the original slots, and the SP automatically frees the hot spare to serve as a hot spare again.
Note: The SP finishes rebuilding the disk module before it begins copying
data, even if you replace the failed disk during the rebuild process.
20
A hot spare is most useful when you need the highest data availability. It eliminates the time and effort needed for someone to notice that a module has failed, find a suitable replacement module, and insert it.
Y ou can have one or mor e hot spar es per storage system. Any module in the storage system can be configured as a hot spare except for modules A0, B0, C0, D0, E0, and A3, because they may store the licensed internal code or serve as the storage system’s cache vault.
For example, assume that the modules in slots A0-E0 are a RAID-5 group, those in slots A1 and B1 are a RAID-1 mirrored pair, and the module in A2 is a hot spare, as shown in Figure 1-13. If module D0 fails, the SP immediately begins rebuilding the RAID-5 group using the hot spare. When it finishes, the RAID-5 group consists of disk modules A0, B0, C0, A2, and E0.
Using the CHALLENGE RAID Command Line Interface
When you replace the failed module in D0, the SP starts copying the structure on A2 to D0. When it finishes, the RAID-5 group once again consists of modules A0-E0 and the hot spare becomes available for use if any other module fails. A similar sequence would occur if, for example, module A1 in the mirrored pair failed.
Deskside
A1 B1
Figure 1-13 Hot Spare Example
A0 B0
C0 D0 E0 A2
Hot spare
Using the CHALLENGE RAID Command Line Interface
Run the command line interface, /usr/raid5/raid5, in an IRIX window on your CHALLENGE server to
bind (group) or unbind physical disks into a RAID-0, RAID-1,
RAID-1_0, or RAID-5 unit or hot spare
Rackmount
A0 B0 C0 D0 E0
A1 B1
A2
change parameters on a currently bound group (logical unit number, or
LUN)
get names of devices controlled by the SP
change or get information about the caching environment
get information about the SP’s configuration
21
Chapter 1: Features of the CHALLENGE RAID Storage System
get information on all CRUs (customer-replaceable units)
display status information on disks
display the SP log
display information about a group of disks
perform housekeeping operations, such as clearing the error log or updating firmware
Note: Although the directory and command are raid5, the command is valid
for all RAID levels.
The relevant parameters of the command line interface are explained for each task in the rest of this guide. Appendix B is a complete guide to the command line interface.
22
Chapter 2
2. Storage System Configurations
This chapter explains the various CHALLENGE RAID configurations. Use it to plan your storage system or whenever you contemplate changes in your storage system or physical disk configuration.
A CHALLENGE RAID storage system is configured on two levels:
Storage-system configuration: number of storage-control processors and
SCSI-2 interfaces
Disk configuration within the storage system
Before you can plan your disk configuration, you must understand storage system configuration. Several storage system configurations are available for CHALLENGE RAID storage systems. Table 2-1 lists the hardware components making up each configuration and summarizes the features of each.
This chapter discusses these configurations in separate sections. Each section explains the error recovery features of the configuration.
23
Chapter 2: Storage System Configurations
Table 2-1 CHALLENGE RAID Configurations
Configuration Host SCSI-2 Interface SCSI-2
Bus
SPs Feature
Basic 1 1 1 1 Applications can continue after failure of any disk module,
but cannot continue after failure of SCSI-2 interface or SP.
Dual-interface/ dual-processor
1 2 2 2 Provides highest availability and best storage system
performance for single-host configurations. Applications can continue after any disk module fails.
Split-bus 2 2 (1 per server) 2
(1 per server)
2 Resembles two basic configurations side by side. Each host
and its applications can continue after any disk module fails. The host using a failed SCSI-2 interface or SP cannot continue after failure, but the other host can. If one host, SCSI-2 adapter, or SP fails, the other host can take over the failed host’s disks with system operator intervention.
Dual-bus/ dual-initiator
2 4 (2 per server) 2
(1 per server)
2 Provides highest availability and best storage-system
performance for dual-host configurations. With RAID of any level other than 0, applications can continue after failure of any disk module. If one host, SCSI-2 adapter , or SP fails, the other host can take over the failed host’s disk units with system operator intervention.
This configuration is required for Silicon Graphics FailSafe™ and Oracle Parallel Server™ (OPS™).
24
Basic Configuration
Basic Configuration
The basic configuration has one host with one SCSI-2 interface connected by a SCSI-2 bus to the SP in the storage system.
The system can survive failure of a disk module within a redundant RAID group, but it cannot continue after failure of a SCSI-2 interface or SP. Table 2-2 lists the error recovery features of the basic configuration.
Table 2-2 Error Recovery: Basic Configuration
Failing Component Continue After
Failure?
Disk module Yes Applications continue running. System operator replaces
Storage-control processor
Fan module Yes Applications continue running. System operator replaces
Power supply Yes If redundant power supply module is present, applications
SCSI-2 interface No I/O operations fail to storage system disk units. Authorized
SCSI-2 cable No I/O operations fail to storage system. System operator
No Storage system fails. System operator replaces SP and restarts
Recovery
module.
operating system.
module.
continue running; otherwise, storage system fails. Service provider replaces power supply.
service provider replaces interface, and system operator restarts operating system and applications.
replaces cable, and restarts operating system and applications.
25
Chapter 2: Storage System Configurations
Dual-Interface/Dual-Processor Configuration
The dual-interface/dual-processor configuration has one host with two SCSI-2 interfaces, each connected by a SCSI-2 bus to a different SP in the storage system.
For better performance with this configuration, you can bind some physical disk units on one SP and some other physical disk units on the other SP. The SP that binds a physical disk unit is the default owner of that physical disk unit.
The storage system can continue running after failure of a disk module within a redundant RAID group. It cannot continue after a SCSI-2 interface or an SP fails unless you manually transfer disk ownership. Table 2-3 lists these features.
Table 2-3 Error Recovery: Dual Interface/Dual-Processor Configuration
Failing Component Continue After
Failure?
Disk module Yes Applications continue running. System operator replaces
Storage-control processor
Fan module Yes Applications continue running. Silicon Graphics SSE or other
Power supply Yes If redundant power supply module is present, applications
Y es I/O operations fail to disk units owned by a failing SP. System
Recovery
module.
operator can transfer control of the failed SP’s disk units to the working SP, shut down the host, power off and on the storage system, reboot the host, and, when convenient, replace the SP and transfer control of disk units to the replacement SP.
authorized service provider replaces module.
continue running; otherwise, storage system fails. Service provider replaces power supply.
26
Dual-Interface/Dual-Processor Configuration
T able 2-3 Error Recovery: Dual Interface/Dual-Processor Configuration
Failing Component Continue After
Failure?
Recovery
SCSI-2 interface Y es I/O operations fail to storage system disk units owned by the
SP attached to the failed interface. System operator can transfer control of the failed SP’s disk units to the SP on the surviving interface, shut down the host, power off and on the storage system, and reboot the host. When convenient, the Silicon Graphics SSE or other authorized service provider can replace the interface and the system operator can transfer control of disk units to the replacement SP.
SCSI-2 cable Yes I/O operations fail to storage-system disk units owned by the
SP attached to the failed cable. System operator can transfer control of these disk units to the other SP, shut down the host, power off and on the storage system, reboot the host, replace the cable, and transfer control of disk units to the replacement SP .
In the example diagrammed in Figure 2-1, one group of five disk modules is bound by storage-control processor A (SP A) and another group of five disk modules is bound by SP B.
Disk modules owned by SP B (LUN 1)
SCSI-2 interface
SCSI-2 interface
CHALLENGE
SCSI-2 bus
SCSI-2 bus
SP A
SP B
CHALLENGE RAID
Figure 2-1 Dual-Interface/Dual-Processor Configuration Example
Disk modules owned by SP A (LUN 0)
27
Chapter 2: Storage System Configurations
Split-Bus Configuration
In this example, if one SP or SCSI-2 interface fails, stored data in either LUN is available through the alternate path. Automatic path switching in the event of an SP or SCSI-2 interface failure is possible if XLV volumes and applicable patches are used. For information on XLV volumes, see Getting Started With XFS Filesystems.
Note: Only qualified Silicon Graphics System Service Engineers can replace
SPs or SCSI-2 interfaces.
The split-bus configuration has two hosts, each with a SCSI-2 interface connected by a SCSI-2 bus to a storage-control processor in the storage system. Each host uses its own disks in the storage system independently.
The split-bus configuration resembles two basic configuration systems side by side. This configuration can be used for sites requiring high availability because either host can continue after failure of any disk module within a disk array, and a host can take over a failed host’s disks. A host cannot continue after a SCSI-2 interface or an SP fails unless you manually transfer disk ownership.
28
Table 2-4 lists the error recovery features for this configuration.
Table 2-4 Error Recovery: Split-Bus Configuration
Failing Component Continue After
Failure?
Disk module Yes Applications continue running. System operator replaces
Storage-control processor
Fan module Yes Applications continue running. Silicon Graphics SSE or other
Y es I/O operations fail to disk units owned by a failing SP. System
Recovery
module.
operator can transfer control of failed SP’s disk units to the working SP, shut down the host, power off and on the storage system, and reboot the host. Silicon Graphics SSE or other authorized service provider replaces the SP and transfers control of disk units to the replacement SP.
authorized service provider replaces the module.
Split-Bus Configuration
T able 2-4 (continued) Error Recovery: Split-Bus Configuration
Failing Component Continue After
Failure?
Power supply Yes If redundant power supply module is present, applications
SCSI-2 interface Y es I/O operations fail to storage-system disk units owned by the
SCSI-2 cable Yes I/O operations fail to storage system disk units owned by the
Recovery
continue running; otherwise, storage system fails. Service provider replaces power supply.
SP attached to the failed interface. System operator can transfer control of the failed SP’s disk units to the SP on the interface in the other host, shut down the other host, power off and on the storage system, and reboot the other host. Silicon Graphics SSE or other authorized service provider replaces the interface.
SP attached to the failed cable. System operator can transfer control of these disk units to the other SP, shut down the host, power off and on the storage system, reboot the host, replace the cable, and transfer control of disk units to the replacement SP .
In the example diagrammed in Figure 2-2, one group of five disk modules is bound by storage-control processor A (SP A), which is connected via a SCSI-2 bus to one CHALLENGE server; another group of five disk modules is bound by SP B, which is connected by a different SCSI-2 bus to the second CHALLENGE server.
29
Chapter 2: Storage System Configurations
SCSI-2 interface
Disk modules owned by SP B (LUN 1)
SCSI-2 bus
Disk modules owned by SP A (LUN 0)
SP A
CHALLENGE 1
SCSI-2 bus
SCSI-2 interface
CHALLENGE 2
SP B
CHALLENGE RAID
Figure 2-2 Split-Bus Configuration Example
Caution: This configuration does not afford failover capability.
If one SP fails or if the SCSI-2 connection from one host is broken, that host does not have access to the CHALLENGE RAID storage system until the SP is replaced or the SCSI-2 connection is repaired. The host using the remaining SCSI-2 connection and remaining operational SP still has full access to its own data.
The storage-control processor that binds a disk module is the default owner of the disk module. The route through the SP that owns a disk module is the primary route to the disk module. The route through the other SP is the secondary route to the disk module.
30
In a dual-interface system, either CHALLENGE server can use any of the disk modules in the storage system, but only one CHALLENGE server at a time can use a disk module.
Dual-Bus/Dual-Initiator Configuration
The dual-bus/dual-initiator configuration provides the highest availability. Each host has two SCSI-2 adapters, each of which connects by a separate SCSI-2 bus to a separate SP in the storage system. Since this configuration protects against a SCSI-bus cable failure, it provides higher availability than the dual-initiator configuration. It is for enterprises requiring the highest level of availability, such as the Oracle Parallel Server and FailSafe products.
For better performance with this configuration, you can bind some physical disk units on one SP and the other physical disk units on the other SP. The SP that binds a physical disk unit is its default owner . The route thr ough the SP that owns a physical disk unit is the primary route to the physical disk unit. The route through the other SP is the secondary route to the physical disk unit, and is available if a component in the primary route fails. T able 2-5 lists the error recovery features of the dual-bus/dual-initiator configuration.
Caution: Because both hosts can access the same disk modules simultaneously, the danger exists that one host can overwrite data stored by the other. This configuration requires specific hardware and software (such as a database lock manager) to protect the integrity of the stored data.
Dual-Bus/Dual-Initiator Configuration
Table 2-5 Error Recovery: Dual-Bus/Dual-Initiator Configuration
Failing Component Continue After
Failure?
Disk module Yes With RAID levels specified at any level other than 0,
Storage-control processor
Fan module Yes Applications continue running. System operator replaces
Yes I/O operations fail to disk units owned by the failing SP.
Recovery
applications continue running. System operator replaces module.
System operator can transfer control of the failed SP’s disk units to the surviving SP, shut down the host, power the storage system off and on, and reboot both hosts. When convenient, Silicon Graphics SSE or other authorized service provider replaces the SP, and the system operation can transfer control of the disk units to the replacement SP.
module.
31
Chapter 2: Storage System Configurations
Table 2-5 Error Recovery: Dual-Bus/Dual-Initiator Configuration
Failing Component Continue After
Failure?
Power supply Yes If redundant power supply module is present, applications
SCSI-2 interface Yes The host with the failed adapter cannot access the disks
SCSI-2 cable Yes I/O operations fail to storage-system disk units owned by the
Host Yes Operations continue on the surviving host.
Recovery
continue running; otherwise, storage system fails. Service provider replaces power supply.
owned by the SP connected to the failed adapter. System operator can transfer control of these disks to the SP connected to the working adapter, shut down both hosts, power off and on the storage system, and reboot both hosts. When convenient, the Silicon Graphics SSE or other authorized service provider can replace the interface, and the system operator can transfer control of disk units to replacement SP.
SP connected to the failed cable. System operator can transfer control of these disk units to the other SP, shut down the host, power off and on the storage system, reboot the host, replace the cable, and transfer control of the disk units to the replacement SP.
In the example diagrammed in Figure 2-3, some modules are bound to one SP, which is their primary owner, and the other disk modules are bound to the other SP, which is their primary owner. Either host can use any of the physical disk units in the storage system, but only one host at a time can use a physical disk unit.
32
Dual-Bus/Dual-Initiator Configuration
SCSI-2
interface 1
SCSI-2
interface 2
CHALLENGE 1
CHALLENGE RAID
Host 2: Mirrored pair
for user directories;
moderate access time
Figure 2-3 Dual-Bus/Dual-Initiator Configuration Example
Caution: Because both hosts have access to the all disk modules and their
data in this configuration, it is possible for one host to overwrite the other’s data unless appropriate filesystem configuration and failsafe software is installed.
SCSI-2 bus
In Out
SP A
Host 1: RAID-5
SCSI-2 bus
In Out
SP B
group for
fast-access
database
Host 2: Accounts on 6 disks bound as RAID-1_0
Host 1: Mirrored pair for user directories; moderate access time
Host 1: Hot spare
SCSI-2
interface 1
SCSI-2
interface 2
CHALLENGE 2
33
Chapter 3
3. Operating the Storage System
This chapter describes how to run CHALLENGE RAID after you have configured it. The chapter explains:
checking storage system status
shutting down the CHALLENGE RAID storage system
restarting the CHALLENGE RAID storage system
This chapter introduces the /usr/raid5/raid5 command (command line interface, or CLI). Use raid5 with its parameters in an IRIX shell on CHALLENGE to get names of devices controlled by the storage-control processor (SP), display status information on disk modules, disk module groups (LUNs), SPs, and other system components, and display the storage processor log, in which error messages are stored.
Note: Although the directory and command are raid5, the command is valid
for all RAID levels.
Other chapters in this guide explain how to use the raid5 command to bind (group) physical disks into RAID units and unbind them, set up caching, and accomplish other tasks.
35
Chapter 3: Operating the Storage System
Checking CHALLENGE RAID Storage System Status
T o check storage system status, you may find it easiest to look at the storage system cabinet to see if the amber service light is lit.
Look for two lights at the right of the disk modules (deskside storage system) or above the disk modules (chassis in rack). The green light indicates whether the unit is powered on; the amber light indicates a fault. See Figure 3-1.
Power light (green)
Service light (amber)
36
Deskside RackFront of storage system
Figure 3-1 CHALLENGE RAID Indicator Lights
The amber service light comes on when
an SP is reseated
the CHALLENGE RAID is powered off and on
the battery backup unit has not finished recharging (if battery backup unit is present in the system)
If the service light is lit, look for a disk-module fault light that is lit. Then you can either explore status further using the raid5 command in an IRIX shell, as explained in this section:
using the raid5 command
getting the device name with getagent
Checking CHALLENGE RAID Storage System Status
getting general system information
getting information about disks
getting information about other components
displaying the CHALLENGE RAID unsolicited event log
Using the
raid5
Command
The raid5 command sends storage management and configuration requests to an application programming interface (API) on the CHALLENGE server. For the raid5 command to function, the agent—an interpreter between the command line interface and the CHALLENGE RAID storage system—must be running.
The synopsis of the raid5 command is
raid5 [-vp] [-d device] parameter [optional_arguments]
In this syntax, variables mean:
-v Enables verbose return.
-p Parses the raid5 command without calling the API. If the
string does not parse correctly, an error message is printed to stderr; otherwise there is no output.
-d device Target RAID device. Use raid5 getagent for a list of RAID
devices. This switch must be present for all raid5 management and configuration commands unless the environment variable indicates otherwise. This switch overrides an environment variable.
Note: Appendix B is a complete alphabetical listing of raid5 parameters.
37
Chapter 3: Operating the Storage System
Getting Device Names With
getagent
Use the getagent parameter with raid5 to display information on devices controlled by the API:
raid5 getagent
Following is a sample output for one device; normally, the output would give information on all devices.
Name: Disk Array Desc: RAID5 Disk Array Node: sc4d210 Signature:0xf3b51700 Peer Signature: 0x657e0a00 Revision: 7.12.4 SCSI ID: 1 Prom Rev: 0x0076100 SP Memory: 64 Serial No: 94-7240-808
Table 3-1 summarizes entries in the raid5 getagent output.
Table 3-1 Output of raid5 getagent
Entry Meaning
Name ASCII string found in the agent configuration file which assigns a
name to the node being accessed (see Node description below).
38
Desc ASCII string found in the agent configuration file which describes
the node being accessed (see Node description below).
Node The /dev/scsi entry which the agent uses as a path to the actual
SCSI device. This value must be entered by the user for every CLI
command (except getagent). Signature Unique 32-bit identifier for the SP being accessed through Node. Peer Signature Unique 32-bit identifier for the other SP in the chassis; 0 if no
additional SP is present. Revision Revision of firmware currently running on the SP. Prom Rev PROM revision present on the SP.
Checking CHALLENGE RAID Storage System Status
Table 3-1 (continued) Output of raid5 getagent
Entry Meaning
SP Memory Amount of DRAM present on the SP. Serial No 12-digit ASCII string that uniquely identifies this subsystem.
Getting General System Information
To get general system information, use
raid5 -d device getcontrol
A possible output of this command follows:
System Fault LED: OFF Statistics Logging: ON System Cache: ON Max Requests: 23 Average Requests: 5 Hard errors: 0 Total Reads: 18345 Total Writes: 1304 Prct Busy: 25 Prct Idle: 75 System Date: 5/5/1995 Day of the week: Friday System Time: 12:43:54
Getting Information About Disks
For information about all bound disks in the system, use this command in an IRIX shell:
/usr/raid5/raid5 -d device getdisk
For information on a particular disk, use
/usr/raid5/raid5 -d device getdisk <diskposition>
39
Chapter 3: Operating the Storage System
In this command, diskposition has the format bd, where b is the bus the disk is located on (a through e; be sure to use lower case) andd is the device number (0 through 3). Figure 3-2 diagrams disk module locations.
Deskside Chassis assembly in rack
A1 B1 C1 D1 E1 A3 B3 C3 D3 E3
A0
B0 C0 D0
E0
A2
B2 C2 D2
E2
5 to 20 disk modules in groups of 5
Figure 3-2 Disk Module Locations
A0 B0 C0 D0
A1 B1 C1 D1 E1
E0 A2 B2 C2 D2
A3 B3 C3 D3 E3
For example, the following command gets information about disk A2.
raid5 -d scsi4d210 getdisk a2
A sample output of this command follows:
A0 Vendor Id: SEAGATE A0 Product Id: ST15150N A0 Lun: 0 A0 State: Bound and Not Assigned A0 Hot Spare: NO A0 Prct Rebuilt: 100 A0 Prct Bound: 100 A0 Serial Number: 032306 A0 Capacity: 0x000f42a8 A0 Private: 0x00009000 A0 Bind Signature: 0x1c4eb2bc A0 Hard Read Errors: 0
E2
40
Checking CHALLENGE RAID Storage System Status
A0 Hard Write Errors: 0 A0 Soft Read Errors: 0 A0 Soft Write Errors: 0 A0 Read Retries: 0 A0 Write Retries: 0 A0 Remapped Sectors: 0 A0 Number of Reads: 1007602 A0 Number of Writes: 1152057
Table 3-2 interprets items in this output.
Table 3-2 Output of raid5 getdisk
Output Meaning
Vendor Id Manufacturer of disk drive Product Id 2.1 GB disk: ST32550N
4.2 GB disk: ST15150N Lun Logical unit number to which this disk is bound State Removed: disk is physically not present in the chassis or has been
powered off Off: disk is physically present in the chassis but is not spinning Powering Up: disk is spinning and diagnostics are being run on it Unbound: disk is healthy but is not part of a LUN Bound and Not Assigned: disk is healthy, part of a LUN, but not being
used by this SP Rebuilding: disk is being rebuilt Enabling: disk is healthy, bound, and being used by this SP Binding: disk is in the process of being bound to a LUN
Formatting: disk is being formatted Hot Spare YES or NO Prct Rebuilt Percentage of disk that has been rebuilt Prct Bound Percentage of disk that has been bound Serial Number Serial number from disk inquiry command Capacity Actual disk capacity in blocks Private Amount of physical disk reserved for private space
41
Chapter 3: Operating the Storage System
Output Meaning
Bind Signature Unique value assigned to each disk in a logical unit at bind time Hard Read Errors Number of hard errors encountered on reads for this disk Hard Write Errors Number of hard errors encountered on writes for this disk Soft Read Errors Number of soft errors encountered on reads for this disk Soft Write Errors Number of soft errors encountered on writes for this disk Read Retries Number of retries occurring during reads Write Retries Number of retries occurring during writes Remapped Sectors Number of sectors that have been remapped Number of Reads Number of reads this disk has seen Number of Writes Number of writes this disk has seen
Table 3-2 (continued) Output of raid5 getdisk
Getting Information About Other Components
42
For state information on other components—field-replaceable units—in the CHALLENGE RAID storage system besides disk modules, use
raid5 -d device getcrus
A sample output of this command follows:
FANA State: Present FANB State: Present VSCA State: Present VSCB State: Present VSCC State: Present SPA State: Present SPB State: Present BBU State: Present
Note: In this output, information on the power supplies is shown under VSC
(voltage semi-regulated converter), SP information is shown under SP, and battery backup unit information is shown under BBU.
Checking CHALLENGE RAID Storage System Status
Table 3-3 interprets items in this output.
Table 3-3 Output of raid5 getcrus
Output Meaning
FANA, FANB Fan banks A and B. VSCA Power supply (voltage semi-regulated converter). VSCB Optional third power supply. SPA Storage-control processor. SPB Optional second storage-control processor. BBU Battery backup unit, which has three states: Faulted (removed),
Charging, and Present (fully charged or charging). If the battery backup unit takes longer than an hour to charge, it
shuts itself off and transitions to the “Faulted” state.
Displaying the CHALLENGE RAID Unsolicited Event Log
The storage-control processor maintains a log of event messages in pr ocessor memory. These events include hard errors, startups, and shutdowns involving disk modules, fans, SPs, power supplies, and the battery backup unit. Periodically , the SP writes this log to disk to maintain it when SP power is off. The log can hold over 2,000 event messages; it has a filter feature that lets you select events by device or error message.
The event messages are in chronological order, with the most recent messages at the end. To display the entire log, use
raid5 -d device getlog
To display the newest n entries in the log, starting with the oldest entry, use
raid5 -d device getlog +n
To display the oldest n entries in the log, starting with the oldest entry, use
raid5 -d device getlog -n
43
Chapter 3: Operating the Storage System
Output of the command raid5 -d device getlog +5 might be
12/17/94 09:59;51 A3: (A07) Cru Powered Down [0x47] 12/17/94 09:59;51 A3: (608) Cru Ready [0x0] 12/17/94 09:59;51 A3: (603) Cru Rebuild Started [0x0] 12/17/94 09:59;51 A3: (604) Cru Rebuild Complete [0x0] 12/17/94 09:59;51 A3: (602) Cru Enabled [0x0]
These entries show that a field-replaceable unit (disk module, fan unit, battery backup unit, SP) has failed, been replaced, been rebuilt, and been enabled.
At the tail of each log entry is an error code in brackets (for example,[0x47]) that gives diagnostic information when it is available. See “getlog” in Appendix B for explanations of these codes.
To clear the event log, use
raid5 -d device clearlog
Note: You must be root to use the clearlog parameter.
Shutting Down the CHALLENGE RAID Storage System
Follow these steps to shut down the CHALLENGE RAID storage system:
1. If you are using storage system caching, make sure that it is disabled; use raid5 getcache to check status. If necessary, disable it with raid5 setcache disable, as explained in Chapter 7.
2. Turn off the power switch on the back of the CHALLENGE RAID storage system, as shown in Figure 3-3.
Note: You do not need to disable the power for the SP(s).
44
Deskside RackBack of storage system
Figure 3-3 Turning Off (On) Power (Back of CHALLENGE RAID Chassis)
Restarting the CHALLENGE RAID Storage System
To start the CHALLENGE RAID storage system, follow these steps:
1. Turn on the storage system’s power; see Figure 3-3.
Restarting the CHALLENGE RAID Storage System
The green power light on the front of the storage system turns on (see Figure 3-4) and the fans rotate.
Power light (green)
Service light (amber)
Deskside RackFront of storage system
Figure 3-4 CHALLENGE RAID Indicator Lights
45
Chapter 3: Operating the Storage System
2. If the busy light on none of the drive modules lights up, make sure that the power for each SP is enabled. Move the fan module’s latch to the UNLOCK position, as indicated in Figure 3-5.
Deskside Rack
Figure 3-5 Unlocking the Fan Module
3. Swing open the fan module.
46
Caution: To prevent thermal shutdown of the system, never leave the
fan module open more than two minutes.
SP B
SP A
Restarting the CHALLENGE RAID Storage System
4. Move the SP’s power switch to the enable position, as shown in
Figure 3-6.
Deskside Rack
SP A SP B
Figure 3-6 Enabling an SP’s Power
5. Close the fan module by closing the fan module and moving the
module’s latch to the LOCK position.
6. Power on the CHALLENGE server(s).
47
4. Configuring Disks
Binding Disks Into RAID Units
Chapter 4
This chapter explains
binding disks into RAID units
getting disk group (LUN) information
changing LUN parameters
The chapter concludes with information on dual processors, load balancing, and device names.
The physical disk unit number is also known as the logical unit number, or LUN. (The unit is a logical concept, but is recognized as aphysical disk unit by the operating system; hence, the seemingly contradictory names.) The LUN is a hexadecimal number between 0 and F (15 decimal).
Unlike standard disks, physical disk unit numbers (LUNs) lack a standard geometry. Disk capacity is not a fixed quantity between disk-array LUNs. The effective geometry of a disk-array LUN depends on the type of physical disks in the array and the number of physical disks in the LUN.
T o group physical disks into RAID-0, RAID-1, RAID-1_0, or RAID-5 units or to create a hot spare, use as root:
raid5 -d device bind raid-type lun-number disk-names [optional-args]
Variables in this syntax are explained below.
-d device Target RAID device, as returned by raid5 getagent; see
“Getting Device Names With getagent,” in Chapter 3.
49
Chapter 4: Configuring Disks
raid-type Choices are
r0: RAID-0
r1: RAID-1
r1_0: RAID-1_0
r5: RAID-5
hs: hot spare
lun-number Logical unit number to assign the unit (a hexadecimal
number between 0 and F).
disk-names Indicates which physical disks to bind, in the format bd,
whereb is the physical bus name (a through e; be sure to use lower case) and d is the device number on the bus (0 through 3). For example, a0 represents the device 0 on bus A, and e2 represents device 2 on bus E.
A RAID-0 bind requires a minimum of three (maximum 16) disks.
A RAID-1 bind requires two disks on separate buses.
50
A RAID-1_0 bind requires a minimum of four disks per bus. The maximum is 16 disks, grouped in sets of four on different buses. A RAID-1_0 bind requires separate buses for each member of the image pair. Select the disks in this order: p1, s1, p2, s2, p3, s3, and so on.
A RAID-5 bind also requires separate buses, each with five disk modules. Legal RAID-5 bind configurations are
a0 b0 c0 d0 e0
a1 b1 c1 d1 e1
a2 b2 c2 d2 e2
a3 b3 c3 d3 e3 A hot spare bind requires one disk, which cannot be A0, B0,
C0, D0, E0, or A3. The capacity of the hot spare must be at least as great as the capacity of the largest disk module it might replace.
Binding Disks Into RAID Units
The optional arguments are as follows:
-r rebuild-time Maximum time in hours to rebuild a replacement disk.
Default is 4 hours; legal values are any number greater than or equal to 0.
A rebuild time of 2 hours rebuilds the disk more quickly but degrades response time slightly. A rebuild time of 0 hours rebuilds as quickly as possible but degrades performance significantly.
If your site requires fast response time and you want to minimize degradation to normal I/O activity, you can extend the rebuilding process over a longer period of time, such as 24 hours. You can change the rebuild time later without damaging the information stored on the physical disk unit.
-s stripe-size Number of blocks per physical disk in a RAID stripe.
Default is 128; legal values are any number greater than 0. The smaller the stripe element size, the more efficient the
distribution of data read or written. However, if the stripe size is too small for a single host I/O operation, the operation requires accessing two stripes, thus causing the hardware to read and/or write from two disk modules instead of one. Generally, it is best to use the smallest stripe element size that will rarely force access to another stripe. The default stripe element size is 128 sectors. The size should be an even multiple of 16 sectors (8 KB).
Note: This value is ignored for hot spares.
-c cache-flags Values are:
none: no caching
read: read caching
write: write caching
rw: read and write caching The default is none.
51
Chapter 4: Configuring Disks
Note: Although bind returns immediate status for a RAID device, the bind
itself does not complete for 45 to 60 minutes, depending on system traffic. Use getlun to monitor the progress of the bind; getlun returns the percent bound. When the bind is complete, each disk is noted as “Bound But Not Assigned” in the getlun output.
The following example binds disks A0, B0, C0, D0, and E0 into a RAID-5 logical unit with a logical unit number of 3, a four-hour maximum rebuild time, and a 128-block stripe size per physical disk, with read cache enabled.
raid5 -d sc4d2l0 bind r5 3 a0 b0 c0 d0 e0 -r 4 -s 128 -c read
The following example binds A2 and B2 into a RAID-1 logical unit with a LUN number of 2 and a four-hour maximum rebuild time, with read cache enabled.
raid5 -d sc4d2l0 bind r1 2 a2 b2 -r 4 -c read
The following example binds disks A1, B1, C1, and D1 into a RAID-1_0 logical unit with a LUN number of 1, a four-hour maximum rebuild time, and a 128-block stripe size per physical disk, with read cache enabled.
raid5 -d sc4d2l0 bind r1_0 1 a1 b1 c1 d1 -r 4 -s 128 -c read
52
The following example binds A3, B3, C3, D3, and E3 into a RAID-0 logical unit with a LUN number of 3, and a 128-block stripe size per physical disk, with read cache enabled.
raid5 -d sc4d2l0 bind r0 3 a3 b3 c3 d3 e3 -s 128 -c read
The following example binds disk E3 as a hot spare with a LUN number of 7.
raid5 -d sc4d2l0 bind hs 7 e3
There is no output for raid5 with the bind parameter. Errors are printed to stderr.
Note: For complete messages, it is recommended that you use the -v option.
Getting Disk Group (LUN) Information
To display information on a logical unit and the components in it, use the getlun parameter:
raid5 getlun lun-number
The following example displays information about LUN 3.
raid5 -d sc4d2l0 getlun 3
Following is truncated output for a RAID-5 group of five disks.
Note: Information on individual disks is not displayed unless statistics
logging is enabled with raid5 getcontrol. See “Getting Information About Other Components” in Chapter 3 of this guide.
Type: RAID5 Stripe size: 128 Capacity: 0x10000 Current owner: YES Auto-trespass: Disabled Auto-assign; Enabled Write cache: Disabled Read cache: Disabled Idle Threshold: 0 Idle Delay Time: 20 Write Aside Size: 2048 Default Owner: YES Rebuild Time: 0 Read Hit Ratio: 0 Write Hit Ratio: 0 Prct Reads Forced Flushed: 0 Prct Writes Forced Flushed: 0 Prct Rebuilt: 100 Prct Bound: 100
Getting Disk Group (LUN) Information
A0 Enabled A0 Reads: 62667 A0 Writes: 29248 A0 Blocks Read: 3212517 A0 Blocks Written: 471642 A0 Queue Max: 26 A0 Queue Avg: 1 A0 Avg Service Time: 14
53
Chapter 4: Configuring Disks
Entry Meaning
A0 Prct Idle: 100 A0 Prct Busy: 0 A0 Remapped Sectors: 0 A0 Read Retries: 50 A0 Write Retries: 0
B0 Enabled [etc.]
C0 Enabled [etc.]
D0 Enabled [etc.]
E0 Enabled [etc.]
Table 4-1 summarizes entries in the raid5 getlun output.
Table 4-1 Output of raid5 getlun
54
Type RAID0, RAID1, RAID10, RAID5, or SPARE Stripe Size Sectors per disk per stripe with which the unit was bound Capacity Number of sectors total for use by user Current owner YES if this SP owns the unit; NO if it does not
Set with chglun; see “Dual Interfaces, Load Balancing, and Device
Names,” later in this chapter Auto-trespass Always Disabled Auto-assign Always Enabled Write Cache Enabled means this LUN is write caching; otherwise, Disabled Read Cache Enabled means this LUN is read caching; otherwise, Disabled Idle Threshold Maximum number of I/Os outstanding; used to determine cache
flush start time; set with chglun Idle Delay Time Amount of time in 100-ms intervals that unit is below idle
threshold; set with chglun
Getting Disk Group (LUN) Information
Table 4-1 (continued) Output of raid5 getlun
Entry Meaning
W rite Aside Size Smallest write-request size in blocks that can bypass the cache and
go directly to the disk; set with chglun
Default Owner YES if this SP is the default owner (not necessarily current owner)
of this LUN, otherwise, NO
Rebuild Time Amount of time in hours in which a rebuild should be performed.
0 means rebuild as fast as possible, but means a degradation in host I/O performance
Read Hit Ratio Percentage of read requests to the controller that can be satisfied
from the cache without requiring disk access
Write Hit Ratio Percentage of write requests to the cache that can be satisfied with
the cache without requiring a disk access Prct Reads Forced Flushed Percentage of read requests that flushed the cache Prct Writes Forced Flushed Percentage of write requests that flushed the cache Prct Rebuilt Percentage complete during a rebuild Prct Bound Percentage complete during a bind Diskname State Enabled, Binding, etc. (same as for getdisk) Diskname Reads Total number of reads this disk has done Diskname Writes Total number of writes this disk has done Diskname Blocks Read Total number of blocks this disk has read Diskname Blocks Written Total number of blocks this disk has written Diskname: Queue Max Maximum number of I/Os queued up to this drive Diskname: Queue Avg Average number of I/Os queued up to this drive Diskname: Avg Service Time Average service time in milliseconds Diskname: Prct Idle Percentage of time disk is not servicing request Diskname: Prct Busy Percentage of time disk is servicing request Diskname: Remapped Sectors Number of remaps that have occurred on this disk
55
Chapter 4: Configuring Disks
Entry Meaning
Diskname Read Retries Number of read retries that have occurred on this disk Diskname Write Retries Number of write retries that have occurred on this disk
Changing LUN Parameters
Table 4-1 (continued) Output of raid5 getlun
To change parameters for a logical unit, use
raid5 -d device chglun -l lun [ -c cache-flags] [-d default-owner] [-r rebuild-time] [-i idle-thresh] [-t idle-delay-time] [-w write-aside]
Note: Only root can use the chglun parameter.
In this syntax, variables mean the following:
-l lun Logical unit number to be changed
-c cache-flags Values are:
56
none: no caching
read: read caching
write: write caching
rw: read and write caching The default is none.
-d default-owner Values are
1: change storage-control processor ownership of LUN
0: don’t change ownership If your storage system has dual SPs, see “Dual Interfaces,
Load Balancing, and Device Names,” later in this chapter.
-r rebuild-time Maximum time in hours to rebuild a replacement disk. Default is 4 hours; legal values are any number greater than or equal to 0.
Dual Interfaces, Load Balancing, and Device Names
-i idle-thresh Maximum number of I/Os that can be outstanding to a LUN and have the LUN still be considered idle. Used to determine cache flush start time. Legal values are any number greater than or equal to 0.
-t idle-delay-time Amount of time in 100-ms intervals that a unit must be below idle-thresh to be considered idle. Once a unit is considered idle, any dirty pages in the cache can begin idle time flushing. Legal values are any number greater than or equal to 0.
-w write-aside The smallest write request size, in blocks, that will bypass the cache and go directly to the disks. Legal values are any number greater than or equal to 0.
The following example changes LUN 3 to perform write caching and rebuild in four hours; it does not change the default owner.
raid5 -d sc4d2l0 chglun -l 3 -c write -d 0 -r 4
There is no output for the chglun parameter. Errors are printed to stderr.
Dual Interfaces, Load Balancing, and Device Names
If your storage system has two SPs (split-bus, dual-bus/dual-initiator, or dual-interface/dual-processor), you can choose which disks to bind on each SP. This flexibility lets you balance the load on your SCSI-2 interfaces and SPs.
The SP on which you bind a physical disk unit is its default owner . The route through the SP that owns a physical disk unit is the primary route to the disk unit, and determines the device name of the disk unit. The route through the other SP is the secondary route to the disk unit.
When the storage system is running, you change the primary route to a physical disk unit by transferring ownership of the physical disk unit from one SP to another . The change in ownership and the new r oute take ef fect at the next power-on.
57
Chapter 5
5. Maintaining Disk Modules
This chapter describes
identifying and verifying a failed disk module
setting up the workplace
replacing a disk module
installing an add-on disk module array
Identifying and Verifying a Failed Disk Module
If you have determined that a module has failed by examining the cabinet fault light or by using the raid5 getdisk or raid5 getcrus command, as explained in Chapter 3 in this guide, you can replace the defective module and rebuild your data without powering off the CHALLENGE RAID storage system or interrupting user applications.
Caution: Removing the wrong drive module can introduce an additional fault that shuts down the physical disk containing the failed module. Before removing a disk module, verify that the suspected module has actually failed.
59
Chapter 5: Maintaining Disk Modules
The fault indicator on a disk module does not necessarily mean that the drive itself has failed. Failure of a SCSI bus, for example, lights the fault indicator on each disk module on that bus.
To verify a suspected disk-module failure, follow these steps:
1. Look for the module with its amber fault light on. Figure 5-1 shows the fault indicator light and other lights on a disk module.
Fault light (amber)
Deskside
Busy light (green)
Figure 5-1 Disk Module Status Lights
Ready light (green)
Ready light (green)
Busy light (green)
Fault light (amber)
Rack
2. Determine the failed module’s ID; use Figure 5-2.
Caution: Use only CHALLENGE RAID disk modules to replace failed disk modules. Order them from the Silicon Graphics hotline: 1-800-800-4SGI (1-800-800-4744). CHALLENGE RAID disk modules contain proprietary firmware that the storage system requires for corr ect functioning. Using any other disks, including those from other Silicon Graphics systems, can cause failure of the storage system. If you replace a disk module, you must update the firmware as explained in “Updating the Disk Module Firmware,” later in this chapter.
60
Identifying and Verifying a Failed Disk Module
Deskside Chassis assembly in rack
A1 B1 C1 D1 E1 A3 B3 C3 D3 E3
A0 B0 C0 D0 E0 A2 B2 C2 D2 E2
5 to 20 disk modules in groups of 5
Figure 5-2 Disk Module Locations
A0 B0 C0 D0
A1 B1 C1 D1 E1
E0 A2 B2 C2 D2
A3 B3 C3 D3 E3
3. If you have not already checked the module status with raid5 getdisk, do
so now; see “Getting Information About Disks” in Chapter 3.
4. If you have not already checked the unsolicited error log for a message
about the disk module, as explained in “Displaying the CHALLENGE RAID Unsolicited Event Log” in Chapter 3, do so now.
E2
A message about the disk module contains its module ID (such as A0 or B3). Check for any other messages that indicate a related failure, such as failure of a SCSI bus or a general shutdown of a chassis, that might mean the disk module itself has not failed.
Note: If you are using storage system caching, the system uses modules
A0, B0, C0, D0, and E0 for its cache vault. If one of these modules fails, the storage system dumps its cache image to the remaining modules in the vault; then it writes all dirty (modified) pages to disk and disables caching. The cache status changes, as indicated in the output of the raid5 getcache command. Caching remains disabled until you insert a replacement module and the storage system rebuilds the module into the physical disk unit. For information on caching, see Chapter 7.
61
Chapter 5: Maintaining Disk Modules
Caution: Although you can remove a disk module without damaging the disk data, do this only when the disk module has actually failed. Never remove a disk module unless absolutely necessary, and only when you have its replacement available. Never replace more than one disk module at a time; use only correct disk modules available from Silicon Graphics, Inc.
Setting Up the Workplace for Replacing or Installing Disk Modules
T o avoid discharging electrostatic char ge that has accumulated on your body discharges through the circuits, follow these guidelines for setting up the work area before you replace disk modules or install additional arrays.
If the air in the work area is very dry, running a humidifier in the work area will help decrease the risk of electrostatic charge damage (ESD).
Provide enough room to work on the equipment. Clear the work site of any unnecessary materials or materials that naturally build up electrostatic charge, such as foam packaging, foam cups, cellophane wrappers, and similar materials.
The disk module is extremely sensitive to shock and vibration. Even a slight jar can severely damage it.
62
Do not remove disk modules from their antistatic packaging until the exact moment that you are ready to install them.
Before removing a disk module from its antistatic bag, place one hand firmly on an unpainted metal surface of the chassis, and at the same time, pick up the disk module while it is still sealed in its antistatic bag. Once you have done this, do not move around the room or contact other furnishings, personnel, or surfaces until you have installed and secured the disk module in the equipment.
After you remove a disk module or filler module, avoid moving away from the work site; otherwise, you may build up an electrostatic char ge.
If you must move around the room or touch other surfaces before securing the disk module in the storage system, first place the disk module back in the antistatic bag. When you are ready again to install the disk module, repeat these procedures.
Replacing a Disk Module
Replacing a Disk Module
This section explains
ordering replacement disk modules
unbinding the disk
removing the failed disk module
installing the replacement disk module
updating the disk module firmware
Ordering Replacement Disk Modules
Order replacement disk modules only from the Silicon Graphics, Inc. hotline:
1-800-800-4SGI (1-800-800-4744)
Use Table 5-1 as a guide to ordering replacement disk modules.
Table 5-1 Ordering Replacement Disk Modules
Unit Silicon Graphics
Marketing Code
Replacement 2.1 GB drive P-S-RAID5-1X2 Replacement 4.2 GB drive P-S-RAID5-1X4
Caution: Use only CHALLENGE RAID disk modules as replacements; only they contain the correct device firmware. Other disk modules, even those from other Silicon Graphics equipment, will not work. Do not mix disk modules of different capacities within one array.
63
Chapter 5: Maintaining Disk Modules
Unbinding the Disk
When you change a physical disk configuration, you change the bound configuration of a physical disk unit. Physical disk unit configuration changes when you add or remove a disk module, or physically move one or more disk modules to different slots in the chassis.
Caution: Unbinding destroys all the data on a physical disk unit. Before unbinding any physical disk unit, make a backup copy of any data on the unit that you want to retain.
T o unbind a disk, use theunbind parameter with the raid5 command. Follow these steps:
1. In an IRIX window, use raid5 getagent to get the device name (node number):
raid5 getagent
2. If necessary, use raid5 getdisk to verify which the position of the failed disk.
3. Use raid5 unbind to unbind the failed disk:
raid5 -d device unbind lun-number [-o]
64
In this syntax, device is the device name as returned by getagent and lun-number is the number of the logical unit to unbind. The -o flag
specifies that the user is not prompted for permission. The unbind parameter has no output.
The following example destroys LUN 3 without prompting for permission:
raid5 -d sc4d2l0 unbind 3 -o
The disks in LUN 3 are now free to be reconfigured. Use raid5 bind to configure disks, as explained in Chapter 4.
Replacing a Disk Module
Removing a Failed Disk Module
You can replace a failed disk module while the storage system is powered on. If necessary, you can also replace a disk module that has not failed, such as a module that has reported many “soft” errors. When replacing a module that has not failed, you must do so while the storage system is powered up so that the SP knows the module is being replaced.
Caution: To maintain pr oper cooling in the storage system, never remove a disk module until you are ready to install a replacement. Never remove more than one disk module at a time.
To remove a disk module, follow these steps:
1. Verify that the suspected module has actually failed.
Caution: If you remove the wrong disk module, you introduce an additional fault that shuts down the physical disk containing the failed module. In this situation, the operating system software cannot access the physical disk until you initialize it again.
2. Read “Setting Up the Workplace for Replacing or Installing Disk
Modules,” earlier in this chapter.
3. Locate the disk module that you want to remove; see Figure 5-2 if
necessary.
4. Position the new disk module in its antistatic packaging within reach of
the storage system.
5. If you are using an ESD wrist strap, attach its clip to the ESD bracket at
the bottom of the storage system. Figure 5-3 shows where to attach the clip on a deskside storage system.
65
Chapter 5: Maintaining Disk Modules
ESD bracket
Clip and wire
of ESD band
Figure 5-3 Attaching the ESD Clip to the ESD Bracket on a Deskside Storage
System
Figure 5-4 shows where to attach the clip on a rack storage system.
ESD bracket
Clip and wire of
ESD band
Figure 5-4 Attaching the ESD Clip to the ESD Bracket on a Rack Storage System
66
6. Put the wrist band around your wrist with the metal button against your skin.
7. Make sure the disk has stopped spinning and the heads have unloaded.
ESD wrist band
Replacing a Disk Module
8. Grasp the disk module by its handle and pull it part of the way out of
the cabinet, as shown in Figure 5-5.
Deskside Rack
ESD wrist band
Figure 5-5 Pulling Out a Disk Module
Caution: Never remove more than one disk module at a time.
Warning: When removing a disk module from an upper chassis
assembly in a CHALLENGE RAID rack system, make sure that you adequately balance the weight of the disk module.
9. Supporting the disk module with your free hand, pull it all the way out
of the cabinet, as shown in Figure 5-6.
67
Chapter 5: Maintaining Disk Modules
Deskside Rack
Figure 5-6 Removing a Disk Module
Caution: When removed from the chassis, the disk modules are
extremely sensitive to shock and vibration. Even a slight jar can severely damage them.
68
10. If the label on the side of the disk module does not show the ID number for the compartment from which you removed the drive, write it on the label; for example, A1.
For the compartment ID numbers, refer to Figure 5-2 or the slot matrix attached to the storage system when it was installed.
11. Put the failed disk module in an antistatic bag and store it in a place where it will not be damaged.
Caution: Before installing a replacement module, wait at least 15 seconds after removing the failed module to allow the SP time to recognize that the module has been removed. If you insert the replacement module too soon, the SP may report the replacement module as defective.
Replacing a Disk Module
Installing a Replacement Disk Module
To install the replacement disk module, follow these steps:
1. Touch the new disk module’s antistatic packaging to discharge it and
the drive module. Remove the new disk module from its packaging. Caution: The disk module is extremely sensitive to shock and vibration.
Even a slight jar can severely damage it.
2. On the label on the side of the disk module, write the ID number for the
compartment into which the drive is going; for example, A3.
3. Engage the disk module’s rail in the chassis rail slot, as shown in
Figure 5-7.
Rail slot
Rail slot
Disk module’s rail
Deskside Rack
Figure 5-7 Engaging the Disk Module Rail
Disk module’s
rail
4. Engage the disk module’s guide in the chassis guide slot, as shown in
Figure 5-8.
69
Chapter 5: Maintaining Disk Modules
Deskside Rack
Guide slot
Disk module’s guide
Figure 5-8 Engaging the Disk Module Guide
5. Insert the disk module, as shown in Figure 5-9. Make sure it is completely seated in the slot.
Guide slot
Disk module’s
guide
70
Deskside Rack
ESD wrist band
Figure 5-9 Inserting the Replacement Disk Module
Replacing a Disk Module
6. Remove and store the ESD wrist band, if you are using one.
The SP formats and checks the new module, and then begins to reconstruct the data. While rebuilding occurs, you have uninterrupted access to information on the physical disk unit.
The default rebuild period is 4 hours.
Note: During the rebuild period, performance might degrade slightly,
depending on the rebuild time specified and on I/O bus activity.
For more information on changing the default rebuild period, see “Binding Disks Into RAID Units” in Chapter 4.
Updating the Disk Module Firmware
After replacing a failed unbound disk module (A0, B0, C0, or A3), update the firmware on the CHALLENGE RAID SP. Follow these steps:
1. Quiesce the bus, disabling all applications. Make sure that only the
RAID agent is running.
2. Type as root:
raid5 -d device firmware /usr/raid5/flarecode.bin
Caution: You must use this command every time you replace a failed unbound disk module (A0, B0, C0, or A3).
The image in the file given in the command contains microcode that runs on the SP and also a microcode image destined for the SP PROM, which runs the power-on diagnostics.
71
Chapter 5: Maintaining Disk Modules
Installing an Add-On Disk Module Array
As your organization’s needs change, you may need to add to or change the configurations of your storage system’s physical disk units. For example, you might want to add disk module arrays to unused slots or change the ownership of a physical disk unit from one SP to another.
If the storage system has unused disk module slots containing only filler modules, you can increase the available storage capacity by installing additional disk module arrays. Normal processing can continue while you install disk modules in arrays of five.
This section explains
ordering add-on disk module arrays
inserting the new disk module array
creating device nodes and binding the disks
Ordering Add-On Disk Module Arrays
72
Call the Silicon Graphics, Inc., hotline to order add-on disk module arrays:
1-800-800-4SGI (1-800-800-4744)
Use Table 5-2 as a guide to ordering add-on disk module arrays.
Table 5-2 Ordering Add-On Disk Module Sets
Unit Marketing Code
Add-on five 2.1 GB drives P-S-RAID5-5X2 Add-on five 4.2 BG drives P-S-RAID5-5X4 Base array with five 2 GB drives P-S-RAID5-B5X2 Base array with five 4 GB drives P-S-RAID5-B5X4 Replacement 2.1 GB drive P-S-RAID5-1X2 Replacement 4.2 GB drive P-S-RAID5-1X4
Installing an Add-On Disk Module Array
Caution: Use only CHALLENGE RAID disk modules as replacements; only they contain the correct device firmware. Other disk modules, even those from other Silicon Graphics equipment, will not work. Do not mix disk modules of different capacities within one array. Do not remove disk modules from bus 0 (slots A0, B0, C0, D0, and E0) for use in other disk module positions.
Installing Add-On Disk Modules
Follow these steps:
Caution: Leave the filler modules installed until the add-on replacement modules are available. Never remove more than one disk module or disk filler module at a time.
1. Read “Setting Up the Workplace for Replacing or Installing Disk
Modules,” earlier in this chapter.
2. Position the new disk modules in their antistatic packaging within
reach of the storage system.
3. If you are using a wrist band, attach its clip to the ESD bracket on the
bottom of the storage system, as shown in Figure 5-3. Put the wrist band around your wrist with the metal button against your skin.
4. Locate the slots where you will install the add-on disk modules; see
Figure 5-2.
Warning: Although you need not complete each chassis assembly
that is partially filled before installing more disk modules in the next chassis assembly, avoid making the rack top-heavy.
Fill the slots in this order:
first, in this order: A0, B0, C0, D0, E0
next, in this order: A1, B1, C1, D1, E1
next, in this order: A2, B2, C2, D2, E2
next, in this order: A3, B3, C3, D3, E3
5. Grasp the filler module for the first slot and pull it out of the cabinet; set
it aside for possible future use. If you cannot grasp the module, use a medium-size flat-blade screwdriver to pry it out gently.
73
Chapter 5: Maintaining Disk Modules
6. Touch the new disk module’s antistatic packaging to discharge it and the drive module. Remove the new disk module from its packaging.
7. On the label on the side of the disk module, write the ID number for the compartment into which the drive is going. You can either write the slot position on the label in the corresponding place on the matrix or make a check mark in the position to indicate the slot that the disk module occupies. Figure 5-10 shows these two ways of labeling disk module A0.
For deskside chassis
TOWER
A0
RACK
TOWER
or
RACK
RACK
A0
TOWER
Figure 5-10 Marking the Label for Disk Module A0
For reference, Figure 5-11 diagrams all disk module locations.
For chassis assembly in rack
RACK
or
TOWER
74
Installing an Add-On Disk Module Array
Deskside Chassis assembly in rack
A1 B1 C1 D1 E1 A3 B3 C3 D3 E3
A0 B0 C0 D0 E0 A2 B2 C2 D2 E2
A0 B0 C0 D0
A1 B1 C1 D1 E1
5 to 20 disk modules in groups of 5
E0 A2 B2 C2 D2
A3 B3 C3 D3 E3
Figure 5-11 Disk Drive Locations
8. Engage the disk module’s rail in the chassis rail slot, as shown in
Figure 5-12. Caution: Disk modules are extremely sensitive to shock and vibration.
Even a slight jar can severely damage them.
E2
Rail slot
Disk module’s rail
Deskside Rack
Figure 5-12 Engaging the Disk Module Rail
Rail slot
Disk module’s
rail
75
Chapter 5: Maintaining Disk Modules
9. Engage the disk module’s guide in the chassis guide slot, as shown in Figure 5-13.
Deskside Rack
Guide slot
Disk module’s guide
Figure 5-13 Engaging the Disk Module Guide
10. Insert the disk module, as shown in Figure 5-14. Make sure it is completely seated in the slot.
Guide slot
Disk module’s
guide
Deskside Rack
76
Figure 5-14 Inserting a Disk Module
Installing an Add-On Disk Module Array
11. Repeat steps 4 through 10 until all add-on modules are installed.
12. When you are finished installing add-on modules, remove and store the
ESD wrist band, if you are using one.
Creating Device Nodes and Binding the Disks
If you are adding disk arrays to a storage system that already has at least one LUN configured, the SPs must be made aware of the new disks. This section explains how to accomplish this without rebooting. Also, in a system with two storage-control SPs which are used for primary and secondary paths, both SPs must be made aware of the new disks. Also, the new disks must be bound into LUNs.
Follow these steps:
1. Change to the /dev directory:
cd /dev
2. Type
./MAKE_VLUNS <controller-number> <target-number>
This command creates the device nodes for the new disks.
3. Bind the newly installed modules into one or more physical disk units,
as described in “Binding Disks Into RAID Units,” in Chapter 4 in this guide.
77
Chapter 6
6. Identifying Failed System Components
This chapter describes how to identify a failed system component (field-replaceable unit) other than a disk module: a fan module, power supply, optional battery backup unit, or storage-control processor board. Read this chapter after you have determined that one of these components has failed as indicated by the cabinet fault light, or in the output of the raid5 getcrus command, as explained in “Getting Information About Other Components,” in Chapter 3.
Note: These components can be replaced only by qualified Silicon Graphics
System Service Engineers or other qualified service providers. Only disk modules are owner-replaceable or end user -replaceable; Chapter 5 pr ovides instructions.
Call the Silicon Graphics hotline to order a replacement module:
1-800-800-4SGI (1-800-800-4744)
T able 6-1 lists Silicon Graphics marketing codes for replacement units for the CHALLENGE RAID storage system.
Table 6-1 Field-Replaceable Units
Unit Silicon Graphics
Marketing Code
Power supply module P-S-RAID5-PWR Storage-control processor (SP) P-S-RAID5-SP Battery backup unit P-S-RAID5-BBU Fan module with 6 fans P-S-RAID5-FAN Add-on 8 MB SIMM cache kit P-S-RAID5-C8 Add-on 64 MB SIMM cache kit P-S-RAID5-C64
79
Chapter 6: Identifying Failed System Components
Note: Failure of the AC distribution system (line cord, utility power , and so
on) shuts down the entire chassis immediately.
Power Supply
The CHALLENGE RAID storage system has two or three redundant power supplies, or VSCs (voltage semi-regulated converters): VSC A, VSC B, and, optionally, VSC C. If the storage system has three power supplies, it can recover from power supply component failures and provide uninterrupted service while the defective component is replaced.
In a storage system with two power supplies, a failed power supply unit can fail or be removed without affecting the disk-drive modules. If a second power supply fails, the entire chassis shuts down immediately.
Note: When the storage system shuts down, the operating system loses
contact with the physical disk units. When the storage system starts up automatically, you may need to reboot it to let the operating system access the physical disk units.
Fan Module
80
Each CHALLENGE RAID storage system has one fan module, containing six fans. If any fan fails, the fan fault light on the back of the fan module turns on, and all other fans speed up to compensate. The storage system can run after one fan fails; however, if another fan failure occurs and temperature rises, the storage system shuts down after two minutes.
If the fault light comes on, or if you see a fan fault in the raid5 getcrus command output, have the entire fan unit replaced as soon as possible by a Silicon Graphics System Service Engineer.
If the temperature within a CHALLENGE RAID chassis reaches an unsafe level, the power system shuts down the chassis immediately.
Battery Backup Unit
Battery Backup Unit
When the optional battery backup unit fails, the following events occur:
The battery backup unit’s service light turns on.
If the storage system is using the cache, the storage system’s
performance may become sluggish while it writes any modified pages from the cache to disk.
Caching is disabled; the cache state as shown in the raid5 getcache
command is “Disabled.” Caching is not re-enabled until the battery backup unit is replaced with a fully charged one.
The raid5 getcrus output has thr ee possible states for the battery backup unit: Faulted (removed), Charging, and Present (fully char ged or charging). If the battery backup unit takes longer than an hour to charge, it shuts itself off and transitions to the “Faulted” state.
If the fault light comes on or if the battery backup unit state is shown as “Faulted” in theraid5 getcrus command output, have the battery backup unit replaced as soon as possible by a Silicon Graphics System Service Engineer.
Storage-Control Processor
This section discusses
diagnosing SP failure
using the auto-reassign capability
Diagnosing SP Failure
When an SP fails, the storage system’s performance may become sluggish if the storage system is using caching. You may find that one or more physical disk units are inaccessible. The failed SP’s service light turns on, and the service light on the front of the storage system turns on.
81
Chapter 6: Identifying Failed System Components
For SP failure, the unsolicited event log contains the message “Peer Controller: Removed.” You can also use raid5 getcrus to verify the failed SP, as explained in Chapter 3. The getcrus output indicates the failed SP, for example, “SPA State: Not Present.”
Using the Auto-Reassign Capability
If one of two SPs becomes physically unavailable to the system, that is, if an SP fails or becomes disconnected, the CHALLENGE RAID storage system can automatically reassign the LUNs it controls to the remaining SP.
Note the following points:
Auto-reassign capability cannot be used in the case of problems caused by a defective CHALLENGE-to-CHALLENGE RAID connection, such as a damaged cable.
To enable auto-reassign, power off the failed SP. You do not need to power down the CHALLENGE RAID storage system.
Caution: Do not power off the SP under any circumstances other than to enable auto-reassign.
82
Powering off an SP requires opening the fan module at the back of the storage system. Because the fan module must not be left in the open position for more than two minutes, make sure you can complete this process within that time limit.
Follow these steps:
Caution: To pr event thermal shutdown of the storage system, never operate the storage system for more than two minutes with the fan module in the open position. Do not use this procedure under any circumstances other than to enable auto-reassign.
Storage-Control Processor
1. On the back of the CHALLENGE RAID storage system, move the fan
module’s latch to the UNLOCK position, as shown in Figure 6-1.
Deskside Rack
Figure 6-1 Unlocking the Fan Module
2. Swing open the fan module, as shown in Figure 6-2.
Deskside Rack
Figure 6-2 Opening the Fan Module
83
Chapter 6: Identifying Failed System Components
3. Disable the failed SP’s power by sliding the switch to the Disable position, as shown in Figure 6-3.
Deskside Rack
SP B
SP A
SP A SP B
84
Figure 6-3 Disabling an SP’s Power
Caution: Do not power off the SP under any circumstances other than
to enable auto-reassign.
4. Immediately close and lock the fan module to let the SP cool.
5. Have the SP replaced as soon as possible by qualified service personnel. Leave the SP powered off in the meantime.
Chapter 7
7. Caching
Storage-system caching requires
two storage-control processors, each with the same amount of cache
memory
fully charged storage system backup battery
minimum of 8 MB of memory for each SP
cache enabling using the raid5 setcache command, as explained in this
chapter
disk modules in slots A0, B0, C0, D0, and E0 as a fast repository for
cached data
Caching cannot occur unless all these conditions are met.
This chapter explains
setting cache parameters
viewing cache statistics
upgrading CHALLENGE RAID to support caching
changing cache unit parameters
85
Chapter 7: Caching
Setting Cache Parameters
The cache parameters you specify for the entire storage system are the cache size of 8 or 64 MB, depending on the amount of memory the SPs have, and the cache page size, as 2, 4, 8, or 16 KB.
To set up caching, use the raid5 setcache command:
raid5 -d device setcache enable | disable [-u usable] [-p page] [-l low] [-h high]
In this syntax, variables mean: enable | disable Enter 1 to enable caching or 0 to disable caching.
-u usable Size in megabytes to use for caching, not greater than the SP
memory size as displayed by raid5 getagent (see “Getting Device Names With getagent,” in Chapter 3). Valid values are 0, 8, and 64:
0: user has selected to disable the cache
other values: user has selected to enable the cache The command line interface does not let you specify more
memory than you have. If you specify less than you have, the remaining memory is unused.
86
Note: For caching, both SPs must have the same amount of
cache memory in order for caching to be preserved in the event of shutdown or other power loss.
-p page Size in KB of pages into which to partition the cache. Valid
sizes are 2, 4, 8, and 16. The default is 2, regardless of whether caching is enabled or disabled.
-l low Percentage of cache full that discontinues flushing. Valid
values are 0 through 100; the default is 50, regardless of whether caching is enabled or disabled.
-h high Percentage of cache full that initiates flushing. Valid values
are 0 through 100; the setting must be greater than the low watermark. The default is 75, regardless of whether caching is enabled or disabled.
This command has no output.
Loading...