Digital PDFs

AA-Q6TGD-TE

April 1995

83 pages

Original

0.3MB

Document:	StorageWorks RAID Array 200 Online Management Utility for Digital UNIX User's Guide
Order Number:	AA-Q6TGD-TE
Revision:	0
Pages:	83
Original Filename:

OCR Text

StorageWorks RAID Array 200
Online Management Utility
for Digital UNIX
User's Guide
AA–Q6TGD–TE

Digital Equipment Corporation
Maynard, Massachusetts

Fourth Printing, April 1995
The information in this document is subject to change without notice and should
not be construed as a commitment by Digital Equipment Corporation.
Digital Equipment Corporation assumes no responsibility for any errors that
might appear in this document.
The software, if any, described in this document is furnished under a license and
may be used or copied only in accordance with the terms of such license. No
responsibility is assumed for the use or reliability of software or equipment that
is not supplied by Digital Equipment Corporation or its affiliated companies.
Copyright  1995 by Digital Equipment Corporation. All rights reserved.
Printed in the U.S.A.
The following are trademarks of Digital Equipment Corporation:
AXP, Digital UNIX, StorageWorks, SWXCR, and the Digital logo.
The following are third-party trademarks:
Microsoft and MS-DOS are registered trademarks and Windows is a trademark
of Microsoft Corporation.
Intel, Intel486, and i486 are trademarks of Intel Corporation.
All other trademarks and registered trademarks are the property of their
respective holders.

Contents
Revision Record..................................................................................................................vii
About This Guide ...............................................................................................................ix
1

Installing the RAID Array 200 Online Management Utility
for Digital UNIX
1.1

Introduction.............................................................................................................1-1

1.2

Before Installing the Utility .....................................................................................1-2

1.3

Files Added to Your System During Installation ......................................................1-2

1.4 Loading the Utility ..................................................................................................1-4
1.4.1 Verifying the Installation...................................................................................1-7
1.4.2 Post Installation.................................................................................................1-8
1.5 Accessing the Array.................................................................................................1-9
1.5.1 Creating Device Special Files ............................................................................1-9
1.5.1.1 Determining Controller Numbers.......................................................................1-9
1.5.1.2 Determining Device Special File Names..........................................................1-10
1.5.1.3 Invoking MAKEDEV ......................................................................................1-11
1.5.2 Reconfiguring the Kernel.................................................................................1-12
1.5.3 Preparing Logical RAID Drives for Access by Digital UNIX Filesystem .........1-14
1.5.3.1 Creating the Partitions on a Logical RAID Drive Using Disklabel ...................1-14
1.5.3.2 Creating a Filesystem on a Logical RAID Drive ..............................................1-16
1.5.3.3 Mounting the Filesystem .................................................................................1-16

Using the RAID Array 200 Online Management Utility
for Digital UNIX Command Line
2.1

Monitoring the RAID Subsystem.............................................................................2-1

2.2 Command Flags for swxcrmon ...............................................................................2-3
2.2.1 Notification of Events to the System Console ....................................................2-3
2.2.2 Notification of Events to Log Files ....................................................................2-3
2.2.3 Notification of Events to Mail ...........................................................................2-4
2.2.4 Frequency Status Update ...................................................................................2-5
2.2.5 Events Reported by the Monitor ........................................................................2-6
2.2.6 Repeating Notification Events ...........................................................................2-7
2.3

Performing Parity Check and Repair on the RAID Subsystem .................................2-7

AA–Q6TGD–TE

iii

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

Using the RAID Array 200 Online Management Utility
for Digital UNIX Command Line (cont.)
2.4 Command Qualifiers for swxcrcheck .......................................................................2-8
2.4.1 Repairing the Logical RAID Drive Parity ..........................................................2-8
2.4.2 Notification of Bad Blocks ................................................................................2-8
2.5 Altering Rebuild Rate on the RAID Subsystem...................................................... 2-10
2.5.1 Viewing the Rebuild Rate ................................................................................2-10
2.5.2 Setting the Rebuild Rate ..................................................................................2-11

Command Line Messages
3.1

Monitor Process Messages.......................................................................................3-1

3.2

Parity Check Messages ............................................................................................3-8

Using the RAID Array 200 Online Management Utility
for Digital UNIX Graphical User Interface (GUI)
4.1

Overview of the RAID Array 200 Online Management Utility for Digital UNIX
Graphical User Interface (GUI)................................................................................4-1

4.2

Accessing the Online Management Utility for Digital UNIX GUI............................4-2

4.3 Using the Online Management GUI .........................................................................4-4
4.3.1 Using the Mouse................................................................................................4-4
4.3.2 Using Windows .................................................................................................4-5
4.3.3 Using Menus .....................................................................................................4-6
4.3.4 Using Help ........................................................................................................4-6
4.3.5 Exiting from the Online Management GUI ........................................................4-6
4.4 Viewing the Subsystem from the Online Management Utility..................................4-7
4.4.1 Selecting a Subsystem .......................................................................................4-7
4.4.2 Viewing Disk Drives .......................................................................................4-10
4.4.3 Viewing Logical RAID Drives.........................................................................4-14
4.5

Printing the Configuration .....................................................................................4-17

Maintaining the Subsystem with the GUI
5.1

Maintaining the Subsystem......................................................................................5-1

5.2 Replacing a Failed Drive .........................................................................................5-2
5.2.1 Hot Spare ..........................................................................................................5-3
5.2.2 Hot Swap...........................................................................................................5-3
5.2.3 Marking a Drive as Failed .................................................................................5-4
5.2.3.1 Rebuilding a Drive ............................................................................................5-6
5.2.3.2 Marking a Drive as Optimal ..............................................................................5-8
iv

AA–Q6TGD–TE

Contents

Maintaining the Subsystem with the GUI (cont.)
5.3

Making a Hot Spare.................................................................................................5-9

5.4

Setting Rebuild Rate..............................................................................................5-10

5.5

Viewing the Bad Block Table ................................................................................5-11

5.6

Checking Parity .....................................................................................................5-12

Index...............................................................................................................................Index-1
Readers Comment Form ........................................................................ Inside Back Cover
Figures
4-1

The Base Window of the Online Management Utility for Digital UNIX GUI,
Displaying a 1-Channel Configuration .....................................................................4-2
4-2 The Base Window of the Online Management Utility for Digital UNIX GUI,
Displaying a 3-Channel Configuration .....................................................................4-3
4-3 The Adapter Selected Buttons..................................................................................4-7
4-4 The Subsystem Displayed after Selecting EISA Slot 8 .............................................4-8
4-5 The Subsystem Displayed after Selecting PCI Slot 6 ...............................................4-9
4-6 The Disk Drive Matrix for a 1-Channel Subsystem................................................4-10
4-7 The Disk Drive Matrix for a 3-Channel Subsystem................................................4-12
4-8 The Logical Drive Table........................................................................................4-14
4-9 Printing the Configuration .....................................................................................4-17
5-1 Fail Drive Confirmation Window ............................................................................5-5
5-2 The Rebuild Drive Confirmation Window ...............................................................5-6
5-3 The Rebuild Drive Confirmation Window ...............................................................5-7
5-4 The Rebuild Completed Message.............................................................................5-7
5-5 The Make Optimal Confirmation Window...............................................................5-9
5-6 Set Rebuild Rate Window......................................................................................5-10
5-7 The Parity Repair Option Window.........................................................................5-12
5-8 The Parity Progress Window .................................................................................5-13
5-9 Parity Check Completes Message ..........................................................................5-13
5-10 Parity Check Status Report Appears When Parity Check Completes ......................5-14
5-11 Parity Errors Detected Message .............................................................................5-14

AA–Q6TGD–TE

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

Tables
1-1
1-2
2-1
2-2
3-1
3-2
4-1
4-2
4-3
4-4
4-5
4-6

Command Line Files Added During Installation ......................................................1-3
GUI Files Added During Installation .......................................................................1-3
Summary of swxcrmon Flags...................................................................................2-2
Summary of swxcrcheck Flags ................................................................................2-7
Monitor Process Messages.......................................................................................3-1
Parity Check Messages ............................................................................................3-8
Mouse Functions .....................................................................................................4-4
Window Elements ...................................................................................................4-5
Drive Matrix Information ......................................................................................4-11
Disk Drive Status...................................................................................................4-13
Logical Drive Table Information ...........................................................................4-15
Logical Drive Status ..............................................................................................4-16

AA–Q6TGD–TE

Revision Record
This Revision Record provides a concise publication history of this manual. It lists the
manual revision levels, release dates, and reasons for the revisions.

The following revision history lists all revisions of this publication and their effective
dates. The publication part number is included in the Revision Level column, with the
last entry denoting the latest revision. This publication supports StorageWorks RAID
Array 200 Subsystems in the Digital UNIX operating system environment.
Revision Level

Date

Summary of Changes

AA–Q6TGA–TE

August 1994

Original release. This release
supports the StorageWorks RAID
Array 200 Online Management
Utility for Digital UNIX command
line only.

AA–Q6TGC–TE

January 1995

This release supports the
StorageWorks RAID Array 200
Online Management Utility for
Digital UNIX command line and
Graphical User Interface (GUI).

AA–Q6TGD–TE

April 1995

This release reflects the change of
DEC OSF/1 to Digital UNIX.

AA–Q6TGB–TE

AA–Q6TGC–TE

vii

About This Guide
This section identifies the audience of this guide and describes the contents
(chapter by chapter) and structure. In addition, this section includes a list of
associated documents and the conventions used in this guide.

The StorageWorks RAID Array 200 Online Management Utility for
Digital UNIX User's Guide
This guide describes how to install the StorageWorks RAID Array 200 Online
Management Utility for Digital UNIX and how to use the RAID Array 200
Online Management Utility for Digital UNIX command line and Graphical User
Interface (GUI).

Intended Audience
This guide supports users of the StorageWorks RAID Array 200 Subsystems in a
Digital UNIX environment. You should be familiar with StorageWorks RAID
terminology, Digital UNIX and the procedures for adding a device to your
system.

Organization
This guide contains the following:
Chapter 1: Installing the RAID Array 200 Online Management Utility for
Digital UNIX
Installing the RAID Array 200 Online Management Utility for Digital UNIX lists
the files added to your system by this installation, and discusses running the setld
utility to install the command line and graphical user interface; the procedures
for adding a device to your system; and preparing logical RAID drives for access
by the Digital UNIX filesystem.

AA–Q6TGD–TE

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

Chapter 2: Using the RAID Array 200 Online Management Utility for
Digital UNIX Command Line
Using the RAID Array 200 Online Management Utility for Digital UNIX
Command Line discusses monitoring the RAID subsystem through the SWXCR
monitor, methods of event notification; and using the command line to run parity
check and repair, and to set the rebuild rate.
Chapter 3: Command Line Messages
Command Line Messages describes the informational, and error messages
displayed by the command line monitor and parity check processes.
Chapter 4: Using the RAID Array 200 Online Management Utility for
Digital UNIX Graphical User Interface (GUI)
Using the RAID Array 200 Online Management Utility for Digital UNIX
Graphical User Interface (GUI) describes how to access and use the GUI.
Chapter 5: Maintaining the Subsystem with the GUI
Maintaining the Subsystem with the GUI discusses how to use the GUI to:
replace a failed drive, including: failing, rebuilding and marking a drive optimal;
create a hot spare, set rebuild rate, view bad block table, and check parity.

Conventions
This guide uses the following conventions:
Style

Meaning

boldface monospace type
plain monospace type
italic type
screens,

To be input by the user
Screen text
For emphasis, manual titles, utilities, menus,
and filenames

AA–Q6TGD–TE

About this Guide

Special Notices
This guide uses in the following to emphasize specific information.
WARNING
WARNING indicates the presence of a hazard that
can cause personal injury if the hazard is not
avoided.

CAUTION
CAUTION indicates the presence of a hazard that
might damage hardware or corrupt software.

NOTE
Notes provide additional information.

Related Documentation
Refer to the StorageWorks RAID Array 200 Subsystems Controller Installation
and Standalone Configuration Utility User's Guide for information on the
controller and on configuring your subsystem.

AA–Q6TGD–TE

1
Installing the Storage Works RAID Array 200
Online Management Utility
for Digital UNIX
After you install and configure the RAID Subsystem, you can install the StorageWorks
RAID Array 200 Online Management Utility for Digital UNIX. This software consists of the
SWXCR Monitor utility, SWXCR Parity Check utility, SWXCR Rebuild Rate utility, and
SWXCR Graphical User Interface (GUI).

1.1

Introduction
The RAID Array 200 Online Management Utility for Digital UNIX consists of a
command line and a Graphical User Interface (GUI).
The command line provides:
•

Monitoring

•

Parity check and repair

•

Rebuild control (of redundant logical RAID drives)

•

Viewing help

The GUI utility provides:
•

Management of the RAID subsystem

•

Rebuild control (of redundant logical RAID drives)

•

Parity Check

AA–Q6TGD–TE

1–1

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

1.2

Before Installing the Utility
Prior to installing the RAID Array 200 Online Management Utility for Digital
UNIX on your subsystem, you should have completed the following:
•

Run the EISA Configuration Utility (for EISA controllers only)

•

Set controller options and parameters

•

Installed the controller and subsystem

•

Connected the subsystem to the controller

•

Configured logical RAID drives on the subsystem with the standalone
configuration utility

In addition, we recommend that you back up your system before you attempt to
install the utility for Digital UNIX.

1.3

Files Added to Your System During Installation
The installation of the online management command line and Graphical User
Interface (GUI) utilities for Digital UNIX create the files listed in Tables 1-1 and
1–2 on your system.

1–2

AA–Q6TGD–TE

Chapter 1. Installing
Table 1-1 Command Line Files Added During Installation
File

Description

/usr/opt/swxcr/bin/swxcrmon

Monitor utility executable

/usr/opt/swxcr/man/man8/swxcrmon.8

swxcrmon manpage

/usr/opt/swxcr/.X11Startup

Sample shell script

/usr/opt/swxcr/S99swxcr_monitor

Sample rc3 script shell

/usr/bin/swxcrmon

Symbolic link

/usr/share/man/man8/swxcrmon.8

Symbolic link

/usr/opt/swxcr/bin/swxcrcheck

Parity check utility executable

/usr/opt/swxcr/bin/swxcrrate

Rebuild rate utility executable

/usr/share/man/man8/swxcrcheck.8

swxcrcheck manpage

/usr/share/man/man8/swxcrrate.8

swxcrrate manpage

/usr/bin/swxcrcheck

Symbolic link

/usr/bin/swxcrrate

Symbolic link

/usr/share/man/man8/swxcrcheck.8

Symbolic link

/usr/share/man/man8/swxcrrate.8

Symbolic link

Table 1-2 GUI Files Added During Installation
/usr/opt/swxcrmgr/bin/swxcrmgr

Management utility
executable

/usr/bin/swxcrmgr

Symbolic link

/usr/opt/swxcrmgr/lib/swxcrmgr

GUI Resource file

/usr/lib/x11/app-defaults/swxcrmgr

Symbolic link

/usr/opt/swxcrmgr/man/man8/swxcr.8

swxcr manpage

/usr/opt/swxcrmgr/man/man8/swxcrmgr.8

swxcrmgr manpage

AA–Q6TGD–TE

1–3

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

1.4

Loading the Utility
To load the online management utility for Digital UNIX command line and GUI,
you need to run the setld utility.
The procedure for loading the online management utility for Digital UNIX takes
about 5 minutes to complete.
To load and configure the online management utility for Digital UNIX, follow
these steps:
1.

Load the distribution diskette into your floppy drive.

Mount the device. Type this command:
# mount -o ro /dev/fd0c mount_point
where mount_point is a valid mount point on your system, for example, /mnt.
(Make sure that /mnt is not in use.)

From the Bourne shell root account, type this command to run setld and
install the utilities:
# setld -l /mount_point

The installation procedure requires you specify and confirm whether you want to
install the command line, the GUI or both. The installation then does the
following:

1–4

•

Checks for the necessary file system space.

•

Displays copyright information.

•

Displays the approximate installation time (5 minutes) and disk space
requirements.

•

Installs the StorageWorks RAID Array 200 Online Management Utility for
Digital UNIX V1.0.0.

•

Creates symbolic links.

AA–Q6TGD–TE

Chapter 1. Installing

The following is an example of an installation.
The subsets listed below are optional:
There may be more optional subsets than can be presented
on a single screen. If this is the case, you can choose
subsets screen by screen or all at once on the last
screen. All of the choices you make will be collected for
your confirmation before any subsets are installed.
1) StorageWorks RAID Array 200 Management Utility for DEC
OSF/1 V1.0.0
2) StorageWorks SWXCR Utility for Digital UNIX V1.2.0
Or you may choose one of the following options:
3) ALL of the above
4) CANCEL selections and redisplay menus
5) EXIT without installing any subsets
Enter your choices or press RETURN to redisplay menus.
Choices (for example, 1 2 4-6): 3
You are installing the following optional subsets:
StorageWorks RAID Array 200 Online Management Utility
for Digital UNIX V1.0.0
StorageWorks SWXCR Utility for Digital UNIX V1.2.0
Is this correct? (y/n):y
Checking file system space required to install selected
subsets:
File system space checked OK.
************************************************************
*
© Digital Equipment Corporation 1994.

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Press return to continue...
The installation of the StorageWorks RAID Array 200
Management Utility for Digital UNIX V1.0.0 will take about 5
minutes and requires approximately 850 Kbytes of disk space
both during and after installation.
StorageWorks RAID Array 200 Management Utility for Digital
UNIX V1.0.0
Copying from . (disk)
Verifying
************************************************************
*
© Digital Equipment Corporation 1994.

Restricted Rights: Use, duplication, or disclosure by the
U.S. Government is subject to restrictions as set forth in
subparagraph (c) (1) (ii) of DFARS 252.227-7013, or in FAR
52.227-19, or in FAR 52.227-14 Alt. III, as applicable.
This software is proprietary to and embodies the
confidential technology of Digital Equipment Corporation.
Possession, use or copying of this software and media is
authorized only pursuant to a valid written license from
Digital or an authorized sublicensor.
************************************************************
*
Press return to continue...
The installation of the StorageWorks SWXCR Utility for
Digital UNIX V1.2.0 will take about 5 minutes and requires
approximately 450 Kbytes of disk space both during and after
installation.
StorageWorks SWXCR Utility for Digital UNIX V1.2.0
Copying from . (disk)
Verifying
Configuring "StorageWorks RAID Array 200 Management
Utility for Digital UNIX V1.0.0" (SWXCRMGR100)
Now creating links from /usr/bin files to kit executable
files.

1–6

AA–Q6TGD–TE

Chapter 1. Installing
Now creating links from /usr/lib/X11/app-defaults files to
kit resource files.
Now creating links from /usr/share/man/man8 files to kit
man pages.
The StorageWorks RAID Array 200 Management Utility for
Digital UNIX Installation is now complete.
Configuring "StorageWorks SWXCR Utility for Digital UNIX
V1.2.0" (SWXCR120)
Now creating links from /usr/bin files to kit executable
files.
Now creating links from /usr/share/man/man8 files to kit
man pages.
**********************************************************
**
The SWXCR Utility will create log files in /usr/opt/swxcr,
unless the SWXCR_LOGPATH environmental variable is
defined. If you wish to use an alternate directory for
log files, you must create that directory and set
SWXCR_LOGPATH to the desired path.
**********************************************************
**
The StorageWorks SWXCR Utility for Digital UNIX Installation
is now complete.

1.4.1 Verifying the Installation
You can verify that all files were installed in the correct directories by typing the
following commands:
# setld -v SWXCR120
# setld -v SWXCRMGR100
This executes the installation verification procedure (IVP).

AA–Q6TGD–TE

1–7

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

1.4.2 Post Installation
After the installation completes, you have the option to create a log file directory
in Korn/Bourne shells as follows:
In Korn/Bourne shells enter: SWXCR_LOGPATH=log_file.directory
export SWXCR_LOGPATH

Then, you can issue env command to examine defined path variable
In csh enter: setenv SWXCR_LOGPATH log_file.directory
For example, setenv SWXCR_LOGPATH /logfiles
Then, if you define the SWXCR_LOGPATH environment variable as the path to
the log file directory, the SWXCR utility writes any log files to that specified
directory. If you do not define the SWXCR_LOGPATH environment variable,
any log files are written to /usr/opt/swxcr.
To start the SWXCR Monitor automatically when you start up the system, place
an rc3 command script containing the swxcrmon command in the /sbin/rc3.d
directory. This command script then causes the SWXCR Monitor utility to be
invoked when entering a multiuser run level. Refer to the rc3 manpages for more
information about the multiuser run level. There is a sample command script
provided in /usr/opt/swxcr/S99swxcr_monitor. You can copy this file directly to
/sbin/rc3.d/S99swxcr_monitor. However, the command flags in this script and
the xcr controller name should be modified as desired. See Chapter 2 for more
information about monitor commands. See Section 1.5, Accessing the Array to
determine the xcr controller name.
If you are running the system in a Windows environment and want to start up the
SWXCR monitor automatically, you should place a .X11Startup command script
in the home directory of the system manager account. For example, if the system
manager uses the root account the command script will have to be placed in
/.X11Startup. This command script will only be run if the system manager logs
into the system using the Windows interface. A sample script is provided by the
installation in /usr/opt/swxcr/.X11Startup and should be modified as desired.

1–8

AA–Q6TGD–TE

Chapter 1. Installing

1.5

Accessing the Array
When you first installed the subsystem, you used the standalone utility to create
logical RAID drives. In order to access these drives through the Digital UNIX
operating system, you need to do the following as described in the following
Section 1.5.1 through 1.5.3:
•

Create device special files

•

Reconfigure the kernel

•

Prepare logical RAID drives for access by Digital UNIX filesystem

1.5.1 Creating Device Special Files
Create block and character device special files in /dev for each of the logical
drives that you defined.
You need the following information to create the device special files:
•

Controller numbers

•

Device special file names

1.5.1.1 Determining Controller Numbers
The operating system assigns controller numbers at boot after scanning the slots
in the system for the SWXCR controller. The operating system scans in lowest to
highest slot number order, first the EISA slots, then the PCI slots. The SWXCR
controllers appear as xcrn devices, where n is the controller number that the
operating system assigns. Digital UNIX assigns xcr0 to the first controller it
encounters, xcr1 to the second, and so on.
For example, suppose you have 3 SWXCR controllers in a system, an EISA
controller in slot 1, an EISA controller in slot 7, and a PCI controller in slot 4.
The operating system will assign xcr0 to the EISA controller in slot 1, xcr1 to the
EISA controller in slot 7, and xcr2 to the PCI controller in slot 4.
View the console log in /usr/var/adm/messages to see the controller numbers that
the operating system assigned to the controllers in your system at boot time.
Record the controller numbers on a piece of paper in order to have them
available to determine the device special file names.

AA–Q6TGD–TE

1–9

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

1.5.1.2 Determining Device Special File Names
The names of the block and character device special files, which appear in /dev
as: /dev/rex (block devices) and /dev/rrex (character devices), include the device
number, x, which you need to calculate.
Calculate the device number using the following equation:
controller number * 8 + logical RAID drive number
(8 is a constant, representing the total number of logical RAID drives it is
possible to have on a controller)
Example 1: Suppose you have a single EISA controller in your system, and you
created a single logical RAID drive on it using the standalone utility. Digital
UNIX will assign the name xcr0 to that controller on boot. The logical RAID
drive number will be 0, because you created only one logical RAID drive
(logical RAID drives are numbered from 0-7 in ascending order for each
controller).
Therefore:
controller number you will use for the calculation in this case is 0
logical RAID drive number will be 0.
Calculate the device number as:
controller number * 8 + logical RAID drive number, or 0 * 8 + 0
Then:
device number = 0.
block device special file = re0.
character device special file = rre0.

1–10

AA–Q6TGD–TE

Chapter 1. Installing

Example 2: Assume you have a PCI controller in your system in addition to the
EISA controller configuration described in Example 1. The EISA controller is
xcr0, and the PCI controller is xcr1. You created 2 logical RAID drives on your
PCI controller, logical RAID drive 0 and logical RAID drive 1. You created 2
logical RAID drives, so you will need to calculate two device numbers for the
PCI controller. Therefore:
controller number you will use for the calculation in this case is 1
logical RAID drive numbers will be 0 and 1
Calculate the device numbers as:
controller number * 8 + logical RAID drive number, or 1 * 8 + 0 and 1 * 8 + 1
Then:
device numbers = 8 and 9.
device special file names = re8 and re9.
character device special file names = rre8 and rre9.
Now, calculate the device numbers and determine the device and character block
special file names for every logical drive on your subsystem. Then go to section
1.5.1.3 to create the device special files.
1.5.1.3 Invoking MAKEDEV
After you determine the device special file names, to create the device special
files, follow these steps:
1. Invoke the MAKEDEV utility in /dev using the name of the device special
file you want to create. For example, type: /dev/MAKEDEV re0
2. Repeat Step 1 to create the device special file for each of the device special
file names you determined.
Go to Section 1.5.2 to reconfigure the kernel.

AA–Q6TGD–TE

1–11

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

1.5.2 Reconfiguring the Kernel
Reconfigure the kernel so that your system will recognize and display the logical
RAID drives on the console during boot. The device special files must be created
using the steps detailed in Sections 1.5.1.1 to 1.5.1.3 prior to reconfiguring the
kernel.
To add device entries to the configuration file in /sys/conf, follow these steps:
1. Change to /sys/conf, type: # cd /sys/conf
2. Save the original configuration file (the configuration file is usually the
system name in capital letters. In this example, the configuration file is
/sys/conf/NODENM), type: # cp NODENM NODENM.old
3. Open the configuration file with an editor and check for any entries for the
SWXCR controllers in your system appearing as follows:
The EISA RAID Controller will appear in the configuration file similar to:
controller xcr0 at eisa0 slot 4 vectr xcrintr
The PCI RAID controller will appear in the configuration file similar to:
controller xcr1 at pci0 slot 4 vectr xcrintr

NOTE
If you configured your controller with the
genvmunx kernel and no entry for it appears in
your custom kernel configuration file, then you will
need to add the entry in the configuration file using
the format for either the EISA or PCI RAID
controller shown above. View the console log file
in /usr/var/adm/messages to see how the
controllers were assigned at boot time.

1–12

Add all the device special files that you created in /dev to the configuration
file as follows:
Type: device disk re0 at xcr0 drive 0
where re0 is the block device special file name, xcr0 is the controller, and
drive 0 is the device special file number.
For example, suppose we have an EISA controller in slot 7 and a PCI
controller in slot 3. We have defined 2 logical RAID drives on each
controller, logical RAID drives 0 and 1 on each.
The entries in the configuration file would look as follows:
AA–Q6TGD–TE

Chapter 1. Installing

controller xcr0 at eisa0 slot 7 vectr xcrintr
device disk re0 at xcr0 drive 0
device disk re1 at xcr0 drive 1
controller xcr1 at pci0 slot 3 vectr xcrintr
device disk re8 at xcr1 drive 8
device disk re9 at xcr1 drive 9
5. Once you type all the device special file entries in the configuration file,
type:
# doconfig -c NODENM
where NODENM is the configuration file name
The system will copy the current configuration file to a backup file, and ask
you if you want to edit the configuration file.
6. Select No to edit the configuration file. Digital UNIX creates a new kernel
/sys/NODENM/vmunix
7. Save the old version of vmunix in the root directory (/) and copy this new
vmunix kernel into vmunix, type:
# cp vmunix vmunix.old
# mv /sys/NODENM/vmunix /vmunix
The next time the system is rebooted, it will display the configured logical
RAID drives on the console.
You have completed reconfiguring your kernel.

AA–Q6TGD–TE

1–13

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

1.5.3 Preparing Logical RAID Drives for Access by Digital UNIX
Filesystem
Digital UNIX treats a logical RAID drive much like a SCSI device; therefore, to
prepare your logical RAID drive for access by the Digital UNIX filesystem you
will need to do the following:
•

Create the partitions on the logical RAID drive using disklabel

•

Create a filesystem on the logical RAID drive

•

Mount the filesystem to copy files to it

1.5.3.1 Creating the Partitions on a Logical RAID Drive Using Disklabel
You create the partitions on a logical RAID drive by issuing a disklabel
command. The disklabel partitions the logical RAID drive for access by the
Digital UNIX filesystem. Digital UNIX defines only partitions a, b, c, and g for
the SWXCR array controller. In addition, the size of the logical RAID drive must
be at least 300 MB to support the a and b partition. If you try to label a logical
RAID drive with less than 300 MB you will get partition errors. Refer to the re
and disktab man pages for more information about disklabel.
To create the read/write partitions on a logical RAID drive using the standard
partition default sizes, enter the following:

1–14

AA–Q6TGD–TE

Chapter 1. Installing

# disklabel -rw

devicepartition SWXCR
device is the character
device special filename as
defined in Section 1.5.1.2
partition is the letter
representing the
partition, a, b, c, or
g on which you want
to create a
filesystem (usually c
for the whole drive).
SWXCR defines this logical
RAID drive as attached to
the SWXCR array
controller. Always use the
device label SWXCR,
regardless of what drive
types make up the logical
RAID drive.

For example, to create partitions on character device, rre0, enter:
# disklabel -rw rre0 SWXCR
To view a logical RAID drive partition, enter: # disklabel -r device

AA–Q6TGD–TE

1–15

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

1.5.3.2 Creating a Filesystem on a Logical RAID Drive
Use the newfs command to create a filesystem on a logical RAID drive the same
way that you would create a filesystem on a disk device by entering the
following:
# newfs

devicepartition
device is the block device name as defined in
Section 1.5.1.2
partition is the letter representing the
partition, a, b, c, or g on which you want
to create a filesystem (usually c for the
whole drive)

For example, to create a filesystem on partition c of block device,
re0c, enter the following:
# newfs re0c
This creates a filesystem on the c partition of logical RAID drive 0 of
the xcr0 controller.

1.5.3.3 Mounting the Filesystem
To access the logical RAID drive, mount it as a device filesystem to a mount
point. For example:
# mount /dev/re0c /mnt
To view the mounted filesystem, enter: df
The logical RAID drive is now accessible to the filesystem much as a disk device
would be. The filesystem can not see the RAID functionality and number of
physical devices attached to the SWXCR array controller. This device appears as
a single logical RAID drive or "disk" to the user as viewed by the filesystem.
To copy files to this device, use the cp command as follows:
# cp filename /mnt
filename is the name of the file to be copied
/mnt is the mount point for the filesystem
1–16

AA–Q6TGD–TE

Chapter 1. Installing

This completes the installation of the StorageWorks RAID Array 200 Online
Management Utility for Digital UNIX. Go on to Chapter 2 to start the monitor, and then
to Chapter 4, if you want to start the graphical user interface.

AA–Q6TGD–TE

1–17

2
Using the RAID Array 200 Online Management
Utility for Digital UNIX
Command Line
The RAID Array 200 Online Management Utility for Digital UNIX provides the SWXCR
Monitor online utility, to notify you of subsystem status changes, the SWXCR parity
check utility, to perform parity check and repair functions on RAID logical drives, and the
SWXCR Rebuild Rate utility, to modify the rebuild and parity check rate for a given
controller.

2.1

Monitoring the RAID Subsystem
The monitor function scans the array for nonoptimal device status and reports the
status to you. You must invoke a separate monitor process for each SWXCR
RAID controller you have on your subsystem. If you do not start a monitor
process, you are not notified of events that occur on your subsystem.
To invoke a monitor process, ensure that your $PATH environment variable path
includes /usr/bin. Then type the swxcrmon command as follows:
# swxcrmon ctlrname
where ctlrname is the name of the xcr controller. Otherwise, you can type the
command as follows:
# /usr/bin/swxcrmon ctlrname
For example, if you specify xcr0 as the controller name, the monitor process
monitors all logical RAID drives configured on the xcr0 controller.
Refer to Post Installation Section 1.4.2 of Chapter 1 for more information about
starting the SWXCR Monitor utility automatically.
The swxcrmon command creates a separate background process.

AA–Q6TGD–TE

2–1

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

Table 2–1 is a summary of the swxcrmon command flags. Refer to the following
sections for a detailed description of the flags.

Table 2–1 Summary of swxcrmon Flags
Action

Flag

Send notification of subsystem events to the
system console.

-r

Write the event text to a log file.

-l

Scan for status once every n minutes, where n is
an interval of 1 to 10 minutes.

-i 1 ...10

Report an event only the first time it occurs. If -n
is not specified, then events are reported every
time they are encountered.

-n

Send a mail message notifying users of
subsystem events. Send to users listed in the
user-created file
/usr/opt/swxcr/swxcrmon.maillist if it exists, or
root if no file is present.

-m

When you issue the swxcrmon command, the Monitor utility displays the
following text on successful startup:
SWXCR Monitor Utility V1.2
Digital Equipment Corporation 1994. All rights reserved.

2–2

AA–Q6TGD–TE

Chapter 2. Using the Command Line

2.2

Command Flags for swxcrmon
The following sections contain detailed descriptions of the swxcrmon command
flags.

2.2.1 Notification of Events to the System Console
You can specify whether notification of subsystem events are sent to the system
console.
# swxcrmon xcrn -r
If neither -l nor -r is specified, then events are sent to the system console.

2.2.2 Notification of Events to Log Files
You can specify that events are written to the product-specific log file using the l flag as follows:
# swxcrmon xcrn -l
If you do not specify the -l flag, no log file is created.
When the StorageWorks RAID Array 200 Online Management Utility for Digital
UNIX is installed, this directory is created:
/usr/opt/swxcr
If you specify the -l flag when you start the monitor process, a log file is created
in the /usr/opt/swxcr directory. However, if you define SWXCR_LOGPATH, the
log file is created in the directory specified by SWXCR_LOGPATH. The log file
name is as follows:
dd-Mon-hh:mm_XCRn.log
where dd is the current date, Mon is the current month, hh is the current hour,
mm is the current minute, and XCRn is the name of the specified controller. For
example, the log file for the XCR0 controller might be 05-May18:42_XCR0.log.

AA–Q6TGD–TE

2–3

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

2.2.3 Notification of Events to Mail
You have the option to have the monitor notify a specified list of users of
subsystem events by means of the Mail utility. To send notification of events to a
group of users, enter the command as follows:
$ swxcrmon xcrn -m
This will search for the file /usr/opt/swxcr/swxcrmon.maillist. The file, which
you create, contains a list of users who are to receive mail messages. If this file is
not found, mail will be sent to the root user on the system. The default (when -m
is not used) is not to send mail. A sample swxcrmon.maillist file follows:
root
lucy
bob@emubird
peter@stone.dec.com
It is important to note that Mail is sent from the account of the user who starts
the monitor process.
When notification of events by mail is specified, events are collected and sent
out in one (or more) Mail message(s). There is not one Mail message per event.
If there is at least one event to report in a given scan of the array, at least one
Mail message is sent when that array scan is complete. The number of events and
the length of the event text determine the number of Mail messages sent. More
than one Mail message may be sent (each with unique events listed in the order
in which the events were encountered).
If you specify the -m flag, the swxcrmon command parses the swxcrmon.maillist
file, and sends a test message to each user on the list. If the test message fails, an
error message is displayed and the monitor process is not created. Note that
specifying an invalid user (one that does not exist) will not necessarily cause an
error. In this case, mail may be returned to the account that started the monitor. If
you notice this, you should edit the swxcrmon.maillist file (this can be done
while the monitor is running) to correct the problem. An example of the initial
test message which is sent is as follows:
Received: by abcdef.dec.com; id AA01658; Tue, 12 Jul 1994 09:53:01 -0400
Date: Tue, 12 Jul 1994 09:53:01 -0400
From: system PRIVILEGED account <root>
Message-Id: <9407121353.AA01658@abcdef.dec.com>
Apparently-To: root

2–4

AA–Q6TGD–TE

Chapter 2. Using the Command Line
Subj:
SWXCR XCR0 Event Notification from node abcdef.dec.com
This is a test message from SWXCRMON, the StorageWorks SWXCR RAID array subsystem
monitor process. You will be receiving MAIL notification of events.

The following is an example of a Mail message sent from the swxcrmon process.
Received: by abcdef.dec.com; id AA01674; Tue, 12 Jul 1994 09:53:02 -0400
Date: Tue, 12 Jul 1994 09:53:02 -0400
From: system PRIVILEGED account <root>
Message-Id: <9407121353.AA01674@abcdef.dec.com>
Apparently-To: root
Subj:
SWXCR XCR0 Event Notification from node abcdef.dec.com
Logical RAID drive 0 is degraded.
Logical RAID drive 1 is degraded.
The hard disk at channel 0, target 2 is failed.

2.2.4 Frequency of Status Updates
You can specify the frequency (in minutes) that the disk array is scanned for
status. Use the -i flag as follows:
# swxcrmon xcrn -i n
where n is a value of 1 to 10 minutes. If the -i flag is not specified, the default is
a 5-minute interval. The interval is the time between completion of one array
scan and the start of the next array scan. Messages from the monitor may not be
exactly n minutes apart.

AA–Q6TGD–TE

2–5

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

2.2.5 Events Reported by the Monitor
The monitor reports the following conditions:
•

Logical RAID drive status changes.

•

SCSI device errors.

•

SCSI device status changes.

•

Rebuild status.

•

SCSI device insertion or removal.

•

Shelf events or errors.

The monitor process may not report on every event, depending on the following:
•

The length of time specified by the -i flag. Refer to the Frequency of Status
Updates Section 2.2.4 of this chapter for detailed information on this flag.
You can use the -i flag to specify the time between array scan operations. If
you specify a relatively short time interval, then the array is scanned
frequently. The more frequent the scan operations, the more likely it is that
an event is captured and reported. If you specify a longer time interval, status
from one event can be overwritten by status from a subsequent event.
For example, if you remove a hard disk and replace it with another disk in
the same target and channel, the disk removal event status is overwritten by
the disk insertion event status. If you specify a short array scan interval, it is
more likely that both the disk removal and the disk insertion events are
reported.

•

2–6

During a rebuild, the SWXCR monitor process reports on each of the
affected logical RAID drives. If you have a relatively small logical RAID
drive and the array scan interval is relatively long, the rebuild of the logical
RAID drive might not be reported. In this example, the rebuild of the logical
RAID drive may have started after an array scan operation and completed
before the next array scan. If this happens, then a rebuild in progress event is
not reported for that logical RAID drive.

AA–Q6TGD–TE

Chapter 2. Using the Command Line

2.2.6 Repeating Notification of Events
By default, an event is reported every time it is encountered (during an array
scan) until it is either fixed or cleared. To specify that events are reported only
once, use the -n flag as follows:
# swxcrmon xcrn -n
Now the event is reported only once, even if the same condition is encountered
on the next array scan.

2.3

Performing Parity Check and Repair on the RAID Subsystem
The parity check function reads the data from the logical RAID drive and
computes the expected parity information for each series of blocks. It compares
the computed parity to the stored parity information on the drives.
Use this function after a power failure or a system crash to verify the integrity of
parity information on logical RAID drives. This utility is only available for
redundant RAID levels (for example, RAID 1, RAID 0+1, and RAID 5).
To invoke the parity check utility, ensure that your $PATH environment variable
path includes /usr/bin. Then, type the swxcrcheck command as follows:
$ swxcrcheck devname
where devname specifies the device name for the logical RAID drive on the
controller.
For example, if you specify re0 as the device name, the parity is checked on
logical RAID drive 0.
Table 2–2 summarizes the default actions of this command. Refer to the
following sections for a detailed description of the qualifiers.
Table 2–2 Summary of swxcrcheck Flags
Action
Repair any inconsistent parity information.
Display a log of bad block information.
Write a copy of the bad block log to a file

AA–Q6TGD–TE

Flag
-r
-l
-o

2–7

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

2.4

Command Qualifiers for swxcrcheck
The following sections contain detailed descriptions of the swxcrcheck command
flags. In each example below, the ren indicates logical RAID drive ren, for
example, re0.

2.4.1 Repairing the Logical RAID Drive Parity
To specify parity repair use the swxcrcheck command with the -r flag as follows:
$ swxcrcheck ren -r
Parity check will automatically repair any inconsistencies that are found. If you
do not specify -r, repair does not take place.

2.4.2 Notification of Bad Blocks
To display a list of bad blocks found, use the swxcrcheck command with the -l
flag as follows:
$ swxcrcheck ren -l
Otherwise, parity check does not display a list of the bad block information.
To save a log of bad block information to a file, use the -o flag as follows:
$ swxcrcheck ren -o filename
When the StorageWorks RAID Array 200 Online Management Utility for Digital
UNIX is installed, the following directory is created:
/usr/opt/swxcr
If you specify the -o flag when the parity check utility is started, a log file is
created in the /usr/opt/swxcr directory unless you have previously defined
SWXCR_LOGPATH; in this case, the log file is created in the directory
specified by SWXCR_LOGPATH. If the filename argument is specified and
includes a directory path, the file is created in the directory path specified in the
filename. Note that if a directory path is specified in the filename, the filename
must also be included (it is not defaulted).
If you specify the -o flag without a filename argument, the filename is as
follows:
dd-Mon-hh:mm_ren.log
where dd is the current date, Mon is the current month, hh is the current hour,
mm is the current minute, and ren is the name of the specified device. For
example, the log file for the re0 device might be 06-Jun-17:19_re0.log.
2–8

AA–Q6TGD–TE

Chapter 2. Using the Command Line

If both the -l and the -o flags are specified, then a log is written to the terminal
and to the output log file. If the specified output file cannot be opened, the log is
written to the terminal instead.
The following is a sample output from a parity check process with the -l flag
specified. In this case, bad blocks were found. The list specifies the block number
where a bad block was found and the number of bad blocks found at that address.
$ swxcrcheck re2 -l
 Digital Equipment Corporation 1994. All rights reserved.
Parity check complete.
The following bad blocks were found:

Block

# Bad

Number

Blocks

----------

000000576

0000000016

0000000992

0000000016

0000001472

0000000016

0000001984

0000000016

0000002176

0000000016

0000002208

0000000016

0000002976

0000000016

0000003680

0000000016

1. The controller transfers data to the disks in multiple-block chunks. In this
example, the number of blocks is 16 (or 8 Kbytes). If a bad block is found,
the whole chunk is marked bad, not just the one bad block. As a result, all
errors show a size of 16 blocks.
2. Note that the numbers in the left column are all divisible by the chunk size
(in this case, 16). This is because the controller marks an entire chunk as
bad whenever an error is found. In this example, if block 580 is a bad
block, then this represents the chunk from 576-591. If more than one error
is found in a chunk, the controller reports that chunk as bad only once. Note
that if a set of bad blocks spans across two chunks (for example, 591-592),
the controller reports both chunks as bad (block 576 for 32 rather than 16)
blocks. Note also that the list may not be in strict numerical order due to
the way the controller processes commands.
AA–Q6TGD–TE

2–9

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

2.5

Altering Rebuild Rate on the RAID Subsystem
The rebuild rate function allows you to view, or select the rate at which a rebuild
or parity check will complete. A rebuild or parity check will slow down system
I/O performance and system I/O will slow down rebuild or parity check
performance. Therefore, you can use the rebuild rate function to raise or lower
the priority of a rebuild or parity check and thus speed up or slow down its
performance.
NOTE
Before invoking the rebuild rate utility, ensure that
your $PATH environment variable includes
/usr/bin.

2.5.1 Viewing the Rebuild Rate
To view the rebuild rate, invoke the rebuild rate utility by entering the command
as follows:
$ swxcrrate devname
where devname is the name of the controller, xcr0, for example.
The utility displays a message such as the following:
Rebuild Rate is currently set to 50.

2–10

AA–Q6TGD–TE

Chapter 2. Using the Command Line

2.5.2 Setting the Rebuild Rate
You can change the rebuild rate while a rebuild or parity check is in progress. If
you find that either process takes longer than you would like, you can increase
the rebuild rate, or if you find that system I/O performance suffers because of the
parity check or rebuild, you can lower the rate.
To set the rebuild rate, invoke the rebuild rate utility by entering the command as
follows:
$ swxcrrate devname rate
where devname is the name of the controller, xcr0, for example, and rate is the
rate you wish to set. Acceptable values for rate are any integer between 0 and 50.
The slowest rate is 0 (all priority is given to I/O and least priority to the rebuild
or check). The fastest rate is 50 (equal priority is given to rebuild or check). The
utility displays a message such as:
Rebuild Rate set to 50 (was 40).

AA–Q6TGD–TE

2–11

3
Command Line Messages
3.1

Monitor Process Messages
The monitor process displays the general informational and error messages
described in Table 3–1 through the reporting mechanisms you set up.
Table 3–1 Monitor Process Messages
Message

Meaning

Controller checksum error; monitor
process terminated.

The monitor encountered a checksum error
when trying to read the controller
configuration. In this case, the controller did
not return any configuration to use so the
monitor process will terminate with an error.

NVRAM and EEPROM configurations
do not match.

The monitor process determined, when
reading the configuration from EEPROM on
the controller, that the NVRAM and
EEPROM configurations do not match. The
monitor process was able to continue
because the EEPROM configuration was
transferred, but the user should run the
configuration utility to determine which
configuration is correct. (It is not clear
whether the EEPROM configuration or the
NVRAM configuration is the correct one).

AA–Q6TGD–TE

3–1

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Table 3-1 Monitor Process Messages (continued)
Message

Meaning

NVRAM configuration does not
match.

The monitor process determined, when
reading the configuration from EEPROM. that
the EEPROM configuration and the actual
configuration do not match. This is because
of a startup error, such as a storage
enclosure was not found, because it's power
was not turned on, or a device failed. Check
your storage enclosure, or run the
configuration utility to determine what caused
the startup error.

Replacement table full. Monitor can
no longer report on rebuilds. Monitor
process is terminating!!!

The controller maintains a list of changes to
the configuration caused by standby
replacements. The monitor process attempts
to clear this list after processing a rebuild
start, but it may not be able to clear the table
if the SWXCR configuration continually
changes. While the monitor makes repeated
attempts to clear the table, it is possible (but
highly unlikely) that the monitor cannot do so.
In this case, the controller can continue to
function, but the monitor process can no
longer correctly report on standby
replacements. Therefore, if the list is full and
the monitor process cannot clear it, the
monitor process will terminate. Reboot your
system.

3–2

AA–Q6TGD–TE

Chapter 3. Command Line Messages
Table 3-1 Monitor Process Messages (continued)
Message

Meaning

Internal log structures getting full.

The SWXCR controller maintains an internal
log of changes in device states within a
session (power-cycle or reset to the next
power-cycle or reset). This log is a fixed-size
list, and when this list is full, the monitor
process reports the second message. When
the list has less than three entries left, the
monitor process reports the first message.

OR
Internal log structures are full!

If you see either message, you must shut
down the system, cycle power, and reboot.
This clears the internal log structure.
The hard disk at channel <chn>,
target <tgt> is being rebuilt. This
affects the status of logical RAID
drive(s) <drive-list>.

AA–Q6TGD–TE

A failed hard disk at the specified channel
and target is being rebuilt. <chn> is the
channel number; <tgt> is the SCSI target ID.
The <drive-list> shows the logical RAID
drives to be rebuilt as part of the rebuild of
the specified hard disk.

3–3

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Table 3-1 Monitor Process Messages (continued)
Message

Meaning

The hard disk at channel <old-chn>,
target <old-chn> is being replaced by
the hard disk at channel <new-chn>,
target <new-tgt>. This affects the
status of logical RAID drive(s)
<drive-list>.

When a hot spare rebuild is in progress, the
monitor reports this message. It indicates the
channel and target of both the hard disk that
failed and the hot spare hard disk which is
replacing it.
The <old-chn> is the channel number of the
hard disk being replaced; <old-tgt> is the SCSI
target id of the hard disk being replaced;
<new-chn> is the channel number of the hot
spare device; <new-tgt> is the SCSI target id
of the hot spare device.
The <drive-list> shows the logical RAID drive
numbers that have blocks on the disk being
replaced. Each of these logical RAID drives
becomes critical and is rebuilt as part of the
rebuild of the hard disk. The monitor reports
on the progress of the rebuild, including the
rebuild of each logical RAID drive.
If the rebuild is successful, the monitor reports
that each of the logical RAID drives is now on
line. If a rebuild operation fails, the monitor
reports that the logical RAID drives are still
degraded.

Logical RAID drive <drive-num> is
<drive-status>

Monitor is reporting on a logical RAID drive
status. If -nonrepeat is specified, only changes
in logical RAID drive status are reported.
Otherwise, any changes in status are reported,
and any logical RAID drive status which is not
'optimal' is reported. <drive-num> is the logical
RAID drive unit number and <drive-status> is
one of the following: "Optimal," "Degraded,"
"Failed," or "Unknown."
.

3–4

AA–Q6TGD–TE

Chapter 3. Command Line Messages
Table 3-1 Monitor Process Messages (continued)
Message

Meaning

The hard disk at channel <chn>, target
<tgt> had a <err-type> error.

A SCSI device at the specified channel and
target had one of the specified type of errors
or the device had the specified number of the
specified type of error. The number of errors
is the increment since the last report of the
specified error type not the total accumulation
of that type of error on the hard disk. <chn>
is the channel number and <tgt> is the SCSI
target id. <count> is the number of errors that
occurred; <err-type> is one of the following:
"parity," "soft," "hard," or "miscellaneous."

OR
The hard disk at channel <chn>, target
<tgt> had <count> <err-type> errors

The hard disk at channel <chn>, target
<tgt> is <dev-sts>.

Monitor is reporting on the status of the hard
disk at the specified channel and target.
SCSI disk status changes are reported only
once, regardless of the presence or absence
of the -nonrepeat flag. <chn> is the channel
number and <tgt> is the SCSI target id. <devsts> is one of the following: "failed, a hot
spare," "write only," or "optimal."

A new hard disk has been inserted at
channel <chn>, target <tgt>.

These two messages notify the user of
insertion and removal of drives. <chn> is the
channel number and <tgt> is the SCSI target
id. These messages are reported only once,
regardless of the presence or absence of the
-nonrepeat flag.

OR
The hard disk at channel <chn>, target
<tgt> has been removed.

The hard disk at channel <chn>, target
<tgt> is not being automatically
replaced by a hot spare. Please
replace the device and rebuild it.

AA–Q6TGD–TE

The hard disk (whose status will have already
been reported) has failed or been removed
and is not being automatically replaced by a
hot spare. User action is required to replace
the drive. <chn> is the channel number and
<tgt> is the SCSI target id.

3–5

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Table 3-1 Monitor Process Messages (continued)
Message

Meaning

Logical RAID drive <drive-num> is
rebuilding.

When a logical RAID drive starts rebuilding,
The monitor displays this message. If norepeat is specified, the first message
displays. Otherwise, the second message
displays, and includes the percent of that
disk that has been rebuilt. <drive-num> is the
logical RAID drive unit number. <percent> is
the percentage complete.

OR
Logical RAID drive <drive-num> is
rebuilding (<percent>% complete).

The rebuild (on hard disk at channel
<chn>, target <tgt>) has successfully
completed.
OR
The rebuild (on hard disk at channel
<chn>, target <tgt>) completed with
some errors.

These messages are used to report on
rebuild status. When a hard disk is rebuilt,
the monitor reports on each affected system
drive as it is being rebuilt. When the disk is
completely rebuilt, or the disk rebuild fails,
one of these messages is used to indicate
the completion status. If possible, the failure
reason is indicated. <chn> is the channel
number and <tgt> is the SCSI target id.

OR
The rebuild (on hard disk at channel
<chn>, target <tgt>) failed because
the new drive failed.
OR
The rebuild (on hard disk at channel
<chn>, target <tgt>) failed because
the source drive failed.
OR
The rebuild (on hard disk at channel
<chn>, target <tgt>) failed because of
bad blocks in the source drive(s).

3–6

AA–Q6TGD–TE

Chapter 3. Command Line Messages
Table 3-1 Monitor Process Messages (continued)
Message

Meaning

Write cache error has occurred.

Some write data in the controller cache was
not successfully updated to disk due to disk
failure.

Shelf failure on channel <chn>.

A shelf failure occurred on the specified
channel. <chn> is the channel number. A
shelf failure may indicate that a power failure
occurred on the shelf.

Shelf error on channel <chn>.

A shelf error occurred on the specified
channel. <chn> is the channel number. A
shelf error may indicate that a drive is not
seated properly or that a blower or redundant
power supply failed.

Monitor processing terminating. Fatal
error encountered: <err_text>.

The monitor process encountered the
specified fatal error. User action depends on
the error specified in <err_text>.

AA–Q6TGD–TE

3–7

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

3.2

Parity Check Messages
The parity check utility displays the informational and error messages described
in Table 3–2 through the reporting mechanisms you set up.
Table 3–2 Parity Check Messages
Message

Meaning

The bad block table is full. There may
be more bad blocks on the specified
system drive than those specified
above.

The bad block table is a fixed size. If too
many bad block are found, the table fills up
and some bad blocks may not get reported.
This message displays at the end of the /LOG
display.

Invalid logical RAID drive specified.

The device name specified in the command
line does not correspond to a logical RAID
drive.

Error in parity check command
occurred.

Controller returned unknown status from the
Parity Check command.

Parity check cannot be performed on
RAID 0 (nonredundant) logical RAID
drive.

The logical RAID drive specified is a
nonredundant RAID level (for example, RAID
0) and therefore parity check cannot be
performed.

OR
Parity check cannot be performed on
JBOD (nonredundant) logical RAID
drive.

3–8

AA–Q6TGD–TE

Chapter 3. Command Line Messages
Table 3–2 Parity Check Messages (cont.)
Message

Meaning

One or more SCSI disks in this logical
RAID drive has failed.

Parity check cannot be performed because
one or more SCSI disks in the logical RAID
drive is dead; parity checks can be performed
only on logical RAID drives in the optimal
state.
The disk should be replaced and the logical
RAID drive rebuilt.

A rebuild or parity check is already in
progress.

AA–Q6TGD–TE

Only one rebuild or parity check at a given
time is allowed by the controller. This error
indicates that either a rebuild or parity check
is currently in progress. The user will have to
wait until the function is complete, and then
reissue the swxcrcheck command.

3–9

4
Using the RAID Array 200 Online Management
Utility for Digital UNIX
Graphical User Interface
This chapter presents the functionality available for monitoring and maintaining
StorageWorks RAID Array 200 Subsystems with the Online Management Utility
Graphical User Interface (GUI) for Digital UNIX

4.1

Overview of the RAID Array 200 Online Management Utility for
Digital UNIX Graphical User Interface (GUI)
The RAID Array 200 Online Management Utility for Digital UNIX Graphical
User Interface (GUI) provides the following functionality to maintain a RAID
Array 200 Subsystem in a Digital UNIX environment:
•

View the status of the disk drives in the array

•

View the logical drives

•

Check and repair parity of redundant logical drives

•

View bad blocks of a logical drive

•

Manually replace a drive after a disk failure using rebuild (for redundant
configurations), or make optimal (for non-redundant configurations

•

Set the rebuild rate

•

Create hot spares

This chapter describes the GUI and how to use it to monitor the status of a RAID
Array 200 subsystem. Chapter 5 describes how to use the GUI to manage the
array.

AA–Q6TGD–TE

4–1

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

4.2

Accessing the Online Management Utility for Digital UNIX GUI
To start the Online Management GUI, type the following: swxcrmgr
The online management GUI base window appears similar to the one shown in
Figure 4–1 for a 1-channel subsystem, or similar to the one shown in Figure 4–2
for a 3-channel subsystem.
Figure 4–1 The Base Window of the Online Management Utility for Digital UNIX
GUI, Displaying a 1–Channel Configuration

4–2

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface
Figure 4–2 The Base Window of the Online Management Utility for Digital UNIX
GUI, Displaying a 3–Channel Configuration

The base window displays:
•

Menus

•

A matrix of the disk drives installed in your subsystem

•

A table of the logical drives configured on your subsystem

•

The current function selected

•

The current adapter selected

•

Action buttons

AA–Q6TGD–TE

4–3

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

4.3

Using the Online Management GUI
The online management GUI is compliant with the Motif style guidelines. It is a
window and menu driven environment designed to be used with a mouse.

4.3.1 Using the mouse
Table 4–1 describes the functions of each of the mouse buttons for a righthanded, three-button mouse.
Table 4–1 Mouse Functions
To:

Press:

Select an item

Left mouse button
For example, double click on a
drive icon to display drive
parameters

Access a menu

Left mouse button
For example, point to a menu
button on the screen, and click the
left mouse button to open the
menu and display its options

4–4

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface

4.3.2 Using Windows
Table 4–2 describes the elements that may appear on the GUI windows.
Table 4–2 Window Elements
Elements

Description

Point to a menu on the screen and click your left mouse button to display menu
items

Title

Displays name of application you accessed, StorageWorks(TM) RAID Array 200
Management Utility for Digital UNIX(TM) V1.0.0 Copyright Digital Equipment
Corporation 1994

Resize
corner

Allows you to change window size from any corner of window
NOTE: The contents of the base window are NOT resizeable

Control area

Displays menus

Action
Buttons

Point and click to an action button to start or cancel the selected function, or to
refresh the information displayed in the window

AA–Q6TGD–TE

4–5

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

4.3.3 Using Menus
From the online management GUI base window, you can access the following
menus:
•

Management

•

Help

The Management window provides the following items in the order shown:
Default Rebuild Rate
Rebuild Drive
Parity Check
Fail Drive
Make Optimal
Make Hot Spare
Bad Block Table
Print Configuration
Exit
The Help window provides the following items in the order shown:
On Menu Options
About this utility
To view the items on a menu, point to the menu and click your left mouse
button.

4.3.4 Using Help
To get help while using the online management for Digital UNIX GUI base
window, select Help from the Help menu.

4.3.5 Exiting from the Online Management GUI
When you finish using the online management GUI, choose the Exit item from
the Management menu. The Exit swxcrmgr window appears. Click OK to exit, or
click Cancel to remain in the utility.

4–6

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface

4.4

Viewing the Subsystem from the Online Management GUI
A subsystem consists of a controller and the storage enclosures attached to it.
The operating condition of a RAID Array 200 subsystem depends upon the status
of each individual disk drive installed in the storage enclosures, and the status of
each of the logical drives configured with those disk drives. You can view the
status of the disk drives and logical drives in each of the storage enclosures
attached to each of your controllers from the base window.
To monitor a subsystem, do the following:
•

Select a controller

•

View the Disk Drives

•

View the Logical Drives

4.4.1 Selecting a Subsystem
The base window lists the controllers installed in your system under Adapter
Selected. The base window displays controllers by the system expansion slot,
[Expansion Board Type] [Slot] [Slot Number], in which you installed them, as
shown in Figure 4–3.
To view a subsystem, click on the slot number of a controller. The base window
then displays the logical drive information for the subsystems connected to the
selected controller (adapter), as shown in Figures 4–4 and 4–5.
Figure 4–3 The Adapter Selected Buttons

AA–Q6TGD–TE

4–7

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Figure 4–4 The Subsystem Displayed after Selecting EISA Slot 8

4–8

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface
Figure 4–5 The Subsystem Displayed after Selecting PCI Slot 6

AA–Q6TGD–TE

4–9

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

4.4.2 Viewing Disk Drives
The GUI displays disk drives in a disk drive matrix on the base window, similar
to the drive matrixes shown in Figure 4–6 for a 1-channel subsystem and Figure
4–7 for a 3-channel subsystem.
The drive matrix displays all of the drives connected to each channel of the
controller installed in the selected slot.
Table 4–3 describes all the information the drive matrix displays, except status.
Table 4–4 describes disk drive status.
Figure 4–6 The Disk Drive Matrix for a 1-Channel Subsystem

SCSI ID

The drive group
designator. Specifies
Each disk drive in a
the drive group
drive group is numbered
this disk drive is a
from 0 - 7.
member of.

Disk drives, such as hot spares,
which are not configured into a drive group,
do not have drive group
numbers associated with them.

Channel number

This box displays the status
of the disk drive. It also
changes color to indicate
when you have selected
this disk drive.

4–10

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface
Table 4–3 Drive Matrix Information
Element

Description

Channel

Each channel on your controller is an
independent SCSI bus supporting up to 7
devices. Additional channels provide you with
additional disk capacity (see SCSI ID).
The RAID controller can access only one drive
per channel at a time, but can simultaneously
access up to 3-channels at a time. If you have a
3-channel controller, you can divide up the disk
drives in your subsystem on different channels,
and thus increase the subsystem performance
by allowing the controller to simultaneously
access one disk per each channel.
You can configure drive groups on one channel
of a controller, or across channels of a
controller.

SCSI ID

Each drive on the channel (an independent SCSI
bus) must have a unique SCSI ID. You can have
up to 7 disk drives per channel. If you want disk
capacity above seven drives, you need
additional channels.

Disk Drive Icons

Each drive installed in your subsystem(s) is
represented in the drive matrix by a drive icon.
The drive icon displays the status of the drive
(Table 4–4 describes each drive status), and
changes color to indicate when you have
selected this drive.

AA–Q6TGD–TE

4–11

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Table 4–3 Drive Matrix Information (cont.)
Element

Description

Drive Group Designator

The drive group designator is a letter that
appears at the top of the drive icon to identify
each drive that belongs to the same drive group.
You can easily determine drives that are not
configured into a drive group, because they do
not have drive group designators.

Drive Number

The drive number appears at the top of the drive
icon. The online management GUI numbers
each drive, so that you can determine easily
how many drives belong to a drive group.
Drives that are not configured into a drive group
do not require a drive number.

Figure 4–7 The Disk Drive Matrix for a 3-Channel Subsystem

4–12

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface
Table 4–4 Disk Drive Status
Status Code

Status

Description

RDY

Ready

A disk drive installed in your subsystem, but not configured
into a drive group or as a hot spare
All disk drives have this status when you first install the
subsystem

HSP

Hot Spare

A drive that you leave unused in your subsystem in the
event that one of the disks of a logical drive fails
If one of the disks of a logical drive fails, the subsystem
automatically begins using the hot spare drive in place of
the failed drive

OPT

Optimal

A disk drive configured into a drive group
A drive with this status is in good working condition

FLD

Failed

A disk drive configured into a drive group that the
subsystem can not access because of some mechanical or
other type of failure
You need to replace a drive with this status
See Chapter 3 for information on how and when to replace
and rebuild failed drives

WOL

Write Only

A disk drive that the system is rebuilding
The data may not be good on a drive being rebuilt;
therefore, the subsystem can not read data from a disk that
is being rebuilt; the subsystem can only write to it

AA–Q6TGD–TE

4–13

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

4.4.3 Viewing Logical RAID Drives
The GUI displays logical RAID drives in a Logical Drive Table on the base
window, similar to the Logical Drive Table shown in Figure 4–8.
The Logical Drive Table displays all of the drives connected to the channel or
channels of the controller in the selected slot.
Table 4–5 describes all the information the Logical Drive Table displays, except
status. Table 4–6 describes logical drive status.
Figure 4–8 The Logical Drive Table
Drive group on
which the
logical drive
was configured.

Type of RAID
implemented
by the
logical Drive.

Operating condition
of the logical drive.

Logical drive
icon

Logical drive
number/Identifier

4–14

Capacity
of the
logical drive.

Caching method
this logical drive uses.

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface
Table 4–5 Logical Drive Table Information
Element

Description

Logical Drive Icon

The GUI represents each logical drive
configured on your subsystem by a logical drive
icon in the Logical Drive Table.

Logical Drive Number

Each logical drive configured on your subsystem
has a unique number/identifier between 0 and 7.
The controller assigns the logical drive number
when you create the logical drive with the
standalone utility.

Drive Group

The Logical Drive Table lists the drive group
used to configure the logical drive. This way,
you can quickly locate in the drive matrix the
drive group associated with a logical drive.

RAID

Indicates the RAID level of the logical drive. The
RAID level determines how the logical drive
stores data.

Size

Indicates the storage capacity of the logical
drive.
The storage capacity of the drive group and the
RAID level used for the logical drive determine
the storage capacity of the logical drive.

Write Policy

Indicates the caching policy selected for this
logical drive, write thru, or write back. See the
StorageWorks RAID Array 200 Subsystem
Controller Installation and Standalone
Configuration Utility User's Guide to learn about
the caching options.

Status

The status of the logical drive is dependent upon
the status of the disk drives in the drive group
used to configure the logical drive. See Table 4–
6 for a description of logical drive status.

AA–Q6TGD–TE

4–15

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Table 4–6 Logical Drive Status
Status

Description

Optimal

All of the disk drives in the drive group associated with this logical drive
have a status of optimal.
The logical drive is in good operating condition.

Degraded

The controller can not access a disk drive in the drive group on which this
logical drive was configured.
This status appears for RAID 5, RAID 0 + 1, and RAID 1 logical drives only,
since these RAID levels have redundancy and can be rebuilt.
See Section 5.2 on replacing a failed drive.

Dead

The controller can not access two or more disk drives in the drive group
associated with a RAID 5, RAID 0 +1, or RAID 1 logical drive. You must
restore data from backup media.
or
The controller can not access a disk drive in the drive group associated with
a RAID 0 logical drive, or JBOD configuration. You replace the failed drive
and restore data from backup media.
See Section 5.2 on replacing a failed drive.

4–16

AA–Q6TGD–TE

Chapter 4. Using the Graphical User Interface

4.5

Printing the Configuration
Use the Print Configuration option to output configuration information to a text
file suitable for printing. The file will contain information such as number of
drive groups, and number, type, and size of logical drives.
To Print the Configuration, follow these steps:
1. Select Print Configuration from the Management menu. A window appears,
similar to the one shown in Figure 4–9, containing a field for the name of the
file to save the information to. The default is SWXCR.asc, which will appear
in the current directory (/) from where the GUI was invoked.
2. Type a filename in the field (or accept the default), and click OK in the
window. The GUI creates a file for the configuration information.
Figure 4–9 The Print Current Configuration Window

Note
The GUI will overwrite old versions of the file;
therefore, when printing multiple configurations,
make sure to save each one under separate file
names.

AA–Q6TGD–TE

4–17

5
Maintaining the Subsystem
with the GUI
This chapter presents the functionality available for monitoring and maintaining
StorageWorks RAID Array 200 Subsystems with the Online Management Utility for
Digital UNIX GUI.

5.1

Maintaining the Subsystem
The RAID Array 200 Online Management Utility for Digital UNIX GUI allows
you to do the following to maintain a RAID Array 200 subsystem:
•

Replace failed drives in redundant and nonredundant configurations

•

Create hot spares

•

Set the rebuild rate

•

Check and repair parity for redundant configurations

•

View bad block table

AA–Q6TGD–TE

5–1

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

5.2

Replacing a Failed Drive
One of the major benefits of a RAID subsystem is the capability to continue
accessing the data stored on the subsystem in the event of a single disk failure.
Although redundant RAID configurations can withstand disk failures, most can
withstand only a single disk failure. For many redundant RAID configurations, if
two disks fail at the same time, all the information stored on the logical RAID
drives with which those disks are associated is lost.
If a drive experiences severe enough errors, the controller will automatically fail
the drive. However, if you know or suspect that one of the drives in your storage
enclosure is bad (some hints include a high rate of errors, or degraded system
performance) you will want to replace it with another drive as soon as possible.
Your RAID Array 200 subsystem and RAID Array 200 Online Management for
Digital UNIX GUI provides you with the following methods for replacing failed
drives:
•

Hot spare

•

Hot swap

•

Mark as failed/Rebuild (redundant configurations only)

•

Mark as failed/Make optimal (non-redundant configurations only)

Sections 5.2.1 to 5.2.3 describe when and how to use each of the methods for
replacing a failed drive.

5–2

AA–Q6TGD–TE

Chapter 5. Maintaining the Subsystem with the GUI

5.2.1 Hot Spare
You can use this method to replace a drive if:
•

The drive you want to replace contains only redundant RAID configurations
(RAID 1, 5, 0+1)

•

You defined one or more hot spare drives

See section 5.3 for information on how to create a hot spare drive.
To replace a drive with a hot spare, remove the bad or suspect drive from the
storage enclosure and the RAID controller automatically begins to recreate the
data stored on the removed disk onto the hot spare drive disk.

5.2.2 Hot Swap
You can use this method to replace a drive if:
•

The drive you want to replace contains only redundant RAID configurations
(RAID 1, 5, 0+1)

•

Your storage enclosure supports hot swaps (StorageWorks storage enclosures
all support hot swap capability)

•

You enabled the Fault Management option when first configuring the
subsystem (see the Controller Installation and Standalone Configuration
Utility User's Guide for information on how to enable this option)

To replace a drive using hot swap, remove the bad or suspect drive from the
storage enclosure, and wait 20 seconds (so the controller recognizes the drive
change) before inserting a new drive in the same slot from which you removed
the bad/suspect drive. The RAID controller automatically begins to recreate the
data stored on the removed disk onto the disk drive inserted into the storage
enclosure in its place.

AA–Q6TGD–TE

5–3

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

5.2.3 Marking a drive as Failed
Use the Mark as Failed option in conjunction with the Rebuild or Mark as
Optimal options to replace a drive if:
•

The hot spare or hot swap options are not supported on your subsystem

•

To replace a drive from a non-redundant configuration (RAID 0, JBOD)

•

To replace a drive, which is logging errors
NOTE
Use caution when failing drives. If the drive you
fail is part of a non-redundant logical RAID drive
(RAID 0 or JBOD), all the data on that logical
RAID drive will be lost. Backup a drive before you
mark it failed.

To replace a drive using Mark as Failed, follow these steps:
1. Select Fail Drive from the Management menu. Fail Drive appears as the
Current Selection at the top of the base window.
2. Click on any drive icon with a status of OPT or WOL, and then click OK.
3. A window appears, as shown if Figure 5–1 for you to confirm that you want
to fail the selected drive. Click OK. The GUI displays FLD status under the
icon for that drive. The amber light on the drive flashes, becomes solid, and
the drive spins down.

5–4

AA–Q6TGD–TE

Chapter 5. Maintaining the Subsystem with the GUI
Figure 5–1 Fail Drive Confirmation Window

4.
5.

Locate the disk with the solid amber light in the storage enclosure and
remove it.
Insert a replacement disk into the same slot of the storage enclosure from
which you removed the disk that you failed.
If the drive that you failed contained a RAID 1, 5, or 0 + 1 redundant
configuration go to section 5.2.3.1 to rebuild the drive. If the drive you
failed contained RAID 0, or JBOD non-redundant configurations go to
Section 5.2.3.2 to mark the drive optimal and restore data from backup.

AA–Q6TGD–TE

5–5

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

5.2.3.1 Rebuilding a Drive
Use Rebuild a Drive when replacing a failed drive which contained a RAID 1, 5,
or 0 + 1 redundant configuration. The rebuild operation reconstructs the data
which the failed drive had contained onto the replacement drive.
To rebuild a drive, after replacing a drive you failed in the storage enclosure with
a drive in good working order, follow these steps:
1. Select Rebuild Drive from the Management menu. Rebuild Drive appears as
the Current Selection at the top of the base window.
2. Click on the icon of the drive with a status of FLD (failed) that you replaced,
and then click OK. A window appears, similar to the one shown in Figure
5–2, for you to confirm that you want to rebuild the selected drive. Click
OK.
Figure 5–2 The Rebuild Drive Confirmation Window

5–6

A window appears, similar to the one shown in Figure 5–3, with a slider bar
showing the progress of the rebuild on the current logical drive. The
controller rebuilds each logical RAID drive associated with this disk drive.
A message appears, as shown in Figure 5–4 indicating when the controller
finishes rebuilding all of the logical RAID drives. Click OK and the base
window appears.

AA–Q6TGD–TE

Chapter 5. Maintaining the Subsystem with the GUI
Figure 5–3 The Rebuild Drive Confirmation Window

Figure 5–4 The Rebuild Completed Message

Note
The status of the drive changes to WOL (writeonly) during the rebuild. Upon completion of the
rebuild, the status of the drive and all logical
drives associated with it will change to optimal.

AA–Q6TGD–TE

5–7

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

5.2.3.2 Marking a Drive as Optimal
Use the Mark as Optimal option when replacing a failed drive which contained
RAID 0, or JBOD non-redundant configurations. Mark as Optimal spins up a
drive and changes the status of the drive to optimal.
Note
Mark as Optimal does not fix any problems with a
drive; it merely makes the system use the drive. If
a drive is prone to fail, it will fail again. Also, data
will not be written to the drive when it is made
optimal. If you find errors with a drive, fail it and
replace it.

You can only mark optimal a drive with a status of FLD. If you have a drive with
a status of WOL, you must first change its status to FLD, before you can mark it
optimal. See Section 5.2.3 to fail a drive.
To make drives with a status of HSP or RDY optimal, you must use the standalone configuration utility. See the StorageWorks RAID Array 200 Subsystem
Controller Installation and Standalone Configuration Utility User's Guide.
To make a drive optimal, after replacing a suspect/bad drive in the storage
enclosure with a drive in good working order, follow these steps:
1. Select Make Optimal from the Management menu. Make Optimal appears in
Current Selection at the top of the base window.
2. Click on the icon of the drive with a status of FLD (failed) that you replaced,
and then click OK. A window appears, similar to the one shown in Figure
5–5, for you to confirm that you want to mark the selected drive as optimal.
Click OK. The status of the drive changes to OPT. The amber light on the
drive, lit when the drive is FLD, will go off when the drive becomes optimal.
The status of all the logical drives contained in this drive also change to
optimal.

5–8

AA–Q6TGD–TE

Chapter 5. Maintaining the Subsystem with the GUI
Figure 5–5 The Make Optimal Confirmation Window

5.3

Making a Hot Spare
Make one or more hot spare disk drives on which the RAID controller can
automatically start to rebuild information when a drive fails which contained a
RAID 1, 5, or 0 + 1 redundant configuration.
You can make a hot spare out of any disk drive, which is not in a drive group.
Use a disk drive with a capacity equal to or greater than the largest disk capacity
of any drive in a group.
To make a hot spare, follow these steps:
1. Select Make Hot Spare from the Management menu. Make Hot Spare
appears in Current Selection at the top of the base window.
2. Click on any drive icon with a status of RDY (ready), and then click OK.
The drive status changes to HSP (Hot Spare). The controller will now use
this drive as a hot spare, if a drive configured into a drive group fails.
If you decided that you no longer want to use a disk as a hot spare, you can
return its status to ready. To return a hot spare disk to ready status, follow these
steps:
1. Select Make Hot Spare from the Management menu. Make Hot Spare
appears in Current Selection at the top of the base window.
2. Click on any drive icon with a status of HSP (Hot Spare), and then click OK.
The drive status changes to RDY (ready). The drive is now available for you
to use in a drive group (through the standalone configuration utility), or to
remove from the subsystem.

AA–Q6TGD–TE

5–9

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

5.4

Setting Rebuild Rate
Use the Default Rebuild Rate option to select how quickly a rebuild or parity
check will complete.
A rebuild or parity check will slow down system I/O performance and system I/O
will slow down rebuild or parity check performance. Therefore, you can use the
rebuild rate function to raise or lower the priority of a rebuild or parity check and
thus speed up or slow down its performance.
You can assign a rebuild rate value between 0 and 50. The slowest rate is 0 (most
priority is given to I/O and least to the rebuild). The fastest rate is 50 (more
priority given to the rebuild). 50 is the default rebuild rate.
To change the rebuild and parity check rate, follow these steps:
1. Select Default Rebuild Rate from the Management menu. A pop-up window
will appear with the current rebuild rate displayed on a slider bar as shown in
Figure 5–6.
Figure 5–6 Set Rebuild Rate Window

5–10

To exit without changing the rate, click the Cancel button in the pop-up
window. To change the rate, drag the slider bar with the mouse to the value
desired, and then click OK. The base window appears.

AA–Q6TGD–TE

Chapter 5. Maintaining the Subsystem with the GUI

Note that you can change the rebuild rate while a rebuild or parity check is in
progress. If you find that the rebuild or check is taking longer than desired, you
can increase the rebuild rate, or if you find that system I/O performance is
suffering because of the rebuild or check, you can lower the rate.

5.5

Viewing the Bad Block Table
The controller keeps track of two bad block tables, one for rebuild and one for
write-back cache.
The rebuild bad block table tracks the location of areas on a physical disk which
are inaccessible during a rebuild, or contain incorrect data during a parity check.
The write-back bad block table tracks the location of areas on a physical disk
where the controller encounters errors writing data.
The entries in either bad block table show the logical drive containing the error,
the location (starting logical block address) of the error, and the length (number
of sectors) of the error. Remember this table displays the logical block addresses
for the logical drive, not the physical disk.
The GUI automatically displays the bad block table upon completion of a parity
check/repair if it found errors. At other times, to view the bad block table, follow
these steps:
1. Select Bad Block Table from the Management menu. The Bad Block Table
window appears, as shown in Figure 5–10.
2. Select either Rebuild or Write Back from the button on the right side of the
table. A list will appear (with scroll bar if necessary) showing any errors by
logical drive, starting block, and number of blocks bad.
3. Click OK in the Bad Block Table window to close it when you finish
viewing.

AA–Q6TGD–TE

5–11

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

5.6

Checking Parity
Run Parity Check to validate parity information for logical drives with parity.
Logical drives on the subsystem with parity include RAID 1, RAID 0 + 1, and
RAID 5 logical drives.
The Parity Check detects any internal inconsistencies between data and parity
information, and optionally makes repairs.
To run Parity Check, follow these steps:
1. Select Parity Check from the Management menu. The Parity Repair Option
Window appears, as shown in Figure 5–7.
Figure 5–7 The Parity Repair Option Window

2.
3.

5–12

Select either the Repair Parity or No Repair mode by clicking on the
appropriate button, and then clicking OK.
Click on the Logical drive for which you want to check parity and click OK.
The Parity Check Progress Window appears, as shown in Figure 5–8,
indicating parity check started, and percent completed.
If the utility detected no errors, when parity check completes, a screen
similar to the one shown in Figure 5–9 appears. Click OK to close the Parity
Check Progress window.
If the utility detects errors, when parity check completes, a bad block table
appears as shown in Figure 5–10.

AA–Q6TGD–TE

Chapter 5. Maintaining the Subsystem with the GUI
Figure 5–8 The Parity Progress Window

Figure 5–9 Parity Check Completes Message

AA–Q6TGD–TE

5–13

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX
Figure 5–10 Parity Status Report Appears When Parity Check Completes

Click OK to close the bad block table after viewing it. A Parity Check
Progress window appears, as shown in Figure 5–11.

Figure 5–11 Parity Errors Detected Message

5–14

Click OK to close the Parity Check Progress window.

AA–Q6TGD–TE

Index
A

accessing array, 1–9

Digital UNIX
preparing logical RAID drives for
access by, 1–14

action buttons, 4–3
adapter selected, 4–7

B
bad block
notification of, 2–8
table, 5–11

C
channel, 4–11
command line, 1–1
files, 1–3
loading, 1–4
monitoring, 2–1
parity check and repair, 2–7
configuration
printing, 4–17
redundant, 5–3, 5–5

devices
create special files for, 1–9
MAKEDEV, 1–11
mounting, 1–4
device special files
creating, 1–9, 1–11
determining controller numbers, 1–9
determining names, 1–10
disk drives
hot spare, 5–3, 5–9
icons, 4–11
marking as failed, 5–4
matrix, 4–12
number, 4–12
rebuilding, 5–6
replacing, failed, 5–2
SCSI id, 4–11
status, 4–13
viewing, 4–10

controller
numbers, 1–9
choosing, 4–7

disklabel, 1–14

current function selected, 4–3

drive group designator, 4–12

expansion slot, 4–7

AA–Q6TGD–TE

Index–1

StorageWorks RAID Array 200 Online Management Utility for Digital UNIX

filesystem, 1–14, 1–16

logical RAID drive
creating filesystem on, 1–16
creating partitions on, 1–14
drive group, 4–15
icon, 4–15
number, 4–15
preparing for access by Digital UNIX
filesystem, 1–14
size, 4–15
status, 4–15, 4–16
viewing, 4–14
write policy, 4–15

flags, 2–2
function selected, 4–3

G
graphical user interface (GUI), 1–1
accessing, 4–2
exiting, 4–6
files, 1–3
help, 4–6
loading, 1–4
menus, 4–6
parity check, 5–12
overview, 4–1
using, 4–4
window elements, 4–5

M
marking drive
failed, 5–4
optimal, 5–8

MAKEDEV, 1–11

hot spare, 5–3
making, 5–9

monitoring, 2–1
invoking monitor, 2–1
messages, 3–1
status update, 2–5

hot swap, 5–2, 5-3, 5-4

Motif, 4–4
mount device, 1–4

Installing the utility, 1–2, 1–4
sample, 1–5

mounting filesystem, 1–16

N
newfs, 1–16
notification of events
by monitor, 2–6
repeating, 2–7
to log files, 2–3
to mail, 2–4
to system console, 2–3

Index–2

AA–Q6TGD–TE

Index

P
parity
check, 5–12
check and repair, 2–7
flags, 2–7
messages, 3–8
notification of bad blocks, 2–8
repairing, 2–8, 5–12

subsystem
1-channel, 4–2, 4–10
3-channel, 4–3, 4–12
maintaining, 5–1
monitoring, 2–1
performing parity check, 2–7
selecting, 4–7
viewing from GUI, 4–7

partitions, 1–14 to 1–16

swxcr.asc, 4–17

printing configuration, 4–17

swxcrcheck, 1–3, 2–8

swxcrmgr, 1–3

RAID, 4–15, 5–3, 5–5
rebuilding, 5–6
rebuild rate
viewing, 2–10, 5–10
setting, 2–11, 5–10
reconfiguring the kernel, 1–12
replacing failed drive 5–2

S
setld utility, 1–4
-l, 1–4
-v, 1–7
StorageWorks RAID Array 200 Online
Management utility for Digital UNIX
files, 1–3
GUI overview, 4–1
loading, 1–4
installing, 1–2
post installation, 1–8
verifying installation, 1–7

AA–Q6TGD–TE

swxcrmon, 1–3, 2–1
automatic starting, 1–8
flags, 2–2
messages, 3–1
status update frequency, 2–5
swxcrrate, 1–3, 2–10, 2–11

V
viewing
bad block table, 5–11
disk drives, 4–10
logical RAID drive, 4–14
subsystem, 4–7
verifying installation, 1–7

X
x11startup, 1–8

Index–3

Reader’s Comments
Manual Order Number:
________________________
Digital is committed to providing the best products and services. Since our manuals are
important components of our products, we value your comments, corrections, and
suggestions for improvements. Please take a few minutes to fill out and return this form.
Attach additional sheets, if necessary. Thank you.
Manual Rating
Excellent
Accuracy (correct presentation of facts)
[ ]
Completeness (adequate information)
[ ]
Clarity (easy to understand)
[ ]
Organization (logical sequence of information) [ ]
Layout (easy to follow subject matter)
[ ]
Indexing (easy to locate desired information) [ ]

Good
[ ]
[ ]
[ ]
[ ]
[ ]
[ ]

Fair
[ ]
[ ]
[ ]
[ ]
[ ]
[ ]

Poor
[ ]
[ ]
[ ]
[ ]
[ ]
[ ]

Errors Noted (include page, paragraph, table or figure number) __________________
______________________________________________________________________
Most-Liked Features:____________________________________________________
______________________________________________________________________
Least-Liked Feature_____________________________________________________
______________________________________________________________________
Suggestions for Improvement _____________________________________________
______________________________________________________________________

Return Address:
Digital Equipment Corporation

Name

Phone

Title

Customer Research
Response Center
ATTN: Nan Andrews Amish
334 South Street, SHR3-2/W3

Company
Street Address
Mail Stop

Shrewsbury, MA 01545
City
Country (if other than USA)

State

ZIP