1999 Sun Microsystems,Inc., 901San AntonioRoad, Palo Alto,California 94303-4900U.S.A. Allrights reserved.
This product ordocument isprotected by copyrightand distributedunder licensesrestrictingits use,copying, distribution,and decompilation.
No part ofthis productor documentmaybe reproducedin anyformby anymeans withoutprior writtenauthorization of Sunand itslicensors,
if any.Third-party software, including fonttechnology,is copyrightedand licensed from Sun suppliers.
Parts of the product may be derived fromBerkeley BSD systems, licensed from the University of California. UNIX is a registeredtrademark in
the U.S. and other countries, exclusively licensed through X/Open Company,Ltd.
Sun, Sun Microsystems, the Sun logo, AnswerBook, docs.sun.com, StorEdge, OpenBoot, and Solaris are trademarks, registeredtrademarks, or
service marks of Sun Microsystems, Inc. in the U.S. and other countries. All SPARCtrademarks areused under license and are trademarks or
registeredtrademarks of SPARCInternational, Inc. in the U.S. and other countries. Products bearing SPARCtrademarks are based upon an
architecturedeveloped by Sun Microsystems, Inc.
The OPEN LOOK and Sun™ Graphical User Interface was developed by Sun Microsystems, Inc. for its users and licensees. Sun acknowledges
the pioneering efforts of Xerox in researching and developing the concept of visual or graphical user interfaces for the computer industry.Sun
holds a non-exclusive license from Xerox to the Xerox Graphical User Interface, which license also covers Sun’s licensees who implement OPEN
LOOK GUIs and otherwise comply with Sun’s written license agreements.
RESTRICTED RIGHTS: Use, duplication, or disclosure by the U.S. Government is subject to restrictions of FAR52.227-14(g)(2)(6/87) and FAR
52.227-19(6/87), or DFAR 252.227-7015(b)(6/95) and DFAR227.7202-3(a).
DOCUMENTATIONIS PROVIDED “AS IS” AND ALL EXPRESS OR IMPLIED CONDITIONS, REPRESENTATIONSAND WARRANTIES,
INCLUDING ANY IMPLIED WARRANTYOF MERCHANTABILITY,FITNESS FOR A PARTICULARPURPOSE OR NON-INFRINGEMENT,
ARE DISCLAIMED, EXCEPT TO THE EXTENT THATSUCH DISCLAIMERS ARE HELD TO BE LEGALLYINVALID.
Copyright 1999 Sun Microsystems, Inc., 901 San Antonio Road, Palo Alto, Californie 94303 Etats-Unis. Tousdroits réservés.
Ce produit ou document est protégé par un copyright et distribué avec des licences qui en restreignent l’utilisation, la copie, la distribution, et la
décompilation. Aucune partie de ce produit ou document ne peut être reproduite sous aucune forme, par quelque moyen que ce soit, sans
l’autorisation préalable et écrite de Sun et de ses bailleurs de licence, s’il y en a. Le logiciel détenu par des tiers, et qui comprend la technologie
relativeaux polices de caractères, est protégé par un copyright et licencié par des fournisseurs de Sun.
Des parties de ce produit pourront être dérivées des systèmes Berkeley BSD licenciés par l’Université de Californie. UNIX est une marque
déposée aux Etats-Unis et dans d’autres pays et licenciée exclusivement par X/Open Company,Ltd.
Sun, Sun Microsystems, le logo Sun, AnswerBook, docs.sun.com, StorEdge, OpenBoot, et Solaris sont des marques de fabrique ou des marques
déposées, ou marques de service, de Sun Microsystems,Inc. aux Etats-Unis et dans d’autres pays. Toutesles marques SPARCsont utilisées sous
licence et sont des marques de fabrique ou des marques déposées de SPARCInternational, Inc. aux Etats-Unis et dans d’autres pays. Les
produitsportant les marques SPARCsont basés sur une architecture développée par Sun Microsystems, Inc.
L’interfaced’utilisation graphique OPEN LOOK et Sun™ a été développée par Sun Microsystems, Inc. pour ses utilisateurs et licenciés. Sun
reconnaîtles efforts de pionniers de Xerox pour la recherche et le développement du concept des interfaces d’utilisation visuelle ou graphique
pour l’industrie de l’informatique. Sun détient une licence non exclusive de Xerox sur l’interface d’utilisation graphique Xerox,cette licence
couvrant également les licenciés de Sun qui mettent en place l’interface d’utilisation graphique OPEN LOOK et qui en outrese conforment aux
licences écrites de Sun.
CETTE PUBLICATION EST FOURNIE "EN L’ETAT"ET AUCUNE GARANTIE, EXPRESSE OU IMPLICITE, N’EST ACCORDEE, Y COMPRIS
DES GARANTIES CONCERNANT LA VALEURMARCHANDE, L’APTITUDEDE LA PUBLICATION A REPONDRE A UNE UTILISATION
PARTICULIERE,OU LE FAITQU’ELLE NE SOIT PASCONTREFAISANTEDE PRODUIT DE TIERS. CE DENI DE GARANTIE NE
S’APPLIQUERAIT PAS,DANS LA MESURE OU IL SERAIT TENU JURIDIQUEMENT NUL ET NON AVENU.
Please
Recycle
Contents
Purpose of Alternate Pathing1
Basic Alternate Pathing Concepts4
Physical Path4
Meta-Disk5
Meta-Network6
Disk Pathgroup7
Network Pathgroup8
Supported Devices and Software Versions9
Example AP Configurations10
AP and Domains11
Managing Copies of the Database13
Locating Databases to Maximize RAS14
Creating and Deleting Databases14
▼To Create a Copy of the AP Database15
▼To Delete a Copy of the AP Database15
Device Nodes for Meta-Disks19
Automatic Switching of Meta-Disks20
Disk Availability and Performance Trade-Offs22
Disk Mirroring Considerations23
Working With Disk Pathgroups and Meta-Disks27
▼To Create a Disk Pathgroup and Meta-Disk27
▼To Switch From the Primary Path to the Alternate Path31
▼To Switch Back to the Primary Path32
▼To Delete Disk Pathgroups and Meta-Disks32
▼To Deconfigure a Meta-Disk34
▼To Reconfigure a Meta-Disk34
Placing the Boot Disk Under AP Control37
▼To Place a Boot Disk under AP Control37
▼To Alternately Path a Mirrored Boot Disk39
▼To Remove a Mirrored Boot Disk From AP Control40
▼To Remove the Boot Disk From AP Control40
AP Boot Sequence40
Using Single-User Mode41
Meta-Network Interfaces43
Working With Network Pathgroups44
▼To Create a Network Pathgroup and Meta-Network44
▼To Switch a Network Pathgroup48
▼To Delete a Network Pathgroup and Meta-Network49
▼To Deconfigure a Meta-Network49
▼To Reconfigure a Meta-Network50
ivSun Enterprise Server Alternate Pathing User’s Guide • May 1999
Alternately Pathing the Primary Network Interface51
▼To Create a Network Pathgroup and Meta-Network for the Primary
Network52
▼To Delete the Network Pathgroup and Meta-Network for the Primary
Network53
▼To Deconfigure the Meta-Network for the Primary Network54
▼To Reconfigure the Meta-Network for the Primary Network55
Using AP and DR Together57
Maintaining the Correct AP State59
Contentsv
viSun Enterprise Server Alternate Pathing User’s Guide • May 1999
Figures
FIGURE 1-1Alternately Pathed I/O Device2
FIGURE 1-2Switching Paths After an I/O Controller Failure 3
FIGURE 1-3Switching Paths for a DR Detach Operation 4
FIGURE 1-4Physical Path 5
FIGURE 1-5Meta-Disk Example 6
FIGURE 1-6Meta-Network 7
FIGURE 1-7Disk Pathgroup 8
FIGURE 1-8Network Pathgroup 9
FIGURE 1-9Typical AP Configuration 10
FIGURE 1-10AP and Disk Mirroring 11
FIGURE 3-1System Boards and Disk Controllers 23
FIGURE 3-2System Boards and Controllers 24
FIGURE 3-3Mirrored Volumes Example 1 24
FIGURE 3-4Mirrored Volumes Example 2 25
FIGURE 3-5Mirrored Volumes Example 3 26
Figuresvii
viiiSun Enterprise Server Alternate Pathing User’s Guide • May 1999
Preface
The Sun Enterprise Server Alternate Pathing User’s Guide describes the Alternate
Pathing (AP) component of the Sun™ Enterprise™ server product line. Some AP
features apply only to the Sun Enterprise 10000 server. These features are noted
throughout this guide.
How This Book Is Organized
This guide contains the following chapters:
Chapter 1 introduces AP.
Chapter 2 covers the AP database operations.
Chapter 3 describes meta-disks and disk pathgroups, and explains how to use them.
Chapter 4 covers unattended system boot issues.
Chapter 5 describes meta-networks and network pathgroups, and explains how to
use them.
Chapter 6 describes how Dynamic Reconfiguration (DR) and AP work together.
Appendix A provides a list of all AP commands.
Appendix B provides an overview of the underlying AP architecture.
Appendix C provides an overview of the underlying AP drivers.
ix
Before You Read This Book
This manual is intended for the Sun Enterprise system administrator, who should
have a working knowledge of UNIX® systems, particularly those based on the
Solaris™ operating environment. If you do not have such knowledge, you should
first read the Solaris User and System Administrator AnswerBook™ documentation
provided with this system, and consider UNIX system administration training.
Using UNIX Commands
This document does not contain information on basic UNIX commands and
procedures such as shutting down the system, booting the system, and configuring
devices.
See one or more of the following for this information:
■ AnswerBook online documentation for the Solaris software environment,
particularly those dealing with Solaris system administration
■ Other software documentation that you received with your system
xSun Enterprise Server Alternate Pathing User’s Guide • May 1999
Typographic Conventions
TABLEP-1Typographic Conventions
Typeface
orSymbolMeaningExamples
AaBbCc123The names of commands, files,
and directories; on-screen
computer output.
AaBbCc123What you type, when
contrasted with on-screen
computer output.
AaBbCc123Book titles, new words or terms,
words to be emphasized.
Command-line variable; replace
with a real name or value.
Edit your .login file.
Use ls -a to list all files.
% You have mail.
% su
Password:
Read Chapter 6 in the User’s Guide.
These are called class options.
You must be root to do this.
To delete a file, type rm filename.
Shell Prompts
TABLEP-2Shell Prompts
ShellPrompt
C shellmachine_name%
C shell superusermachine_name#
Bourne shell and Korn shell$
Bourne shell and Korn shell superuser#
Sun Enterprise 6x00, 5x00, 4x00, 3x00 Dynamic Reconfiguration
User’s Guide
Sun Enterprise 10000 Dynamic Reconfiguration Reference Manual
Sun Enterprise 10000 SSP 3.1.1 User’s Guide
Sun Enterprise 10000 SSP 3.1.1 Reference Manual
Sun Documentation on the Web
The docs.sun.comsmweb site enables you to access Sun technical documentation
on the Web. You can browse the docs.sun.com archive or search for a specific book
title or subject at:
http://docs.sun.com
Sun Welcomes Your Comments
We are interested in improving our documentation and welcome your comments
and suggestions. You can email your comments to us at:
docfeedback@sun.com
Please include the part number of your document in the subject line of your email.
xiiSun Enterprise Server Alternate Pathing User’s Guide • May 1999
CHAPTER
1
Introduction to Alternate Pathing
This chapter describes the basic purpose of Alternate Pathing and provides an
overview of Alternate Pathing concepts and terminology.
Purpose of Alternate Pathing
Alternate Pathing (AP) supports high availability of I/O controllers—the hardware
components that reside on system boards and enable the Sun Enterprise server to
communicate with I/O devices such as disks and networks. With AP, each I/O
device connects to two I/O controllers.
1
Sun Enterprise Ser ver
I/O
I/O controller
Active path
(I/O flows
here)
Passive path
(no I/O)
I/O device
FIGURE 1-1 Alternately Pathed I/O Device
System board
The I/O controllers are part of two separate electrical pathways to the I/O device,
known as alternate paths. AP enables you to set up and use alternate paths on the Sun
Enterprise servers.
There are two purposes for AP. One purpose is to help protect against I/O controller
failures. With AP, if one I/O controller fails, you can switch to the alternate
controller.
2Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
I/O controller
problem
Active path
(unavailable)
Sun Enterprise Ser ver
I/O
FIGURE 1-2 Switching Paths After an I/O Controller Failure
Sun Enterprise Server
I/O
New active
path
For disk controllers, this switch occurs automatically whenever a path failure is
detected during normal operation. For network controllers, you must manually
switch paths (using a single AP command).
The second purpose of AP is to support Dynamic Reconfiguration (DR). DR is used
to logically attach and detach system boards from the operating system without
having to halt and reboot. For example, with DR you can detach a board from the
operating system, physically remove and service the board, and then re-insert the
board and attach it to the operating system again. You can do all of this without
halting the operating system or terminating any user applications.
If you want to detach a board that is connected to an I/O device, and if that I/O
device is alternately pathed, you can first use AP to redirect the I/O flow to a
controller on a different board. You can then use DR to detach the system board
without interrupting the I/O flow. On the Sun Enterprise 10000, the switch occurs
automatically during the DR operation (for both disk and network devices),
assuming a viable alternate controller exists on another board. The following figure
shows the relationship between AP and DR.
Chapter 1Introduction to Alternate Pathing3
Sun Enterprise Server
Sun Enterprise Ser ver
Active path
(unavailable)
I/O
DR Detach,
and hot-swap
FIGURE 1-3 Switching Paths for a DR Detach Operation
I/O
Active path
(new)
Basic Alternate Pathing Concepts
This section discusses basic AP concepts and introduces the terminology that is used
throughout this chapter.
Physical Path
For the purposes of AP, an I/O device is either a disk or network. An I/O controller is
the controller card for an I/O device. An I/O port is a connector on a controller card.
(Often there are two ports per controller card.) A device node is a path in the devices
directory that is used to specify a physical device, for example,
/dev/dsk/c0t0d1s0. The term physical path refers to the electrical path from the
host to a disk or network.
4Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Sun Enterprise Ser ver
I/O
System board
and I/O controllers
I/O device
Physical path
FIGURE 1-4 Physical Path
You reference a physical device by means of a device node, for example, /dev/dsk/
c0t1d1s0.
Meta-Disk
A meta-disk, as illustrated in FIGURE 1-5, is a construct that enables you to access a
disk by using either of two physical paths without having to reference either path
explicitly within your scripts and programs. You reference a meta-disk (in your
scripts and programs) using an AP-specific device node such as /dev/ap/dsk/mc0t1d1s0. (See “Device Nodes for Meta-Disks” on page 19 for more information.)
In the following figure, an AP-specific device node is used to perform disk I/O,
regardless of which pln port (pln2 or pln9) is currently handling I/O.
Chapter 1Introduction to Alternate Pathing5
Sun Enterprise Server
/dev/ap/dsk/mc0t1d1s0
System boards
pln2
I/O
FIGURE 1-5 Meta-Disk Example
SSA
pln9
SPARCstorage™
Array (SSA) controllers
with one or two PLN
ports. (Black PLN ports
are connected to the SSA.)
SSA port
Meta-Network
A meta-network, as illustrated in FIGURE 1-6, is a construct that enables you to access a
network by using either of two physical paths without having to reference either
path explicitly within your scripts and programs. You reference a meta-network (in
your scripts and programs) using a meta-network interface name such as mle1. (See
“Meta-Network Interfaces” on page 43 for more information.)
In the following figure, mle1 is used to access a meta-network, regardless of which
controller (le1 or le6) is currently processing I/O for the meta-network.
6Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Sun Enterprise Ser ver
mle1
System boards
and controllers
le0
le1
le2
le3
le4
le5
le6
le7
I/O
Network
FIGURE 1-6 Meta-Network
Disk Pathgroup
A disk pathgroup, as illustrated in FIGURE 1-7, consists of two physical paths leading to
the same disk array. When a physical path is part of a pathgroup, it is called an
alternate path. An alternate path to a disk can be uniquely identified by the pln port
or sf port that the alternate path uses. Only one alternate path at a time handles
disk I/O. The alternate path that is currently handling I/O is called the activealternate.
Note that whereas a meta-disk provides a means for you to access a disk
(in your scripts and programs), a disk pathgroup provides a means for you to
manipulate the path to that disk (when you run AP commands). For example, to
perform a switch operation (that is, change the active alternate from one alternate
path to another), you reference a disk pathgroup within an apconfig(1M)
command.
One of the alternate paths is designated as the primary path. The primary path is
initially the active alternate. Although the active alternate changes when you
perform a switch operation, the primary path remains constant. You reference a disk
pathgroup by specifying the pln port (for example, pln1)orsf port (for example,
sf1) that corresponds to the primary path. (For information about determining the
pln or sf port name, see “Device Nodes for Meta-Disks” on page 19.)
For example, the following figure shows the results of using the apconfig(1M)
command to switch the active alternate of a disk pathgroup.
Chapter 1Introduction to Alternate Pathing7
Sun Enterprise Ser ver
I/O
System boards
and controllers
Switch
I/O
Alternate path
(primary path)
Disk array
FIGURE 1-7 Disk Pathgroup
Alternate path
(active alternate)
You reference a disk pathgroup (for example, to switch from one path to another) by
specifying the primary path, for example, apconfig -P pln2 -a pln9.
Network Pathgroup
A network pathgroup, as illustrated in the following figure, consists of two network
controllers connected to the same physical network. The terms alternate path, activealternate, primary path, and switch have basically the same meaning as they do for
disk pathgroups.
To specify a network pathgroup, reference the corresponding meta-network interface
name, for example, mle1. (Meta-network interface names are described in “MetaNetwork Interfaces” on page 43.) For example,
the apconfig(1M) command to switch the active alternate of a network pathgroup.
FIGURE 1-8 shows the results of using
8Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Sun Enterprise Ser ver
System boards
and controllers
le0
le1
le2
le3
I/O
le4
le5
I/O
le6
le7
Alternate path
(primary path)
FIGURE 1-8 Network Pathgroup
You reference a network pathgroup (for example, to switch from one path to
another) by specifying the meta-network interface, for example, apconfig -Pmle1 -a le6.
Supported Devices and Software
Versions
AP 2.2 supports the Solaris 7 operating environment (32-bit and 64-bit modes).
AP supports the following disk arrays:
■ Sun StorEdge™ A5000 and A7000 disk arrays attached to sf ports
■ SPARCStorage Array™ (SSA) disk arrays attached to pln ports on FC-AL SBus
Switch
Host Adapters (SOC controllers)
Alternate path
(active alternate)
Network
The network devices and third party software products supported by AP are listed
in the Release Notes Supplement Solaris 7 5/99.
If you alternately path disks, and you also use a volume manager with those disks,
the disks must be known to the volume manager by their AP meta-disk names. This
requirement enables AP to switch the active path without interfering with the
volume manager.
Chapter 1Introduction to Alternate Pathing9
You can place the boot disk and the primary network interface under AP control. AP
makes it possible for the system to boot unattended even if the primary network or
boot disk controller is not accessible, as long as viable alternate paths for these
devices are defined.
Example AP Configurations
FIGURE 1-9 shows how you can use AP to support an Ethernet network and a disk
array.
Backplane
Board 1
SBus
I/F
FIGURE 1-9 Typical AP Configuration
SBus
Ethernet
Board 2
SBus
I/F
SBus
SBus controllers
Fibre channel
In this example, two network controllers—one each on Board 1 and
Board 2—are connected to the same network. Similarly, two SSA controllers on the
two boards are connected to the same SSA. In this situation, if Board 1 is detached
with a DR Detach operation, AP can switch usage to Board 2 without interfering
with any I/O operations that may be in progress.
AP is not equivalent to disk mirroring. Disk mirroring primarily achieves data
redundancy although two paths are available, one for each side of the mirror. AP
achieves true pathing redundancy, by making two paths available for each side of the
10Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
mirror. To use AP and disk mirroring together, you must configure your volume
manager software (such as Sun Enterprise Volume Manager
TM
) so that it uses the AP
meta-disk paths.
The following figure shows an example of how AP can be used in conjunction with
disk mirroring.
Backplane
Board 1
SBus
I/F
SSA
FIGURE 1-10 AP and Disk Mirroring
SBus
Mirrored
Board 2
SBus
I/F
Fibre
channels
SBus
SBus
controllers
SSA
This type of configuration enables you to switch the paths used to implement the
mirror from one board to another board, without disrupting the disk mirroring or
any active I/O.
AP and Domains
The Sun Enterprise 10000 server supports Dynamic System Domains, or simply
“domains.” AP cannot be used across two domains. For example, suppose a board
contains a controller that is part of a pathgroup, and you move that board into a
different domain using DR. (You can do this only if the alternate path on that board
is not currently active.) In this case, you can no longer switch to the alternate path
on that board.
Chapter 1Introduction to Alternate Pathing11
12Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
CHAPTER
2
Alternate Pathing Database
This chapter describes how to create and manage the AP database which maintains
the state of the AP configuration.
Managing Copies of the Database
AP maintains a database that contains information about all defined meta-disks and
meta-networks, and their corresponding alternate paths and properties. One set of
data is maintained for each domain on the Sun Enterprise 10000 server. On other
Sun Enterprise servers, one set of data is maintained for the entire machine. You
should always set up multiple copies of the database. In this way, if a given database
copy is not accessible or becomes corrupted, AP can automatically begin to use a
current, noncorrupted database copy.
You must dedicate an entire disk partition, one that has at least 300 Kbytes, to each
database copy. You can use larger partitions, but doing so wastes disk space. Keep
the following information in mind when selecting disk partitions for the AP
database:
■ You should set up three to five database copies.
■ As configured at the factory, partition four of the root disk is appropriately sized
for the AP database and is not allocated for any other purposes. This partition is
a good choice for an AP database copy, assuming you are not using it for other
purposes.
■ The database copies should have no I/O controllers in common with each other.
Following this rule allows maximum availability in case of controller failure.
■ If you have configured your system to make use of Dynamic Reconfiguration
(DR), the database copies should be hosted by I/O controllers on different system
boards so that a database copy is accessible if one of the system boards is
detached.
13
■ If you want to place an AP database copy in a partition of an alternately pathed
disk, create two copies of the database using each of the physical paths utilized by
the AP meta-disk to access the partition. AP behaves as if two copies of the
database exist, when actually there is only one, since the disk is accessible via two
paths. The behavior does not result in database inconsistencies, however, since
AP always updates and accesses database copies sequentially. This behavior does
not result in performance problems since the AP database is not accessed
frequently.
On the Sun Enterprise 10000 server, a subset of the information in the AP database is
automatically maintained on the SSP for use at boot time. This database contains AP
information for the boot disk.
Locating Databases to Maximize RAS
You should consider how you expect to use the system boards that host the I/O
controllers for the disks where the AP databases will be stored. If you expect to
detach a board often, perhaps because you intend to migrate it between domains,
you should probably not place an AP database on any disk attached to a controller
hosted by that board. If you do find it necessary to detach such a board, error
messages will be sent to the console whenever AP attempts to write to the
unavailable database. This does not pose a serious problem. You can re-attach the
board at any time, and the stale database is re-synchronized immediately. However,
if you attach the board to other domains in the mean time, those domains may write
data to the slice that is reserved for the database.
Creating and Deleting Databases
Note – The following AP command examples assume that your command search
path includes the directory where the commands are installed. See “Using SingleUser Mode” on page 41.
14Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
▼ To Create a Copy of the AP Database
● Use apdb(1M) as follows:
# apdb -c /dev/rdsk/c0t1d0s4 -f
The -c option specifies the raw disk slice (under /dev/rdsk) where you want to
create the database copy. You must dedicate an entire disk partition to each database
copy. The disk partition must have at least 300 Kbytes.
The -f (force) option is only necessary to create the first AP database copy.
▼ To Delete a Copy of the AP Database
● Use apdb(1M) as follows:
# apdb -d /dev/rdsk/c0t1d0s4 -f
# apconfig -D
#
The -d option specifies the raw disk slice (under /dev/rdsk) where the copy of the
database that you want to delete is located.
The -f (force) option is only necessary to delete the second-to-last copy and the last
copy of the database.
In this example, apconfig -D is used after the deletion operation to view
information about the existing AP database copies. Since no information is returned,
the apdb(1M) command must have deleted the last database copy.
Note – If you delete the last copy of the AP database, the contents of the database
still resides in memory (including all committed and uncommitted entries). If you
then create a new copy of the AP database, without rebooting in the interim, the new
database contains the same information as the previous database. However, if you
reboot after deleting the last copy of the database, the database information in
memory is lost. In this case, if you create a new copy of the database, it contains no
data.
Chapter 2Alternate Pathing Database15
Viewing Database Information
You can view information in the database, including information about the database
copies, the disk entries within the database, and the network entries within the
database.
▼ To View Information About Database Copies
● Use apconfig -D as follows:
# apconfig -D
path: /dev/rdsk/c0t1d0s4
major: 32
minor: 12
timestamp: Thu Jul 27 16:24:27 1995
checksum: 687681819
corrupt: No
inaccessible: No
In this example, there is only one AP database. The command shows the path to this
database, along with its major number, minor number, timestamp, and checksum.
The corrupt field indicates whether the database is corrupt. (If corrupt is set to
Yes, the data did not validate properly against the checksum.) The inaccessible
field indicates whether the device that holds the database can be accessed.
Viewing Pathgroup Information
The AP database contains information about disk and network pathgroups. When a
pathgroup is initially defined (as described in Chapter 3 and Chapter 5), its
pathgroup definition is an uncommitted database entry. The meta-disk or metanetwork associated with an uncommitted entry is not available until the pathgroup
definition is committed. Conversely, when a pathgroup definition is deleted, the
deletion must be committed before it takes effect. The two states (uncommitted and
committed) enable you to review the effects of an operation before allowing the
16Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
operation to proceed. To commit the uncommitted database entries, use the
apdb -C command. Note that uncommitted entries remain in the database
indefinitely, until you either commit them or remove them.
▼ To View Uncommitted Disk Entries
● Use apconfig(1M) with the -S and -u options as follows, where S stands for
storage and u stands for uncommitted:
# apconfig -S -u
c1 sf0 P A
c2 sf1
metadiskname(s):
mc1t5d0 U
mc1t4d0 U
mc1t3d0 U
mc1t2d0 U
mc1t1d0 U
mc1t0d0 U
For more information see Chapter 3.
▼ To View Committed Disk Entries
● Use apconfig(1M) with the -S option, as follows, where S stands for storage:
# apconfig -S
c1 pln0 P A
c2 pln1
metadiskname(s):
mc1t5d0 R
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
mc1t0d0
For more information see Chapter 3.
Chapter 2Alternate Pathing Database17
▼ To View Uncommitted Network Entries
● Use apconfig(1M) with the -N and -u options, as follows, where N stands for
network and u stands for uncommitted:
# apconfig -N -u
metanetwork: mle0 U
physical devices:
le2
le0 P A
For more information see Chapter 5.
▼ To View Committed Network Entries
● Use apconfig(1M) with the -N option, as follows:
# apconfig -N
metanetwork: mle3
physical devices:
le4
le3 P A
For more information see Chapter 5.
18Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
CHAPTER
3
Using Meta-Disks and Disk
Pathgroups
You can create meta-disks and disk pathgroups only for disks that are accessible
through two paths. You should generally use two separate controllers on different
system boards.
Note – AP does not modify the data on a disk when that disk is placed under AP
control or when a pathgroup is deleted (except for the data on the slices that contain
AP database copies). AP does not repartition a disk. If a pathgroup is deleted, you
can continue to access the data by using its physical device name.
Device Nodes for Meta-Disks
Here are two examples of physical device nodes for disk devices:
■ /dev/dsk/c0t0d0s0
■ /dev/rdsk/c0t0d0s0
where:
c references the I/O port on the host (not the disk array)
t is the bus within the disk array
d is the target ID of the disk on that bus
s is the slice number on the disk
These physical device nodes represent a particular physical path to a partition on a
disk.
19
For SCSI disks:
■ The c number is the host adapter number.
■ The t number is the target number of a disk tray.
■ The d number is the disk number.
■ The s number is the slice number on the disk.
Each controller port has both a port number (such as c0) and a port name (such as
pln2 or sf3). The port name consists of the port type and instance number. See
/etc/path_to_inst for more information.
When a disk array is connected to two ports, it can potentially be accessed from
either path by using the physical device node, for example, /dev/dsk/c0t0d0s0 or
/dev/dsk/c1t0d0s0.
The device node for a meta-disk is derived from the physical device node of the
primary path for a pathgroup. Here are two examples of meta-disk device nodes:
■ /dev/ap/dsk/mc0t0d0s0
■ /dev/ap/rdsk/mc0t0d0s0
As you can see, an ap directory has been added, and an m (for “meta”) is prepended
to the device specification. The device node for a meta-disk has the ability to access
the underlying physical disk drive from multiple paths.
Automatic Switching of Meta-Disks
Meta-disks can be automatically switched from the active path to the alternate path
in two situations:
■ The active path fails.
■ The board containing the controller for the active path is detached by using a DR
Detach operation. (Automatic switching during a DR detach is available only on
the Sun Enterprise 10000 server.)
20Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
When the active path fails, an automatic switch is attempted only if an alternate path
is available. The failed path is then marked unavailable, or tried. You can identify
the tried paths with apconfig -S:
# apconfig -S
c1 pln0 P A
c2 pln1 T
metadiskname(s):
mc1t5d0
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
mc1t0d0
In this example, the currently inactive path, pln1, is marked with a T, which
indicates that path was tried but failed.
The tried flag is only significant for automatic switch operations (not for manual
switch operations). AP never attempts to automatically switch to a tried path. (This
prevents thrashing in the case that both paths may have failed.)
You can reset the tried flag with any of the following actions:
■ Rebooting the corresponding domain.
■ Performing a DR detach followed by a DR attach of a board that contains the
controller marked as tried.
■ Manually resetting the tried flag for a particular controller.
You can manually reset the tried flag as shown in this example:
# apdisk -w pln1
In this example, pln1 is a controller with the tried flag set to true. The apdisk -w
feature should be used judiciously. This command merely clears the tried flag; it
does not address any potential problems with the controller or device. This
command should only be used in situations where the failed path has been restored
without an intervening DR operation or reboot. Note that you can attempt to
perform a manual switch to a path that is marked as tried.
Chapter 3Using Meta-Disks and Disk Pathgroups21
Disk Availability and Performance
Trade-Offs
Before you configure your disk arrays and controllers, you need to establish your
disk usage priorities. You may be able to achieve greater availability of your disk
resources by trading off performance, or perhaps by investing in more hardware.
Consider a dual-ported SSA disk array. This type of device can be connected to
either one or two Fibre Channel disk controllers (SOC controllers). Within an SSA,
there are multiple targets. Each target contains multiple disks. And each disk is
divided into multiple slices. Depending upon how you set up your system, there can
be several different levels of contention for these disk I/O resources.
■ Disk contention
■ Target contention (I/O bus contention)
■ Controller contention
For example, suppose you divide a disk into four slices, and create a file system out
of the four slices. Although the file system spans several slices, those slices reside on
the same disk and you may as well have simply placed the file system on a single
slice. This is generally a poor configuration and it results in disk-level contention
since every read or write to that file system requires access to the same disk.
You can place a file system across several disks in the same target. In this situation
you have target-level contention, since each read or write to the file system must
access the same target. Target-level contention is not as severe as disk-level
contention, but it is still a poor configuration in most cases.
If you create a file system across too many targets in the same SSA, you will have
controller-level contention. This is because each read or write to the file system must
use the same controller.
Generally, it is best to create a file system across multiple SSA disk arrays (using
multiple controllers). However, there is a trade-off between disk access speed and
system availability. The more disk arrays that you use for your file systems, the
faster your disk access times will be. (This is true up to a point, which can be
determined empirically.) However, if any component fails in any of the disk arrays,
your file systems will become unavailable. If you limit the number of disk arrays for
a file system, perhaps to a single disk array, performance will decrease, but overall
system availability should increase. This is because there are fewer components that
could fail for a given file system.
Suppose you have six disk controllers connected to three dual-ported SSA disk
arrays.
22Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
System Boards and Disk Controllers
SSASSASSA
FIGURE 3-1 System Boards and Disk Controllers
If you want to maximize availability, you can alternately path each SSA using AP.
The advantage of this is that you can use DR to attach and detach system boards
(possibly to service or upgrade those boards) without losing access to the file
systems on the SSAs. Of course, you would want to place the alternate disk
controllers (SOC controllers) on different system boards. One arrangement is to use
two system boards with three disk controllers on each board. This is a simple and
clean arrangement, and it enables you to switch to the controllers on one of the
boards when you need to detach the other board. It also enables you to migrate the
disk resources among domains fairly easily (by detaching and attaching a single
board).
Note that two SOC controllers must be purchased for each SSA. Also, in very large
installations, it is possible to run into limitations in terms of the number of SBus
slots that are available to host all the SOC controllers needed to dual path a large
number of SSAs.
Disk Mirroring Considerations
If you use a product such as Sun Enterprise Volume Manager (SEVM) to mirror your
disks, and you also want to detach system boards with Dynamic Reconfiguration
(DR), you must configure your volumes and mirrors so that they will work properly
with AP and DR.
For example, suppose you have 12 system boards, each of which has one host
adapter (called a “controller” in the following diagram):
Chapter 3Using Meta-Disks and Disk Pathgroups23
System Board 0
Controller 0
System Board 1
Controller 1
System Board 2
Controller 2
System Board 11
Controller 11
FIGURE 3-2 System Boards and Controllers
You may need to create a mirrored volume. Consider the following configuration:
24Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
In this configuration, vol01-01 consists of a four-way slice that is accessed through
four separate controllers which reside on four separate system boards. vol01-01 is
mirrored to vol01-02 which also consists of a four-way slice. For example, Controller
0 Slice 0 is mirrored to Controller 3 Slice 1, and so forth.
Consider the situation where you need to detach a board that contains one of these
four controllers. Before you can detach the board, you must deactivate the half of the
mirror that uses controllers on that board. This is impossible with the configuration
shown above. For example, if you wish to detach Board 0 (which contains Controller
0), you would have to deactivate both sides of the mirror, which makes the file
system inaccessible. The result is that you cannot use DR on any of the boards in the
configuration shown above.
One way around this situation is to mirror your volumes so that controllers from the
same system board do not appear on both sides of the mirror, for example:
In the configuration above, you can detach any board (Board 0 through Board 5), by
first deactivating the half of the mirror that uses a controller on that board. For
example, to detach Board 4 (which hosts Controller 4), simply deactivate vol01-02
first. You do not lose access to the file system, since it is still available through vol01-
01. Later when you re-attach Board 4, you can add vol01-02 to the mirror again.
The problem with this solution is that your system is vulnerable to single points of
failure while the mirror is down. If a disk fails, there is no mirrored backup disk.
You can protect against this by using AP. You might set up the following AP metadevices:
■ mc0 is the meta-device for Controller 0 and Controller 6
■ mc1 is the meta-device for Controller 1 and Controller 7
■ mc2 is the meta-device for Controller 2 and Controller 8
■ mc3 is the meta-device for Controller 3 and Controller 9
■ mc4 is the meta-device for Controller 4 and Controller 10
■ mc5 is the meta-device for Controller 5 and Controller 11
Chapter 3Using Meta-Disks and Disk Pathgroups25
Abbreviations are used above to simplify this discussion. For example, the full metadevice name may be mc0t0d0s0, and it may encapsulate the physical devices
c0t0d0s0 and c6t0d0s0 as the alternate paths.
In this configuration, you can detach any of the boards (Board 0 to Board 11) without
taking down the mirror. This reduces your exposure to single points of failure. For
example, to detach Board 4, which contains Controller 4, you would first switch the
meta-device mc4 so that it uses Controller 10 on Board 10. (You do this with a single
AP command, apconfig -P.)
In this example, as you increase the level of RAS support (that is, the availability of
disk I/O resources and the level of protection against single points of failure), you
have to increase the number of controllers and boards in the configuration. This
amounts to an increase in the cost of the system to provide greater support for RAS
features.
This is a hypothetical example. The main point is that you must consider AP and DR
when you set up your volumes and mirrors. Otherwise, you may end up in a
situation where you cannot use AP and DR. If you use Sun Enterprise Volume
Manager (SEVM), keep track of the physical controllers and slices that make up your
volumes. SEVM can be used in such a way that it chooses the physical components
automatically, but this selection process is not aware of AP and DR considerations.
You should explicitly choose the physical components that make up your volumes if
you want to assure compatibility with AP and DR.
26Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Working With Disk Pathgroups and
Meta-Disks
Note – The example commands in this section use pln ports (for SSA disk arrays).
If you have Sun StorEdge
ports are shown. If you have Sun StorEdge A7000 disk arrays, you specify isp ports
wherever pln ports are shown.
TM
A5000 disk arrays, you specify sf ports wherever pln
▼ To Create a Disk Pathgroup and Meta-Disk
1. Decide which two ports will make up the alternate paths for the pathgroup.
a. You can use the apinst(1M) command to display all ports (for example, pln0
and pln1) and their disk device nodes (for example, /dev/dsk/c1t0d0):
b. You must know your system hardware configuration to recognize when two
ports are connected to the same disk array.
In this example, it is assumed that the SSA contains six disks and two SSA ports.
One SSA port is connected to pln port c1, and the other SSA port is connected to
pln port c2.
Chapter 3Using Meta-Disks and Disk Pathgroups27
2. Use apdisk(1M) with the -c , -p, and -a options to create an uncommitted disk
pathgroup:
# apdisk -c -p pln0 -a pln1
where:
-p specifies the primary path
-a specifies the alternate path
-c specifies that this information is to be created.
This apdisk(1M) command creates a meta-disk name, as well as all of the necessary
information in the AP database for maintaining the two alternate paths for all six
disks.
3. Verify the results:
# apconfig -S -u
c1 pln0 P A
c2 pln1
metadiskname(s):
mc1t5d0 U
mc1t4d0 U
mc1t3d0 U
mc1t2d0 U
mc1t1d0 U
mc1t0d0 U
The apconfig -S -u command lists the uncommitted meta-disks. -S lists storage
devices only (that is, disks rather than networks). -u lists uncommitted devices only.
The U next to each meta-disk name indicates that the meta-disk entry is
uncommitted.
The P next to pln0 indicates that pln0 is the primary path, and the A indicates that
pln0 is the active alternate. Although you can change the active alternate, the
primary path always remains constant. The significance of the primary path is that it
is initially the active alternate, it is used when the meta-disk is named, and it is used
to identify the meta-disk. In this case, c1t0d0 from the primary path name becomes
part of mc1t0d0 in the meta-disk name.
4. If you are satisfied with the results shown in the previous step, use apdb(1M)
with the -C option to commit the uncommitted database entries:
# apdb -C
28Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
5. Before continuing, verify the results by using apconfig -S to view the committed
storage entries in the database:
# apconfig -S
c1 pln0 P A
c2 pln1
metadiskname(s):
mc1t5d0
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
mc1t0d0
If a partition is currently mounted under a physical path name, it should be
unmounted and remounted under the meta-disk path name.
If you do not want to unmount a partition, perhaps because it is heavily used, you
can delay placing the partition under AP control until you are ready to bring the
system down for maintenance and then reboot. In this scenario, you modify the
/etc/vfstab file so that when the system is rebooted, the partition comes up
under an AP device. (If you are placing the boot disk under AP control, you also
need to modify /etc/vfstab using apboot(1M) as described in Chapter 4.)
The apconfig -S command lists the committed storage entries in the database. As
shown in the example, this listing is exactly the same as the previous listing, except
the U no longer appears after the meta-disk names, indicating that the meta-disks are
no longer uncommitted.
6. Use drvconfig(1M):
# drvconfig -i ap_dmd
The drvconfig command rebuilds the devices directory, which represents the
device tree in the kernel. The AP disk meta-driver is a pseudo-device.
Chapter 3Using Meta-Disks and Disk Pathgroups29
7. Use the following command to verify the results:
# ls /devices/pseudo/ap_dmd*
/devices/pseudo/ap_dmd@0:128,blk
/devices/pseudo/ap_dmd@0:128,raw
/devices/pseudo/ap_dmd@0:129,blk
/devices/pseudo/ap_dmd@0:129,raw
/devices/pseudo/ap_dmd@0:130,blk
/devices/pseudo/ap_dmd@0:130,raw
...
As you can see from the listing, drvconfig created minor nodes for the alternately
pathed device.
8. Use apconfig(1M) with the -R option to create symbolic links from the devices
directories, /dev/ap/dsk and /dev/ap/rdsk, to the meta-disk special files
under /devices/pseudo:
# apconfig -R
9. Use the following command to view the symbolic links and verify the results:
The device nodes that you need—under /dev/ap/dsk as well as /dev/ap/rdsk—
are now ready to be used.
10. Modify every reference that uses a physical device node (that is, a path that
begins with /dev/dsk or /dev/rdsk) to use the corresponding meta-disk device
node (that is, a path that begins with /dev/ap/dsk or /dev/ap/rdsk).
30Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
▼ To Switch From the Primary Path to the
Alternate Path
Note – You can perform a switch at any time, even while I/O is occurring on the
device. It is a good idea to experiment with the switching process to verify that you
understand it and that your system is set up properly, rather than wait until a critical
situation occurs.
Caution – When you switch paths, AP does not check to verify that data can be
transferred over the path to which you are switching (although it does determine
whether or not that path is detached or offline). You can verify the status of the path
before switching to it by performing an I/O operation such as prtvtoc(1M).AP
does not produce any error or warning messages if you switch to a path that is not
functioning properly. If you switch to a non-functioning path for your boot disk,
your system may crash if the path is not switched back immediately.
1. Use the apconfig -S command to view the current configuration:
# apconfig -S
c1 pln0 P A
c2 pln1
metadiskname(s):
mc1t5d0
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
In this example, pln0 is the active alternate since it is followed by an A. (It is also
the primary path, since it is followed by a P.)
2. To perform the switch, use apconfig(1M) with the -P and -a options:
# apconfig -P pln0 -a pln1
-P specifies the primary path and thereby identifies the pathgroup for which you want
to change the active alternate. Thus, -P pln0 in the example above identifies the
pathgroup for which pln0 is the primary path. -a specifies the alternate that you
want to make active.
Chapter 3Using Meta-Disks and Disk Pathgroups31
3. You can verify the results using apconfig(1M) with the -S option to view the
committed meta-disks in the database:
# apconfig -S
c1 pln0 P
c2 pln1 A
metadiskname(s):
mc1t5d0
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
The active alternate has been switched to pln1.
Note that you do not have to commit a switch operation.
▼ To Switch Back to the Primary Path
● You can switch back to the primary path using the following commands:
# apconfig -P pln0 -a pln0
# apconfig -S
c1 pln0 P A
c2 pln1
metadiskname(s):
mc1t5d0
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
The first apconfig command, above, switches the active alternate for the pathgroup
that has the primary controller pln0. The active alternate becomes pln0.
▼ To Delete Disk Pathgroups and Meta-Disks
1. Convert meta-disk references to physical device references.
32Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
a. If your boot disk is under AP control, use apboot(1M) to remove it from AP
control, as described in “To Remove the Boot Disk From AP Control” on
page 40.
You do not need to unmount any file systems mounted from the boot disk, since
apboot(1M) places those file systems on top of physical devices without
requiring you to unmount them.
b. Unmount any file systems that are built on top of AP meta-disks (other than file
systems mounted from the boot disk).
c. Your scripts and programs may contain references to meta-disks of the form
/dev/ap/dsk/mc?t?d?s? and /dev/ap/rdsk/mc?t?d?s?. These references
must be converted to references of the form /dev/dsk/c?t?d?s? and
/dev/rdsk/c?t?d?s?, respectively.
Typically, references to meta-disks exist in the following locations:
■ /etc/vfstab
■ /etc/system
■ /etc/dumpadm.conf
■ Any application or script that references disks
2. Use apdisk(1M) with the -d option to specify the primary path of the path group
you intend to delete:
# apdisk -d pln0
3. To verify the results, use apconfig(1M) with the -S option to view the
committed disk entries in the database:
# apconfig -S
c1 pln0 P A
c2 pln1
metadiskname(s):
mc1t5d0 D
mc1t4d0 D
mc1t3d0 D
mc1t2d0 D
mc1t1d0 D
mc1t0d0 D
If the pathgroup was not previously committed, the apdisk -d command deletes it
from the database. However, if the pathgroup was previously committed, the
apdisk -d command simply marks it as deleted, but the deletion is not completed
Chapter 3Using Meta-Disks and Disk Pathgroups33
until the next time you commit the entries in the database. In the example above, the
pln0 pathgroup was previously committed, so the letter D indicates that it is marked
for deletion.
4. Use apdb(1M) to commit the database entries, thereby completing the deletion:
# apdb -C
5. You can verify the deletion with apconfig -S:
# apconfig -S
Note – You can undo a deletion if the deletion is uncommitted. To undo a deletion,
use apdisk -z, specifying the same port that you previously specified.
▼ To Deconfigure a Meta-Disk
● Your scripts and programs may contain references to meta-disks of the form
/dev/ap/dsk/mc?t?d?s? and /dev/ap/rdsk/mc?t?d?s?. These references
must be converted to references of the form /dev/dsk/c?t?d?s? and
/dev/rdsk/c?t?d?s?, respectively.
Typically, references to meta-disks exist in the following locations:
■ /etc/vfstab
■ /etc/system
■ /etc/dumpadm.conf
■ Any application or script that references disks
▼ To Reconfigure a Meta-Disk
● Convert physical device references back into meta-disk references.
34Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Note – This procedure assumes that you previously created a disk pathgroup and
meta-disk, and subsequently deconfigured the meta-disk references. If you simply
want to reconfigure the meta-disk interface, use this procedure.
Your scripts and programs may contain references to physical devces of the form
/dev/ap/dsk/mc?t?d?s? and /dev/ap/rdsk/mc?t?d?s?, respectively.
Typically, references to disk devices exist in the following locations:
■ /etc/vfstab
■ /etc/system
■ /etc/dumpadm.conf
■ Any application or script that references disks
Chapter 3Using Meta-Disks and Disk Pathgroups35
36Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
CHAPTER
4
Using AP Boot Devices
This chapter describes how you can alternately path the boot disk.
Placing the Boot Disk Under AP Control
On the Sun Enterprise 10000 server, you can allow for unattended system boot, even
if the controller for the boot disk fails, by placing your the boot disk under AP
control.
On all Sun Enterprise servers, you can make it possible to use Dynamic
Reconfiguration (DR) to detach a system board, even if that system board hosts a
controller for the boot disk. To do this, you must alternately path the boot disk using
controllers from two different system boards. Note, however, if a controller for the
primary network is hosted on the same system board as a controller for the boot
disk, you must alternately path the primary network as well. Otherwise, you will
not be able to use DR to detach that board.
▼ To Place a Boot Disk under AP Control
1. Create an AP pathgroup for the boot disk.
This process is described in Chapter 3.
37
2. Use apboot(1M) to define the new AP boot device.
apboot(1M) modifies /etc/vfstab and /etc/system. For example:
# apboot mc2t0d0
where mc2t0d0 is the meta-disk name of the boot disk. The apboot(1M) command
examines /etc/vfstab and replaces the physical device name of the disk (for
example, /dev/dsk/c2t0d0* or /dev/dsk/c1t0d0*) with the meta-disk name
(for example, /dev/dsk/mc2t0d0*). The apboot(1M) command also edits /etc/system so that the kernel drivers required for AP boot disk usage are loaded at the
proper time.
Do not manually replace the physical devices in /etc/vfstab with meta-disks for
the boot disk. Instead, use apboot(1M) to ensure that all required changes are
made.
In addition, apboot(1M) checks /etc/vfstab to determine if the swap device is to
be changed to a meta-device. If so, it issues the appropriate apboot(1M)
commands. Similarly, apboot(1M) checks the configuration of the dump device,
and calls dumpadm(1M) if necessary to configure the dump device as a metadevice.
3. Set the OpenBoot™ PROM (OBP) devalias variable boot-device to the physical
path most likely to be used for booting.
For example:
ok setenv boot-device \
/sbus@68,0/SUNW,soc@0,0/SUNW,pln@a0000000,78cab4/ssd@0,2
4. Define a devalias for the alternate boot device path as a convenience in case you
need to perform a manual boot.
At this point, reboot the system to begin using the AP boot device.
Normally, the file systems that are mounted as part of the boot process are split
across two separate disks (because of disk space requirements). If you place the boot
disk under AP control (using apboot(1M)), you must manually edit the /etc/vfstab file to also place other file systems that are mounted during the boot process
under AP control. In the /etc/vfstab file, you must change the device to mount
and device to fsck paths for all other mount points that you want to place under
AP control.
If you want to create a new AP database copy after you have placed the boot disk
under AP control, and that database copy is to be located on a partition controlled
by a controller port that does not control any of the current AP database partitions,
38Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
you must first remove the boot disk from AP control. Make sure that the new AP
database has been created. Then, place the boot disk under AP control again. Failure
to follow this procedure may cause the database to become inaccessible during boot.
▼ To Alternately Path a Mirrored Boot Disk
Mirroring the boot disk is primarily a function of your disk management software.
The purpose of this procedure is to notify AP about a mirrored boot disk. When you
use mirrored boot disks that are also alternately pathed, you have four potential
physical paths to the boot disk, two for each side of the mirror. (This is the suggested
configuration if you want to maximize protection against controller failure.) There
are two benefits to performing the following procedure:
■ Once you perform the following procedure, AP assures that the appropriate
alternate path is always designated as the active path, even if you boot using a
different boot device path.
■ On the Sun Enterprise 10000 server (which supports automatic switching of the
boot device path at boot time when an error is detected with the currently active
boot device path), the following procedure assures that all four paths are
available as alternates should an automatic switch be required at boot time.
1. Place the boot disk under AP control, as described in “To Place a Boot Disk under
AP Control.
2. Create an AP pathgroup for the mirror of the boot disk.
This process is described in Chapter 3.
3. Notify AP about the boot disk mirror.
# apboot -m mc3t0d0
In this example, mc3t0d0 is the meta-disk for the mirror of the boot disk.
4. Create a mirror of your boot disk (using the two meta-disks) with your disk
management software.
Chapter 4Using AP Boot Devices39
▼ To Remove a Mirrored Boot Disk From AP
Control
● Use apboot(1M) to undefine the AP mirrored boot device.
# apboot -u mc3t0d0
▼ To Remove the Boot Disk From AP Control
● Use apboot(1M) to specify an appropriate physical device node.
# apboot c2t0d0
In the above command, c2t0d0 is the physical device node of an alternate path for
the boot disk (as currently specified in /etc/vfstab). apboot also edits the
/etc/system file to remove force loading of AP kernel driver modules, since they
are no longer needed when the boot disk is not an AP device. In addition,
apboot(1M) reconfigures swap and dump devices to use appropriate alternate
paths if necessary.
Caution – If you place the boot disk under AP control, and later decide to remove
the AP package (using pkgrm(1M)), you must first use apboot(1M) to remove the
boot disk from AP control. If you do not first remove the boot disk from AP control,
the configuration using that disk becomes unbootable.
AP Boot Sequence
This subsection briefly describes the flow of events that occur when the Sun
Enterprise 10000 server is booted on an alternately pathed boot disk. This sequence
of events illustrates how auto-switching of the boot disk controller is achieved
during the boot process, if such a switch is necessary. The boot sequence proceeds as
follows:
1. By default, the system is booted from the device specified by the OBP devalias
boot-device. Note that this device may be different from the last active
alternate for the boot disk.
40Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
2. OBP stores the path to the boot disk on the SSP.
3. If a failure occurs, it is detected after a few minutes. (The default is 10 minutes.)
Then, the AP Reboot Host program is executed.
Note – Several minutes may pass before action is taken, so do not immediately
enter new commands if you notice that the boot process has failed. If you attempt a
manual recovery from a boot failure, be aware that the automatic reboot recovery
process will still be executing and may override your manual recovery command.
4. AP Reboot Host retrieves the path stored earlier by OBP, and communicates the
path to the AP SSP daemon.
5. The AP SSP daemon looks up the alternate path for the boot disk in the AP SSP
database, and retries the boot process with that alternate path.
6. After the reboot succeeds, AP determines the alternate from which the system
was booted, and makes it the active alternate.
Using Single-User Mode
Normally, when the Sun Enterprise server is fully booted, you use the AP command
versions located in /usr/sbin. However, if your server comes up in single-user
mode because the boot process did not fully complete, you can use the commands in
/sbin. The command versions under /sbin do not rely on the AP daemon services
(which are not available in single-user mode). If the system enters single-user mode
because of a problem related to AP, you may be able to resolve the problem by using
the /sbin commands to perform needed AP operations.
Two AP-related problems may cause the system to come up in single-user mode:
■ If two paths are supposed to lead to the same disk (according to the AP SSP
database) but those paths actually lead to different disks, and if that disk needs to
be mounted during the boot process. (This can only happen if you change the
physical configuration of the pathgroup without running AP commands to
update the database.)
■ If an active alternate for a disk turns out to be inaccessible and that disk is
required during the boot process. A disk is required during the boot process if it
has file systems that are mounted during the boot process (that is, it has entries in
the /etc/vsftab file).
These situations arise only with respect to disks, not networks. In either case, you
may be able to use the AP commands under /sbin to resolve the problem.
Chapter 4Using AP Boot Devices41
42Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
CHAPTER
5
Using Meta-Networks and
Network Pathgroups
To use AP, both physical networks within a network pathgroup must be of the same
type. For example, a network pathgroup could consist of two le networks or two qe
networks, but not one of each.
Both alternates in a network pathgroup should be physically connected to the same
network. For example, Ethernet controllers should be connected to the same subnet.
While multiple physical network connections exist, only one controller at a time is
active. The controllers should be on different system boards so that DR operations
(such as DR Detach) can be performed without affecting all potential active
alternates.
The AP switch procedures in this section show how to switch the active alternate.
Meta-Network Interfaces
A meta-network interface name is derived from the name of the primary alternate
for that meta-network. A meta-network interface name has the form mxxx, where xxx
is the primary interface name such as le0.
You must use two network interfaces of the same type when creating a metanetwork interface. For example, you can use le0 and le4,ornf0 and nf1.
However, you can not use le0 and nf1. Some examples follow:
■ LE Ethernet meta-network names have the form mle#, where # is the instance
number. For example, assume the network controllers le0 and le1 connect to the
same Ethernet subnet. A meta-network mle0 can encompass these two controllers
43
(if the primary controller is le0). Similarly, QE Ethernet meta-network names
have the form mqe#. Note that you cannot mix le and qe networks within the
same pathgroup.
■ FDDI meta-network names have the form mf# and mnf#. nf networks can be
either SAS or DAS. SAS and DAS configurations may be mixed when creating a
meta-network interface.
Working With Network Pathgroups
▼ To Create a Network Pathgroup and Meta-
Network
Note – This procedure should not be used for the primary network. To alternately
path the primary network, see “Alternately Pathing the Primary Network Interface”
on page 51.
1. Use apnet(1M) with the -c option:
# apnet -c -p le0 -a le2
# apconfig -N -u
metanetwork: mle0 U
physical devices:
le2
le0 P A
This apnet(1M) command creates the network pathgroup as well as the metanetwork interface name mle0 for the two physical devices le0 and le2. The metanetwork interface name is derived from the primary controller name (specified by
-p).
The apconfig(1M) command lists the uncommitted network entries in the
database. -N specifies that network database entries should be listed. -u specifies
that uncommitted entries should be listed.
44Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
2. If you are satisfied with the network pathgroup, commit the entry:
# apdb -C
# apconfig -N
metanetwork: mle0
physical devices:
le2
le0 P A
apdb -C commits the database entries. apconfig -N lists the committed network
entries in the database. The listing appears exactly as it did before, except that the U
no longer appears after mle0.
3. Remove all direct usage of both members of the pathgroup (see ifconfig(1M)).
a. If the physical interface is currently plumbed, and it is not the primary network
interface, and it is not the interface that you will be using as you run
commands to configure the meta-network, unplumb the physical interface.
Note – If the interface you will be configuring down is the primary network
interface, or if it is the interface that you will be using as you use commands to
configure the meta-network, follow one of the procedures in “Alternately Pathing
the Primary Network Interface” on page 51.
You can unplumb the physical interface as shown in the following example:
# ifconfig le0 down; ifconfig le0 unplumb
Usually network interfaces are configured during system boot via the file /etc/
hostname.xxx, where xxx is the interface name (such as le0). This file contains
the IP address or the hostname associated with the interface. You should remove
or rename the /etc/hostname.xxx for all interfaces that have been made AP
alternates, since direct usage of the alternate must not occur.
b. Create an /etc/hostname.mxxx file (such as /etc/hostname.mle0) for any
meta-networks that you want to configure at system reboot.
This file should contain the meta-network IP address or the hostname for the
interface. You can simply rename /etc/hostname.le0 to
/etc/hostname.mle0.
The normal operating state of a network interface is plumbed up when in use, and
unplumbed when not in use. When you automatically configure network interfaces
through /etc/hostname.*, the interfaces are left in one of these states. It is
Chapter 5Using Meta-Networks and Network Pathgroups45
possible to leave a network interface in a transitory state of plumbed when you
manually configure your network interface. As this is not a normal operational
mode, it is unlikely that network interfaces will be left in this state. Do not leave
meta-networks in this state during AP network configuration. A network metadevice may be deleted only if it and all other network meta-devices of that device
type are either in the unplumbed state or the plumbed up state. Otherwise, AP
ignores the delete request, and depending on your configuration, may display
warning messages of the following form:
WARNING:mnf_setphyspath: APUNSET busy
WARNING:ap_db_commit: mnf3 not deleted, metadevice returned error
16
Note – If you are using FDDI, you must specify a unique MACID for the meta-
network. The MACID is set through the ether parameter to the ifconfig(1M)
command. First examine the MACID for each of the alternates. You can do this by
bringing up each alternate and examining the ether field. Then, fabricate a MACID
that does not match any of the alternates.
The allocation of Media Access Control Identifiers (MACIDs) is described in
RFC1340, “Assigned Numbers”, July 1992. When generating a MACID for your AP
network interface, the new 48-bit hardware address should be acquired from the
IEEE Standards Office, 345 East 47th Street, New York, N.Y. 10017. However, it is
possible to “create” a number by transposing digits on an existing MACID of one of
the meta-interface alternate elements. After creating a number, it is important to
verify that there is no other hardware on the same subnet that is a legitimate user of
the created address.
This meta-network MACID is used to configure the active physical interface of the
meta-network. The use of this MACID is necessary to prevent duplication of
MACIDs on the network when combining AP switching of interfaces and DR board
insertion activities. The meta-network defaults to the MACID of the active alternate
at boot time. To ensure that the MACID is set properly at boot time, place
ifconfig(1M) commands in the /etc/rcs.d/S30rootusr.sh start-up script.
The following is an example. Note that the ## characters are used to indicate the
46Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
lines to be added; the leading ## characters should not be added. You should
replace MACID_mnf2 and MACID_mle0 with the actual numbers (for example,
8:0:20:68:6d:62).
/etc/rcS.d/S30rootusr.sh:
[....]
interface_names="`echo /etc/hostname.*[0-9] 2>/dev/null`"
if [ "$interface_names" != '/etc/hostname.*[0-9]' ]; then
(
echo 'configuring network interfaces:\c'
IFS="${IFS}."
set -- $interface_names
while test $# -ge 2; do
shift
if [ "$1" != "xx0" ]; then
addr=`shcat /etc/hostname\.$1`
/sbin/ifconfig $1 plumb
###
### Set the MACid for AP
###
##if [ "$1" = "mnf2" ] ; then
##/sbin/ifconfig $1 ether MACID_mnf2
##fi
##if [ "$1" = "mle0" ] ; then
##/sbin/ifconfig $1 ether MACID_mle0
##fi
##
if [ -n "$addr" ]; then
/sbin/ifconfig $1 inet "$addr" \
netmask + broadcast + -trailers up \
2>&1 >/dev/null
fi
echo " $1\c"
fi
shift
done
echo '.'
)
fi
[....]
Chapter 5Using Meta-Networks and Network Pathgroups47
4. Bring up the meta-network in the usual manner, but use the meta-network name
instead of the physical network name. You can do this by either rebooting the
machine or manually configuring the network as in the following example:
# ifconfig mle0 plumb
# ifconfig mle0 inet 136.162.65.30 up netmask + broadcast +
Setting netmask of mle0 to 255.255.255.0
# ifconfig -a
lo0: flags=849<UP,LOOPBACK,RUNNING,MULTICAST> mtu 8232
inet 127.0.0.1 netmask ff000000
mle0: flags=843<UP,BROADCAST,RUNNING,MULTICAST> mtu 4352
inet 136.162.65.30 netmask ffffff00 broadcast 136.162.65.255
ether 0:0:be:0:8:c5
At this point, the device node, such as /dev/mle, can be used to access the network
from Solaris commands such as snoop(1M).
▼ To Switch a Network Pathgroup
You can switch a network pathgroup even while the network sustains traffic.
Caution – This procedure requires that you reboot the machine. If you are not ready
to reboot the machine, do not perform this procedure.
● Use the apconfig(1M) command:
# apconfig -P mle0 -a le2
# apconfig -N
metanetwork: mle0
physical devices:
le2 A
le0 P
The -P option specifies the pathgroup and -a specifies the alternate that you want
to become active. The listing above shows that the active alternate has been switched
to le2, as indicated by the A following le2.
You do not have to commit a switch operation.
48Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
▼ To Delete a Network Pathgroup and Meta-
Network
1. Remove all usage of the corresponding meta-network, and use apnet -d:
In the listing produced by apconfig -N,aD follows mle0, indicating that the
pathgroup is marked as deleted.
2. Commit the entries in the database using apdb -C:
# apdb -C
# apconfig -N
#
The apconfig -N command produces no listing, indicating that the network
pathgroup (the only one that existed previously in this example) has been deleted.
You can undo a deletion if the deletion is uncommitted. To undo a deletion, use
apnet -z, specifying the same meta-network interface that you previously deleted.
When an apnet -m -r or apnet -m -a command is used, AP marks the
current pathgroup configuration as deleted and creates a new uncommitted
pathgroup definition. Once the database change is committed with apdb -C, the
new definition replaces the old.
3. Remove the /etc/hostname.mxxx file, as described in “To Deconfigure a MetaNetwork.
▼ To Deconfigure a Meta-Network
Caution – This procedure requires you to reboot the machine. If you are not ready
to reboot the machine, do not perform this procedure.
Chapter 5Using Meta-Networks and Network Pathgroups49
1. Verify that the primary network interface is mqe0 (in this example):
3. Remove the configuration file for the meta network interface:
# rm -f /etc/hostname.qe0
4. Reboot.
Alternately Pathing the Primary
Network Interface
The primary network interface between your Sun Enterprise server and the other
machines on the network is the Ethernet interface that is on the same subnet as the
SSP. One way to identify the primary network is to look in the /etc/hostname.xxx
files until you find the one that contains the IP name that matches the IP name found
in the file /etc/nodename. The corresponding xxx network (for example, qe0)is
the primary network.
You can alternately path the primary network, if you want. The primary network is
the only network interface that can be auto-switched at boot-time. During the boot
process on the Sun Enterprise 10000 server (but not on other Sun Enterprise servers),
if the active alternate for the primary network fails, the system attempts to find a
working alternate. Note that the AP database on your SSP is used for this purpose.
A subset of the AP database that resides on the SSP is used at boot time to
automatically switch to a functional path for the primary network. By the time the
system is ready to start using the network, the file systems on the host are already
up and running, so the main AP database can be used.
When you configure an alternately pathed network, you must not configure the
meta-network while the underlying driver is still active. When you configure AP for
a network that you are currently using, the transition period between configuring
Chapter 5Using Meta-Networks and Network Pathgroups51
the physical interface down and the AP interface up generates a loss of network
service for your Sun Enterprise server. To perform this transition, you should use
one of the following procedures, which are shown in order of preference:
■ Create the appropriate AP database entries, create a new /etc/hostname.mxxx
file, remove (or rename) the corresponding /etc/hostname.xxx file, and then
reboot your Sun Enterprise server.
■ Set up a script file to perform the transition on your Sun Enterprise server.
■ Log into your Sun Enterprise server via another network interface to enable
commands to be issued when the network service is lost on the network interface
that you are bringing up under AP.
The following example shows how to reconfigure the primary network using the
first approach. This example assumes you have an Sun Enterprise server named hmb
with a primary network interface on qe0, and that you want to have a meta-network
interface composed of qe0 and qe4. (If you are unsure which network interfaces
should be combined into a meta-network, you can use snoop -d to determine which
of your configured networks are on the same subnet.)
▼ To Create a Network Pathgroup and Meta-
Network for the Primary Network
Caution – This procedure requires you to reboot the machine. If you are not ready
to reboot the machine, do not perform this procedure.
1. Verify that the primary network interface is qe0:
Chapter 5Using Meta-Networks and Network Pathgroups55
3. Remove the configuration file for the meta network interface:
# rm -f /etc/hostname.qe0
4. Reboot.
56Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
CHAPTER
6
Interaction Between AP and DR
This chapter describes the relationship between AP and DR.
Using AP and DR Together
Dynamic Reconfiguration (DR) and Alternate Pathing (AP) are designed to work
closely together. DR enables you to attach and detach system boards without halting
the operating system, as described in the Sun Enterprise 10000 DynamicReconfiguration User’s Guide. AP enables you to switch usage away from a controller
on a board that is being detached or possibly to a controller on a board that has been
attached.
On the Sun Enterprise 10000 server, AP automatically switches each disk and network
meta-device that has an active controller on a board being detached (assuming a
viable alternate path exists on another board). Also, on the Sun Enterprise 10000
server, AP prevents you from manually switching to controllers on a board that is in
the drain state of a DR detach operation.
On Sun Enterprise servers other than the Sun Enterprise 10000 server, you must
manually switch disk and network meta-devices (if necessary) before detaching a
board.
57
The following AP command shows that the pln1 controller is on a board that is
detached (as indicated by the DE flag) and, thus, you cannot switch to that controller:
# apconfig -S
c1 pln0 P A
c2 pln1 DE
metadiskname(s):
mc1t5d0
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
Similarly, the following AP command shows that the pln1 controller is on a board
that is in the drain state (as indicated by the DR flag) and, thus, you cannot switch to
that controller:
# apconfig -S
c1 pln0 P A
c2 pln1 DR
metadiskname(s):
mc1t5d0
mc1t4d0
mc1t3d0
mc1t2d0
mc1t1d0
AP is notified that a board is in the DR drain state only on the Sun Enterprise 10000
server.
When you plan to detach a board that hosts the active controller of a pathgroup, you
can manually switch to a controller on another board before the DR Detach
operation, or even during the DR Detach operation. On machines other than the Sun
Enterprise 10000 server, however, you must perform any such switch before the
detach operation attempts to complete, or the detach operation will fail. In this case,
you can perform the switch and then retry the detach operation.
For more information about DR, see the Sun Enterprise 10000 Dynamic Reconfiguration
User’s Guide or the Sun Enterprise 6x00, 5x00, 4x00, and 3x00 Systems Dynamic
Reconfiguration User’s Guide.
58Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Maintaining the Correct AP State
On machines other than the Sun Enterprise 10000 server, if you attach or detach a
board that hosts an I/O controller for a disk or network pathgroup, you must run
apconfig -F. This command sets or clears the the detach flag (DE) for that board so
that it correctly indicates whether the board is attached or detached. (On the Sun
Enterprise 10000 server, you do not need to use apconfig -F after an attach or
detach operation, since the DE flag is automatically set or cleared after the DR
operation completes.)
If you detach a board that hosts a network controller, and that network device has
not been used since the previous boot, you must run apconfig -F to notify the
system that the network device is no longer available.
The apconfig -N command may incorrectly indicate that a network controller
resides on a board that is detached (or incorrectly indicate that it resides on a board
that is present) if the corresponding AP meta-driver has not been loaded. Use
apconfig -F to ensure the correct information is shown by apconfig -N.
Chapter 6Interaction Between AP and DR59
60Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
APPENDIX
A
AP Commands
The AP man pages are in the Alternate Pathing Reference Manual portion of your Sun
Enterprise server documentation set, and online as well (once you have installed the
AP packages). This chapter introduces the AP commands and files. Unless noted
otherwise, AP commands are executable on the Sun Enterprise server, not in the SSP
environment. (The SSP is part of the Sun Enterprise 10000 system, and is not
available with other Sun Enterprise servers.) Following is a list of the AP man pages:
■ ap(1M) – alternate pathing overview
■ ap_reboot_host(1M) – fast alternate path boot. This command is executed on
the SSP by other commands. Do not execute it manually. (The SSP is part of the
Sun Enterprise 10000 system, and is not available with other Sun Enterprise
servers.)
■ ap_daemon(1M) – alternate pathing daemon
■ ap_ssp_daemon(1M) – AP SSP daemon. This daemon runs on the SSP. (The SSP
is part of the Sun Enterprise 10000 system, and is not available with other Sun
Enterprise servers.)
■ apboot(1M) – define an AP boot device
■ apcheck(1M) – determine accessibility of AP SCSI meta–device
■ apconfig(1M) – display and manage AP configuration
■ apdb(1M) – manage replicas of AP database
■ apdisk(1M) – manage AP for SCSI disks
■ apinst(1M) – identify SCSI bus controller
■ apnet(1M) – manage AP for networks
■ apssp(1M) – client of AP SSP daemon. This command is executed on the SSP by
other commands. Do not execute it manually. (The SSP is part of the Sun
Enterprise 10000 system, and is not available with other Sun Enterprise servers.)
■ ap(7) – AP driver
■ ap_dmd(7) – AP disk meta-driver
■ ap_nmd(7) – AP network meta-driver
61
62Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
APPENDIX
B
AP Components
AP consists of the following components:
■ AP daemon – ap_daemon(1M) runs on the Sun Enterprise server and receives
user requests through the AP commands executed on the server. The daemon acts
as an intermediary between the commands and the AP librarian, ap(7). It passes
requests received through RPCs to the librarian by invoking ioctls. The actual
work of maintaining the database is performed by the librarian.
■ AP librarian – ap(7) manages the AP database and interacts with the meta-
drivers as necessary. It receives requests through ioctls and acts on them by
updating the database or calling entry points in the meta-drivers.
■ AP meta-drivers – The low-level capability for re-routing I/O accesses to alternate
paths is implemented in the meta-drivers.
All application I/O requests that use the appropriate meta-disk go through a metadriver that passes them to the physical device drivers. As a result, the meta-drivers
can decide which physical path to use, whether a given path is no longer
functioning, and so forth. The information on which the meta-drivers base their
decisions comes from the AP librarian and AP database.
■ AP SSP daemon – ap_ssp_daemon(1M) executes on the SSP and receives RPC
requests from the AP daemon whenever the AP database changes. (Note that the
SSP is part of the Sun Enterprise 10000 system; the SSP is not available with any
other Sun Enterprise servers.) This daemon is responsible for maintaining a file
on the SSP containing alternate pathing information for booting.
63
64Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
APPENDIX
C
Driver Layers
The following figure illustrates the driver layers that are used when AP controls disk
devices.
User process
/dev/ap/dsk/meta-device
AP disk meta-driver
(ap_dmd)
Two alternates
Disk driver
(e.g., ssd for SSA)
Nexus driver
(e.g., pln/SOC for SSA)
Device
(e.g., SSA disk array)
FIGURE C-1 AP Disk Driver Layers
65
A user process references a meta-disk, which provides access to the AP disk metadriver. The AP disk meta-driver controls two instances of the physical disk driver
which, in turn, controls two instances of the nexus driver (or controller driver). The
nexus driver controls the physical device.
FIGURE C-2 illustrates the driver layers that are used when AP controls networks.
User process
/dev/mxxx
Stream head
Read
processing
Read
processing
Write
processing
Write
processing
ReadWrite
Driver routines
Physical network
interface
FIGURE C-2 AP Network Driver Layers
AP network meta-driver
(mxx)
Stream end
(xx driver)
66Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
A user process references a meta-network, which provides access to the stream head.
The AP network meta-driver is inserted into the stream between the high-level
read/write processing components and the physical driver routines.
Appendix CDriver Layers67
68Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Glossary
active alternateThe alternate path that is currently handling I/O for a pathgroup.
AP database (or simply
database)A database that is maintained by the AP subsystem. The AP database contains
alternate pathOne of the physical paths within a pathgroup.
committed database
entryAn entry in the AP database that is currently being used by AP to manage
meta-diskA disk abstraction that provides access to an underlying group of two physical
meta-networkA network abstraction that provides access to an underlying group of two
pathgroupA set of two alternate paths that provide access to the same device or set of
physical pathThe electrical path from the host to a disk or network.
pln portAn optical link connection (OLC) module on an SSA controller that can be
all of the information needed to maintain the configuration alternate paths.
access to a disk or network. (Compare with uncommitted database entry.)
paths to a disk.
physical paths to a network.
devices.
connected to an SSA port.
primary pathThe alternate path in a pathgroup that is initially the active alternate. The name
of the primary path is used when constructing the name of the meta-disk or
meta-network. The primary path does not change when a switch occurs.
SSAA SPARCstorage Array, a collection of disks within a hardware peripheral. The
SSA provides access to each of its housed disks via two ports.
SSA portAn optical link connection (OLC) module on an SSA that can be connected to a
pln port.
SSA controllerA controller that resides on the host system and has one or two pln ports (also
called a SOC controller).
Glossary69
switchThe act of establishing a new active path as the path to be used for a given
uncommitted database
entryAn entry in the AP database that has not been committed and is therefore not
pathgroup. Note that switching does not change the primary path.
currently in effect. If a pathgroup has been created but the database entry has
not been committed, that pathgroup is not currently used by AP to manage
access to a disk or network. If a previously committed pathgroup has been
deleted, but that database entry has not been committed, that pathgroup is still
being used by AP to manage access to the disk or network.
70Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
Index
A
A (active alternate indicator), 28
active alternate, 7
indicator (A), 28
alternate path, 2, 7
identifying, 7
specifying, 28
alternate pathing, 61
alternately pathed network, configuring, 52
alternately pathing primary network, 52
AP / DR interaction, 57
AP and domains, 11
AP and Dynamic Reconfiguration (DR), 3
AP boot sequence, 40
AP commands, /usr/sbin vs. /sbin, 41
AP configuration, typical, 10
AP Daemon, 63
AP Librarian, 63
AP meta-driver, 63
AP SSP daemon, 63
AP state, managing, 59
AP vs. disk mirroring, 10
ap_dmd* in devices directory, 30
apboot example, 38, 40
apboot -m example, 39
apboot -u example, 40
apconfig -D example, 16
apconfig -N example, 18, 45, 48, 49, 53
apconfig -N -u example, 18, 44
apconfig -N, ensuring it displays correct
information, 59
apconfig -P -a example, 31, 32, 48
apconfig -R example, 30
apconfig -S example, 17, 21, 29, 31, 32, 33, 34
apconfig -S -u example, 17, 28
apconfig(1M), 16
apdb -C example, 28, 34, 45, 49, 52
apdb -c -f example, 15
apdb -d -f example, 15
apdb(1M), 15
apdisk -c -p -a example, 28
apdisk -d example, 33
apdisk -w example, 21
apdisk -z, 34
apinst example, 27
apnet and undoing deletion, 49
apnet -c -p -a example, 44, 52
apnet -d example, 49
attaching boards and AP, 3
auto switching at boot time, overview, 40
automatic failover, 3
automatic switching during DR, 57
automatic switching of meta-disks, 20
B
bin, /usr/sbin vs. /sbin, 41
boot disk
Index71
AP and boot disk, 37
mirroring and AP, 39
boot disk, remove from AP control, 40
boot sequence, 40
boot time, auto switching, 40
boot, unattended, 10
boot-device and AP, 38
boot-device devalias, 38
bringing up network, 48
C
clearing DE (detached) flag, 59
commands, /usr/sbin vs. /sbin, 41
commands, list of, 61
committed database entries, 16
deleting, 33
disk entries, viewing, 29
network entries, viewing, 18
viewing, 17
configuration, typical, 10
configuring alternately pathed network, 52
controller (defined), 4
corrupt database, determining if, 16
create, 30
creating database, 14, 15
creating database after boot disk under AP
device node references, modifying for AP, 30
devices directory, listing AP entries, 30
devices directory, rebuilding, 29
devices supported by AP, 9
disk
automatic failover, 3
automatic switching, 20
boot disk under AP, 37
boot disk, AP and mirror, 39
disk devices supported by AP, 9
72Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
disk pathgroup, 7
meta-disk, 5
mirrored boot disk, remove from AP control, 40
path for meta-disk, 5
remove boot disk from AP control, 40
disk mirroring vs. AP, 10
disk mirroring, example, 11
disk pathgroup, 7
disk pathgroup vs. meta-disk, 7
domains and AP, 11
DR (drain state) flag, 58
DR / AP interaction, 13, 57
DR and automatic switching, 57
DR drain state, switching paths and, 57
drain state (DR) flag, 58
driver, AP meta-driver, 63
drvconfig example, 29
Dynamic Reconfiguration (DR), 3
Dynamic Reconfiguration (DR) and AP, 57
alternately pathed I/O device, 2
AP and disk mirroring, 11
disk pathgroup, 8
meta-disk, 6
network pathgroup, 9
typical AP configuration, 10
inaccessible database, determining if, 16
information about database, viewing, 16
interaction between AP and DR, 57
interface, meta-network interface, 43
introduction to AP, 1
L
LE meta-network names, 43
librarian, AP Librarian, 63
links, devices directory to meta-disk special
files, 30
listing AP devices directory entries, 30
P (primary path indicator), 28
packages, removing AP packages, 40
partition for database, recommended, 13
path
switch
switching during DR, 3
verifying before switching to, 31
path for a meta-disk, 5
path for meta-disk, 20
74Sun Enterprise Server Alternate Pathing User’s Guide • May 1999
R
raw disk slice for creating database, 15
raw disk slice for deleting database, 15
rebuilding devices directory, 29
references to device nodes, modifying for AP, 30
remove alternate pathing of primary network, 49,
50, 53, 54, 55
remove boot disk from AP control, 40
remove mirrored boot disk from AP control, 40
removing AP packages, 40
removing configuration files for physical network
interfaces, 50, 51, 53, 54, 55, 56
removing direct usage of physical paths (for
networks), 45
repartition, not done by AP, 19
resetting tried flag, 21
S
single user mode and AP, 41
single user mode, reasons it is invoked, 41
Solaris version supported, 9
SSAs supported by AP, 9
SSP database copy, 14
SSP, AP SSP daemon, 63
state of AP, managing, 59
supported devices, 9
switch, 31
automatic switch at boot time, 40
automatic switching and DR, 57
during DR drain state, 57
example (for disks), 31
switch operation (defined), 7
switching meta-disks, automatic, 20
switching network pathgroup (Ethernet or
FDDI), 48
switching path during DR, 3
symbolic links, devices directory to meta-disk