0% found this document useful (0 votes)

70 views147 pages

Solaris 10 ZFSEssentials

Uploaded by

Nazım Bilir

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

70 views147 pages

Solaris 10 ZFSEssentials

Uploaded by

Nazım Bilir

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 147

Copyright................................................................................................................................ 1
Preface.................................................................................................................................. 10
Acknowledgments................................................................................................................. 14
About the Author.................................................................................................................. 16
Chapter 1. Introducing ZFS File Systems............................................................................... 18
Section 1.1. Overview of ZFS.......................................................................................................................................................................................................... 18
Section 1.2. Fast and Simple Storage............................................................................................................................................................................................ 23
Section 1.3. ZFS Commands.......................................................................................................................................................................................................... 24
Chapter 2. Managing Storage Pools...................................................................................... 26
Section 2.1. ZFS Pool Concepts..................................................................................................................................................................................................... 26
Section 2.2. Creating a Dynamic Stripe........................................................................................................................................................................................ 28
Section 2.3. Creating a Pool with Mirrored Devices..................................................................................................................................................................... 30
Section 2.4. Creating a Pool with RAID-Z Devices....................................................................................................................................................................... 32
Section 2.5. Creating a Spare in a Storage Pool............................................................................................................................................................................ 34
Section 2.6. Adding a Spare Vdev to a Second Storage Pool........................................................................................................................................................ 35
Section 2.7. Replacing Bad Devices Automatically....................................................................................................................................................................... 36
Section 2.8. Locating Disks for Replacement............................................................................................................................................................................... 39
Section 2.9. Example of a Misconfigured Pool............................................................................................................................................................................. 40
Chapter 3. Installing and Booting a ZFS Root File System..................................................... 42
Section 3.1. Simplifying (Systems) Administration Using ZFS.................................................................................................................................................... 42
Section 3.2. Installing a ZFS Root File System............................................................................................................................................................................. 43
Section 3.3. Creating a Mirrored ZFS Root Configuration........................................................................................................................................................... 47
Section 3.4. Testing a Mirrored ZFS Root Configuration............................................................................................................................................................. 48
Section 3.5. Creating a Snapshot and Recovering a ZFS Root File System................................................................................................................................. 49
Section 3.6. Managing ZFS Boot Environments with Solaris Live Upgrade................................................................................................................................ 52
Section 3.7. Managing ZFS Boot Environments (beadm)............................................................................................................................................................ 60
Section 3.8. Upgrading a ZFS Boot Environment (beadm)......................................................................................................................................................... 60
Section 3.9. Upgrading a ZFS Boot Environment (pkg)............................................................................................................................................................... 61
Section 3.10. References................................................................................................................................................................................................................ 63
Chapter 4. Managing ZFS Home Directories......................................................................... 64
Section 4.1. Managing Quotas and Reservations on ZFS File Systems........................................................................................................................................ 64
Section 4.2. Enabling Compression on a ZFS File System........................................................................................................................................................... 70
Section 4.3. Working with ZFS Snapshots.................................................................................................................................................................................... 72
Section 4.4. Sharing ZFS Home Directories................................................................................................................................................................................. 76
Section 4.5. References.................................................................................................................................................................................................................. 77
Chapter 5. Exploring Zpool Advanced Concepts.................................................................... 78
Section 5.1. X4500 RAID-Z2 Configuration Example.................................................................................................................................................................. 78
Section 5.2. X4500 Mirror Configuration Example..................................................................................................................................................................... 86
Section 5.3. X4500 Boot Mirror Alternative Example.................................................................................................................................................................. 91
Section 5.4. ZFS and Array Storage............................................................................................................................................................................................... 91
Chapter 6. Managing Solaris CIFS Server and Client............................................................. 92
Section 6.1. Installing the CIFS Server Packages.......................................................................................................................................................................... 92
Section 6.2. Configuring the SMB Server in Workgroup Mode................................................................................................................................................... 95
Section 6.3. Sharing Home Directories......................................................................................................................................................................................... 96
Chapter 7. Using Time Slider............................................................................................... 100
Section 7.1. Enabling Time Slider Snapshots.............................................................................................................................................................................. 100
Section 7.2. Enabling Nautilus Time Slider................................................................................................................................................................................ 102
Section 7.3. Modifying the Snapshot Schedule........................................................................................................................................................................... 104
Section 7.4. Setting the Snapshot Schedule per File System...................................................................................................................................................... 108
Chapter 8. Creating a ZFS Lab in a Box................................................................................ 110
Section 8.1. Creating Virtual Disks with Virtual Media Manager............................................................................................................................................... 110
Section 8.2. Registering a CD Image with Virtual Media Manager............................................................................................................................................ 114
Section 8.3. Creating a New Virtual Machine.............................................................................................................................................................................. 116
Section 8.4. Modifying the New Virtual Machine....................................................................................................................................................................... 120
Section 8.5. Installing an OS on a Virtual Machine.................................................................................................................................................................... 123
Section 8.6. Installing Virtual Box Tools..................................................................................................................................................................................... 128
bvdindexIndex.................................................................................................................... 136

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Onur Bingul
3690014

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Essentials

Scott Watanabe

Sun Microsystems Press

Upper Saddle River, NJ • Boston • Indianapolis • San Francisco

New York • Toronto • Montreal • London • Munich • Paris • Madrid
Capetown • Sydney • Tokyo • Singapore • Mexico City

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Prepared for [email protected], Onur Bingul
Copyright © 2010 Sun Microsystems, Inc.. This download file is made available for personal use only and is subject to the Terms of Service. Any other use requires prior written consent
from the copyright owner. Unauthorized use, reproduction and/or distribution are strictly prohibited and violate applicable laws. All rights reserved.
Many of the designations used by manufacturers and sellers to distinguish their products are claimed as
trademarks. Where those designations appear in this book, and the publisher was aware of a trademark claim,
the designations have been printed with initial capital letters or in all capitals.
The author and publisher have taken care in the preparation of this book, but make no expressed or implied warranty of
any kind and assume no responsibility for errors or omissions. No liability is assumed for incidental or consequential
damages in connection with or arising out of the use of the information or programs contained herein.
Sun Microsystems, Inc., has intellectual property rights relating to implementations of the technology described in this
publication. In particular, and without limitation, these intellectual property rights may include one or more U.S. patents,
foreign patents, or pending applications. Sun, Sun Microsystems, the Sun logo, J2ME, J2EE, Solaris, Java, Javadoc, Java
Card, NetBeans, and all Sun and Java based trademarks and logos are trademarks or registered trademarks of
Sun Microsystems, Inc., in the United States and other countries. UNIX is a registered trademark in the United States and
other countries, exclusively licensed through X/Open Company, Ltd.
THIS PUBLICATION IS PROVIDED “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED,
INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. THIS PUBLICATION COULD INCLUDE
TECHNICAL INACCURACIES OR TYPOGRAPHICAL ERRORS. CHANGES ARE PERIODICALLY ADDED TO THE
INFORMATION HEREIN; THESE CHANGES WILL BE INCORPORATED IN NEW EDITIONS OF THE
PUBLICATION. SUN MICROSYSTEMS, INC., MAY MAKE IMPROVEMENTS AND/OR CHANGES IN THE
PRODUCT(S) AND/OR THE PROGRAM(S) DESCRIBED IN THIS PUBLICATION AT ANY TIME.
The publisher offers excellent discounts on this book when ordered in quantity for bulk purchases or special
sales, which may include electronic versions and/or custom covers and content particular to your business,
training goals, marketing focus, and branding interests. For more information, please contact:
U.S. Corporate and Government Sales
(800) 382-3419
[email protected]
For sales outside the United States please contact:
International Sales
[email protected]
Visit us on the Web: informit.com/ph
Library of Congress Cataloging-in-Publication Data
Watanabe, Scott.
Solaris 10 ZFS essentials / Scott Watanabe.
p. cm.
Includes index.
ISBN 978-0-13-700010-4 (pbk. : alk. paper)
1. Solaris (Computer ﬁle) 2. File organization (Computer science) 3. Database management.
I. Title.
QA76.9.F5W38 2009
005.74—dc22
2009044423

Copyright © 2010 Sun Microsystems, Inc.

4150 Network Circle, Santa Clara, California 95054 U.S.A.
All rights reserved.
Printed in the United States of America. This publication is protected by copyright, and permission must
be obtained from the publisher prior to any prohibited reproduction, storage in a retrieval system, or
transmission in any form or by any means, electronic, mechanical, photocopying, recording, or likewise.
For information regarding permissions, write to:
Pearson Education, Inc.
Rights and Contracts Department
501 Boylston Street, Suite 900
Boston, MA 02116
Fax: (617) 671-3447
ISBN-13: 978-0-13-700010-4
ISBN-10: 0-13-700010-3
Text printed in the United States on recycled paper at Courier in Stoughton, Massachusetts.
First printing, December 2009

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Preface ix
Acknowledgments xiii
About the Author xv

Chapter 1 Introducing ZFS File Systems 1

1.1 Overview of ZFS 1
1.1.1 Advantages of Using ZFS 2
1.1.2 A Top-Level View of ZFS 3
1.2 Fast and Simple Storage 6
1.3 ZFS Commands 7

Chapter 2 Managing Storage Pools 9

2.1 ZFS Pool Concepts 9
2.2 Creating a Dynamic Stripe 11
2.3 Creating a Pool with Mirrored Devices 13
2.4 Creating a Pool with RAID-Z Devices 15
2.5 Creating a Spare in a Storage Pool 17
2.6 Adding a Spare Vdev to a Second Storage Pool 18
2.7 Replacing Bad Devices Automatically 19
2.8 Locating Disks for Replacement 22
2.9 Example of a Misconﬁgured Pool 23

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Chapter 3 Installing and Booting a ZFS Root File System 25

3.1 Simplifying (Systems) Administration Using ZFS 25
3.2 Installing a ZFS Root File System 26
3.2.1 ZFS Root File System Requirements 27
3.2.2 ZFS Root Pool Recommendations 27
3.2.3 Interactive ZFS Root File System Installation
Example 28
3.3 Creating a Mirrored ZFS Root Conﬁguration 30
3.4 Testing a Mirrored ZFS Root Conﬁguration 31
3.5 Creating a Snapshot and Recovering a ZFS Root
File System 32
3.5.1 Creating a Snapshot of the ZFS Root Storage Pool 32
3.5.2 Sending the ZFS Root Pool Snapshots to Storage 33
3.5.3 Restoring the ZFS Root Pool Snapshots to a
New Disk 33
3.6 Managing ZFS Boot Environments with Solaris
Live Upgrade 35
3.6.1 Migrating a UFS Root File System to a
ZFS Root File System 35
3.6.2 Patching a ZFS Boot Environment with
Solaris Live Upgrade 40
3.7 Managing ZFS Boot Environments (beadm) 43
3.8 Upgrading a ZFS Boot Environment (beadm) 43
3.9 Upgrading a ZFS Boot Environment (pkg) 44
3.10 References 46

Chapter 4 Managing ZFS Home Directories 47

4.1 Managing Quotas and Reservations on
ZFS File Systems 47
4.1.1 Setting the quota and refquota Properties 48
4.1.2 The reservation and refreservation Settings 51
4.2 Enabling Compression on a ZFS File System 53
4.3 Working with ZFS Snapshots 55
4.3.1 Managing the Snapshot Directory 55
4.3.2 Recovering Files from Snapshots 57
4.4 Sharing ZFS Home Directories 59
4.5 References 60

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Chapter 5 Exploring Zpool Advanced Concepts 61

5.1 X4500 RAID-Z2 Conﬁguration Example 61
5.2 X4500 Mirror Conﬁguration Example 69
5.3 X4500 Boot Mirror Alternative Example 74
5.4 ZFS and Array Storage 74

Chapter 6 Managing Solaris CIFS Server and Client 75

6.1 Installing the CIFS Server Packages 75
6.1.1 Installing the Server Packages with
Package Manager 75
6.1.2 Installing the Server Packages from the
Command Line 77
6.2 Conﬁguring the SMB Server in Workgroup Mode 78
6.3 Sharing Home Directories 79

Chapter 7 Using Time Slider 83

7.1 Enabling Time Slider Snapshots 83
7.2 Enabling Nautilus Time Slider 85
7.3 Modifying the Snapshot Schedule 87
7.3.1 Snapshot Basics 89
7.3.2 Changing the Period 89
7.3.3 Changing the Number of Snapshots Kept 91
7.4 Setting the Snapshot Schedule per File System 91

Chapter 8 Creating a ZFS Lab in a Box 93

8.1 Creating Virtual Disks with Virtual Media Manager 93
8.2 Registering a CD Image with Virtual Media Manager 97
8.3 Creating a New Virtual Machine 99
8.4 Modifying the New Virtual Machine 103
8.5 Installing an OS on a Virtual Machine 106
8.6 Installing Virtual Box Tools 111

Index 119

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

systems, snapshots, directory entries, devices, and more. And ZFS implements an
improvement on RAID-5 by introducing RAID-Z, which uses parity, striping, and
atomic operations to ensure reconstruction of corrupted data. It is ideally suited
for managing industry-standard storage servers.

Books in the Solaris System Administration Series

The following sections talk more about the other two books in the series.

Solaris™ 10 System Administration Essentials

Solaris™ 10 System Administration Essentials covers all the breakthrough features
of the Solaris 10 operating system in one place. It does so in a straightforward way
that makes an enterprise-level operating system accessible to system administra-
tors at all levels.
The book provides a comprehensive overview along with hands-on examples of
the key features that have made Solaris the leading UNIX operating system for
years, and it covers the signiﬁcant new features of Solaris 10 that put it far ahead
of its competitors, such as Zones, the ZFS ﬁle system, Fault Management Architec-
ture, Service Management Facility, and DTrace (the dynamic tracing tool for trou-
bleshooting OS and application problems on production systems in real time).

Solaris™ 10 Security Essentials

Solaris™ 10 Security Essentials covers all the security features and technologies in
the Solaris 10 operating system that make it the OS of choice for IT environments
that require optimum security.
The book explains the strengths of Solaris operating system security—its scal-
ability and adaptability—in a simple, straightforward way. It explains how secu-
rity features in Solaris can protect a single-user system with login authentication
as well as how those features can protect Internet and intranet conﬁgurations.

Intended Audience
The three books in the Solaris System Administration Series can beneﬁt anyone
who wants to learn more about the Solaris 10 operating system. They are written
to be particularly accessible to system administrators who are new to Solaris, as

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

well as to people who are perhaps already serving as administrators in companies

running Linux, Windows, and other UNIX systems.
If you are not currently a practicing system administrator but want to become
one, this series, starting with Solaris™ 10 System Administration Essentials, pro-
vides an excellent introduction. In fact, most of the examples used in the books are
suited to or can be adapted to small learning environments such as a home setup.
So, even before you venture into corporate system administration or deploy Solaris
10 in your existing IT installation, these books will help you experiment in a small
test environment.

OpenSolaris

In June 2005, Sun Microsystems introduced OpenSolaris, a fully functional Solaris

operating system release built from open source. Although the books in this series
focus on Solaris 10, they often incorporate aspects of OpenSolaris. Now that
Solaris has been open-sourced, its evolution has accelerated even beyond its nor-
mally rapid pace. The authors of this series have often found it interesting to intro-
duce features or nuances that are new in OpenSolaris. At the same, many of the
enhancements introduced into OpenSolaris are ﬁnding their way into Solaris 10.
So, whether you are learning Solaris 10 or already have an eye on OpenSolaris, the
books in this series are for you.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Any good book always results from more than just the effort of the attributed author,
and this book is far from being an exception. I have beneﬁted from the tremendous
support of Cindy Swearingen and Jeff Ramsey. Cindy provided insight and guidance
without which this book simply would not have been possible. She critiqued my
ideas and writing diplomatically, and she offered great ideas of her own.
Jeff took my rough drawings and created detailed illustrations, upon which the
ﬁgures in this book are based. If they are not clear in any way, that is no fault of
Jeff’s. As always, he was a pleasure to work with.
I would like to thank Todd Lowry, Uzzal Dewan, and the rest of the staff at
Sun’s RLDC for allowing me access to equipment for testing and showing me how
they use ZFS in their operation. Thanks to Dean Kemp for being my technical
sounding board and giving the opportunity to use my knowledge of ZFS to resolve
some real-world problems.
A big thank you to Jim Siwila for jumping in with editorial support at critical
points during challenging times at Sun. Thanks to Judy Hall at Sun for her sup-
port and presenting me with the opportunity to write this book.
To my family and friends, thank you for your understanding and support.

x i ii

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Scott Watanabe is a freelance consultant with more than twenty-ﬁve years in the
computer/IT industry. Scott has worked for Sun Microsystems for more than
eleven years and has worked with Sun technology since the mid-1980s. Scott has
also worked as a systems administrator for more than ten years and as a systems
manager for a few of those years. While at Sun, Scott worked as a chief architect
for Utility Computing, a backline engineer in Network Technical Support, and a
lead course developer for Internal Technical Training.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

This chapter introduces ZFS, a new kind of ﬁle system from Sun Microsystems that
is elegant in its simplicity, particularly in the way that you can easily manage data
and storage. The ZFS ﬁle system is available in the following Solaris releases:

Solaris 10 6/06 release (ZFS root and install features available in Solaris 10
10/08 release)
OpenSolaris 2008.11 release (includes ZFS root and install features)

1.1 Overview of ZFS

ZFS is an all-purpose file system that is built on top of a pool of storage devices.
ZFS doesn’t use a middle layer of virtualized volumes like other volume manage-
ment products. This means you can easily aggregate your storage components with
a few ZFS commands.
File systems that are created from a storage pool share space with the other file
systems in the pool. You don’t have to allocate storage space based on the intended
size of a file system because file systems grow automatically within the space that
is allocated to the storage pool. When new storage devices are added, all file sys-
tems in the pool can immediately use the additional space.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

ZFS provides the following redundancy conﬁgurations:

Mirror
RAID-Z (single parity), also called RAID-Z1
RAID-Z2 (double parity)

RAID-Z is similar to RAID-5 but uses a variable-stripe width to eliminate the

RAID-5 write hole (stripe corruption due to a loss of power between data and par-
ity updates). All RAID-Z writes are full-stripe writes. See Jeff Bonwick’s blog on
RAID-Z at http://blogs.sun.com/bonwick/entry/raid_z.
Additionally, 256 checksums on data blocks protect data against corruption.
Using disk scrubbing on live-disk ZFS can detect errors that otherwise go undetec-
ted and that can be corrected before the data is lost. See Eric Lowe’s entry on
silent data corruption at http://blogs.sun.com/elowe/entry/zfs_saves_
the_day_ta.
ZFS is a smarter file system. If it detects damaged data, it will go back and
retrieve another copy of the data. You can enable another layer of redundancy in
the ZFS file system by increasing the number of copies of data written to disk in
the file system. This feature was added to Solaris 10 10/08.
ZFS also provides built-in compression. Compression reduces the amount of I/O
to the disk by two to three times by reducing the amount of data written. You can
select different compression algorithms depending on the workload and data
stored.
Additionally, ZFS supports raw volumes. Now applications that require raw
devices can take advantage of ZFS. You can use ZFS for swap devices, and it is
installed by default when selecting ZFS as the root file system.

1.1.1 Advantages of Using ZFS

The advantages of using ZFS are numerous:

Data inconsistency detection and recovery are provided by ZFS redundancy

features. ZFS guarantees data integrity from disk to memory and from mem-
ory to disk. See Bonwick’s blog at http://blogs.sun.com/bonwick/en_
US/entry/zfs_end_to_end_data.
It is easy to set up a mirrored ZFS root pool configuration so that you don’t
need additional volume management products to provide this feature.
ZFS boot environments are created instantly and can be used for patching
and upgrading.
ZFS makes it easy to create and manage file systems without needing
complex command strings or editing multiple configuration files.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

ZFS makes it easy to set up file system quotas or reservations, turn compre-
sion on or off, or set other characteristics per file system.
ZFS provides constant-time snapshots so that an administrator or users can
roll back copies of the data.
ZFS makes it possible to delegate administration directly to users. Using the
RBAC facility, ZFS administration can be delegated to users.
ZFS provides automation of mundane tasks. Once a ZFS file system is cre-
ated, Solaris will remember to mount the file system on reboot.

Licensed by
ZFS provides a history of commands used on a pool or ﬁle system.

1.1.2 A Top-Level View of ZFS

Onur Bingul
To get a better understanding of ZFS, it helps to see the whole before talking about
the parts of ZFS. ZFS can be split into two major parts: the ZFS pool and the ZFS
datasets. The zpool command manages the ZFS pool, and the zfs command man-
ages the ZFS datasets. Each of these commands has subcommands that manage

3690014
each part from cradle to grave.
The fundamental base of ZFS is the ZFS pool. A pool is the primary storage ele-
ment that supports the ZFS datasets, as illustrated in Figure 1.1. The pool acts
much like the memory of a computer system. When a dataset needs another
chunk, ZFS will allocate more storage from the pool. When a ﬁle is deleted, the free
storage returns to the pool for all datasets in the pool to use.

Pool

Figure 1.1 The base of ZFS: the pool

A ZFS pool can contain many ZFS datasets, as shown in Figure 1.2. Each
dataset can grow and shrink as needed. Using snapshots, quotas, and reservations
can affect how much each dataset can consume of the ZFS pool.
ZFS supports three datasets:

ZFS file system: A file system that is mounted for normal usage (such as
home directories or data storage)
Volume: Raw volumes that can be used for swap and dump partitions in a
ZFS boot configuration
Clone: A copy of a ZFS file system or volume

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Pool

File Systems

Figure 1.2 The ZFS pool with three ZFS datasets

At a minimum, a single virtual device (vdev) is needed for a ZFS pool (see
Figure 1.3). A vdev can comprise a ﬁle, a slice of a disk, the whole disk, or even a
logical disk from a hardware array or another volume product such as Solaris Vol-
ume Manager. A vdev supports three redundancy conﬁgurations: mirror, RAID-Z,
and RAID-Z2.

Vdev
Pool

File Systems

Figure 1.3 ZFS pool with a single vdev

In Figure 1.4, the vdev contains a mirrored array made up of two whole disks.
ZFS can handle disks of similar size and different disk geometries. This feature is
good for a home system, where if a disk fails, the replacement disk doesn’t have to
be of the same make and model. The only requirement of a replacement disk is
that it must be the same capacity or larger.

Vdev
Pool

Disks
File Systems

Figure 1.4 A vdev with a mirrored array supporting a ZFS pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Increasing the storage of a ZFS pool is simple. You just add another vdev. In
Figure 1.5, a second mirrored vdev is concatenated to the pool. You can increase a
ZFS pool at any time, and you don’t need to adjust any of the datasets contained
within the pool.

Vdev

Pool

Disks

File Systems

Disks

Expanded Pool

Vdev

Figure 1.5 Concatenation of a second mirrored vdev to the ZFS pool

Once a ZFS pool is expanded, the data sets in the pool can take advantage of the
increased size. ZFS will take advantage of the expanded storage and distribute the
new writes across all the disk spindles (see Figure 1.6).

Vdev
Pool

Disks
File Systems

Disks

Vdev

Figure 1.6 File system expanding in the larger ZFS pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1.2 Fast and Simple Storage

ZFS was designed to be fast and simple with only two commands to remember:
zpool and zfs. You can create a ZFS pool and have a writable ﬁle system in one
step. To create the conﬁguration in Figure 1.4, you just execute the following
command:

$ time pfexec zpool create mpool mirror c5t0d0 c5t1d0

real 0m6.155s
user 0m0.216s
sys 0m1.456s

By using the time command, you can see it took six seconds to create a new
ZFS pool. In this example you used the pfexec command to execute the zpool
command as root. You can then view the status of the new ZFS pool by running the
following command:

$ zpool status mpool

pool: mpool
state: ONLINE
scrub: none requested
config:

NAME STATE READ WRITE CKSUM

mpool ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t0d0 ONLINE 0 0 0
c5t1d0 ONLINE 0 0 0

errors: No known data errors

With the previous code, you have created a new pool called mpool with a new
virtual device that consists of two disks in mirror conﬁguration. At this point, you
can write to the newly mounted ﬁle system:

$ df -h | grep mpool
mpool 85M 18K 85M 1% /mpool
$ pfexec touch /mpool/j
$ ls -l /mpool/j
-rw-r--r-- 1 root root 0 2009-04-24 01:20 /mpool/j

Just by using one command line, you created a mirrored pool and mounted it.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1.3 ZFS Commands

ZFS has only two commands. The zpool command creates, modifies, and destroys
ZFS pools. The zfs command creates, modifies, and destroys ZFS file systems. In
this book, Chapters 2 and 5 are dedicated to administration of ZFS pools, and the
other chapters use the zfs command as part of the administrative tasks.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

ZFS storage pools are the basis of the ZFS system. In this chapter, I cover the basic
concepts, configurations, and administrative tasks of ZFS pools. In Chapter 5, I
cover advanced configuration topics in ZFS pool administration.
The zpool command manages ZFS storage pools. The zpool command creates,
modifies, and destroys ZFS pools.
Redundant configurations supported in ZFS are mirror (RAID-1), RAID-Z (simi-
lar to RAID-5), and RAID-Z2 with a double parity (similar to RAID-6). All tradi-
tional RAID-5-like algorithms (RAID-4, RAID-6, RDP, and EVEN-ODD, for
example) suffer from a problem known as the RAID-5 write hole. If only part of a
RAID-5 stripe is written and power is lost before all blocks have made it to disk, the
parity will remain out of sync with the data (and is therefore useless) forever, unless
a subsequent full-stripe write overwrites it. In RAID-Z, ZFS uses variable-width
RAID stripes so that all writes are full-stripe writes. This design is possible only
because ZFS integrates file system and device management in such a way that the
file system’s metadata has enough information about the underlying data redun-
dancy model to handle variable-width RAID stripes. RAID-Z is the world’s first
software-only solution to the RAID-5 write hole.

2.1 ZFS Pool Concepts

ZFS pools are comprised of virtual devices. ZFS abstracts every physical device
into a virtual device. A vdev can be a disk, a slice of a disk, a ﬁle, or even a logical

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

volume presented by another volume manager such as Solaris Volume Manager

(SVM) or a LUN from a hardware RAID device.
These are the virtual devices types:

Dynamic stripe: A dynamic stripe is a nonredundant conﬁguration of a sim-

ple disk or concatenation of disks.
Redundant group (mirror, RAID-Z1, or RAID-Z2): A mirror can be a two-
way or three-way mirror. RAID-Z groups are recommended to have up to nine
disks in the group. If there are more, then multiple vdevs are recommended.
Two disks minimum are needed for RAID-Z, and three disks at a minimum
are needed for RAID-Z2. (Note that RAID-Z and RAID-Z1 are interchange-
able terms. With the introduction of the RAID-Z2 feature, the term RAID-Z
evolved into RAID-Z1 to differentiate it from RAID-Z2. I use both terms in
this book.)
Spare: A spare vdev is for a hot standby disk replacement.
Log: A log vdev is for ZFS Intent Log (ZIL). The ZIL increases the write per-
formance on ZFS. Only dynamic stripe and mirrored vdev conﬁgurations are
supported for this vdev type.
Cache: A cache vdev is used to speed up random reads from a RAID-Z-conﬁgured
pool. Its intended use is for read-heavy workloads. There is no redundancy support
at this point for this vdev type. If there is a read error, then ZFS will read from the
original storage pool.

Log and cache vdevs are used with solid-state disks (SSDs) in the Sun Storage
7000 series in hybrid storage pools (HSPs; see http://blogs.sun.com/ahl/
entry/fishworks_launch).
Best-practice guidelines for ZFS pools include the following:

Mirrored conﬁguration beyond a three-way mirror should not be used. Think

about using a RAID-Z conﬁguration instead.
Use RAID-Z or RAID-Z2 virtual device groups with fewer than ten disks in
each vdev.
Using whole disks is best. ZFS works best with “just a bunch of disks”
(JBOD).
Use slices for vdev groups only for boot disks.
Use disks of a terabyte or less for boot devices.
Use matched-capacity disks (mixed geometry is OK) for the best maximum
storage results.
Use matching sizes of vdevs in a ZFS pool. Match the number of disks and
redundancy groups in each vdev in a pool for best performance.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Creating/adding new vdevs to a ZFS pool is the most unforgiving part about ZFS.
Once committed, some operations cannot be undone. The zpool command will warn
you, however, if the operation is not what’s expected. There is a force option in
zpool to bypass any of the warnings, but it is not recommended that you use the
force option unless you are sure you will not need to reverse the operation.
These are the rules for ZFS pools:

Once a normal (dynamic stripe) vdev is added to a ZFS pool, it cannot be

removed.
Only the special-use vdevs can be removed: spares, log, and cache.
Disks the same size or larger can be replaced within a vdev.
Disks can be added to a single disk or mirrored vdev to form a mirror or a
three-way mirror.
New disks cannot be added to an existing RAID-Z or RAID-Z2 vdev
conﬁguration.

2.2 Creating a Dynamic Stripe

A dynamic stripe is the most basic pool that can be created. There is no redun-
dancy in this conﬁguration. If any disk fails, then the whole pool is lost. Any pool
created with multiple vdevs will dynamically stripe across each vdev or physical
device.
You can use the zpool command with the subcommand to create a dynamic
stripe. After the create subcommand is the name of the new ZFS pool, dstripe,
and the disks that will comprise the pool.
# zpool create dstripe c5t0d0 c5t1d0
The following listing presents the results of creating the ZFS pool dstripe. On
line 2, zpool list is executed to list all the ZFS pools on the system. Line 3
starts a list of the available pools. The command gives you general information
about the ZFS pools.

1 # zpool create dstripe c5t0d0 c5t1d0

2 # zpool list
3 NAME SIZE USED AVAIL CAP HEALTH ALTROOT
4 dstripe 234M 75K 234M 0% ONLINE -
5 rpool 15.9G 3.21G 12.7G 20% ONLINE -

Next on line 6, zpool status is issued to inquire about the status of the ZFS
pools. Starting at line 7, the status of the ZFS pool dstripe is displayed, with a nor-
mal status. Reading the config: section of the output starting at line 10, the pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

dstripe is shown as two concatenated disks. Lines 14 and 15 list the vdevs (c5t0d0
and c5t1d0) that belong to the pool dstripe. The second pool listed is made of a sin-
gle disk called rpool, conﬁgured as a dynamic stripe with only a single vdev
(c3d0s0). It was created as part of the OS installation process.

6 # zpool status
7 pool: dstripe
8 state: ONLINE
9 scrub: none requested
10 config:
11
12 NAME STATE READ WRITE CKSUM
13 dstripe ONLINE 0 0 0
14 c5t0d0 ONLINE 0 0 0
15 c5t1d0 ONLINE 0 0 0
16
17 errors: No known data errors
18
19 pool: rpool
20 state: ONLINE
21 scrub: none requested
22 config:
23
24 NAME STATE READ WRITE CKSUM
25 rpool ONLINE 0 0 0
26 c3d0s0 ONLINE 0 0 0
27
28 errors: No known data errors

Figure 2.1 illustrates the resulting dynamic pool with its two vdevs of single
nonredundant disks. Any problems with the disks (sector errors or disk failure)
may result in the loss of the whole pool or data.

Vdev
Pool dstripe

Disk

Vdev

Figure 2.1 Dynamic stripe pool with two vdevs

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

2.3 Creating a Pool with Mirrored Devices

The following command creates a ZFS mirrored pool called mpool with a mirrored
vdev. As expected, the new pool mpool is about half the capacity of dstripe. The
pool dstripe is a concatenation of two disks of the same capacity, and mpool is a
mirror of a disk of the same capacity.
The following command line shows how to use the zpool command with the
subcommand create to create a pool with mirrored vdevs. After the create sub-
command is the name of the new ZFS pool, mpool. The mirror subcommand will

Licensed by
create a mirrored vdev with the disks c5t2d0 and c5t3d0.
# zpool create mpool mirror c5t2d0 c5t3d0
The following output is the creation of a mirrored ZFS pool called mpool. Using

Onur Bingul
the zpool list command, you now can see the capacity of the newly created pool
on line 5. Notice it is half the capacity of the dstripe pool.

3690014
1 # zpool create mpool mirror c5t2d0 c5t3d0
2 # zpool list
3 NAME SIZE USED AVAIL CAP HEALTH ALTROOT
4 dstripe 234M 75K 234M 0% ONLINE -
5 mpool 117M 73.5K 117M 0% ONLINE -
6 rpool 15.9G 3.21G 12.7G 20% ONLINE -

Starting at line 20 (in the following part of the listing) is the status information of
mpool. The disks c5t2d0 and c5t3d0 are configured as a mirror vdev, and the mirror
is part of mpool. This is an important concept in reading the status information of
ZFS pools. Notice the indentations on the pool name notations. The first is the name
of the ZFS pool. Then at the first indentation of the name are the vdevs that are part
of the pool. In the case of a dynamic stripe, this is the physical device. If the pool is
created with redundant vdev(s), the first indentation will be mirror, raidz1, or
raidz2. Then the next indentation will be the physical devices that comprise the
redundant vdev. On lines 13 and 25 are the names dstripe and mpool, respectively.
On lines 14 and 15 are the disks that belong to dstripe on the first indentation. On
line 25 is the mirror configuration, and the next indented line lists the disks belong-
ing to the mirror configuration. Compare Figures 2.1 and 2.2 for a graphical repre-
sentation of each pool’s configuration.

7 # zpool status
8 pool: dstripe
9 state: ONLINE
10 scrub: none requested
11 config:

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

12
13 NAME STATE READ WRITE CKSUM
14 dstripe ONLINE 0 0 0
15 c5t0d0 ONLINE 0 0 0
16 c5t1d0 ONLINE 0 0 0
17
18 errors: No known data errors
19
20 pool: mpool
21 state: ONLINE
22 scrub: none requested
23 config:
24 NAME STATE READ WRITE CKSUM
25 mpool ONLINE 0 0 0
26 mirror ONLINE 0 0 0
27 c5t2d0 ONLINE 0 0 0
28 c5t3d0 ONLINE 0 0 0
29
30 errors: No known data errors
31
32 pool: rpool
33 state: ONLINE
34 scrub: none requested
35 config:
36
37 NAME STATE READ WRITE CKSUM
38 rpool ONLINE 0 0 0
39 c3d0s0 ONLINE 0 0 0
40 errors: No known data errors

Figure 2.2 illustrates the results of creating the ZFS pool mpool with one mir-
rored vdev.

Vdev Mirror
Pool mpool

Disks

Figure 2.2 Pool mpool with a mirrored vdev

The following sequence adds a second mirror vdev to the pool, doubling the size
of pool mpool. Notice that after the addition of the second mirror vdev, the dstripe
and mpool pools are the same capacity. Line 30 is the mirrored vdev with the phys-
ical disks c5t4d0 and c5t5d0 indented.

1. # zpool add mpool mirror c5t4d0 c5t5d0

2. # zpool list
3. NAME SIZE USED AVAIL CAP HEALTH ALTROOT
4. dstripe 234M 75K 234M 0% ONLINE -

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

5. mpool 234M 78K 234M 0% ONLINE -

6. rpool 15.9G 3.21G 12.7G 20% ONLINE -
7. # zpool status
8. pool: dstripe
9. state: ONLINE
10. scrub: none requested
11. config:
12.
13. NAME STATE READ WRITE CKSUM
14. dstripe ONLINE 0 0 0
15. c5t0d0 ONLINE 0 0 0
16. c5t1d0 ONLINE 0 0 0
17.
18. errors: No known data errors
19.
20. pool: mpool
21. state: ONLINE
22. scrub: none requested
23. config:
24.
25. NAME STATE READ WRITE CKSUM
26. mpool ONLINE 0 0 0
27. mirror ONLINE 0 0 0
28. c5t2d0 ONLINE 0 0 0
29. c5t3d0 ONLINE 0 0 0
30. mirror ONLINE 0 0 0
31. c5t4d0 ONLINE 0 0 0
32. c5t5d0 ONLINE 0 0 0
33.
34. errors: No known data errors
35.
36. pool: rpool
37. state: ONLINE
38. scrub: none requested
39. config:

40. NAME STATE READ WRITE CKSUM

41. rpool ONLINE 0 0 0
42. c3d0s0 ONLINE 0 0 0

43. errors: No known data errors

Figure 2.3 shows the results of adding a second mirrored vdev to ZFS pool
mpool.

2.4 Creating a Pool with RAID-Z Devices

In ZFS you can also create redundant vdevs similar to RAID-5, called RAID-Z. To
create a pool with double parity, you would use RAID-Z2 instead.
The following command line will create a ZFS pool named rzpool with two
RAID-Z1 vdevs, each with four disks:
# zpool create rzpool raidz1 c5t6d0 c5t7d0 c5t8d0 c5t9d0 raidz1
c5t10d0 c5t11d0 c5t12d0 c5t13d0
The ﬁrst RAID-Z1 vdev consists of disks c5t6d0, c5t7d0, c5t8d0, and c5t9d0, and
the second RAID-Z1 vdev has c5t6d0, c5t7d0, c5t8d0, and c5t9d0 as members.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Vdev Mirror
Pool mpool

Disks

Vdev Mirror

Figure 2.3 Pool mpool with the added mirror vdev

The zpool list command shows the summary status of the new pool rzpool.
The output of zpool status, on lines 16 and 21, shows the RAID-Z1 virtual
devices and physical disk devices that make up each RAID-Z1 virtual device.

1 # zpool create rzpool raidz1 c5t6d0 c5t7d0 c5t8d0 c5t9d0

raidz1 c5t10d0 c5t11d0 c5t12d0 c5t13d0
2 # zpool list
3 NAME SIZE USED AVAIL CAP HEALTH ALTROOT
4 dstripe 234M 75K 234M 0% ONLINE -
5 mpool 234M 78K 234M 0% ONLINE -
6 rpool 15.9G 3.21G 12.7G 20% ONLINE -
7 rzpool 936M 138K 936M 0% ONLINE -
8 # zpool status
…lines deleted…
9 pool: rzpool
10 state: ONLINE
11 scrub: none requested
12 config:
13
14 NAME STATE READ WRITE CKSUM
15 rzpool ONLINE 0 0 0
16 raidz1 ONLINE 0 0 0
17 c5t6d0 ONLINE 0 0 0
18 c5t7d0 ONLINE 0 0 0
19 c5t8d0 ONLINE 0 0 0
20 c5t9d0 ONLINE 0 0 0
21 raidz1 ONLINE 0 0 0
22 c5t10d0 ONLINE 0 0 0
23 c5t11d0 ONLINE 0 0 0
24 c5t12d0 ONLINE 0 0 0
25 c5t13d0 ONLINE 0 0 0
26
27 errors: No known data errors

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 2.4 has two RAID-Z1 vdevs with four physical disks to each vdev. The
vdevs are concatenated to form the ZFS pool rzpool.

RAID-Z1 Vdev
Pool rzpool

Disks

RAID-Z1 Vdev

Figure 2.4 Pool rzpool using two RAID-Z1 redundant vdevs

2.5 Creating a Spare in a Storage Pool

You can add spare vdevs to a pool to increase its reliability and maintain perfor-
mance. In case of disk failure, ZFS can replace failed disks with spare disks auto-
matically. This gives the administrator time to make repairs without compromising
the reliability and performance of the pool. Spares can be shared among multiple
pools. The following example adds a spare disk device to the pool mpool. On line 22,
the spare vdev is now listed as part of the pool mpool.

1 # zpool add mpool spare c5t14d0

2 # zpool list
3 NAME SIZE USED AVAIL CAP HEALTH ALTROOT
4 dstripe 234M 75K 234M 0% ONLINE -
5 mpool 234M 84K 234M 0% ONLINE -
6 rpool 15.9G 3.20G 12.7G 20% ONLINE -
7 rzpool 936M 141K 936M 0% ONLINE -
8 # zpool status mpool
9 pool: mpool
10 state: ONLINE
11 scrub: none requested

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

12 config:
13
14 NAME STATE READ WRITE CKSUM
15 mpool ONLINE 0 0 0
16 mirror ONLINE 0 0 0
17 c5t2d0 ONLINE 0 0 0
18 c5t3d0 ONLINE 0 0 0
19 mirror ONLINE 0 0 0
20 c5t4d0 ONLINE 0 0 0
21 c5t5d0 ONLINE 0 0 0
22 spares
23 c5t14d0 AVAIL
24
25 errors: No known data errors

In Figure 2.5, an additional vdev is added to the ZFS pool mpool. A spare vdev
can be shared with multiple ZFS pools. The spares must have at least the same
capacity of the smallest disk in the other vdev devices.

Vdev Mirror
Pool mpool

Disks

Spare Vdev
Vdev Mirror

Figure 2.5 Pool mpool with a spare vdev added

2.6 Adding a Spare Vdev to a Second Storage Pool

As indicated earlier in the chapter, you can share a spare vdev with other pools on
the same system. In the following example, the spare vdev (c5t14d0) disk is shared
with ZFS pool rzpool. After you add the spare to the rzpool, the spare disk now

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

appears in ZFS pools mpool and rzpool, namely, on lines 16 and 17 and lines 38
and 39.

1 # zpool add rzpool spare c5t14d0

2 # zpool status mpool rzpool
3 pool: mpool
4 state: ONLINE
5 scrub: none requested
6 config:
7
8 NAME STATE READ WRITE CKSUM
9 mpool ONLINE 0 0 0
10 mirror ONLINE 0 0 0
11 c5t2d0 ONLINE 0 0 0
12 c5t3d0 ONLINE 0 0 0
13 mirror ONLINE 0 0 0
14 c5t4d0 ONLINE 0 0 0
15 c5t5d0 ONLINE 0 0 0
16 spares
17 c5t14d0 AVAIL
18
19 errors: No known data errors
20
21 pool: rzpool
22 state: ONLINE
23 scrub: none requested
24 config:
25
26 NAME STATE READ WRITE CKSUM
27 rzpool ONLINE 0 0 0
28 raidz1 ONLINE 0 0 0
29 c5t6d0 ONLINE 0 0 0
30 c5t7d0 ONLINE 0 0 0
31 c5t8d0 ONLINE 0 0 0
32 c5t9d0 ONLINE 0 0 0
33 raidz1 ONLINE 0 0 0
34 c5t10d0 ONLINE 0 0 0
35 c5t11d0 ONLINE 0 0 0
36 c5t12d0 ONLINE 0 0 0
37 c5t13d0 ONLINE 0 0 0
38 spares
39 c5t14d0 AVAIL
40
41 errors: No known data errors

In Figure 2.6, the spare vdev is shared between the mpool and rzpool ZFS pools.
Each pool can use the spare when needed.

2.7 Replacing Bad Devices Automatically

ZFS has the capability to replace a disk in a pool automatically without interven-
tion by the administrator. This feature, known as autoreplace, is turned off by
default. This feature will allow ZFS to replace a bad disk with a spare from the
spares pool, automatically allowing the pool to operate at optimum performance.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Vdev Vdev
Pool Pool
mpool rzpool

Spare Vdev

Figure 2.6 Pools mpool and rzpool sharing a spare vdev

This allows the administrator to replace the failed drive at a later time. The man-
ual disk replacement procedure is covered later in this chapter.
If you list the properties of the ZFS pool, you can see that the autoreplace fea-
ture is turned off using the get subcommand.

1 # zpool get all mpool

2 NAME PROPERTY VALUE SOURCE
3 mpool size 234M -
4 mpool used 111K -
5 mpool available 234M -
6 mpool capacity 0% -
7 mpool altroot - default
8 mpool health DEGRADED -
9 mpool guid 11612108450022771594 -
10 mpool version 13 default
11 mpool bootfs - default
12 mpool delegation on default
13 mpool autoreplace off default
14 mpool cachefile - default
15 mpool failmode wait default
16 mpool listsnapshots off default

To turn on the autoreplace feature, use the following command line:

# zpool autoreplace=on mpool
To simulate a bad device, I shut down the system and removed c5t4d0 from the
system. The system now cannot contact the disk and has marked the removed disk
as unavailable. With the system rebooted, you can examine the output of the
zpool status command:

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1 $ zpool status mpool

2 pool: mpool
3 state: DEGRADED
4 status: One or more devices could not be opened. Sufficient replicas exist for
5 the pool to continue functioning in a degraded state.
6 action: Attach the missing device and online it using 'zpool online'.
7 see: http://www.sun.com/msg/ZFS-8000-2Q
8 scrub: resilver completed after 0h0m with 0 errors on Mon Apr 6 00:52:36 2009
9 config:
10
11 NAME STATE READ WRITE CKSUM
12 mpool DEGRADED 0 0 0
13 mirror ONLINE 0 0 0
14 c5t2d0 ONLINE 0 0 0
15 c5t3d0 ONLINE 0 0 0
16 mirror DEGRADED 0 0 0
17 spare DEGRADED 0 0 0
18 c5t4d0 UNAVAIL 0 89 0 cannot open
19 c5t14d0 ONLINE 0 0 0 31K resilvered
20 c5t5d0 ONLINE 0 0 0 31K resilvered
21 spares
22 c5t14d0 INUSE currently in use
23
24 errors: No known data errors

On line 3, the state of the pool has been degraded, and line 4 tells you that the
pool can continue in this state. On lines 6 and 7, ZFS tells you what actions you will
need to take, and by going to the Web site, a more detailed message tells you how to
correct the problem. Line 19 tells you that the spare disk has been resilvered with
disk c5t5d0. Line 22 now gives you the new status of the spare disk c5t14d0.
The original deﬁnition of resilver is the process of restoring a glass mirror with
a new silver backing. In ZFS, it is a re-creation of data by copying from one disk to
another. In other volume management systems, the process is called resynchroni-
zation. Continuing the example, you can shut down the system and attach a new
disk in the same location of the missing disk. The new disk at location c5t4d0 is
automatically resilvered to the mirrored vdev, and the spare disk is put back to an
available state.

$ zpool status mpool

pool: mpool
state: ONLINE
scrub: resilver completed after 0h0m with 0 errors on Mon Apr 6 02:21:05 2009
config:

NAME STATE READ WRITE CKSUM

mpool ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t2d0 ONLINE 0 0 0
c5t3d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t4d0 ONLINE 0 0 0 23K resilvered
c5t5d0 ONLINE 0 0 0 23K resilvered
spares
c5t14d0 AVAIL

errors: No known data errors

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

If the original disk is reattached to the system, ZFS does not handle this case
with the same grace. Once the system is booted, the original disk must be detached
from the ZFS pool. Next, the spare disk (c5d14d0) must be replaced with the origi-
nal disk. The last step is to place the spare disk back into the spares group.

$ pfexec zpool detach mpool c5t4d0

$ pfexec zpool replace mpool c5t14d0 c5t4d0
$ pfexec zpool add mpool spare c5t14d0
$ zpool status mpool
pool: mpool
state: ONLINE
scrub: resilver completed after 0h0m with 0 errors on Mon Apr 6 02:50:25 2009
config:

NAME STATE READ WRITE CKSUM

mpool ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t2d0 ONLINE 0 0 0
c5t3d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t4d0 ONLINE 0 0 0 40.5K resilvered
c5t5d0 ONLINE 0 0 0 40.5K resilvered
spares
c5t14d0 AVAIL

errors: No known data errors

2.8 Locating Disks for Replacement

Locating the correct disk for replacement is crucial. If the wrong disk is replaced,
it could corrupt the whole ZFS pool. There is a simple process for locating a disk in
an array using the format command. This technique works best when the disks
are relatively quiet and have disk-access LEDs that can be seen while in operation.
If the disk is still accessible to Solaris, follow these steps:

1. Start the format command, and select the disk you want to locate by select-
ing the number before the drive.
2. Type analyze, and hit the Enter/Return key.
3. Select read test by typing read, and then hit Enter/Return.
4. Look at the array, and ﬁnd the disk LED with constant access.
5. Once the disk is located, stop the read test by pressing Ctrl+C.
6. To exit the format utility, type quit and hit Return, and then type quit and
hit Return again.
7. Replace the disk according to manufacturer’s instructions.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

If the damaged disk is not seen by format, try to light up the LEDs of the disks
above and below the target number of the damaged disk. For example, if you were
looking to ﬁnd c5t6d0 and it was not seen by the format command, you would ﬁrst
locate c5t5d0 by using the format read test and then locate c5t7d0 by using the
same method. The drive in between should be c5t6d0.

2.9 Example of a Misconﬁgured Pool

Licensed by
The following is an example of a ZFS pool that started as a RAID-Z pool and
needed more space. The administrator just added the new disks to the pool. The
status output follows.

$ zpool status -v
pool: mypool
state: ONLINE
Onur Bingul
status: One or more devices has experienced an error resulting in data

3690014
corruption. Applications may be affected.
action: Restore the file in question if possible. Otherwise restore the entire pool
from backup.
see: http://www.sun.com/msg/ZFS-8000-8A
scrub: resilver completed with 0 errors on Tue Nov 18 17:05:10 2008
config:

NAME STATE READ WRITE CKSUM

mypool ONLINE 0 0 18
raidz1 ONLINE 0 0 0
c2t0d0 ONLINE 0 0 0
c2t1d0 ONLINE 0 0 0
c2t2d0 ONLINE 0 0 0
c2t3d0 ONLINE 0 0 0
c2t4d0 ONLINE 0 0 0
c2t5d0 ONLINE 0 0 0
c2t9d0 ONLINE 0 0 0
c2t11d0 ONLINE 0 0 0
c2t12d0 ONLINE 0 0 18
c2t10d0 ONLINE 0 0 0
c2t13d0 ONLINE 0 0 0
spares
c2t8d0 AVAIL

errors: Permanent errors have been detected in the following files:

mypool/dataset/u01:<0x2a07a>
mypool/dataset/u01:<0x2da66>

c2t9d0, c2t10d0, c2t11d0, c2t12d0, and c2t13d0 are single-disk vdevs. They have
no redundancy. The spare vdev will not help because there is nothing to resilver
with when one of the single-disk vdevs fails. In this case, ZFS has detected perma-
nent errors in mypool/dataset/u01 and has suggested an action of restoring the
ﬁles from backup or restoring the whole pool from backup.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

The ZFS pool is in danger of a single disk failure that may destroy the whole pool.
The only way to get this pool to be redundant is to mirror all the single-vdev drives.
Figure 2.7 represents the ZFS pool mypool. The figure shows five single-disk
vdevs with a single RAID-Z vdev. Any damage to the single disks will cause ZFS to
lose data, or, worse, the pool will fail.
You can correct this configuration for pool redundancy in two ways:

If there are enough disks, you can mirror the single-disk vdevs.
You can back up all the ﬁle systems and re-create the pool. Then do a full
restore of the ﬁle systems.

Vdev

Pool mypool

Figure 2.7 The ZFS pool mypool has a single RAID-Z vdev and ﬁve single-disk vdevs

The procedure to replace a single-disk vdev in the preceding example is to mir-

ror the drive in question and then detach the bad drive. What follows is the com-
mand-line procedure:

1. Mirror the drive with bad sector problems:

# zpool replace mypool <baddisk> <gooddisk>
2. Detach the bad disk:
# zpool detach mypool <baddisk>

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

This chapter describes how to install and boot a ZFS root ﬁle system and how to
recover from a root pool disk failure. Upgrading and patching ZFS boot environ-
ments are also covered.
The capability to install and boot a ZFS root ﬁle system is available in the fol-
lowing Solaris releases:

Solaris 10 10/08 release: The ZFS root ﬁle system is optional.

OpenSolaris 2008.11: The ZFS root ﬁle system is installed by default.

3.1 Simplifying (Systems) Administration Using ZFS

ZFS installation and boot features simplify the administration process. Currently,
ZFS ﬁle systems need a partition table prior to the installation of the Solaris OS.
With traditional ﬁle systems, the following questions typically need to be
answered:

How large does the root (/) ﬁle system need to be?
How large does /usr need to be?
How large does the swap partition need to be?
How large does the /var partition need to be?

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

How much disk space needs to be allocated to use Solaris Live Upgrade?
Will the current requirements change in 6 to 12 months?

With ZFS, answering the preceding questions at installation time is unnecessary.

In the past, a UNIX OS could be booted with just the root (/) file system, and
disk reliability was not as good as today. By creating separate partitions, a system
administrator could boot from the root (/) file system and repair a system with a
few tools available in the root (/) file system. Today, it is not possible to boot with-
out the /usr file system or directory.
In a ZFS boot environment, a separate /var file system is a good idea because
you can set a quota so that if a runaway process exists or a denial-of-service attack
occurs, the problem can be limited to that file system.
Disk partition sizes do not need to be decided at installation time when you use
a ZFS root file system. Swap, dump, and boot environments can be dynamically
allocated as needed. Swap and dump areas can be dynamically resized as require-
ments change. If more RAM is added to a system, swap and dump areas can be
adjusted to accommodate new hardware.
Boot environments are managed by either Solaris Live Upgrade (Solaris 10) or
the beadm command (OpenSolaris releases). The beadm command and the pkg
commands simplify upgrading and patching the OpenSolaris releases. The advan-
tages of using ZFS boot environments are as follows:

Quick rollbacks
Less downtime
Quick cloning
Safer patching
Safer upgrades
Simple administration

3.2 Installing a ZFS Root File System

Starting with the Solaris 10 10/08 release, a specialized installer provides the
option to install a ZFS root file system. In addition, both the Solaris Live
Upgrade and JumpStart installation methods support a ZFS root file system.
The Solaris Live Upgrade feature enables an administrator to upgrade a system
with minimal downtime, often just the time it takes to reboot. Should a problem
arise that cannot be fixed quickly, the system can be rebooted to the previous
state. JumpStart is a Solaris network installation facility that supports two

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

modes: a normal install and a binary image install called Flash. Currently, Flash
installation is not supported for ZFS. An unsupported workaround is provided at
http://blogs.sun.com/scottdickson/entry/a_much_better_way_to.
A supported but limited ZFS and Flash solution should be available in an
upcoming Solaris 10 release.
By default, the OpenSolaris release 2008.11 installs a ZFS root ﬁle system. No
option exists for installing a UFS root ﬁle system.

3.2.1 ZFS Root File System Requirements

The following are the requirements for installing a ZFS root ﬁle system. Most of
the requirements are the same as those for installing a Solaris release. The mini-
mum requirement for disk space should be followed carefully.

Memory requirements
768MB RAM minimum
1GB RAM recommended for better overall ZFS performance
Disk requirements
16GB is the minimum required to install a ZFS boot OS.
Maximum swap and dump sizes are based on physical memory and kernel
needs at boot time but are generally sized no more than 2GB by the instal-
lation program.
The ZFS storage pool must be a nonredundant disk slice conﬁguration or a
mirrored disk slice conﬁguration with the following characteristics:
• There are no RAID-Z or RAID-Z2 vdevs and no separate log devices.
• Disks must be labeled SMI.
• Disks must be less than 1TB.
• Disks on x86 systems must have an fdisk partition.
Compression can be enabled. The default is the lzjb algorithm. The gzip
algorithm is also available, and compression is selectable.

3.2.2 ZFS Root Pool Recommendations

Review the following root pool recommendations:

Use a mirrored root pool to reduce downtime due to hardware failures.

Use one root pool for all data on small systems such as laptops and systems
with two to four disks.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Consider using one root pool and one nonroot pool for data management on
larger systems where the performance of faster hardware is a factor in deter-
mining the best conﬁguration.
Use Solaris Live Upgrade to upgrade and patch ZFS boot environments.
Keep root pool snapshots for recovery purposes.

3.2.3 Interactive ZFS Root File System Installation Example

The following example illustrates using a Solaris 10 10/08 x86 DVD to install a
ZFS root ﬁle system. This example includes only the ZFS installation information.

1. Boot from the installation DVD or the network.

2. Select Solaris from the GRUB menu (see Figure 3.1).

Figure 3.1 GRUB menu from DVD boot

3. Select the Solaris Interactive Text installation option from the menu, either
selection 3 or 4 (see Figure 3.2).
4. Continue answering questions according to your environment.
5. Select ZFS as the root ﬁle system to be installed (see Figure 3.3).
6. In the form shown in Figure 3.4, adjust the size of the swap and dump
devices, if necessary.
7. Select the /var partition on a separate dataset option.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 3.2 Solaris installation menu

Figure 3.3 File system selection menu

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 3.4 Form to adjust swap and dump devices and to select a
separate /var partition

A separate /var partition is a best practice for security and backup/restore

reasons. Separating /var and setting a quota on the dataset can limit
denial-of-service attacks in which a rogue process ﬁlls the entire root disk,
causing the system to crash. By excluding /var during backups, less data
needs to be recorded and restored.
8. Continue with the installation.

3.3 Creating a Mirrored ZFS Root Conﬁguration

You can create a mirrored ZFS root conﬁguration by following these steps:

1. You can create a mirrored ZFS root pool in three basic steps after installation
is complete. Alternatively, a mirrored root pool can be installed from a Jump-
Start installation or during the initial installation process by selecting the
disks to be components of the mirrored ZFS root pool. Copy the partition
information from the source disk to the target disk.
a. Label the disk as SMI. EFI labels are not supported for boot disks.
# fdisk –B /dev/rdsk/c3d1p0

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

b. Partition the boot disk.

If the source and target disks are the same, use syntax similar to the fol-
lowing by substituting your disk devices:
# prtvtoc /dev/rdsk/c3d0s2 | fmthard –s - /dev/rdsk/c3d1s2
If the source and target disks are different sizes, use the format com-
mand, and create a slice 0 from the end of the boot slice (usually 1/0/0) to
the end of the disk.
2. Attach the new disk to the root pool:
# zpool attach rpool c3d0s2 c3d1s2
3. After the new disk is resilvered and you can check the status with the zpool
status command, install the boot block on the newly added disk:
# installgrub /boot/grub/stage1 /boot/grub/stage2 /dev/rdsk/
c3d1s0

3.4 Testing a Mirrored ZFS Root Conﬁguration

Testing a mirrored ZFS root conﬁguration is important to ensure the conﬁguration

works before you really need to boot from the newly added disk. Follow these steps
to test the conﬁguration:

1. Shut down the system:

# init 5
2. Remove or disconnect the primary boot disk.
3. Power on the system, and boot.
4. Log in to the system, and run the following command:
# zpool status rpool
You will see output that looks like Figure 3.5.
5. Shut down the system.
6. Reconnect the primary boot disk.
7. Power on the system, and boot.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 3.5 Output of status of degraded pool rpool

3.5 Creating a Snapshot and Recovering a ZFS Root

File System
A ZFS snapshot is a read-only copy of a file system. You can create and save snapshots
of the ZFS root file system components for recovery in the event of a system failure.
Snapshot and recovery of the ZFS root file system should be planned and prac-
ticed. If you practice these steps in your environment, you will know approxi-
mately how much time it will take to perform a root file system recovery.
The examples in the following sections describe how to snapshot and recover
from a local disk to save time and gain practice. Sending a snapshot to tape or to
another server is preferred but can take more time.

3.5.1 Creating a Snapshot of the ZFS Root Storage Pool

The following steps describe how to create a pool for snapshots and then how to
snapshot the root pool. Although this example uses the OpenSolaris ZFS boot envi-
ronment, the steps are identical for the Solaris 10 release.

1. Create a storage pool for snapshots:

# zpool create –f bpool c5d29
2. Create a ZFS ﬁle system in the snapshot pool:
# zfs create bpool/snap
3. Create a directory for the OpenSolaris boot components:
# mkdir /bpool/snap/opensolaris

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

4. Snapshot the root pool.

# zfs snapshot –r rpool@backup

# zfs list –t snapshot
NAME USED AVAIL REFER MOUNTPOINT
rpool@backup 17K - 72K -
rpool/ROOT@backup 0 - 18K -
rpool/ROOT/opensolaris@install 140M - 2.21G -
rpool/ROOT/opensolaris@backup 80.5M - 2.35G -
rpool/dump@backup 16K - 844M -
rpool/export@backup 15K - 19K -
rpool/export/home@backup 15K - 19K -

Licensed by
rpool/export/home/watanabe@backup 258K - 22.4M -
rpool/swap@backup 0 - 16K -
rzpool/export/home/watanabe@tx 95.8K - 11.0M -

Onur Bingul
3.5.2 Sending the ZFS Root Pool Snapshots to Storage
Sending root pool snapshots to storage can take some time to complete. Depending
on disk, server, or network speeds, this process can consume a few hours. The fol-

3690014
lowing example shows how to send the snapshot to a different ZFS storage pool on
the same system:

# zfs send -v rpool@backup > /bpool/snap/opensolaris/rpool.backup

# zfs send -vR rpool/ROOT/opensolaris@backup >
/bpool/snap/opensolaris/rpool.R.opensolaris.backup
sending from @ to rpool/ROOT/opensolaris@install
sending from @install to rpool/ROOT/opensolaris@backup
# zfs send -vR rpool/export/home@backup > /bpool/snap/opensolaris/export.backup
sending from @ to rpool/export/home@backup
sending from @ to rpool/export/home/watanabe@backup

3.5.3 Restoring the ZFS Root Pool Snapshots to a New Disk

Restoring root pool snapshots to a new disk is very straightforward. Practice these
steps in your environment to learn approximately how much time it will take to
restore from snapshots.

1. Boot from the installation DVD or the network.

2. Exit the installation program, and open a terminal window.
3. Import the snapshot pool:
# zpool –f import bpool
4. Label the disk as SMI:
# fdisk –B /dev/rdsk/c3d1p0

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

5. Partition the boot disk.

Use the format command, and create a slice 0 from the end of the boot slice
(usually 1/0/0) to the end of the disk.
6. Create a new directory to mount the restored root pool:
# mkdir /a
7. Install the boot block on the root pool disk:

# installgrub /boot/grub/stage1 /boot/grub/stage2

/dev/rdsk/c3d1s0

8. Create the new root pool to restore the snapshots:

# zpool create -f -o failmode=continue -R /a -m

legacy -o cachefile=/etc/zfs/zpool.cache rpool c3d1s0

9. Receive each snapshot in order from the top level:

# cat /bpool/snap/opensolaris/rpool.backup | zfs

receive -dvF rpool
# cat /bpool/snap/opensolaris/rpool.R.opensolaris.backup| zfs receive -dvF rpool
# cat /bpool/snap/opensolaris/export.backup| zfs
receive -dvF rpool

10. Set the bootfs property on the root pool:

# zpool set bootfs=rpool/ROOT/opensolaris rpool
11. Create a 2GB dump device:
# zfs create -V 2G rpool/dump
12. Create a 2GB swap device with a 4KB block size:
# zfs create -V 2G -b 4k rpool/swap
13. Reboot the system with the new disk:
# init 6

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

3.6 Managing ZFS Boot Environments with

Solaris Live Upgrade

You can use Solaris Live Upgrade to manage boot environments in the Solaris 10
release. Solaris Live Upgrade copies the current operating environment, and then
you can apply an OS upgrade or a patch cluster to this boot environment. Once the
upgrade process is completed, the new boot environment is activated and booted
with minimal downtime. You can roll back to the previous environment by activat-
ing the previous boot environment and rebooting. Starting in the Solaris 10 10/08
release, you can use Solaris Live Upgrade to migrate a UFS root ﬁle system to a
ZFS root ﬁle system.

3.6.1 Migrating a UFS Root File System to a ZFS Root File System
The following procedure describes how to migrate to a ZFS root file system by
using Solaris Live Upgrade. The one-way migration is supported for the UFS
root file system components only. To migrate nonroot UFS file systems, use the
ufsdump and ufsrestore commands to copy UFS data and restore the data
into a ZFS file system.
The ZFS root file system migration involves three major steps:

1. Preparing the disk and creating a ZFS storage pool

2. Creating a new ZFS boot environment
3. Activating the ZFS boot environment

The following example uses a UFS root ﬁle system on an x86-based system. The
primary disk is c0d0s0, and the target update disk is c0d1.

1. With the disk labeled as SMI, create a new s0 partition and a new ZFS pool:

bash-3.00# fdisk -B /dev/rdsk/c0d1p0

bash-3.00# format
Searching for disks...done

AVAILABLE DISK SELECTIONS:

0. c0d0 <DEFAULT cyl 2085 alt 2 hd 255 sec 63>
/pci@0,0/pci-ide@1,1/ide@0/cmdk@0,0
1. c0d1 <DEFAULT cyl 2085 alt 2 hd 255 sec 63>
/pci@0,0/pci-ide@1,1/ide@0/cmdk@1,0
Specify disk (enter its number): 1
selecting c0d1

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Controller working list found

[disk formatted, defect list found]

FORMAT MENU:
disk - select a disk
type - select (define) a disk type
partition - select (define) a partition table
current - describe the current disk
format - format and analyze the disk
fdisk - run the fdisk program
repair - repair a defective sector
show - translate a disk address
label - write label to the disk
analyze - surface analysis
defect - defect list management
backup - search for backup labels
verify - read and display labels
save - save new disk/partition definitions
volname - set 8-character volume name
!<cmd> - execute <cmd>, then return
quit
format> p

PARTITION MENU:
0 - change `0' partition
1 - change `1' partition
2 - change `2' partition
3 - change `3' partition
4 - change `4' partition
5 - change `5' partition
6 - change `6' partition
7 - change `7' partition
select - select a predefined table
modify - modify a predefined partition table
name - name the current table
print - display the current table
label - write partition map and label to the disk
!<cmd> - execute <cmd>, then return
quit
partition> p
Current partition table (original):
Total disk cylinders available: 2085 + 2 (reserved cylinders)

partition> 0
Part Tag Flag Cylinders Size Blocks
0 unassigned wm 0 0 (0/0/0) 0

Enter partition id tag[unassigned]: root

Enter partition permission flags[wm]:
Enter new starting cyl[3]:

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Enter partition size[0b, 0c, 3e, 0.00mb, 0.00gb]: 2082c

partition> p
Current partition table (unnamed):
Total disk cylinders available: 2085 + 2 (reserved cylinders)

Part Tag Flag Cylinders Size Blocks

0 root wm 3 - 2084 15.95GB (2082/0/0) 33447330
1 unassigned wm 0 0 (0/0/0) 0
2 backup wu 0 - 2084 15.97GB (2085/0/0) 33495525
3 unassigned wm 0 0 (0/0/0) 0
4 unassigned wm 0 0 (0/0/0) 0
5 unassigned wm 0 0 (0/0/0) 0
6 unassigned wm 0 0 (0/0/0) 0
7 unassigned wm 0 0 (0/0/0) 0
8 boot wu 0 - 0 7.84MB (1/0/0) 16065
9 alternates wm 1 - 2 15.69MB (2/0/0) 32130

partition> label
Ready to label disk, continue? y

partition> q

2. Create a new ZFS storage pool called rpool:

bash-3.00# zpool create rpool c0d1s0

3. Create a new ZFS boot environment.

bash-3.00# lucreate -n 10u6be-zfs -p rpool

Checking GRUB menu...
Analyzing system configuration.
No name for current boot environment.
INFORMATION: The current boot environment is not named -
assigning name <c0d0s0>.
Current boot environment is named <c0d0s0>.
Creating initial configuration for primary boot environment <c0d0s0>.
The device </dev/dsk/c0d0s0> is not a root device for any
boot environment; cannot get BE ID.
PBE configuration successful: PBE name <c0d0s0> PBE Boot
Device </dev/dsk/c0d0s0>.

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Comparing source boot environment <c0d0s0> file systems with

the file system(s) you specified for the new boot environment. Determining which
file systems should be in the new boot environment.
Updating boot environment description database on all BEs.
Updating system configuration files.
The device </dev/dsk/c0d1s0> is not a root device for any boot environment;
cannot get BE ID.
Creating configuration for boot environment <10u6be-zfs>.
Source boot environment is <c0d0s0>.
Creating boot environment <10u6be-zfs>.
Creating file systems on boot environment <10u6be-zfs>.
Creating <zfs> file system for </> in zone <global> on <rpool/ROOT/10u6be-zfs>.
Populating file systems on boot environment <10u6be-zfs>.
Checking selection integrity.
Integrity check OK.
Populating contents of mount point </>.
Copying.
Creating shared file system mount points.
Creating compare databases for boot environment <10u6be-zfs>.
Creating compare database for file system </rpool/ROOT>.
Creating compare database for file system </>.
Updating compare databases on boot environment <10u6be-zfs>.
Making boot environment <10u6be-zfs> bootable.
Updating bootenv.rc on ABE <10u6be-zfs>.
File </boot/grub/menu.lst> propagation successful
Copied GRUB menu from PBE to ABE
No entry for BE <10u6be-zfs> in GRUB menu
Population of boot environment <10u6be-zfs> successful.
Creation of boot environment <10u6be-zfs> successful.
bash-3.00# lustatus
Boot Environment Is Active Active Can Copy
Name Complete Now On Reboot Delete Status
------------------- -------- ------ --------- ------ -------
c0d0s0 yes yes yes no -
10u6be-zfs yes no no yes -
bash-3.00# zfs list
NAME USED AVAIL REFER MOUNTPOINT
rpool 4.93G 10.7G 24K /rpool
rpool/ROOT 3.92G 10.7G 18K /rpool/ROOT
rpool/ROOT/10u6be-zfs 3.92G 10.7G 3.92G /
rpool/dump 518M 11.2G 16K -
rpool/swap 518M 11.2G 16K -

4. Activate the new ZFS boot environment:

bash-3.00# luactivate -n 10u6be-zfs

Generating boot-sign, partition and slice information for PBE <c0d0s0>
A Live Upgrade Sync operation will be performed on startup of boot
environment <10u6be-zfs>.

Generating boot-sign for ABE <10u6be-zfs>

NOTE: File </etc/bootsign> not found in top level dataset for BE <10u6be-zfs>
Generating partition and slice information for ABE <10u6be-zfs>
Boot menu exists.
Generating multiboot menu entries for PBE.
Generating multiboot menu entries for ABE.
Disabling splashimage
Re-enabling splashimage

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

No more bootadm entries. Deletion of bootadm entries is complete.

GRUB menu default setting is unaffected
Done eliding bootadm entries.

************************************************************

The target boot environment has been activated. It will be used when you reboot.
NOTE: You MUST NOT USE the reboot, halt, or uadmin commands. You MUST USE either the
init or the shutdown command when you reboot. If you do not use either init or shutdown,
the system will not boot using the target BE.

************************************************************

In case of a failure while booting to the target BE, the

following process needs to be followed to fallback to the currently working
boot environment:

1. Boot from the Solaris failsafe or boot in Single User mode from Solaris
Install CD or Network.

2. Mount the Parent boot environment root slice to some

directory (like /mnt). You can use the following command to mount:

mount -Fufs /dev/dsk/c0d0s0 /mnt

3. Run <luactivate> utility without any arguments from the Parent boot
environment root slice, as shown below:

/mnt/sbin/luactivate

4. luactivate, activates the previous working boot environment and

indicates the result.

5. Exit Single User mode and reboot the machine.

************************************************************

Modifying boot archive service

Propagating findroot GRUB for menu conversion.
File </etc/lu/installgrub.findroot> propagation successful
File </etc/lu/stage1.findroot> propagation successful
File </etc/lu/stage2.findroot> propagation successful
File </etc/lu/GRUB_capability> propagation successful
Deleting stale GRUB loader from all BEs.
File </etc/lu/installgrub.latest> deletion successful
File </etc/lu/stage1.latest> deletion successful
File </etc/lu/stage2.latest> deletion successful
Activation of boot environment <10u6be-zfs> successful.
bash-3.00# lustatus
Boot Environment Is Active Active Can Copy
Name Complete Now On Reboot Delete Status
------------------- -------- ------ --------- ------ -------
c0d0s0 yes yes no no -
10u6be-zfs yes no yes no -
bash-3.00# init 6
updating /platform/i86pc/boot_archive
propagating updated GRUB menu
Saving existing file </boot/grub/menu.lst> in top level dataset
for BE <10u6be-zfs> as <mount-point>//boot/grub/menu.lst.prev.
File </boot/grub/menu.lst> propagation successful
File </etc/lu/GRUB_backup_menu> propagation successful
File </etc/lu/menu.cksum> propagation successful
File </sbin/bootadm> propagation successful

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Notice the difference in the lustatus output. After the luactivate operation,
the Active On Reboot ﬂag changed for boot environment 10u6be-zfs.

3.6.2 Patching a ZFS Boot Environment with Solaris Live Upgrade

Patching a boot environment by using Solaris Live Upgrade minimizes downtime
and is a simpler and safer way to patch a system. Any patching has risks of not
working or making a system unbootable. You can also use Solaris Live Upgrade to
add new packages or upgrade a system to a new release level.
After a patch is applied, the new boot environment can be booted. If any prob-
lems occur, the system can be rebooted to the original boot environment.
To patch a ZFS boot environment with Solaris Live Upgrade, follow these steps:

1. Create a new boot environment to be patched:

bash-3.00# lucreate -n 10u6be-1

Checking GRUB menu...
System has findroot enabled GRUB
Analyzing system configuration.
Comparing source boot environment <10u6be-zfs> file
systems with the file
system(s) you specified for the new boot environment.
Determining which
file systems should be in the new boot environment.
Updating boot environment description database on all BEs.
Updating system configuration files.
Creating configuration for boot environment <10u6be-1>.
Source boot environment is <10u6be-zfs>.
Creating boot environment <10u6be-1>.
Cloning file systems from boot environment <10u6be-zfs> to
create boot
environment <10u6be-1>.
Creating snapshot for <rpool/ROOT/10u6be-zfs> on
<rpool/ROOT/10u6be-zfs@10u6be-1>.
Creating clone for <rpool/ROOT/10u6be-zfs@10u6be-1> on
<rpool/ROOT/10u6be-1>.
Setting canmount=noauto for </> in zone <global> on
<rpool/ROOT/10u6be-1>.
Saving existing file </boot/grub/menu.lst> in top level
dataset for BE
<10u6be-1> as <mount-point>//boot/grub/menu.lst.prev.
File </boot/grub/menu.lst> propagation successful
Copied GRUB menu from PBE to ABE
No entry for BE <10u6be-1> in GRUB menu
Population of boot environment <10u6be-1> successful.
Creation of boot environment <10u6be-1> successful.

2. Unpack the downloaded patch cluster to the /var/tmp directory:

bash-3.00# unzip 10_x86_Recommended.zip

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

3. Upgrade the new ZFS boot environment:

bash-3.00# luupgrade -t -n 10u6be-1 -s /var/tmp/10_x86_Recommended

System has findroot enabled GRUB

No entry for BE <10u6be-1> in GRUB menu
Validating the contents of the media
</var/tmp/10_x86_Recommended>.
The media contains 148 software patches that can be added.
All 148 patches will be added because you did not specify
any specific patches to add.
Mounting the BE <10u6be-1>.
Adding patches to the BE <10u6be-1>.
Validating patches...

Loading patches installed on the system...

Done!

Loading patches requested to install.

...Deleted Output…

Unmounting the BE <10u6be-1>.

The patch add to the BE <10u6be-1> completed.

4. Activate the new ZFS boot environment:

bash-3.00# luactivate -n 10u6be-1

System has findroot enabled GRUB
Generating boot-sign, partition and slice information for
PBE <10u6be-zfs>
A Live Upgrade Sync operation will be performed on startup
of boot
environment <10u6be-1>.

Generating boot-sign for ABE <10u6be-1>

Saving existing file </etc/bootsign> in top level dataset for BE
<10u6be-1> as <mount-point>//etc/bootsign.prev.
Generating partition and slice information for ABE <10u6be-1>
Copied boot menu from top level dataset.
Generating multiboot menu entries for PBE.
Generating multiboot menu entries for ABE.
Disabling splashimage
Re-enabling splashimage
No more bootadm entries. Deletion of bootadm entries is complete.
GRUB menu default setting is unaffected
Done eliding bootadm entries.

************************************************************

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

In case of a failure while booting to the target BE, the following process needs
to be followed to fallback to the currently working boot environment:

1. Boot from Solaris failsafe or boot in single user mode from the Solaris Install
CD or Network.

2. Mount the Parent boot environment root slice to some directory (like /mnt).
You can use the following command to mount:

mount -Fzfs /dev/dsk/c0d1s0 /mnt

3. Run <luactivate> utility without any arguments from the Parent boot
environment root slice, as shown below:

/mnt/sbin/luactivate

4. luactivate, activates the previous working boot environment and

indicates the result.

5. Exit Single User mode and reboot the machine.

************************************************************

Modifying boot archive service

5. Check the status of the Active On Reboot ﬂag for the new ZFS boot
environment:

bash-3.00# lustatus
Boot Environment Is Active Active Can Copy
Name Complete Now On Reboot Delete Status
------------------- -------- ------ --------- ------ -------
c0d0s0 yes no no yes -
10u6be-zfs yes yes no no -
10u6be-1 yes no yes no -

6. Shut down the system:

bash-3.00# init 0
updating /platform/i86pc/boot_archive
propagating updated GRUB menu
Saving existing file </boot/grub/menu.lst> in top level
dataset for BE
<10u6be-1> as <mount-point>//boot/grub/menu.lst.prev.
File </boot/grub/menu.lst> propagation successful
File </etc/lu/GRUB_backup_menu> propagation successful
File </etc/lu/menu.cksum> propagation successful
File </sbin/bootadm> propagation successful

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

3.7 Managing ZFS Boot Environments (beadm)

Managing ZFS boot environments with the beadm command simpliﬁes upgrades
and rollbacks. The basic features of the beadm command enable you to create a
boot environment, activate a boot environment, and destroy a boot environment.
You can use the beadm and pkg commands to update an OpenSolaris release to
a current build. Use caution when upgrading to a new development build. These
builds are not as well tested as the release candidates and are unsupported. Using
the beadm command to manage and upgrade your boot environments can reduce

Licensed by
risk by enabling you to revert to the prior boot environment.

3.8 Upgrading a ZFS Boot Environment (beadm)

Onur Bingul
Upgrade a ZFS boot environment using beadm by following these steps:

1. Create a new boot environment named opensolaris1:

3690014
$ pfexec beadm create opensolaris1
2. Mount the new boot environment. For example, mount the new boot environ-
ment at /mnt:

$ pfexec beadm mount opensolaris1 /mnt

$ pfexec beadm list
BE Active Mountpoint Space Policy Created
-- ------ ---------- ----- ------ -------
opensolaris NR / 2.38G static 2009-01-29 23:32
opensolaris1 - /mnt 64.0K static 2009-01-30 01:43
Set pkg repository to OpenSolaris development

3. Set the pkg authority to the OpenSolaris development repository:

$ pfexec pkg set-authority -P -O http://pkg.opensolaris.org/dev opensolaris.org

$ pfexec pkg authority
AUTHORITY URL
opensolaris.org (preferred) http://pkg.opensolaris.org/dev/

4. Update the image of the new boot environment:

$ pfexec pkg -R /mnt image-update -v

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

5. Activate the new boot environment.

After activating the opensolaris1 boot environment, the Active On Reboot
ﬂag R has moved (line 12). The current active boot environment ﬂag is repre-
sented by N.

1 $ beadm list
2 BE Active Mountpoint Space Policy Created
3 -- ------ ---------- ----- ------ -------
4 opensolaris NR / 2.46G static 2009-01-29 23:32
5 opensolaris1 - /mnt 1.41G static 2009-01-30 01:43
6 watanabe@opensolaris:~$
7 $ pfexec beadm activate opensolaris1
8 $ beadm list
9 BE Active Mountpoint Space Policy Created
10 -- ------ ---------- ----- ------ -------
11 opensolaris N / 68.50M static 2009-01-29 23:32
12 opensolaris1 R /mnt 3.80G static 2009-01-30 01:43

6. Reboot the system:

$ pfexec init 6

7. Log in and verify the new image.

In the following example, the new boot environment is at build 105:

$ cat /etc/release
OpenSolaris 2009.04 snv_105 X86
Copyright 2009 Sun Microsystems, Inc. All Rights Reserved.
Use is subject to license terms.
Assembled 22 December 2008
$ beadm list
BE Active Mountpoint Space Policy Created
-- ------ ---------- ----- ------ -------
opensolaris - - 122.34M static 2009-01-29 23:32
opensolaris1 NR / 3.95G static 2009-01-30 01:43

3.9 Upgrading a ZFS Boot Environment (pkg)

The OpenSolaris release has made upgrades easier to do. The pkg image-update
command creates a boot environment and upgrades and activates it without the –R
ﬂag. Follow these steps to upgrade a ZFS boot using pkg:

1. Set the pkg authority to the OpenSolaris development repository:

$ pfexec pkg set-authority -P -O http://pkg.opensolaris.org/dev
opensolaris.org

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

2. Update the image of the new boot environment.

The pkg command creates a new boot environment by using the current
name and appending a number to it:

$ pfexec pkg image-update –v

… Deleted output …
$ beadm list
BE Active Mountpoint Space Policy Created
-- ------ ---------- ----- ------ -------
opensolaris N / 17.37M static 2009-01-29 23:32
opensolaris-1 R - 4.04G static 2009-01-30 06:24
opensolaris1 - - 1.50G static 2009-01-30 01:43

Figure 3.6 shows the verbose output of the image update.

Figure 3.6 Verbose output of the image update

3. Reboot to the new boot environment:

$ pfexec init 6

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

3.10 References
You can ﬁnd more information at the following locations:

Solaris ZFS Administration Guide:

http://docs.sun.com/app/docs/doc/819-5461
ZFS Best Practices Guide:
www.solarisinternals.com/wiki/index.php/ZFS_Best_Practices_
Guide
ZFS Troubleshooting Guide:
www.solarisinternals.com/wiki/index.php/ZFS_Troubleshooting_
Guide

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

This chapter describes how to use ZFS ﬁle systems for users’ home directories. Man-
aging home directories with ZFS is very easy:

Quotas and reservations are easier to manage than in other file systems.
Individual users can recover their own files from snapshots, leaving the
administrator to do other work.
Compression on file systems enables better disk management.
File system properties can be assigned depending on users’ needs.

4.1 Managing Quotas and Reservations on ZFS File Systems

Quotas and reservations are storage management tools that control the maximum
and minimum that a file system can consume from a storage pool. Using a file sys-
tem per user, you can control the storage resources each user can consume. This is
a different model from what traditional UNIX operating systems have used in the
past, where the user ID was used to aggregate the consumption of storage blocks.
ZFS is much simpler to administer, because you can create file systems to aggre-
gate the consumption by user.
A best practice is to set up each user or project with its own ZFS file system,
which creates a nice, granular storage management practice. Depending on a
user’s needs, you can set up different ZFS properties on specific home directory file
systems, if required.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Another good management practice for a server is to separate the system root
pool data from any data pools or ﬁle systems. Even if a mistake is made with the
root pool, the data can be recovered easily.

4.1.1 Setting the quota and refquota Properties

quota and refquota are two ZFS properties that control the maximum consump-
tion of storage blocks per file system. quota is the absolute maximum that a file
system can consume, including descendents, snapshots, and clones. refquota lim-
its only the file system, not including descendents, snapshots, and clones.

4.1.1.1 Using the quota Setting

The most basic case is setting a quota on a user’s home directory as a ZFS ﬁle sys-
tem. In this example, a simple quota is set on user layne’s home directory, and he
attempts to write to the ﬁle system:

1 $ zfs list homepool/export/home/layne

2 NAME USED AVAIL REFER MOUNTPOINT
3 homepool/export/home/layne 8.18M 572M 8.18M /export/home/layne
4 $ pfexec zfs set quota=20m homepool/export/home/layne
5 $ zfs list homepool/export/home/layne
6 NAME USED AVAIL REFER MOUNTPOINT
7 homepool/export/home/layne 8.18M 11.8M 8.18M /export/home/layne

On line 3, the availability is 572MB. After setting the quota on line 4 to 20MB,
the available storage to user layne drops to 11.8MB.
Next, user layne adds ﬁles to the home directory until the quota limit is hit:

1 layne@opensolaris:~$ mkfile 2m testfile1;mkfile 2m testfile2

2 layne@opensolaris:~$ mkfile 2m testfile3;mkfile 2m testfile4; mkfile 2m testfile5
3 layne@opensolaris:~$ mkfile 2m testfile6
4 testfile6: initialized 1835008 of 2097152 bytes: Disc quota exceeded
5 layne@opensolaris:~$ ls –lF test*
6 -rw------- 1 layne other 2097152 2009-02-12 07:47 testfile1
7 -rw------- 1 layne other 2097152 2009-02-12 07:47 testfile2
8 -rw------- 1 layne other 2097152 2009-02-12 07:47 testfile3
9 -rw------- 1 layne other 2097152 2009-02-12 07:48 testfile4
10 -rw------- 1 layne other 2097152 2009-02-12 07:48 testfile5
11 -rw------- 1 layne other 2097152 2009-02-12 07:48 testfile6
12 layne@opensolaris:~$ du -sh test*
13 2.1M testfile1
14 2.1M testfile2
15 2.1M testfile3
16 2.1M testfile4
17 2.1M testfile5
18 1.9M testfile6
19 layne@opensolaris:~$ touch testfile5
20 touch: setting times of `testfile5': Disc quota exceeded
21 layne@opensolaris:~$ rm testfile6
22 layne@opensolaris:~$

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

The write fails on line 4 with only 1,835,008 bytes written to disk. The ls out-
put shows that testfile6 is the same size as the rest of the 2MB files written.
The size difference is because of the way that ls calculates the size of the file. The
ls command determines file size by using the difference between the first and last
blocks of the file. To get a more accurate size, the du command is used.
On line 20, the touch fails because of the copy-on-write (COW) feature of ZFS.
Because ZFS would need to copy the file first and no available space exists, the
write fails. Deleting the file testfile6 is no problem, because ZFS does not have
to keep a copy.
ZFS does not write in place like most other file systems. Instead, it will write a
new copy of the data and move the pointers to the data once the operation is com-
plete via COW.
Next, the following example re-creates testfile6 and takes a snapshot of the
file system that is full:

1 layne@opensolaris:~$ mkfile 2m testfile6

2 testfile6: initialized 1835008 of 2097152 bytes: Disc quota exceeded
3 <zfs snapshot taken here>
4 layne@opensolaris:~$ touch testfile6
5 touch: setting times of `testfile6': Disc quota exceeded
6 layne@opensolaris:~$ rm testfile6
7 rm: cannot remove `testfile6': Disc quota exceeded

Again, the touch ﬁle on line 5 fails. On line 7, the ﬁle cannot be removed
because the snapshot needs more room to rewrite the blocks to point to the snap-
shot, but the quota is a hard limit.
You can correct this problem in four ways:

Destroy the snapshot.

Increase the quota.
Remove the quota.
Use refquota to keep the usage exclusive to the ﬁle system and to 20MB.

4.1.1.2 Using the refquota Setting

Using the refquota setting limits the storage used in just the normal file system.
The setting has no effect on any descendents, snapshots, or clones. In this case,
any snapshots or clones are not counted against the quota limits of the user’s home
directory. The following example illustrates how to set refquota to 20MB and
then attempts to write files to the refquota limit.
First, clear out the previous quota and remove the test files for user layne. On
line 1, you remove the snapshot called samplesnap, and next you clear the quota
set in the previous example. On line 3, you retrieve the quota and refquota
values to make sure they are reset to the defaults.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1 $ pfexec zfs destroy homepool/export/home/layne@samplesnap

2 $ pfexec zfs quota=none homepool/export/home/layne
3 $ zfs get quota,refquota homepool/export/home/layne
4 NAME PROPERTY VALUE SOURCE
5 homepool/export/home/layne quota none default
6 homepool/export/home/layne refquota none default

Remove old test ﬁles:

layne@opensolaris:~$ rm test*
Next, set the refquota setting to 20MB for user layne, and check the settings
by issuing the zfs get command on line 2:

1 $ pfexec zfs refquota=20m homepool/export/home/layne

2 $ zfs get quota,refquota homepool/export/home/layne
3 NAME PROPERTY VALUE SOURCE
4 homepool/export/home/layne quota none default
5 homepool/export/home/layne refquota 20M local

Create a couple of 10MB files until the refquota limit is reached and an error mes-
sage is generated by the system. The output of the ls command leads you to believe
that both files are the same size, but the output of the du command tells a different
story. The du command counts blocks used and is accurate concerning the size of the file.

layne@opensolaris:~$ mkfile 10m testfile1

layne@opensolaris:~$ mkfile 10m testfile2
testfile2: initialized 1835008 of 10485760 bytes: Disc quota exceeded
layne@opensolaris:~$ ls -lF test*
-rw------- 1 layne other 10485760 2009-02-12 15:57 testfile1
-rw------- 1 layne other 10485760 2009-02-12 15:57 testfile2
layne@opensolaris:~$ du -h test*
10M testfile1
1.9M testfile2

Next, take a snapshot of the home directory:

$ pfexec zfs snapshot homepool/export/home/layne@samplesnap
Remove a test ﬁle, and check the storage usage:

1 layne@opensolaris:~$ zfs list homepool/export/home/layne

2 NAME USED AVAIL REFER MOUNTPOINT
3 homepool/export/home/layne 20.1M 0 20.1M /export/home/layne
4 layne@opensolaris:~$ rm testfile2
5 layne@opensolaris:~$ zfs list homepool/export/home/layne
6 NAME USED AVAIL REFER MOUNTPOINT
7 homepool/export/home/layne 20.1M 1.82M 18.2M /export/home/layne

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

On line 3, 0 bytes are available for use in layne’s home directory. After remov-
ing testfile2, 1.82MB is available. Notice also that the used storage has not
changed because the file testfile2 is still in the snapshot.
Using a refquota property, the file system continues to grow until the refquota
limit is reached. A file system with a refquota set means that space that is con-
sumed by descendents can grow until a parent file system quota is reached or the
pool’s available capacity is reached.

4.1.1.3 Limiting the Storage Size of Descendents

You can limit the storage size of ﬁle system descendent data by using the quota
setting. If you wanted to limit the home directories of all users to 250MB, set the
quota in the parent directory, for example, homepool/export/home, for all user
home directories:

1 $ zfs list
2 NAME USED AVAIL REFER MOUNTPOINT
3 homepool 108M 560M 26.9K /homepool
4 homepool/export 107M 560M 28.4K /export
5 homepool/export/home 107M 560M 50.9K /export/home
6 homepool/export/home/ken 8.17M 560M 8.17M /export/home/ken
7 homepool/export/home/layne 20.1M 1.82M 18.2M /export/home/layne
8 homepool/export/home/watanabe 78.8M 560M 71.2M /export/home/watanabe
...Output delete...
9 $ pfexec zfs quota=250m homepool/export/home
10 $ zfs get quota homepool/export/home
11 NAME PROPERTY VALUE SOURCE
12 homepool/export/home quota 250M local
13 $ zfs list
14 NAME USED AVAIL REFER MOUNTPOINT
15 homepool 108M 560M 26.9K /homepool
16 homepool/export 107M 560M 28.4K /export
17 homepool/export/home 107M 143M 50.9K /export/home
18 homepool/export/home/ken 8.17M 143M 8.17M /export/home/ken
19 homepool/export/home/layne 20.1M 1.82M 18.2M /export/home/layne
20 homepool/export/home/watanabe 78.8M 143M 71.2M /export/home/watanabe
...Output deleted...

After setting a quota of 250MB on homepool/export/home, line 9, the avail-

able storage dropped to 143MB from 560MB with 107MB used. Compare line 5
with line 17 of the output to see the effect of setting the quota to 250MB. User
layne’s quota has not been affected since his quota is set at 20MB. Users ken and
watanabe now have an upper limit of 143MB for available space.

4.1.2 The reservation and refreservation Settings

reservation and refreservation are two ZFS ﬁle system properties that con-
trol the minimum allocated storage blocks. The reservation property is the abso-
lute minimum that a ﬁle system can consume, including descendents, snapshots,
and clones. The refreservation property limits only the space consumed by the

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

ﬁle system, not including descendents, snapshots, and clones. You can use these
properties to guarantee a minimum amount of storage to a ﬁle system.

4.1.2.1 Using the reservation Setting

The most basic case is setting a reservation on a ﬁle system. This example sets a
simple reservation on a ﬁle system:

1 $ zfs list
2 NAME USED AVAIL REFER MOUNTPOINT
3 homepool 108M 560M 26.9K /homepool
4 homepool/export 107M 560M 28.4K /export
5 homepool/export/home 107M 143M 50.9K /export/home
6 homepool/export/home/ken 8.17M 143M 8.17M /export/home/ken
7 homepool/export/home/layne 20.1M 1.82M 18.2M /export/home/layne
8 homepool/export/home/watanabe 78.8M 143M 71.2M /export/home/watanabe
...Output deleted...
9 $ pfexec zfs reservation=100m homepool/export/home/ken
10 $ zfs get reservation,refreservation homepool/export/home/ken
11 NAME PROPERTY VALUE SOURCE
12 homepool/export/home/ken reservation 100M local
13 homepool/export/home/ken refreservation none default
14 watanabe@opensolaris:~$
15 watanabe@opensolaris:~$ zfs list
16 NAME USED AVAIL REFER MOUNTPOINT
17 homepool 200M 468M 26.9K /homepool
18 homepool/export 199M 468M 28.4K /export
19 homepool/export/home 199M 51.1M 50.9K /export/home
20 homepool/export/home/ken 8.17M 143M 8.17M /export/home/ken
21 homepool/export/home/layne 20.1M 1.82M 18.2M /export/home/layne
22 homepool/export/home/watanabe 78.8M 51.1M 71.2M /export/home/watanabe
...Output deleted...

After setting the reservation for homepool/export/home/ken to 100MB, the

100MB was charged to the parent directory (homepool/export/home/ken) on
line 26. On line 27, no change is made to the available storage for homepool/
export/home/ken. On line 28, homepool/export/home/layne, available
space is unchanged because of the quota set in the previous example. On line 29,
homepool/export/home/watanabe has 51.1MB available. This space is calcu-
lated with the original available (line 8) for user watanabe minus (the reservation
minus the used storage by ken):

143MB - (100MB - 8.17MB) = 51.17MB

4.1.2.2 Using the refreservation Setting

Using the refreservation setting limits the used storage in just the ﬁle system.
The setting has no effect on any descendents, snapshots, or clones. The following
example sets the refreservation property to 5MB.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

First, clear out the previous reservation setting, and set refreservation to
50MB:

1 $ pfexec zfs reservation=none homepool/export/home/ken

2 $ zfs get reservation,refreservation homepool/export/home/ken
3 NAME PROPERTY VALUE SOURCE
4 homepool/export/home/ken reservation none default
5 homepool/export/home/ken refreservation none default
6 $ zfs list
7 NAME USED AVAIL REFER MOUNTPOINT
8 homepool 108M 560M 26.9K /homepool
9 homepool/export 107M 560M 28.4K /export

Licensed by
10 homepool/export/home 107M 143M 50.9K /export/home
11 homepool/export/home/ken 8.17M 143M 8.17M /export/home/ken
12 homepool/export/home/layne 20.1M 1.82M 18.2M /export/home/layne
13 homepool/export/home/watanabe 78.8M 143M 71.2M /export/home/watanabe
...Output deleted...
14 $ pfexec zfs refreservation=50m homepool/export/home/ken

Onur Bingul
15 $ zfs get reservation,refreservation homepool/export/home/ken
16 NAME PROPERTY VALUE SOURCE
17 homepool/export/home/ken reservation none default
18 homepool/export/home/ken refreservation 50M local
19 $ zfs list
20 NAME USED AVAIL REFER MOUNTPOINT

3690014
21 homepool 150M 518M 26.9K /homepool
22 homepool/export 149M 518M 28.4K /export
23 homepool/export/home 149M 101M 50.9K /export/home
24 homepool/export/home/ken 50M 143M 8.17M /export/home/ken
25 homepool/export/home/layne 20.1M 1.82M 18.2M /export/home/layne
26 homepool/export/home/watanabe 78.8M 101M 71.2M /export/home/watanabe
...Output deleted... 844M 9.13G 16K -

After setting the refreservation property to 50MB, on line 31, the used space
charged to homepool/export/home/ken grows by 42MB to 50MB. The parent
directory’s used number grows by 42MB to 149MB from 107MB. The ﬁle system
homepool/export/home/layne again stays the same on line 32. On line 33, the
next ﬁle system, homepool/export/home/watanabe, has the available storage
reduced by 42MB to 101MB.

4.2 Enabling Compression on a ZFS File System

You can reduce the amount of space that is consumed by a ZFS file system by
using the compression feature. This property is disabled by default. When enabled,
the default compression algorithm is lzjb. The lzjb algorithm strikes a balance
between performance and compression. The best compression will consume more
CPU cycles to achieve this and will affect the performance of the file system. Using
a balanced compression algorithm like lzjb, file system performance might actu-
ally increase depending on the data. The other compression algorithm supported is
gzip. This is the same algorithm as that used by the gzip command. You can

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

specify the level of compression using the gzip algorithm from 1 (fastest) to 9
(best compression). By default, gzip-6 is used when gzip is set.
Enabling compression helps manage available space for a file system. When
compression is enabled on a file system, the space usage numbers will differ
depending on the command used to calculate size.
The following example illustrates how to enable compression using the lzjb
and gzip-9 algorithms. The files in Table 4.1 were copied after enabling the com-
pression property on the file system.

$ pfexec zfs compression=on homepool/export/home/judy

$ pfexec zfs compression=gzip-9 homepool/export/home/judy

Table 4.1 File Sizes and Disk Usage Comparisons

Filename Orignal lzjb gzip-9 Original lzjb gzip-9

ls –lh ls –lh ls –lh du -h du -h du -h
1.Millsap2000.01.03- 42KB 42KB 42KB 43KB 37KB 28KB
RAID5.pdf
Anna_Karenina_T.pdf 4.4MB 4.4MB 4.4MB 4.4MB 4.0MB 2.7MB
Crime_and_ 2.5MB 2.5MB 2.5MB 2.6MB 2.3MB 1.6MB
Punishment_T.pdf
Emma_T.pdf 1.9MB 1.9MB 1.9MB 1.9MB 1.8MB 1.2MB
Moby_Dick_NT.pdf 2.3MB 2.3MB 2.3MB 2.3MB 2.1MB 1.4MB
Shakespeare William 259KB 259KB 259KB 387KB 265KB 200KB
Romeo and Juliet.pdf
War_and_Peace_NT.pdf 6.3MB 6.3MB 6.3MB 6.3MB 5.8MB 4.0MB
zfs_last.pdf 3.1MB 3.1MB 3.1MB 3.2MB 3.1MB 2.9MB
zfs_lc_preso.pdf 2.3M 2.3M 2.3M 2.3M 2.3M 1.8M

The listing output is the same in all cases, but the disk usage is less with com-
pression turned on. The best compression is gzip-9.
Another statistic you can use is the compressratio property status of the ZFS
file system. This will show the compression efficiency of the data on the file sys-
tem. Table 4.2 provides an idea of the kind of compression ratios available from
lzjb and gzip-9.
$ zfs get compressratio zfs_filesystem

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Table 4.2 Compression Ratio Output

Compression Compression Ratio

Original (no compression) 1.00
lzjb 1.08
gzip-9 1.47

You need to be careful when moving or copying data from a compressed ﬁle sys-
tem. The compression could cause backup or archive programs to underestimate
the size of the media and the time that is needed to execute the job. Using the
compressratio and disk usage will give an approximate size of the uncom-
pressed data as it moved to an archive. Take the output of du –s of the ﬁle system
or directory, and multiply by the compression ratio to get an approximation of the
size of the data that needs to be copied.

4.3 Working with ZFS Snapshots

Users can manage their own deleted, destroyed, and modiﬁed data by implement-
ing snapshots for home directories. A snapshot is a read-only copy of a ZFS ﬁle
system and is created using the zfs snapshot command.

4.3.1 Managing the Snapshot Directory

Once a snapshot is taken, a .zfs directory is created in the root of the file system.
This directory is hidden by default from most commands unless the .zfs directory
is specifically named. It is best to keep the snapshot directory hidden so that archive
programs, such tar, cpio, and gzip, do not copy the same data multiple times.
For example, listing files in a home directory yields normal files:

cindy@opensolaris:~$ ls -a
. .bash_history Books local.login .profile
.. .bashrc local.cshrc local.profile
cindy@opensolaris:~$ ls .zfs
snapshot

In the example, specifying the .zfs directory with the ls command shows the
snapshot directory.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

You can make the .zfs directory visible by changing the snapdir property
from hidden to visible:

$ zfs get snapdir homepool/export/home/cindy

NAME PROPERTY VALUE SOURCE
homepool/export/home/cindy snapdir hidden local
$ pfexec zfs snapdir=visible homepool/export/home/cindy
$ zfs get snapdir homepool/export/home/cindy
NAME PROPERTY VALUE SOURCE
homepool/export/home/cindy snapdir visible local
???
cindy@opensolaris:~$ ls -a
. .bash_history Books local.login .profile
.. .bashrc local.cshrc local.profile .zfs

When the snapdir property is set to visible, the ls command displays the
.zfs directory.
The following example is the output from a tar archive of a home directory with
the snap directory hidden:

cindy@opensolaris:~$ tar cvf /var/tmp/cindy.hidden.tar .

./
./local.cshrc
./.profile
./.bashrc
./Books/
./Books/Anna_Karenina_T.pdf
./Books/zfs_lc_preso.pdf
./Books/Shakespeare William Romeo and Juliet.pdf
./Books/zfs_last.pdf
./Books/War_and_Peace_NT.pdf
./Books/1.Millsap2000.01.03-RAID5.pdf
./Books/Crime_and_Punishment_T.pdf
./Books/Moby_Dick_NT.pdf
./Books/Emma_T.pdf
./.bash_history
./local.login
./local.profile

The following example is the same tar command with the snap directory visi-
ble. The archive now copies the .zfs directory, doubling the size of the archive.
This ﬁle system has only one snapshot. If the ﬁle system had additional snap-
shots, the archive size could grow larger.

cindy@opensolaris:~$ tar cvf /var/tmp/cindy.visible.tar .

./
./.zfs/
./.zfs/snapshot/
./.zfs/snapshot/sunday/
./.zfs/snapshot/sunday/local.cshrc

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

./.zfs/snapshot/sunday/.profile
./.zfs/snapshot/sunday/.bashrc
./.zfs/snapshot/sunday/Books/
./.zfs/snapshot/sunday/Books/Anna_Karenina_T.pdf
./.zfs/snapshot/sunday/Books/zfs_lc_preso.pdf
./.zfs/snapshot/sunday/Books/Shakespeare William Romeo and Juliet.pdf
./.zfs/snapshot/sunday/Books/zfs_last.pdf
./.zfs/snapshot/sunday/Books/War_and_Peace_NT.pdf
./.zfs/snapshot/sunday/Books/1.Millsap2000.01.03-RAID5.pdf
./.zfs/snapshot/sunday/Books/Crime_and_Punishment_T.pdf
./.zfs/snapshot/sunday/Books/Moby_Dick_NT.pdf
./.zfs/snapshot/sunday/Books/Emma_T.pdf
./.zfs/snapshot/sunday/.bash_history
./.zfs/snapshot/sunday/local.login
./.zfs/snapshot/sunday/local.profile
./local.cshrc
./.profile
./.bashrc
./Books/
./Books/Anna_Karenina_T.pdf
./Books/zfs_lc_preso.pdf
./Books/Shakespeare William Romeo and Juliet.pdf
./Books/zfs_last.pdf
./Books/War_and_Peace_NT.pdf
./Books/1.Millsap2000.01.03-RAID5.pdf
./Books/Crime_and_Punishment_T.pdf
./Books/Moby_Dick_NT.pdf
./Books/Emma_T.pdf
./.bash_history
./local.login
./local.profile
cindy@opensolaris:~$

cindy@opensolaris:~$ ls -lFh /var/tmp/cindy*

-rw-r--r-- 1 cindy other 23M 2009-02-15 18:01 /var/tmp/cindy.hidden.tar
-rw-r--r-- 1 cindy other 46M 2009-02-15 18:01 /var/tmp/cindy.visible.tar

4.3.2 Recovering Files from Snapshots

Users can recover their own files from ZFS snapshots, which frees you from sim-
ple file recovery tasks. When file system snapshots are available, each user has
their own snapshot directory. This means each of the users can securely recover
their own files from a snapshot.
The snapshot files are read-only and cannot be modified in place. To be able to
modify any file in the snapshot directory, the file needs to be copied to a regular
directory that is not in the .zfs snapshot directory. Chapter 7 covers the adminis-
tration and use of Time Slider a GUI-based tool that is designed to recover files
from a snapshot.
You can use the ZFS rollback subcommand to recover whole environments
from a snapshot.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

4.3.2.1 Delegating ZFS Commands

By default a regular user does not have the capability to roll back a file system
snapshot, so you, the administrator, can grant regular users the capability to roll
back their own file system snapshots. For more information about delegating ZFS
permissions to users, see the ZFS man page.
In the following example, user cindy is delegated the permissions to roll back
her file system snapshot. The subcommands that are delegated are mount,
snapshot, and rollback on the user’s home directory.

$ pfexec zfs allow cindy mount,rollback,snapshot homepool/export/home/cindy

$ zfs allow homepool/export/home/cindy
-------------------------------------------------------------
Local+Descendent permissions on (homepool/export/home/cindy)
user cindy mount,rollback,snapshot
-------------------------------------------------------------

4.3.2.2 Rolling Back a File System

In this example, you’ll roll back a ﬁle system to its last known good snapshot by
using the rollback subcommand that was just delegated to user cindy.
In this example, the entire home directory is removed accidentally:

cindy@opensolaris:~$ rm -rf *
cindy@opensolaris:~$ rm -rf .*
rm: cannot remove directory `.'
rm: cannot remove directory `..'
$ ls -Fa
./ ../

Now you, or the home directory owner, must ﬁnd the most recent snapshot and
select it for a rollback. You can do this with the zfs list command:

$ zfs list -t snapshot

NAME USED AVAIL REFER MOUNTPOINT
homepool/export/home/cindy@sunday 23.1M - 23.1M -
homepool/export/home/layne@samplesnap 1.91M - 20.1M -
homepool/export/home/watanabe@today 7.59M - 31.6M -
rpool@today 18K - 75.5K -
rpool/ROOT@today 0 - 18K -
rpool/ROOT/opensolaris@today 0 - 2.67G -
rpool/ROOT/opensolaris-1@install 72.3M - 2.21G -
rpool/ROOT/opensolaris-1@opensolaris1 12.1M - 2.24G -
rpool/ROOT/opensolaris-1@2009-01-30-13:24:22 100M - 2.68G -
rpool/ROOT/opensolaris-1@today 23.4M - 2.74G -
rpool/ROOT/opensolaris-1@today1 32.0M - 2.79G -
rpool/ROOT/opensolaris1@today 3.85M - 2.68G -
rpool/dump@today 16K - 844M -
rpool/swap@today 0 - 16K -
cindy@opensolaris:~$ zfs rollback homepool/export/home/cindy@sunday

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

To identify the snapshots by date and time, use zfs get creation to retrieve
the creation date and time of the snapshot:

$ zfs get creation homepool/export/home/cindy@sunday

NAME PROPERTY VALUE SOURCE
homepool/export/home/cindy@sunday creation Sun Feb 1 15:56 2009 -

Check the home directory for the restored ﬁles:

cindy@opensolaris:~$ ls -Fa
./ .bash_history Books/ local.login .profile
../ .bashrc local.cshrc local.profile

4.4 Sharing ZFS Home Directories

ZFS home directories are shared by using Network File System (NFS) services, as
is done with Unix File System (UFS). You can set up the automounter so that the
home directories appear the same on the server as they do on client systems. When
users log in to the server, they can ﬁnd the home directory using the same pass-
word entry used on the client. Using the /home directory to mount user directo-
ries, the automounter maps server:/export/home/user_name to /home/
user_name. In this example, only ﬁles are used.

1. Edit the /etc/passwd file on the server to change all the user home directo-
ries from /export/home/user_name to /home/user_name. For more infor-
mation on the passwd file, use man –s 4 passwd. Distribute the new
password file to the other client systems.

watanabe:x:101:10:scott watanabe:/home/watanabe:/bin/bash
layne:x:65535:1:Layne :/home/layne:/bin/bash
judy:x:65536:1:Judy:/home/judy:/bin/bash
cindy:x:65537:1:Cindy:/home/cindy:/bin/bash
ken:x:65538:1:Ken:/home/ken:/bin/bash

2. Edit /etc/auto_home, and add the following line:

opensolaris:/home/&
The server name in this example is opensolaris:

# Home directory map for automounter

#
* opensolaris:/export/home/&
+auto_home

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

The star (*) is a wildcard key, and the ampersand (&) takes the key
(user_name) and makes a replacement. Copy this ﬁle to the other clients.
3. Set the sharenfs property to each home directory ﬁle system:

# zfs sharenfs=on homepool/export/home/cindy

# zfs sharenfs=on homepool/export/home/judy
# zfs sharenfs=on homepool/export/home/layne
# zfs sharenfs=on homepool/export/home/ken
# zfs sharenfs=on homepool/export/home/watanabe

4. Reboot the server to start the automounter.

4.5 References

Here are some references for more information:

ZFS Administration Guide:
http://docs.sun.com/app/docs/doc/819-5461
ZFS Best Practices Wiki:
www.solarisinternals.com/wiki/index.php/ZFS_Best_
Practices_Guide

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

In this chapter, you will explore techniques for creating the best possible pool configu-
rations. The examples shown in this chapter are from the ZFS Configuration Guide at
www.solarisinternals.com/wiki/index.php/ZFS_Configuration_Guide. I
will walk you through the commands that create the configuration and explain why it
works.
There are many ways to configure ZFS pools, some better than others, but it is my
hope that from this chapter you will get a better idea for setting up a ZFS pool that
suits your needs.

5.1 X4500 RAID-Z2 Conﬁguration Example

This example uses a Sun x4500 that is fully populated with 250GB disks. The
x4500 is an x64 system with two dual-core AMD Opteron processors: Model 290 at
2.8GHz with 48 hot-pluggable SATA II disks. The single point of failure for the
storage is the motherboard. The disks’ pool is spread over all the controllers,
except for the system disks. This conﬁguration is not designed for speed but for
maximum disk availability with comfortable redundancy. The example creates a
single mirror for the boot/system disk and ZFS pool with seven RAID-Z2 vdevs and
four disk spares.
Six SATA II controllers are capable of handling eight disks per controller. Each
disk has its own line into the controller (see Figure 5.1). If a cable goes bad, only a
single disk is affected by the defect.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

c0
t4

Figure 5.1 SATA controller with eight disks

In Figure 5.2, the drawing is altered to line up the disks with the controller
inline. There are six controllers: c0, c1, c4, c5, c6, and c7. Each controller has eight
target disks, starting from t0 to t7. The conﬁguration has 48 disks total.
The boot disks are described as Solaris Volume Manager (SVM) mirrored. This
now can easily be ZFS boot disks today in Solaris 10. Disks c4t0d0 and c5t0d0 are
mirrored as the boot/system disks in Figure 5.3.
The ﬁrst zpool command will create the new pool called rpool and create four
RAIDZ2 vdevs:

# zpool create rpool \

raidz2 c0t1d0 c1t1d0 c4t1d0 c5t1d0 c6t1d0 c7t1d0 \
raidz2 c0t2d0 c1t2d0 c4t2d0 c5t2d0 c6t2d0 c7t2d0 \
raidz2 c0t3d0 c1t3d0 c4t3d0 c5t3d0 c6t3d0 c7t3d0 \
raidz2 c0t4d0 c1t4d0 c4t4d0 c5t4d0 c6t4d0 c7t4d0

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7

c5
t0 t1 Licensed by
t2 t3 t4 t5 t6 t7

c6
t0 t1
Onur Bingul
t2 t3 t4 t5 t6 t7

3690014
c7
t0 t1 t2 t3 t4 t5 t6 t7

Figure 5.2 X4500 disk controllers and eight target disks per controller

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

Figure 5.3 The boot disks c4t0d0 and c5t0d0 are circled as a mirror created by SVM

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

The command line has been parsed to see the vdev creation for the ZFS pool.
Notice the four raidz2 subcommands with six disks per vdev. Each raidz2 sub-
command uses disk targets in the same column (see Figure 5.4).

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

RAID-Z2 Vdev RAID-Z2 Vdev

Figure 5.4 Creation of a ZFS pool with four RAID-Z2 vdevs

The ﬁrst vdev uses all the disk targets (starting with t1). Each of the next vdevs
spans the next three target numbers: 2, 3, and 4.

1 # zpool status
2 pool: rpool
3 state: ONLINE
4 scrub: none requested
5 config:
6
7 NAME STATE READ WRITE CKSUM
8 rpool ONLINE 0 0 0
9 raidz2 ONLINE 0 0 0
10 c0t1d0 ONLINE 0 0 0

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

11 c1t1d0 ONLINE 0 0 0
12 c4t1d0 ONLINE 0 0 0
13 c5t1d0 ONLINE 0 0 0
14 c6t1d0 ONLINE 0 0 0
15 c7t1d0 ONLINE 0 0 0
16 raidz2 ONLINE 0 0 0
17 c0t2d0 ONLINE 0 0 0
18 c1t2d0 ONLINE 0 0 0
19 c4t2d0 ONLINE 0 0 0
20 c5t2d0 ONLINE 0 0 0
21 c6t2d0 ONLINE 0 0 0
22 c7t2d0 ONLINE 0 0 0
23 raidz2 ONLINE 0 0 0
24 c0t3d0 ONLINE 0 0 0
25 c1t3d0 ONLINE 0 0 0
26 c4t3d0 ONLINE 0 0 0
27 c5t3d0 ONLINE 0 0 0
28 c6t3d0 ONLINE 0 0 0
29 c7t3d0 ONLINE 0 0 0
30 raidz2 ONLINE 0 0 0
31 c0t4d0 ONLINE 0 0 0
32 c1t4d0 ONLINE 0 0 0
33 c4t4d0 ONLINE 0 0 0
34 c5t4d0 ONLINE 0 0 0
35 c6t4d0 ONLINE 0 0 0
36 c7t4d0 ONLINE 0 0 0
37
38 errors: No known data errors

In the output of zpool status, on lines 9, 16, 23, and 30, the RAID-Z2 vdevs
correspond to the RAID-Z2 vdevs in Figure 5.4. Each of the vdev disks has the
same target number and different controller numbers. In this ZFS pool, each of the
vdev disks can be damaged before any data is lost. Also, if a controller is damaged,
the pool can still function.
In the next four zpool commands, you add three more RAID-Z2 vdevs across all
six controllers and four disks to the spare pools:

1 # zpool add rpool raidz2 c0t5d0 c1t5d0 c4t5d0 c5t5d0 c6t5d0 c7t5d0
2 # zpool add rpool raidz2 c0t6d0 c1t6d0 c4t6d0 c5t6d0 c6t6d0 c7t6d0
3 # zpool add rpool raidz2 c0t7d0 c1t7d0 c4t7d0 c5t7d0 c6t7d0 c7t7d0
4 # zpool add rpool spare c0t0d0 c1t0d0 c6t0d0 c7t0d0

In the following ﬁgures, the ZFS pool conﬁguration is shown after each com-
mand line is executed.
In Figure 5.5, a new RAID-Z2 vdev is added to the ZFS pool rpool after execut-
ing line 1. All the disks in this vdev are the target 5 (t5) disks across all six SATA
controllers.
In Figure 5.6, a new RAID-Z2 vdev is added to ZFS pool rpool after executing line 2.
All the disks in this vdev are the target 6 (t6) disks across all six SATA controllers.
In Figure 5.7, a new RAID-Z2 vdev is added to ZFS pool rpool after executing line 3.
All the disks in this vdev are the target 6 (t6) disks across all six SATA controllers.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2 Vdev

RAID-Z2 Vdev RAID-Z2 Vdev

Figure 5.5 After executing line 1, a RAID-Z2 vdev using the target 5 (t5) disks is added
to the ZFS pool

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2 Vdev

RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2v Vdev

Figure 5.6 After executing line 2, a RAID-Z2 vdev using the target 6 (t6) disks is added
to the ZFS pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2 Vdev

RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2v Vdev

Figure 5.7 After executing line 3, a RAID-Z2 vdev using the target 7 (t7) disks is added
to the ZFS pool

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

Spare Pool
RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2 Vdev
RAID-Z2 Vdev RAID-Z2 Vdev RAID-Z2v Vdev

Figure 5.8 After executing line 4, four spares are added to the ZFS pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

After executing lines 1, 2, and 3, the ZFS pool is almost complete. Next, execut-
ing line 4 will add four spare disks to the pool rpool. Figure 5.8 shows the conﬁgu-
ration after the spares are added.
Running a status command on the ZFS pool, you can now see the completed
rpool. The four spares are listed starting at line 48.

1 # zpool status
2 pool: rpool
3 state: ONLINE
4 scrub: none requested
5 config:
6
7 NAME STATE READ WRITE CKSUM
8 rpool ONLINE 0 0 0
9 raidz2 ONLINE 0 0 0
10 c0t1d0 ONLINE 0 0 0
11 c1t1d0 ONLINE 0 0 0
12 c4t1d0 ONLINE 0 0 0
13 c5t1d0 ONLINE 0 0 0
14 c6t1d0 ONLINE 0 0 0
15 c7t1d0 ONLINE 0 0 0
16 raidz2 ONLINE 0 0 0
17 c0t2d0 ONLINE 0 0 0
18 c1t2d0 ONLINE 0 0 0
19 c4t2d0 ONLINE 0 0 0
20 c5t2d0 ONLINE 0 0 0
21 c6t2d0 ONLINE 0 0 0
22 c7t2d0 ONLINE 0 0 0
23 raidz2 ONLINE 0 0 0
24 c0t3d0 ONLINE 0 0 0
25 c1t3d0 ONLINE 0 0 0
26 c4t3d0 ONLINE 0 0 0
27 c5t3d0 ONLINE 0 0 0
28 c6t3d0 ONLINE 0 0 0
29 c7t3d0 ONLINE 0 0 0
30 raidz2 ONLINE 0 0 0
31 c0t4d0 ONLINE 0 0 0
32 c1t4d0 ONLINE 0 0 0
33 c4t4d0 ONLINE 0 0 0
34 c5t4d0 ONLINE 0 0 0
35 c6t4d0 ONLINE 0 0 0
36 c7t4d0 ONLINE 0 0 0
37 raidz2 ONLINE 0 0 0
38 c0t5d0 ONLINE 0 0 0
39 c1t5d0 ONLINE 0 0 0
40 c4t5d0 ONLINE 0 0 0
41 c5t5d0 ONLINE 0 0 0
42 c6t5d0 ONLINE 0 0 0
43 c7t5d0 ONLINE 0 0 0
44 raidz2 ONLINE 0 0 0
45 c0t6d0 ONLINE 0 0 0
46 c1t6d0 ONLINE 0 0 0
47 c4t6d0 ONLINE 0 0 0
48 c5t6d0 ONLINE 0 0 0
49 c6t6d0 ONLINE 0 0 0
50 c7t6d0 ONLINE 0 0 0
51 raidz2 ONLINE 0 0 0
52 c0t7d0 ONLINE 0 0 0
53 c1t7d0 ONLINE 0 0 0

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

54 c4t7d0 ONLINE 0 0 0
55 c5t7d0 ONLINE 0 0 0
56 c6t7d0 ONLINE 0 0 0
57 c7t7d0 ONLINE 0 0 0
58 spares
59 c0t0d0 AVAIL
60 c1t0d0 AVAIL
61 c6t0d0 AVAIL
62 c7t0d0 AVAIL
63
64 errors: No known data errors

Creating the RAID-Z2 vdevs across the target disks columns also has an advan-
tage of spreading the I/O across all the SATA controllers and not causing a single
controller to be a bottleneck.

5.2 X4500 Mirror Conﬁguration Example

Again, in this example, the single point of failure for the storage is the mother-
board. The disks’ pool is spread over multiple controllers. This conﬁguration is
designed for speed and maximum redundancy. Also, the data I/O will be spread
across all six controllers for maximum performance. The example creates a single
mirror for the boot/system disk and ZFS pool with 14 three-way mirrored vdevs
and 4 disk spares.
This example starts with mirrored disks created by Solaris Volume Manager in
Figure 5.9 as in the previous example.

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

Figure 5.9 The boot disks c4t0d0 and c5t0d0 are circled as a mirror created by SVM

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Next you create a ZFS pool named mpool with seven three-way mirrors using
the following command:

# zpool create mpool \

mirror c0t1d0 c1t1d0 c4t1d0 \
mirror c0t2d0 c1t2d0 c4t2d0 \
mirror c0t3d0 c1t3d0 c4t3d0 \
mirror c0t4d0 c1t4d0 c4t4d0 \
mirror c0t5d0 c1t5d0 c4t5d0 \
mirror c0t6d0 c1t6d0 c4t6d0 \
mirror c0t7d0 c1t7d0 c4t7d0

The command line was formatted to make each three-way mirrored vdev clear.
The zpool status command shows the newly created ZFS pool mpool with the
seven vdevs:

# zpool status
pool: mpool
state: ONLINE
scrub: none requested
config:

NAME STATE READ WRITE CKSUM

errors: No known data errors

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Notice the indentation in the output. ZFS pool mpool has seven mirror vdevs,
and each vdev has three disks. Figure 5.10 shows the vdevs in the diagram of the
X4500.
In Figure 5.10, it is easy to see each mirrored vdev using three different SATA
II controllers. A mirrored vdev could lose a maximum of two disks or a controller
without losing any data in the ZFS pool.

Mirror Vdev

c0
t0 t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

Figure 5.10 The ZFS pool with seven three-way mirrored vdevs

The next eight commands ﬁll out the rest of the conﬁguration of the X4500 with
another seven three-way mirrored vdevs and four disks in a spare pool:

# zpool add mpool mirror c5t1d0 c6t1d0 c7t1d0

# zpool add mpool mirror c5t2d0 c6t2d0 c7t2d0
# zpool add mpool mirror c5t3d0 c6t3d0 c7t3d0
# zpool add mpool mirror c5t4d0 c6t4d0 c7t4d0
# zpool add mpool mirror c5t5d0 c6t5d0 c7t5d0
# zpool add mpool mirror c5t6d0 c6t6d0 c7t6d0
# zpool add mpool mirror c5t7d0 c6t7d0 c7t7d0
# zpool add mpool spare c0t0d0 c1t0d0 c6t0d0 c7t0d0

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

The zpool status command shows the ZFS pool mpool now with seven addi-
tional mirrored vdevs and a spare pool with four disks:

# zpool status
pool: mpool
state: ONLINE
scrub: none requested
config:

NAME STATE READ WRITE CKSUM

mpool ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t1d0 ONLINE 0 0 0
c1t1d0 ONLINE 0 0 0
c4t1d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t2d0 ONLINE 0 0 0
c1t2d0 ONLINE 0 0 0
c4t2d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t3d0 ONLINE 0 0 0
c1t3d0 ONLINE 0 0 0
c4t3d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t4d0 ONLINE 0 0 0
c1t4d0 ONLINE 0 0 0
c4t4d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t5d0 ONLINE 0 0 0
c1t5d0 ONLINE 0 0 0
c4t5d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t6d0 ONLINE 0 0 0
c1t6d0 ONLINE 0 0 0
c4t6d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c0t7d0 ONLINE 0 0 0
c1t7d0 ONLINE 0 0 0
c4t7d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t1d0 ONLINE 0 0 0
c6t1d0 ONLINE 0 0 0
c7t1d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t2d0 ONLINE 0 0 0
c6t2d0 ONLINE 0 0 0
c7t2d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t3d0 ONLINE 0 0 0
c6t3d0 ONLINE 0 0 0
c7t3d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t4d0 ONLINE 0 0 0
c6t4d0 ONLINE 0 0 0
c7t4d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t5d0 ONLINE 0 0 0
c6t5d0 ONLINE 0 0 0
c7t5d0 ONLINE 0 0 0

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

mirror ONLINE 0 0 0
c5t6d0 ONLINE 0 0 0
c6t6d0 ONLINE 0 0 0
c7t6d0 ONLINE 0 0 0
mirror ONLINE 0 0 0
c5t7d0 ONLINE 0 0 0
c6t7d0 ONLINE 0 0 0
c7t7d0 ONLINE 0 0 0
spares
c0t0d0 AVAIL
c1t0d0 AVAIL
c6t0d0 AVAIL
c7t0d0 AVAIL

Licensed by
errors: No known data errors

Figure 5.11 shows the X4500 conﬁguration completed.

Onur Bingul Mirror Vdev

c0
t0
3690014
t1 t2 t3 t4 t5 t6 t7

c1
t0 t1 t2 t3 t4 t5 t6 t7

c4
t0 t1 t2 t3 t4 t5 t6 t7
SVM Mirror

c5
t0 t1 t2 t3 t4 t5 t6 t7

c6
t0 t1 t2 t3 t4 t5 t6 t7

c7
t0 t1 t2 t3 t4 t5 t6 t7

Spare Pool
Mirror Vdev

Figure 5.11 The ZFS pool with 14 three-way mirrored vdevs and a spare pool with
4 disks

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

5.3 X4500 Boot Mirror Alternative Example

Using a ZFS boot disk, you can create a normal two-way ZFS mirror and share the
spare pool with the data ZFS pool of the previous two examples. In this way, you
can increase the reliability of the whole system including the system disks.

5.4 ZFS and Array Storage

ZFS also works well with your current storage arrays. Any logical unit number
(LUN) presented to ZFS is used like any other disk. These are some consider-
ations when using storage arrays with ZFS:

ZFS is not a shared ﬁle system. The LUNs presented are exclusive to the
ZFS host.
ZFS is oblivious to disk failures if hardware RAID is used. The LUN pre-
sented will look normal to ZFS, and any disk replacement will be handled at
the array level. The array is responsible for rebuilding the LUN.
A striped LUN is just a bigger disk. A disk failure in a striped LUN will be a
failure of the whole LUN. The failed disk needs to be replaced and the LUN
rebuilt before ZFS resilvering can take place.
ZFS still needs its own redundancy. Even if the LUN is redundant (mirror or
RAID-5), ZFS still needs to be conﬁgured in a mirror or a RAID-Z for best
results.
ZFS is MPxIO safe. In Solaris 10, MPxIO is integrated in the operating sys-
tem. MPxIO provides a multipathing solution for storage devices accessible
through multiple physical paths. See www.sun.com/bigadmin/xperts/
sessions/23_fibre/#7.

Another obvious disadvantage to using a hardware redundant conﬁguration in

ZFS is the available disk space. At each level of redundancy, available storage is
decreased. When a hardware mirror is used with a ZFS mirror, the available space
is 25 percent of the raw capacity.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

OpenSolaris has a built-in native CIFS client and server. The client is installed by
default, and the server packages can be installed after interactive installation.
OpenSolaris can manage CIFS by the command line or a GUI via the GNOME ﬁle
manager. There is no anonymous user in the built-in server. If an anonymous user is
required, then SAMBA would be the better option to conﬁgure.
The CIFS client and server are not available in Solaris 10.

6.1 Installing the CIFS Server Packages

OpenSolaris uses Image Packaging System (IPS). This is the next generation of
package management for Solaris. You need two packages in order to install the
SMB server: SUNWsmbs and SUNWsmbskr. The latter is a kernel package and
will require a reboot before the server can be activated.

6.1.1 Installing the Server Packages with Package Manager

To install server packages with Package Manager, follow these steps:

1. To start Package Manager, select System > Administration > Package

Manager. In the Search box, type smb (see Figure 6.1).
2. Select SUNWsmbs and SUNWsmbkr.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

3. Click the Install/Update button. Package Manager will contact the repository
at OpenSolaris.org and retrieve the packages’ information (see Figure 6.2).
4. Click the Next button. Then wait for the packages to install.
5. Reboot the system.
6. Select System > Shutdown.
7. Click the Restart button to activate SMB in the kernel.

Figure 6.1 Package Manager with SMB packages ﬁltered

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 6.2 Package Manager with the server packages ready to be installed

6.1.2 Installing the Server Packages from the Command Line

Use the pkg command to retrieve and install packages from the OpenSolaris.org
repository:

1 watanabe@opensolaris:~$ pfexec pkg install SUNWsmbs SUNWsmbskr

2 DOWNLOAD PKGS FILES XFER (MB)
3 Completed 2/2 34/34 1.49/1.49
4
5 PHASE ACTIONS
6 Install Phase 80/80
7 PHASE ITEMS
8 Reading Existing Index 9/9
9 Indexing Packages 2/2

Now reboot the system to activate the SMB server in the kernel.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

6.2 Conﬁguring the SMB Server in Workgroup Mode

In this scenario, the clients are primarily laptops (Windows or Mac) that need to
mount the home directories with their laptops. There is no need for Windows Inter-
net Naming Service (WINS) or Active Directory (AD). To create a CIFS server in
AD mode, see the Solaris CIFS Administration Guide at http://docs.sun.com/
app/docs/doc/820-2429 for detailed task instruction.

1. Enable the CIFS service:

# svcadm enable –r smb/server
2. The Solaris CIFS SMB service uses WORKGROUP as the default group. If
the workgroup needs to be changed, use the following command to change the
workgroup name:
# smbadm join –w workgroup-name
3. Next edit the /etc/pam.conf ﬁle to enable encrypted passwords to be used
for CIFS. Add the following line to the end of the ﬁle:
other password required pam_smb_passwd.so.1 nowarn

Figure 6.3 GNOME ﬁle manager with network and server OPENSOLARIS

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

4. Each user currently in the /etc/passwd ﬁle needs to reencrypt to be able to

use the CIFS service:
$ pfexec passwd user-name
In Figure 6.3, you can see that the server OPENSOLARIS is an SMB server.
5. Select the server OPENSOLARIS to ﬁnd any SMB shares that the server has.
In Figure 6.4, no SMB shares are currently available to mount from OPEN-
SOLARIS.

Figure 6.4 Selected OPENSOLARIS and with no available SMB shares

6.3 Sharing Home Directories

Sharing home directories using the OpenSolaris SMB server has been designed for
ease of administration. The system creates a share when a user logs in and
removes it when the user logs out.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1. To share all home directories, create and edit the /etc/smbautohome file. In
the example OpenSolaris installation, all home directories are in /export/
home/username file systems.
2. Edit /etc/smbautohome, and add the following line:
* /export/home/&
The star (*) and the ampersand (&) compose a shortcut to match a username
with its home directory. Otherwise, each user would need their own entry in
the /etc/smbautohome file.
3. Test the new home directory using the GNOME file manager. On the Go To
line, type smb://hostname/username, which is smb://opensolaris/judy
in the following example.
After entering the SMB URL in the Go To box, you are presented with a login
window, as illustrated in Figure 6.5.
4. Log in. In Figure 6.6, user judy is logging in to the workgroup domain
WORKGROUP.

Figure 6.5 The GNOME ﬁle manager logging into an OpenSolaris SMB server

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 6.6 Login window for user judy

As shown in Figure 6.7, user judy is presented with her home directory. She
can see only her own directory and cannot browse other users’ directories.
This makes for a cleaner interface for a user.

Figure 6.7 The ﬁle manager after user judy logs in

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Another way to activate SMB shares is by setting the sharesmb property to on,
using the zfs set command:

$ pfexec zfs set sharesmb=on homepool/export/home

$ pfexec zfs set sharesmb=on homepool/export/home/cindy
$ pfexec zfs set sharesmb=on homepool/export/home/judy
$ pfexec zfs set sharesmb=on homepool/export/home/ken
$ pfexec zfs set sharesmb=on homepool/export/home/layne
$ pfexec zfs set sharesmb=on homepool/export/home/watanabe

Figure 6.8 shows what the ﬁle manager will see if all the ZFS ﬁle systems are
shared using the sharesmb setting.

Figure 6.8 The ﬁle manager showing individual home directory shares

In Figure 6.8, the names listed are the ZFS ﬁle systems and may confuse some
users. You should share common directories via the smbshare property.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Licensed by
Onur Bingul
3690014
Time Slider is a tool that users can use to retrieve deleted, damaged, or previous
versions of a ﬁle or directory. Time Slider has two parts: a front-end GUI in the
Nautilus ﬁle manager and a back end managed by Service Management Facility
(SMF). Time Slider is currently available on OpenSolaris.

7.1 Enabling Time Slider Snapshots

You can enable and configure Time Slider snapshots using the Time Slider Setup
GUI. In this chapter, I will show how to configure Time Slider only for home direc-
tories. The default configuration takes snapshots of all ZFS file systems.
The Time Slider schedule has five periodic snapshots (see Table 7.1). Each
period has a different number of snapshots kept.

Table 7.1 Snapshot Types, Periodic Frequency, and Number Kept

Type of Snapshot Frequency Kept

Frequent 15 minutes 4
Hourly 60 minutes 24
Daily Daily 31
Weekly Weekly (1, 8, 15, 22, 29) 4
st
Monthly Monthly (on the 1 ) 12

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

To enable Time Slider snapshots, follow these steps:

1. Start the Time Slider Setup GUI as the administrative user of the system.
From the GNOME desktop, select System > Administration > Time Slider
Setup from the top menu bar (see Figure 7.1).

Figure 7.1 Starting the Time Slider Setup GUI

2. Next, select the Enable Time Slider check box, as shown in Figure 7.2, and
click OK.

Figure 7.2 Enabling the Time Slider box

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

3. Click the Advanced Options button. Here you’ll use Time Slider on home
directories, so deselect everything but user home directories (see Figure 7.3).
Then click the OK button.

Figure 7.3 Advanced options in Time Slider

7.2 Enabling Nautilus Time Slider

By using the Nautilus file manager and the Time Slider plug-in, a user can recover
files from previous snapshots of the user’s home directory. The recovered file or
directory will be placed on the user’s desktop.

1. Start the Nautilus file manager, and set the mode to List View. From the
Nautilus file manager menu, select View > List to set the List View mode.
The List View mode will display information about the file or directory when
the Time Slider box is selected.
Figure 7.4 shows the Nautilus file manager in the List View mode.
2. Click the Time Slider icon. As shown in Figure 7.5, with the Time Slider icon
selected, the slider appears.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 7.4 Start the Nautilus ﬁle manager, and set the view mode to List View

Figure 7.5 The Nautilus ﬁle manager with the Time Slider icon selected

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

3. In the “Restore information” column only, the file testfile has different ver-
sions available for restoration. Drag the slider to the desired time frame.
When the slider is dragged to an older snapshot, a difference is noted from
the latest version of the file or directory. For example, in Figure 7.6, the file
testfile is compared with the current version of the file. The date time-
stamp and size differences are noted in the “Restore information” column.

Figure 7.6 The Nautilus ﬁle manager with Time Slider and the slider moved to the
desired time frame

4. Select the ﬁle testfile for restore. Then select Edit > Restore to Desktop, as
shown in Figure 7.7.
5. In Figure 7.8, you see the ﬁle testfile restored to the desktop. It appears
right below the Start Here icon.

7.3 Modifying the Snapshot Schedule

You can modify the ZFS snapshot schedule using SMF. You can gain ﬁner control
over the behavior of the autosnapshots by using the command line of SMF.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 7.7 The Nautilus ﬁle manager with ﬁle testfile selected for restoring
to the desktop

Figure 7.8 The testfile ﬁle restored to the desktop

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

7.3.1 Snapshot Basics

Check to see whether all default snapshot services are running. There should be ﬁve
instances running by default for each of the ﬁve periodic snaps listed in Table 7.1.

$ svcs | grep snapshot

online 16:01:38 svc:/system/filesystem/zfs/auto-snapshot:monthly
online 16:01:39 svc:/system/filesystem/zfs/auto-snapshot:daily
online 16:01:41 svc:/system/filesystem/zfs/auto-snapshot:weekly
online 16:01:41 svc:/system/filesystem/zfs/auto-snapshot:hourly
online 16:01:43 svc:/system/filesystem/zfs/auto-snapshot:frequent

Each snapshot is executed from cron as user zfssnap. To check the cron sched-
ule, use the crontab command. Do not edit the crontab for zfssnap. Each SMF
instance manages the cron schedule. See the man page on crontab for more
information on the entry format.

$ pfexec crontab -l zfssnap

0 0 1 1,2,3,4,5,6,7,8,9,10,11,12 * /lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:monthly
0 0 1,2,3,4,5,6,7,8,9,10,11,12,13,14, \
15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 * * \
/lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:daily
0 0 1,8,15,22,29 * * /lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:weekly
0 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15, \
16,17,18,19,20,21,22,23 * * * /lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:hourly
0,15,30,45 * * * * /lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:frequent

7.3.2 Changing the Period

Change the period for the frequent snapshots service to every ten minutes. Use the
svccfg command to change the service properties. First, you list the properties for
the frequent method by using the svccfg command, as in line 1:

1 $ pfexec svccfg -s frequent listprop

2 zfs application
3 zfs/avoidscrub boolean false
4 zfs/backup-lock astring unlocked
5 zfs/backup-save-cmd astring "not set"
6 zfs/fs-name astring //
7 zfs/interval astring minutes

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

8 zfs/label astring frequent

9 zfs/snapshot-children boolean true
10 zfs/verbose boolean true
11 zfs/backup astring none
12 zfs/offset astring 0
13 zfs/period astring 15
14 zfs/keep astring 4
15 general framework
16 general/action_authorization astring solaris.smf.manage.zfs-auto-snapshot
17 general/value_authorization astring solaris.smf.manage.zfs-auto-snapshot
18 general/enabled boolean true
19 restarter framework NONPERSISTENT
20 restarter/logfile astring /var/svc/log/\
system-filesystem-zfs-auto-snapshot:frequent.log
21 restarter/start_pid count 585
22 restarter/start_method_timestamp time 1235948503243160000
23 restarter/start_method_waitstatus integer 0
24 restarter/transient_contract count
25 restarter/auxiliary_state astring none
26 restarter/next_state astring none
27 restarter/state astring online
28 restarter/state_timestamp time 1235948503.264556000

The zfs/period property on line 13 is currently set to 15, and the zfs/
interval property on line 7 is set to minutes. The following command line will set
the period to ten minutes:

$ pfexec svccfg -s frequent setprop zfs/period=10

$ pfexec svccfg -s frequent listprop zfs/period
zfs/period astring 10

Next, reread the instance:

$ pfexec svccfg -s frequent refresh
Restart the frequent service:
$ pfexec svcadm restart frequent
Check the zfssnap crontab to conﬁrm the updated schedule. Now the snap-
shots occur at a 10-minute interval.

1 $ pfexec crontab -l zfssnap

2 0 0 1 1,2,3,4,5,6,7,8,9,10,11,12 * /lib/svc/method/zfs-auto-snapshot\
svc:/system/filesystem/zfs/auto-snapshot:monthly
3 0 0 1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,\
16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31 * *\
/lib/svc/method/zfs-auto-snapshot svc:/system/filesystem/zfs/\
auto-snapshot:daily

continues

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

4 0 0 1,8,15,22,29 * * /lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:weekly
5 0 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23 * * \
* /lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:hourly
6 0,10,20,30,40,50 * * * * /lib/svc/method/zfs-auto-snapshot \
svc:/system/filesystem/zfs/auto-snapshot:frequent

On line 6, the frequency of the snaps is now changed to every 10 minutes start-
ing at the top of the hour.

7.3.3 Changing the Number of Snapshots Kept

To keep the last hour of changes, the number of snapshots that are kept needs to
be increased from four to six. Again, you will use the commands svccfg and
svcadm to make the changes and restart the service:

$ pfexec svccfg -s frequent setprop zfs/keep=6

$ pfexec svccfg -s frequent listprop zfs/keep
zfs/keep astring 6
$ pfexec svccfg -s frequent refresh
$ pfexec svcadm restart frequent

7.4 Setting the Snapshot Schedule per File System

You can set snapshot schedules independently on each file system. In the follow-
ing example, the new file system jan has inherited the property autosnapshots
turned off. The file system needs only hourly and weekly scheduled snapshots. Use
the zfs command to set the autosnapshot properties. The svcadm command
restarts the hourly and weekly autosnapshot methods to inherit the changes in the
user jan file system.

$ pfexec zfs create rpool/export/home/jan

$ zfs get com.sun:auto-snapshot rpool/export/home/jan
NAME PROPERTY VALUE SOURCE
rpool/export/home/jan com.sun:auto-snapshot false inherited from rpool/export/home
$ pfexec zfs com.sun:auto-snapshot:weekly=true rpool/export/home/jan
$ pfexec zfs com.sun:auto-snapshot:hourly=true rpool/export/home/jan
$ pfexec svcadm restart hourly weekly

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

8.1 Creating Virtual Disks with Virtual Media Manager

Virtual Media Manager (VMM) is used to manage hard disks, CD/DVD images,
and ﬂoppy images. It can create new virtual disks or register disk images with
Virtual Box. It will register ISO images or raw ﬂoppy images for VMs to use.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

In this section, you will see how to create two IDE 16GB disks for boot disks and
ﬁve SATA 130MB disks for data. For a Solaris installation, a minimum disk size of
16GB is recommended. You can do the installation on a much smaller disk, but
this chapter will use 16GB. ZFS requires disks to be a minimum of 128MB. The
VMM can create two kinds of virtual disks:

Dynamically expanding storage: A virtual disk that on initialization uses

very little physical disk space but will grow dynamically to the size speciﬁed
as the guest OS uses the disk space
Fixed-size storage: A virtual disk that does not grow and is stored in a ﬁle
the same size as the virtual hard disk

Use the following steps to select the dynamically expanding disks to save on the
physical disk space:

1. Start Virtual Box, and then select File > VMM (see Figure 8.1).

Figure 8.1 Starting Virtual Media Manager from the File menu

2. Once the VMM panel appears, click the New button. Figure 8.2 includes pre-
viously created virtual disk images.
3. The Create New Virtual Disk Wizard appears. Click the Next button to get to
the Hard Disk Storage Type panel (see Figure 8.3).
“Dynamically expanding storage” should be selected by default. If it’s not,
select it before clicking the Next button. The Virtual Disk Location and Size
panel is next (see Figure 8.4).
4. Accept the default location of the new virtual disk. The default location
depends on the OS type of the host system. Enter ZFStest.vdi for the name
of the new disk, and make the size 16GB (see Figure 8.4 (2)).
5. Clicking Next, you get the Summary panel (see Figure 8.5).
6. Click Finish to create the new boot disks for the VM. Repeat the sequence,
and create another 16GB disk named ZFStest2.vdi.

Now that the boot disks are created, you need to create six 130MB data disks
named z0, z1, z2, z3, z4, and z5.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.2 VMM panel with the New button highlighted

Figure 8.3 Hard Disk Storage Type panel with “Dynamically expanding storage” selected

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.4 Virtual Disk Location and Size panel

Figure 8.5 Summary panel with the conﬁguration

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.6 Virtual Disk Location and Size panel

1. Name the ﬁrst disk z0.vdi.

2. Set the size of the disk to 130MB (see Figure 8.6 (2)).
3. After creating the other disks, review them in the VMM panel (see Figure 8.7).

Figure 8.7 lists all the virtual disks registered to Virtual Box. Notice the actual
size of the new data disks of 1.5KB vs. the virtual size.

8.2 Registering a CD Image with Virtual Media Manager

You will need to register the install CD with the VMM. In this case, the CD image
is OpenSolaris 2008.11 downloaded from http://opensolaris.org. Follow these
steps to register it:

1. Click the CD/DVD Images button to get to the registered CD/DVD ISO. Figure 8.8
shows previously registered ISO images.
2. Click the Add button, and locate the installation ISO image. In this example,
the OpenSolaris image is osol-0811.iso. Figure 8.9 shows the new regis-
tered ISO image.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.7 Hard disks in the VMM

Figure 8.8 Registered ISO images in the VMM

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.9 Registered ISO images in the VMM with the new OpenSolaris ISO

8.3 Creating a New Virtual Machine

Creating a new VM is the next step in setting up a ZFS lab system. Follow these
steps to create the new VM:

1. The main Virtual Box panel (see Figure 8.10) contains a New button at the
top left of the panel. The left subpanel is a list of defined systems and status,
and the right subpanel has the configuration of each system.
In the main Virtual Box panel, click the New button to open the Create New
Virtual Machine Wizard (see Figure 8.11).
2. Click the Next button to get to the VM Name and OS Type panel (see Figure 8.12).
For the name of the system, enter ZFStest (see Figure 8.12 (1)). For the
operating system, select Solaris (see Figure 8.12 (2)). For the version,
select OpenSolaris (see Figure 8.12 (3)). Then click the Next button.
3. The next panel configures the memory for the system. Type 1024 in the box for
1GB of physical memory in the VM. Figure 8.13 shows the completed panel.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.10 Virtual Box main panel

4. Click the Next button to get to the Virtual Hard Disk panel. Because you
created the boot disks previously, select “Use existing hard disk.” Then select
ZFStest.vdi from the menu. Figure 8.14 shows the completed panel.
5. Click the Next button to get to the Summary panel. Check the selection. Then
click the Finish button to create the new VM (see Figure 8.15).
After you click the Finish button, the new VM is created and listed on the left
subpanel (see Figure 8.16).

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.11 Create New Virtual Machine Wizard

Figure 8.12 VM Name and OS Type panel completed

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.13 The Memory panel with 1024MB

Figure 8.14 Completed Virtual Hard Disk panel with ZFStest.vdi selected

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Licensed by
Onur Bingul
3690014
Figure 8.15 The Summary panel before creating the new VM

8.4 Modifying the New Virtual Machine

In this section, you will modify the VM, add new virtual disks, and mount a boot
ISO image to the VM:

1. First, select ZFStest on the left panel, and then click the Settings button to
open the settings panel for ZFStest. Figure 8.17 shows the General settings
for the VM.
2. Click the Storage icon to conﬁgure more storage for the VM. Figure 8.18
shows the VM with one disk conﬁgured.
3. Add the remaining disks that were previously created to the VM. First, click
Enable Additional Controller to add a SATA controller to the VM, and then
click the Add Disk icon on the right side of the panel. Match the Slot/Hard
Disk pairs for each disk added (see Table 8.1).
Figure 8.19 shows the completed disk addition and assignments.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.16 The Virtual Box main panel with new ZFStest VM

4. Once the disks have been added to the VM and the SATA assignments have
been done, click the OK button. The SATA port assignments can be random,
but assigning a numbered disk name to a SATA port number will make it eas-
ier to see which disk belongs to which disk target from the OS point of view.
5. Click the CD/DVD-ROM icon. Select Mount CD/DVD Drive. Select ISO Image
File. Then select osol-0811.iso from the drop-down menu. Only registered ISO
images will be listed. See Figure 8.20 for the completed panel. Click the OK
button to complete the conﬁguration.

The ZFStest VM is now completed and ready to boot from the ISO CD image.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.17 The General settings panel for ZFStest VM

Figure 8.18 The Storage settings panel with one disk conﬁgured for VM

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Table 8.1 Slot and Hard Disk Assignments

Slot Hard Disk

IDE Secondary ZFStest2.vdi
SATA Port 0 z0.vdi
SATA Port 1 z1.vdi
SATA Port 2 z2.vdi
SATA Port 3 z3.vdi
SATA Port 4 z4.vdi
SATA Port 5 z5.vdi

Figure 8.19 New hard disks added to the VM

8.5 Installing an OS on a Virtual Machine

This section covers how to install an operating system on a virtual machine.

To get started, turn on the VM by selecting the Start icon at the top of the
main panel to start the boot process. Virtual Box will spawn a new window to

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.20 The CD/DVD-ROM panel with the ISO boot image mounted on the VM

display the VM running. At this point, the VM will grab the mouse and keyboard.
To release both the mouse pointer and keyboard, press the key indicated on the
bottom right of the screen. In the MacBook Pro, it is the left Cmd key. On a Solaris
host it is the right Ctrl key.
Figure 8.21 shows the icons at the bottom of the Virtual Box VM window. The
buttons from left to right are hard disk activity, CD-ROM activity and control, net-
work activity and control, USB control, shared folder, virtualization hardware
acceleration status, mouse capture status, and mouse capture key.

Figure 8.21 Icons at the bottom of the VM window

1. A GRUB menu will appear. Select the default GRUB selection. Figure 8.22
shows the GRUB menu from the installation CD.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.22 The default GRUB menu from the OpenSolaris LiveCD

2. Select your keyboard and language. In this example, just hit the Enter/
Return key twice. Now the LiveCD will drop into a GNOME session. This is a
working OpenSolaris system.
3. Double-click Install OpenSolaris to start the install process, as shown in
Figure 8.23.
4. The first screen is the Welcome screen shown in Figure 8.24. Click the Next
button to continue.
5. The installer asks a few questions before the actual install. The first question
is about the disk. By default the first 16GB disk will be selected. Select “Use
the whole disk.” The completed screen should look like Figure 8.25.
6. Click the Next button to get the Time Zone screen. Select your local time zone
either by using the drop-down menus or by locating a city in your time zone,
in this case Denver, Colorado (see Figure 8.26).

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.23 VM ZFStest in LiveCD mode

7. Click the Next button to move on to the Locale screen. In Figure 8.27, English
is the default language.
8. Click the Next button to get to the Users screen. Select a password for the
root user. Create a username for yourself, and set the password. At this point,
you can change the name of the system or leave it as opensolaris. Figure 8.28
shows the completed Users screen.
9. Click the Next button to move to the Installation review screen. This is the
last chance to review the installation parameters before the installation pro-
cedure begins. See Figure 8.29 for an example.
The installation will take a few minutes to complete, depending on the host
system. You can watch the splash screen for information on the features of
OpenSolaris. The completion bar at the bottom will indicate what percent of
the installation has completed (see Figure 8.30).

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.24 Installer at the Welcome screen

10. Once the install is complete, the Finished screen appears, as shown in
Figure 8.31. Click the Reboot button to restart the VM with the newly
installed OpenSolaris OS.
11. After clicking the Reboot button, the system will reboot from the CD-ROM
ISO image. The GRUB menu will appear. Use the down arrow key to scroll
down to the Boot from Hard Disk item, and press the Return/Enter key to
start the OS from the hard disk. In Figure 8.32 the Boot from Hard Disk item
is selected.
12. When you press Enter/Return to boot the OS, another GRUB screen appears.
This GRUB menu comes from the hard disk installation. Let the default
GRUB selection boot by pressing Enter/Return, or let the timer run down to
boot the OS.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.25 The installer at the Disk screen with “Use the whole disk” selected

8.6 Installing Virtual Box Tools

Installing the Virtual Box tools will complete the installation of the OS. There is a
different VB tool installation for each of the various OSs that VB supports. The VB
tools will enable the VM to seamlessly share the mouse and keyboard and install a
video device driver. The video device driver will enable you to resize the VM screen
on your host system much like any other application.

1. You need to mount the Virtual Box Guest Additions CD image. First, log in to
OpenSolaris as the user deﬁned during the installation. Next, right-click (on
a Mac, Cmd+click) the OpenSolaris CD icon on the desktop. Select Unmount
Volume to unmount the CD image. In Figure 8.33, Unmount Volume is
selected from the drop-down menu.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.26 The installer at the Time Zone screen with Denver, Colorado, selected

2. You need to mount the Virtual Box Guest Additions CD image. Once the CD
image is unmounted, free your mouse from the VM using the mouse capture
key indicated at the bottom-right corner of the window. Now move the mouse
pointer over the CD-ROM icon at the bottom of the screen, and click the icon.
Select CD/DVD-ROM Image, as shown in Figure 8.34.
3. The Virtual Media Manager window is now displayed. Select the
VirtualBoxAdditions.iso image, and click the Select button. In
Figure 8.35, the VMM window is open with the Virtual Box Guest Additions
image selected.
If the image does not mount, go to the CD-ROM icon at the bottom of the VM
window, select the physical CD-ROM, and then repeat the selection of the
image ﬁle.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Licensed by
Onur Bingul
3690014

Figure 8.27 The installer at the Locale screen with English selected

4. When the image is mounted, a warning message box is displayed. Click the
Run button to continue with the mount.
5. The Virtual Box Guest Additions CD image must be installed from the com-
mand line. Open a terminal window by clicking the Terminal icon on the top
menu bar.
6. Change directories to /media/VBOXADDITIONS_2.2.0_45846. The VBOX
directory may be different depending on the build you download from
http://virtualbox.org.
$ cd /media/VBOX*
7. Install the guest package VBoxSolarisAdditions.pkg using the following
command line. Because this install needs root privileges, the command
needs to start with the command pfexec. The user deﬁned in the install

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.28 The installer at the Users screen with the ﬁelds completed

will have root privileges deﬁned by the Solaris Role-Based Access Control
(RBAC) facility. See the man page on RBAC, or go to http://docs.sun.com
for more information on RBAC.
$ pfexec pkgadd –d VBoxSolarisAdditions.pkg
8. Press Enter/Return, and answer y for yes to install the package. Log out of
the system by clicking System on the top menu bar and selecting Log Out.
See Figure 8.36 for the menu.
9. Log in again to activate the VB Guest Additions image. Now your
OpenSolaris system is ready to be your ZFS lab in a box.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.29 The installer at the Installation review screen

Figure 8.30 The installer loading OpenSolaris

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.31 The installer Finished screen after the installation is completed

Figure 8.32 The GRUB menu with the Boot from Hard Disk item selected

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.33 The mouse right-click menu with Unmount Volume selected

Figure 8.34 CD icon drop-down menu with CD/DVD-ROM Image selected

Figure 8.35 The VMM window with Virtual Box Guest Additions image selected
to be mounted

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 8.36 The Log Out action selected on the System menu

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

A upgrading with beadm command, 43–44

AD (Active Directory), 78 upgrading with pkg image-update
Administration command, 44–45
delegation of, 3 x4500 boot mirror example, 74
simplifying systems
administrations, 25–26
Arrays, storage arrays in ZFS, 74 C
Automation, of tasks, 3 Cache vdev
autoreplace, 19–22 overview of, 10
rules for ZFS pools, 11
B CD images
beadm command managing with VMM, 93
boot environment management registering with VMM, 97–99
options, 26 Checksums, 2
managing boot environment with, 43 CIFS
upgrading boot environment with, 43–44 clients, 75
Boot disks configuring SMB server in workgroup
creating, 94 mode, 78–79
mirroring, 62 installing server packages from command
Boot environment line, 77
managing with beadm command, 43 installing server packages with Package
managing with Solaris Live Upgrade, 35 Manager, 75–77
options for managing, 26 overview of, 75
patching with Solaris Live sharing home directories, 79–82
Upgrade, 40–42 Clients, CIFS, 75
references, 46 Clone, 3

119

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Command line, installing server packages Flash, 27

from, 77 Floppy images, 93
Commands
delegation of, 58
G
ZFS file system, 7
Compression gzip command, 53–55
enabling, 53–55
ZFS file system and, 2 H
Create New Virtual Disk Wizard, 94
Create New Virtual Machine, 101–103 Hard disks, 106. See also Disks
crontab command, 89–90 hidden property, 56
Home directories
delegation of command, 58
D enabling compression on ZFS file system,
Data integrity, beneﬁts of ZFS ﬁle system, 2 53–55
Datasets, ZFS supported, 3–4 limiting storage size of descendants, 51
Delegation managing quotas and
of administration, 3 reservations, 47–48
of commands, 58 managing snapshot directory, 55–57
Descendants, limiting storage size of, 51 overview of, 47
Disks quota setting, 48–49
assignments, 106 references, 60
creating virtual disks with VMM, 93–97 refquota setting, 49–51
guidelines for ZFS pools, 10 refreservation setting, 51–53
locating correct disk for reservation setting, 51–52
replacement, 22–23 rolling back files systems, 58–59
replacing bad devices sharing, 59–60, 79–82
automatically, 19–22
root file system disk requirements, 27 I
rules for ZFS pools, 11
Image Packaging System (IPS), 75
SATA, 94
Installation
types of virtual disks, 94
example of interactive installation, 28–30
du command, 50
installing root file system, 26–37
DVD images
references, 46
managing with VMM, 93
IPS (Image Packaging System), 75
registering with VMM, 97–99
ISO images, 93, 97–99
Dynamic stripes
creating, 11–12
overview of, 10 J
rules for ZFS pools, 11 JBOD (just a bunch of disks), 10
JumpStart
configuring mirrored root pool, 30
F
installing root file system, 26–27
File manager, Nautilus, 83, 85–87 Just a bunch of disks (JBOD), 10
File systems
scheduling Time Slider snapshots per
file system, 91 L
UFS. See UFS (Unix File System) Lab, creating ZFS lab
ZFS. See ZFS file system creating new virtual machine, 99–103

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

creating virtual disks with VMM, 93–97 P

installing OSs on virtual Package Manager, 75–77
machines, 106–111 Patches
installing tools for, 111–118 benefits of ZFS file system, 2
modifying virtual machine, 103–106 patching boot environment using Solaris
overview of, 93 Live Upgrade, 40–42
registering CD image with VMM, 97–99 pkg command
list subcommand, zpool command, 16 installing server packages from command
Live Upgrade. See Solaris Live Upgrade line, 77
Log vdev upgrading boot environment, 43-44
overview of, 10 Pools
rules for ZFS pools, 11 adding spare vdev to second storage
Logical unit number (LUN), 74 pool, 18–19
ls command, 55–56 creating dynamic stripes, 11–12
LUN (logical unit number), 74 creating mirrored pools, 13–15
lzjb algorithm, 53–55 creating pool with RAID-Z devices, 15–17
creating spares in storage pools, 17–18
creating ZFS pool with four RAID-Z2
M vdevs, 64
Memory requirements, root file system, 27 example of misconfigured pool, 23–24
Migration, UFR root file system to ZFS root fast and simple storage with ZFS, 6
file system, 35–40 guidelines for, 10
mirror subcommand, zpool command, 13 limiting storage size of descendants, 51
Mirrored disks, 62 locating correct disk for
Mirrored pools replacement, 22–23
creating, 13–15 overview of, 9
creating mirrored root pool configuration, as primary storage element, 3–5
30–31 quotas and reservations, 47–48
testing mirrored root pool configuration, replacing bad devices
31–32 automatically, 19–22
x4500 boot mirror example, 74 root pool recommendations, 27–28
x4500 RAID mirrored configuration rules for, 11
example, 69–73 virtual devices types, 10
mirrors (RAID-Z), 10
Q
N quota setting, on home directory, 48–49
Quotas
Nautilus file manager
benefits of ZFS file system, 3
enabling Nautilus Time Slider, 85–87
managing, 47–48
overview of, 83
quota setting, 48–49
restoring with, 88
refquota setting, 49–51
NFS (Network File System), 59

R
O RAID
OpenSolaris SMB server. See SMB server configurations, 4
OSs (operating systems), installing on ZFS file system options, 2
virtual machines, 106–111 ZFS supported, 9

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

RAID-5 write holes, 9 upgrading boot environment using

RAID-Z devices beadm command, 43–44
creating pool with, 15–17 upgrading boot environment using pkg
creating vdevs, 62 image-update command, 44–45
creating ZFS pool with four RAID-Z2
vdevs, 64
X4500 RAID-Z2 configuration example, S
61–69 SATA
Raw volumes, 2, 3 controllers, 61–63
Redundancy, 2. See also RAID disks, 94
Redundant groups. See Mirrored pools Servers, CIFS
refquota setting, 49–51 configuring SMB server in workgroup
refreservation setting, 52–53 mode, 78–79
reservation setting, 51–52 installing server packages from command
Reservations line, 77
managing, 47–48 installing server packages with Package
refreservation setting, 52–53 Manager, 75–77
reservation setting, 51–52 Service Management Facility (SMF)
Restores modifying snapshot schedule, 87–88
recovering files from snapshots, 57 overview of, 83
restoring from snapshot of root Sharing home directories, 59–60, 79–82
pool, 33–34 Slices, guidelines for ZFS pools, 10
Resynchronization, volume management SMB server
and, 21 configuring in workgroup mode, 78–79
rollback subcommand installing server packages from command
recovering whole environments from line, 77
snapshots, 57 installing server packages with Package
rolling back files systems, 58–59 Manager, 75–77
root ﬁle (/) system sharing home directories, 79–82
boot and installation references, 46 SMF (Service Management Facility)
creating mirrored configuration, 30–31 modifying snapshot schedule, 87–88
creating snapshot of, 32–33 overview of, 83
example of interactive installation, 28–30 Snapshots
installing, 26–30 benefits of ZFS file system, 3
managing boot environment using beadm creating, 32–33
command, 43 managing snapshot directory, 55–57
managing boot environment with Solaris recovering files from, 57
Live Upgrade, 35–42 restoring from snapshot of root
migrating UFR root file system to ZFS pool, 33–34
root file system, 35–40 sending snapshot of root pool to
overview of, 25 storage, 33
patching boot environment using Solaris Snapshots, Time Slider
Live Upgrade, 40–42 basics of, 89
pool recommendations, 27–28 changing number kept, 91
restoring from snapshot, 33–34 changing period of, 89–91
sending snapshot to storage, 33 enabling, 83–85
system administration and, 25–26 enabling Nautilus Time Slider, 85–87
system requirements, 27 modifying schedule, 87–88
testing mirrored configuration, 31–32 scheduling per file system, 91

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris Live Upgrade scheduling snapshots per file system, 91

installation methods, 26 Tools, installing on virtual
managing boot environment with, 35 machines, 111–118
option for managing boot Top-level view, of ZFS ﬁle system, 3–5
environment, 26
patching boot environment using, 40–42
Solid-state disks (SSDs), 10 U
Spare vdev UFS (Unix File System)
adding spare vdev to second storage migrating UFS to ZFS root file system,
pool, 18–19 35–40
creating in storage pools, 17–18 no options for installing, 27

Licensed by
overview of, 10 sharing home directories via NFS, 59
rules for ZFS pools, 11 ufsdump command, 35
SSDs (solid-state disks), 10 ufsrestore command, 35
status subcommand, zpool Unix File System. See UFS (Unix File

Onur Bingul
command, 20–21 System)
Storage Upgrades
arrays, 74 benefits of ZFS file system, 2
pools. See Pools upgrading boot environment using beadm
virtual disks for, 94 command, 43–44
Storage blocks, 51–52
Stripes, dynamic. See Dynamic stripes
Sun x4500
3690014 upgrading boot environment using pkg
image-update command, 44–45
/usr directory, 26
boot mirror example, 74
RAID mirrored configuration example,
69–73 V
RAID-Z2 configuration example, 61–69 /var directory
SUNWsmbs package, 75 example of interactive installation, 28, 30
SUNWsmbskr package, 75 in ZFS boot environment, 26
svcadm command, 91 vdev (virtual device), 4
svccfg command, 89–91 adding spare vdev to second storage
SVM (Solaris Volume Manager) pool, 18–19
mirrored boot disks, 62–63, 69 creating RAID-Z2 vdevs, 62–63
volume management with, 10 types of, 10
System administration, 25–26 ZFS pools comprised of, 9
System requirements, 27 Virtual Box
installing tools on virtual
machines, 111–118
T main panel of, 100, 104
tar archive, 56 overview of, 93
Time Slider Virtual devices. See vdev
basics of snapshots, 89 Virtual disks
changing number of snapshots kept, 91 creating, 96–97
changing period of snapshots, 89–91 location and size, 96
enabling Nautilus Time Slider, 85–87 types of, 94
enabling Time Slider snapshots, 83–85 Virtual machines. See VMs (virtual
modifying snapshot schedule, 87–88 machines)
overview of, 83 Virtual Media Manager. See VMM (Virtual
recovering files from snapshots, 57 Media Manager)

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

visible property, managing directory setting snapshot schedule per file

visibility, 56 system, 91
VMM (Virtual Media Manager) ZFS Configuration Guide, 61
creating virtual disks with, 93–97 ZFS file system
hard disks in, 98 advantages of, 2–3
ISO images in, 98–99 available in Solaris releases, 1
registering CD image with, 97 commands, 7
starting from File menu, 94 enabling compression on, 53–55
VMs (virtual machines) fast and simple storage with, 6
creating new, 99–103 overview of, 1–2
installing OSs on, 106–111 quotas and reservations, 47–48
installing tools on, 106–111 rolling back, 58–59
modifying, 103–106 root file. See root file (/) system
Virtual Box for, 93 top-level view of, 3–5
Volumes zfssnap, 89–90
managing, 10 zpool command, 3
raw volumes, 2, 3 creating dynamic stripes, 11–12
resynchronization and, 21 creating mirrored pools, 13–15
function of, 7
W list subcommand, 16
Workgroup mode, configuring SMB server managing storage pools with, 9
in, 78–79 status subcommand, 20–21
zpool command, advanced concepts
X overview of, 61
storage arrays and, 74
x4500. See Sun x4500 X4500 boot mirror example, 74
X4500 RAID mirrored configuration
Z example, 69–73
zfs command X4500 RAID-Z2 configuration example,
function of, 7 61–69

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 System Administration Essentials is the first book to concisely

yet comprehensively cover all of the breakthrough features of the Solaris 10
operating system. The Solaris OS has a long history of innovation, and the
Solaris 10 OS is a watershed release that includes features such as Zones,
ZFS™, the Fault Management Architecture, the Service Management
Facility, and Dynamic Tracing (DTrace). In addition, the Solaris 10 OS
fully supports 32-bit and 64-bit x86 platforms, as well as the SPARC®
architecture.
The book’s key topics include
• Installing, booting, and shutting down a system • Managing users
• Managing packages and patches (software updates) • Configuring networks
• Controlling system processes • Using printing services
• Managing disks and devices

Solaris™ 10 System Administration Essentials is a practical guide to deploying and managing the Solaris 10
operating system in a business or academic environment. The book is easy to read and rich with examples—
a perfect companion for system administrators who are deploying the Solaris OS for the first time.
informit.com/title/9780137000098

Solaris™ 10 Security Essentials

ISBN-13: 978-0-13-701233-6
Solaris™ 10 Security Essentials describes the various security technologies
contained in the Solaris operating system. The book describes how to make
installations secure and how to configure the OS to the particular needs of
your environment, whether your systems are on the edge of the Internet or
running a data center. The authors present the material in a straightforward
way that makes a seemingly arcane subject accessible to system
administrators at all levels.
The strengths of the Solaris operating system’s security model are its
scalability and its adaptability. It can protect a single user with login
authentication or multiple users with Internet and intranet configurations
requiring user-rights management, authentication, encryption, IP security,
key management, and more. This book is written for users who need to secure their laptops, network
administrators who must secure an entire company, and everyone in between.
The book’s topics include
• Zones virtualization security • Cryptographic services and key management
• System hardening • Auditing
• Trusted Extensions (Multilevel Security) • Network security
• Privileges and role-based access control (RBAC) • Pluggable Authentication Modules (PAM)

Solaris™ 10 Security Essentials is a superb guide to deploying and

managing secure computer environments.
informit.com/title/9780137012336

informit.com/ph

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

InformIT is a brand of Pearson and the online presence

for the world’s leading technology publishers. It’s your source
for reliable and qualified content and knowledge, providing
access to the top brands, authors, and contributors from
the tech community.

LearnIT at InformIT
Looking for a book, eBook, or training video on a new technology? Seek-
ing timely and relevant information and tutorials? Looking for expert opin-
ions, advice, and tips? InformIT has the solution.

• Learn about new releases and special promotions by

subscribing to a wide variety of newsletters.
Visit informit.com /newsletters.

• Access FREE podcasts from experts at informit.com /podcasts.

• Read the latest author articles and sample chapters at
informit.com /articles.

• Access thousands of books and videos in the Safari Books

Online digital library at safari.informit.com.
• Get tips from expert blogs at informit.com /blogs.
Visit informit.com /learn to discover all the ways you can access the
hottest technology content.

Are You Part of the IT Crowd?

Connect with Pearson authors and editors via RSS feeds, Facebook,
Twitter, YouTube, and more! Visit informit.com /socialconnect.

informIT.com THE TRUSTED TECHNOLOGY LEARNING SOURCE

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

FREE TRIAL—GET STARTED TODAY!

www.informit.com/safaritrial

Find trusted answers, fast

Only Safari lets you search across thousands of best-selling books from the top
technology publishers, including Addison-Wesley Professional, Cisco Press,
O’Reilly, Prentice Hall, Que, and Sams.

Master the latest tools and techniques

In addition to gaining access to an incredible inventory of technical books,
Safari’s extensive collection of video tutorials lets you learn from the leading
video training experts.

WAIT, THERE’S MORE!

Keep your competitive edge
With Rough Cuts, get access to the developing manuscript and be among the ﬁrst
to learn the newest technologies.

Stay current with emerging technologies

Short Cuts and Quick Reference Sheets are short, concise, focused content
created to get you up-to-speed quickly on new and cutting-edge technologies.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Register the Addison-Wesley, Exam Registering your products can unlock

Cram, Prentice Hall, Que, and the following benefits:
Sams products you own to unlock • Access to supplemental content,
great benefits. including bonus chapters,
source code, or project files.
To begin the registration process,
• A coupon to be used on your
simply go to informit.com/register
next purchase.
to sign in or create an account.
You will then be prompted to enter Registration benefits vary by product.
the 10- or 13-digit ISBN that appears Benefits will be listed on your Account
on the back cover of your product. page under Registered Products.

About InformIT — THE TRUSTED TECHNOLOGY LEARNING SOURCE

INFORMIT IS HOME TO THE LEADING TECHNOLOGY PUBLISHING IMPRINTS

Addison-Wesley Professional, Cisco Press, Exam Cram, IBM Press, Prentice Hall
Professional, Que, and Sams. Here you will gain access to quality and trusted content and
resources from the authors, creators, innovators, and leaders of technology. Whether you’re
looking for a book on a new technology, a helpful article, timely newsletters, or access to
the Safari Books Online digital library, InformIT has a solution for you.

SAFARI BOOKS ONLINE

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Vsphere Esxi 801 Installation Setup Guide
No ratings yet
Vsphere Esxi 801 Installation Setup Guide
238 pages
Sanet - CD B0713PYBK7
No ratings yet
Sanet - CD B0713PYBK7
206 pages
Solaris 10 Network Administration - Student Guide
100% (1)
Solaris 10 Network Administration - Student Guide
628 pages
Letu Da Notes-Compiled
No ratings yet
Letu Da Notes-Compiled
438 pages
Solaris 10 Scsa Part 2
0% (1)
Solaris 10 Scsa Part 2
213 pages
Veritas Cluster 2.0
No ratings yet
Veritas Cluster 2.0
383 pages
Solaris ZFS Administration Guide: Sun Microsystems, Inc. 4150 Network Circle Santa Clara, CA 95054 U.S.A
No ratings yet
Solaris ZFS Administration Guide: Sun Microsystems, Inc. 4150 Network Circle Santa Clara, CA 95054 U.S.A
262 pages
Solaris 10 For Experienced System Administrators Student Guide
No ratings yet
Solaris 10 For Experienced System Administrators Student Guide
434 pages
Suse Manager Administration Guide
No ratings yet
Suse Manager Administration Guide
230 pages
Solaris Solutions For System A
100% (1)
Solaris Solutions For System A
625 pages
Oracle Data Modeling and Relational Database Design - 9-2010 - Volume II - Student Guide-D56497GC10-Edition 1.0-September 2010-D67008 X
No ratings yet
Oracle Data Modeling and Relational Database Design - 9-2010 - Volume II - Student Guide-D56497GC10-Edition 1.0-September 2010-D67008 X
242 pages
Window Server 2016
No ratings yet
Window Server 2016
292 pages
EFS Systems On A Linux A Linux Base: Getting Started G Started
100% (1)
EFS Systems On A Linux A Linux Base: Getting Started G Started
120 pages
Architectural Research Report Writing Format
67% (3)
Architectural Research Report Writing Format
5 pages
Sun Cluster 3.1 Administration ES-338
No ratings yet
Sun Cluster 3.1 Administration ES-338
540 pages
Oracle Solaris 11-Update
No ratings yet
Oracle Solaris 11-Update
103 pages
Storage Management (ZFS)
No ratings yet
Storage Management (ZFS)
12 pages
SUSE Linux Enterprise Server Security Guide
No ratings yet
SUSE Linux Enterprise Server Security Guide
472 pages
AIX Performance Cmds
No ratings yet
AIX Performance Cmds
4 pages
Srimaan: PG-TRB
No ratings yet
Srimaan: PG-TRB
24 pages
Bhramam Question
No ratings yet
Bhramam Question
300 pages
Solaris/Unix Training
No ratings yet
Solaris/Unix Training
119 pages
Less Known Solaris Features
No ratings yet
Less Known Solaris Features
394 pages
Avnr 1
No ratings yet
Avnr 1
147 pages
Red Hat Enterprise Linux-8-8.3 Release Notes-En-Us
No ratings yet
Red Hat Enterprise Linux-8-8.3 Release Notes-En-Us
154 pages
Ibm Linux Ibm Z 2
No ratings yet
Ibm Linux Ibm Z 2
264 pages
Scsa Part 1 Kamal PDF
No ratings yet
Scsa Part 1 Kamal PDF
216 pages
Ironstream Configuration and Users Guide v2.1 PDF
50% (2)
Ironstream Configuration and Users Guide v2.1 PDF
554 pages
1Z0-821 Oracle Solaris 11 System Administration PDF
No ratings yet
1Z0-821 Oracle Solaris 11 System Administration PDF
6 pages
SAN
100% (1)
SAN
81 pages
Less Known Solaris Features
No ratings yet
Less Known Solaris Features
404 pages
Red Hat Enterprise Linux-8-8.4 Release Notes-En-Us
No ratings yet
Red Hat Enterprise Linux-8-8.4 Release Notes-En-Us
172 pages
Solaris 11 Exam
No ratings yet
Solaris 11 Exam
29 pages
Docker From Zero To Hero Your DevOps Kickstart Build, Deploy, and Manage Containers With Practical Exercises. Perfect For... (Parvin, R.) (Z-Library)
No ratings yet
Docker From Zero To Hero Your DevOps Kickstart Build, Deploy, and Manage Containers With Practical Exercises. Perfect For... (Parvin, R.) (Z-Library)
169 pages
Grade 3 PPT - Q3 - W6 - TIMBRE
No ratings yet
Grade 3 PPT - Q3 - W6 - TIMBRE
47 pages
RHEL 8.5 - Working With DNS in Identity Management
No ratings yet
RHEL 8.5 - Working With DNS in Identity Management
92 pages
10 Years of Linux Security
No ratings yet
10 Years of Linux Security
38 pages
PHP Notes
No ratings yet
PHP Notes
111 pages
Playboy Magazine Edition Croatia January 2016 - Pamela Anderson - Free Poster Calendar 2016 - Plastic Wrap Unopened
No ratings yet
Playboy Magazine Edition Croatia January 2016 - Pamela Anderson - Free Poster Calendar 2016 - Plastic Wrap Unopened
5 pages
Red Hat Enterprise Linux-9-9.1 Release Notes-En-Us
No ratings yet
Red Hat Enterprise Linux-9-9.1 Release Notes-En-Us
158 pages
DAA NOTES UNIT 1 (Design and Analysis of Algorithm)
No ratings yet
DAA NOTES UNIT 1 (Design and Analysis of Algorithm)
18 pages
66-Veritas Cluster Server 6.0 For UNIX - Install and Configure Appendixes B and C
No ratings yet
66-Veritas Cluster Server 6.0 For UNIX - Install and Configure Appendixes B and C
484 pages
HP 3PAR HP-UX Implementation Guide PDF
No ratings yet
HP 3PAR HP-UX Implementation Guide PDF
51 pages
SPARC T7 Servers 2015 Hardware Installation Specialist Online Assessment...
100% (4)
SPARC T7 Servers 2015 Hardware Installation Specialist Online Assessment...
5 pages
Solaris System Performance Management - SA-400 - Student Guide
No ratings yet
Solaris System Performance Management - SA-400 - Student Guide
700 pages
Kubernetes Troubleshooting
No ratings yet
Kubernetes Troubleshooting
16 pages
Flash For Dummies
No ratings yet
Flash For Dummies
53 pages
SPARC T7 Servers Installation and Maintenance Assessment
100% (1)
SPARC T7 Servers Installation and Maintenance Assessment
4 pages
Vsphere Automation SDK 801 Programming Guide
No ratings yet
Vsphere Automation SDK 801 Programming Guide
286 pages
Linux Lab
No ratings yet
Linux Lab
37 pages
What Is NFS?: - NFS Is A Protocol For A Distributed Filesystem
No ratings yet
What Is NFS?: - NFS Is A Protocol For A Distributed Filesystem
69 pages
LDoms Configuration
No ratings yet
LDoms Configuration
34 pages
Red Hat Enterprise Linux-7-Upgrading From Rhel 6 To Rhel 7-En-Us
No ratings yet
Red Hat Enterprise Linux-7-Upgrading From Rhel 6 To Rhel 7-En-Us
26 pages
Network File System
No ratings yet
Network File System
29 pages
ES445 Sun Cluster 3 2 Advanced Administration SG PDF
No ratings yet
ES445 Sun Cluster 3 2 Advanced Administration SG PDF
402 pages
Distributed File Systems: - Objectives - Contents
No ratings yet
Distributed File Systems: - Objectives - Contents
34 pages
Bare Machine Recovery For IBM Spectrum Protect TBMR For Linux
No ratings yet
Bare Machine Recovery For IBM Spectrum Protect TBMR For Linux
31 pages
1Z0-821 Oracle Solaris 11 System Administration
100% (1)
1Z0-821 Oracle Solaris 11 System Administration
6 pages
HP Service Guard Cluster
No ratings yet
HP Service Guard Cluster
2 pages
Boot Pratition Increase
100% (1)
Boot Pratition Increase
2 pages
04-A-Linux Troubleshooting Tools
No ratings yet
04-A-Linux Troubleshooting Tools
18 pages
QUESTION BANK - (Laplace and Fourier Transform - CUTM1002)
No ratings yet
QUESTION BANK - (Laplace and Fourier Transform - CUTM1002)
7 pages
Basic VERITAS Volume Replicator™ 4.1 Commands: Veritas Education Quick Reference
No ratings yet
Basic VERITAS Volume Replicator™ 4.1 Commands: Veritas Education Quick Reference
2 pages
Sym Netbackup PDF
No ratings yet
Sym Netbackup PDF
9 pages
Rosevil DLL Sample December 5 - 9
No ratings yet
Rosevil DLL Sample December 5 - 9
13 pages
ARC List
No ratings yet
ARC List
4 pages
Solaris 10 Quick Reference Basic
100% (7)
Solaris 10 Quick Reference Basic
3 pages
Engineering Science For BS MBS3211: 3. Electricity
No ratings yet
Engineering Science For BS MBS3211: 3. Electricity
64 pages
Assessment Oracle SPARC M5-32 Server Overview (WBT)
No ratings yet
Assessment Oracle SPARC M5-32 Server Overview (WBT)
2 pages
Oracle 1z0-485
No ratings yet
Oracle 1z0-485
46 pages
Sun Explorer User's Guide PDF
No ratings yet
Sun Explorer User's Guide PDF
100 pages
How To Enable SSH As Root On AWS Ubuntu Instance
No ratings yet
How To Enable SSH As Root On AWS Ubuntu Instance
2 pages
SSH Command - Usage, Options, Configuration
No ratings yet
SSH Command - Usage, Options, Configuration
5 pages
Oracle 1z0-027
No ratings yet
Oracle 1z0-027
55 pages
VXDMP Corrupt
No ratings yet
VXDMP Corrupt
5 pages
Concept Paper 3 Eng 10 New
No ratings yet
Concept Paper 3 Eng 10 New
6 pages
Financial Accounting and Reporting Quiz Bowl - Quizizz
No ratings yet
Financial Accounting and Reporting Quiz Bowl - Quizizz
13 pages
Core Services" Help Desk Support Consultant Assessment
No ratings yet
Core Services" Help Desk Support Consultant Assessment
6 pages
Ds LEIAN DCDU 12B Specification
No ratings yet
Ds LEIAN DCDU 12B Specification
9 pages
AFM ER308 Afm Er308L
No ratings yet
AFM ER308 Afm Er308L
9 pages
TDM Chapter 2 Handouts 2020 KKA
No ratings yet
TDM Chapter 2 Handouts 2020 KKA
20 pages
Material Safety Data Sheet: Ephedrine Hydrochloride
No ratings yet
Material Safety Data Sheet: Ephedrine Hydrochloride
6 pages
Anc Assessment
No ratings yet
Anc Assessment
6 pages
Oracle Solaris 11.3 Network Administration Cheatsheet
No ratings yet
Oracle Solaris 11.3 Network Administration Cheatsheet
5 pages
T7 Astro Camera Astronomy Planetary Quick Guide
No ratings yet
T7 Astro Camera Astronomy Planetary Quick Guide
20 pages
SCHEME HND 1 General Computer II 2019-2020
No ratings yet
SCHEME HND 1 General Computer II 2019-2020
5 pages
Banana Fibre Extracting Project
No ratings yet
Banana Fibre Extracting Project
2 pages
Bon - Gas: Reduced Bore Ball Valve For Fuel Gas
No ratings yet
Bon - Gas: Reduced Bore Ball Valve For Fuel Gas
7 pages
ENVI Classic Tutorial: Target Detection
No ratings yet
ENVI Classic Tutorial: Target Detection
18 pages
The Kardashev Scale Measuring Civilizational Advancement
No ratings yet
The Kardashev Scale Measuring Civilizational Advancement
2 pages
Chapter 3
No ratings yet
Chapter 3
10 pages
Thermodynamics Chemistry Difficult NEET Practice Questions, MCQS, Past Year Questions (PYQs), NCERT Questions, Question Bank, CL
No ratings yet
Thermodynamics Chemistry Difficult NEET Practice Questions, MCQS, Past Year Questions (PYQs), NCERT Questions, Question Bank, CL
1 page
Chemical Identity: Material Safety Data Sheet Gasoline/Petrol
No ratings yet
Chemical Identity: Material Safety Data Sheet Gasoline/Petrol
4 pages
Styrofoam Cup: Experiment 3: Energy in Thermal System Objectives
No ratings yet
Styrofoam Cup: Experiment 3: Energy in Thermal System Objectives
3 pages
RM-898 Research Methodology For DM-5, CE&M-4, SE-3 and TN-2 Students Only
No ratings yet
RM-898 Research Methodology For DM-5, CE&M-4, SE-3 and TN-2 Students Only
3 pages
Joan Batayo Profile
No ratings yet
Joan Batayo Profile
2 pages
1st Sem Result
No ratings yet
1st Sem Result
1 page
Mastering Active Directory
From Everand
Mastering Active Directory
VICTOR P HENDERSON
No ratings yet
Neuroscience of Mind Empowerment and Metacognition: Epigenetics, Neuroplasticity, Meditation, and Music Therapy
From Everand
Neuroscience of Mind Empowerment and Metacognition: Epigenetics, Neuroplasticity, Meditation, and Music Therapy
Prof. Anees Akhtar
No ratings yet
BlockChain for Beginners
From Everand
BlockChain for Beginners
Matthew Smith
No ratings yet
The Amazing Biological Revolution and The Amazing New Health Care
From Everand
The Amazing Biological Revolution and The Amazing New Health Care
Bertil Lindmark
No ratings yet
AWS in ACTION Part -1: Real-world Solutions for Cloud Professionals
From Everand
AWS in ACTION Part -1: Real-world Solutions for Cloud Professionals
Poonam Devi
No ratings yet
Software Patterns Made Easy
From Everand
Software Patterns Made Easy
Justice Nanhou
No ratings yet

Solaris 10 ZFSEssentials

Uploaded by

Solaris 10 ZFSEssentials

Uploaded by

Table of Contents

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Sun Microsystems Press

Upper Saddle River, NJ • Boston • Indianapolis • San Francisco

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Copyright © 2010 Sun Microsystems, Inc.

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Chapter 1 Introducing ZFS File Systems 1

Chapter 2 Managing Storage Pools 9

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Chapter 3 Installing and Booting a ZFS Root File System 25

Chapter 4 Managing ZFS Home Directories 47

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Chapter 5 Exploring Zpool Advanced Concepts 61

Chapter 6 Managing Solaris CIFS Server and Client 75

Chapter 7 Using Time Slider 83

Chapter 8 Creating a ZFS Lab in a Box 93

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Books in the Solaris System Administration Series

Solaris™ 10 System Administration Essentials

Solaris™ 10 Security Essentials

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

well as to people who are perhaps already serving as administrators in companies

In June 2005, Sun Microsystems introduced OpenSolaris, a fully functional Solaris

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1.1 Overview of ZFS

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

ZFS provides the following redundancy conﬁgurations:

RAID-Z is similar to RAID-5 but uses a variable-stripe width to eliminate the

1.1.1 Advantages of Using ZFS

Data inconsistency detection and recovery are provided by ZFS redundancy

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1.1.2 A Top-Level View of ZFS

Figure 1.1 The base of ZFS: the pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 1.2 The ZFS pool with three ZFS datasets

Figure 1.3 ZFS pool with a single vdev

Figure 1.4 A vdev with a mirrored array supporting a ZFS pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 1.5 Concatenation of a second mirrored vdev to the ZFS pool

Figure 1.6 File system expanding in the larger ZFS pool

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1.2 Fast and Simple Storage

$ time pfexec zpool create mpool mirror c5t0d0 c5t1d0

$ zpool status mpool

NAME STATE READ WRITE CKSUM

errors: No known data errors

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

1.3 ZFS Commands

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

2.1 ZFS Pool Concepts

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

volume presented by another volume manager such as Solaris Volume Manager

Dynamic stripe: A dynamic stripe is a nonredundant conﬁguration of a sim-

Mirrored conﬁguration beyond a three-way mirror should not be used. Think

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Once a normal (dynamic stripe) vdev is added to a ZFS pool, it cannot be

2.2 Creating a Dynamic Stripe

1 # zpool create dstripe c5t0d0 c5t1d0

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 2.1 Dynamic stripe pool with two vdevs

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

2.3 Creating a Pool with Mirrored Devices

Solaris™ 10 ZFS Essentials. Solaris™ 10 ZFS Essentials, ISBN: 9780137049639

Figure 2.2 Pool mpool with a mirrored vdev

1. # zpool add mpool mirror c5t4d0 c5t5d0