PK8YF A40 Instructorhints PDF

V4.
cover
Front cover
AIX 6 System
Administration I:
Implementation
(Course code AU14)
Instructor Exercises Guide

with hints
ERC 12.0
IBM certified course material

Instructor Exercises Guide with hints
Trademarks
The reader should recognize that the following terms, which appear in the content of this
training document, are official trademarks of IBM or other companies:
IBM is a registered trademark of International Business Machines Corporation.
The following are trademarks of International Business Machines Corporation in the United
States, or other countries, or both:
AIX AIX 5L AIX 6 (logo)
AS/400 CICS/6000 DB2
Domino HACMP iSeries
Language Environment Lotus Magstar
MVS Network Station Notes
OS/2 POWER POWER4
POWER5 POWER6 PowerPC
pSeries Requisite RISC System/6000
RS/6000 SecureWay SP
System p System/370 System Storage
Tivoli 400
VMware and the VMware "boxes" logo and design, Virtual SMP and VMotion are
registered trademarks or trademarks (the "Marks") of VMware, Inc. in the United States
and/or other jurisdictions.
Alerts is a registered trademark of Alphablox Corporation in the United States, other
countries, or both.
PS/2 is a trademark or registered trademark of Lenovo in the United States, other
countries, or both.
PostScript is either a registered trademark or a trademark of Adobe Systems Incorporated
in the United States, and/or other countries.
Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the
United States, other countries, or both.
Microsoft and Windows are trademarks of Microsoft Corporation in the United States, other
countries, or both.
February 2008 edition
The information contained in this document has not been submitted to any formal IBM test and is distributed on an as is basis without
any warranty either express or implied. The use of this information or the implementation of any of these techniques is a customer
responsibility and depends on the customers ability to evaluate and integrate them into the customers operational environment. While
each item may have been reviewed by IBM for accuracy in a specific situation, there is no guarantee that the same or similar results will
result elsewhere. Customers attempting to adapt these techniques to their own environments do so at their own risk.
Copyright International Business Machines Corporation 1997, 2008. All rights reserved.
This document may not be reproduced in whole or in part without the prior written permission of IBM.
Note to U.S. Government Users Documentation related to restricted rights Use, duplication or disclosure is subject to restrictions
set forth in GSA ADP Schedule Contract with IBM Corp.
Intel is a trademark or registered trademark of Intel Corporation or its subsidiaries in the

United States and other countries.
UNIX is a registered trademark of The Open Group in the United States and other
countries.
Linux is a registered trademark of Linus Torvalds in the United States, other countries, or
both.
Other company, product, or service names may be trademarks or service marks of others.
iii AIX System Administration I Copyright IBM Corp. 1997, 2008

Course materials may not be reproduced in whole or in part
without the prior written permission of IBM.
iv AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
TOC Contents
Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Instructor exercises overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
Exercises configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
Exercises description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv
Exercise 1. root login methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

Part 1 - Direct logins to root . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Part 2 - Using the su command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-5
Exercise 2. Using SMIT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

Part 1 - Using SMIT (ASCII interface) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-3
Part 2 - (Optional) Using SMIT (Motif interface). . . . . . . . . . . . . . . . . . . . . . . . 2-8
Exercise 3. System startup and shutdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1

Part 1 - Multi-User mode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-3
Part 2 - Boot using System Management Services . . . . . . . . . . . . . . . . . . . . . 3-4
Part 3 - System Resource Controller (SRC) . . . . . . . . . . . . . . . . . . . . . . . . . . 3-7
Exercise 4. AIX software installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

Part 1 - List and install AIX software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-3
Part 2 - Working with AIX Fixes (Optional) . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
Exercise 5. AIX 6.1 installation and configuration . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1
Exercise 6. Web-based System Manager . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

Part 1: Use the Web-based System Manager . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
Part 2 (Optional): Configure the Web-based System Manager for AIX client access
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-4
Part 3 (Optional): Configure the Web-based System Manager for remote client
access . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6
Exercise 7. Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

Part 1 - List device configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-3
Part 2 - List and change system parameters . . . . . . . . . . . . . . . . . . . . . . . . . . 7-5
Part 3 - Removing and redefining a network adapter device . . . . . . . . . . . . . . 7-6
Part 4 - Changing the status of a disk drive device . . . . . . . . . . . . . . . . . . . . . 7-8
Part 5 - Configure a tape device (optional) . . . . . . . . . . . . . . . . . . . . . . . . . . 7-10
Exercise 8. LVM commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1
Copyright IBM Corp. 1997, 2008 Contents v

Exercise 9. Working with LVM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9-1

Part 1 - Adding and removing a disk from a volume group . . . . . . . . . . . . . . . . 9-3
Part 2 - Creating and removing a volume group . . . . . . . . . . . . . . . . . . . . . . . . 9-5
Part 3 - Exploring your storage environment . . . . . . . . . . . . . . . . . . . . . . . . . . 9-9
Part 4 - Adding a volume group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-13
Part 5 - Adding a logical volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9-14
Exercise 10. Working with file systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-1

Part 1 - Inodes and NBPI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-4
Part 2 - Creating a journaled file system . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-6
Part 3 - Changing the file system size . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-10
Part 4 - Reducing the size of a file system . . . . . . . . . . . . . . . . . . . . . . . . . . 10-13
Part 5 - Removing a file system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-15
Part 6 - Working with mirrors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10-17
Exercise 11. Managing file systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-1

Part 1 - Determining file system usage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-3
Part 2 - Using fragments for disk usage efficiency . . . . . . . . . . . . . . . . . . . . . 11-5
Part 3 - Using JFS compression. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-8
Part 4 - Fixing file system problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-11
Part 5 - Optional exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11-13
Exercise 12. Paging space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-1
Exercise 13. Using backup and restore . . . . . . . . . . . . . . . . . 13-1

Part 1 - Use SMIT to backup a non-rootvg volume group . . ....... .. . . . . . 13-3
Part 2 - Prepare file system for backup/restore . . . . . . . . . ....... .. . . . . . 13-6
Part 3 - Backup and restore by inode (file system) . . . . . . . ....... .. . . . . . 13-8
Part 4 - Backup and restore by name . . . . . . . . . . . . . . . . . ....... .. . . . . 13-15
Part 5 - (Optional) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....... .. . . . . 13-19
Exercise 14. (Optional) Using tar and cpio . . . . . . . . . . . . . . . . 14-1

Part 1 - Using tar. . . . . . . . . . . . . . . . . . . . . . . . . 14-2
Part 2 - Using cpio (Optional) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14-4
Exercise 15. Security files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15-1

Part 1 - Security control files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-2
Part 2 - SUID and sticky bit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15-4
Exercise 16. User administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16-1

Part 1: User administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-3
Part 2 - Group administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-8
Part 3 - Customizing the default .profile file . . . . . . . . . . . . . . . . . . . . . . . . . 16-12
Part 4 - Removing users . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-13
Part 5 - Communicating with users . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-14
vi AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
TOC Part 6 - Examine the security set up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-16

Part 7 - Customizing the login herald . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-17
Exercise 17. Scheduling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-1

Part 1 - Using at . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-3
Part 2 - Using batch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7
Part 3 - Using crontab Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-7
Part 4 - Optional . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-9
Exercise 18. Printers and queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-1

Part 1 - Add a local printer and a queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-3
Part 2 - Add a file device queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-6
Part 3 - Check the queue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-7
Part 4 - Change the characteristics of a queue . . . . . . . . . . . . . . . . . . . . . . . 18-9
Part 5 - Manage jobs in queues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-12
Part 6 - Add a remote print queue (optional) . . . . . . . . . . . . . . . . . . . . . . . . 18-16
Exercise 19. Networking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-1

Part 1 - Deconfigure TCP/IP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-3
Part 2 - Configure TCP/IP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-6
Part 3 - Testing the configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-7
Part 4 - Using telnet. . . . . . . . . . . . . . . . . . . . . . . . 19-8
Part 5 - Using ftp . . . . . . . . . . . . . . . . . . . . . . . . . 19-9
Appendix A. Optional exercise: AIX 5L V5.3 installation . . . . . . . . . . . . . . . . . . . A-1
Appendix B. Serial devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1

Part 1 - Verify a terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-4
Part 2 - Remove a terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-5
Part 3 - Add a terminal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-7
Part 4 - Check terminal communications settings . . . . . . . . . . . . . . . . . . . . . . B-8
Part 5 - Change terminal characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-10
Part 6 - Fixing a hung terminal (optional). . . . . . . . . . . . . . . . . . . . . . . . . . . . B-11
Appendix C. Optional exercise: The System V print subsystem. . . . . . . . . . . . . C-1

Part 1 - Switch print subsystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-3
Part 2 - Add a System V printer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-4
Part 3 - Submit print jobs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C-7
Part 4 - (Optional) Configure a remote print server and client . . . . . . . . . . . . . C-9
Copyright IBM Corp. 1997, 2008 Contents vii

viii AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
TMK Trademarks
The reader should recognize that the following terms, which appear in the content of this
training document, are official trademarks of IBM or other companies:
IBM is a registered trademark of International Business Machines Corporation.
The following are trademarks of International Business Machines Corporation in the United
States, or other countries, or both:
AIX AIX 5L AIX 6 (logo)
AS/400 CICS/6000 DB2
Domino HACMP iSeries
Language Environment Lotus Magstar
MVS Network Station Notes
OS/2 POWER POWER4
POWER5 POWER6 PowerPC
pSeries Requisite RISC System/6000
RS/6000 SecureWay SP
System p System/370 System Storage
Tivoli 400
VMware and the VMware "boxes" logo and design, Virtual SMP and VMotion are
registered trademarks or trademarks (the "Marks") of VMware, Inc. in the United States
and/or other jurisdictions.
Alerts is a registered trademark of Alphablox Corporation in the United States, other
countries, or both.
PS/2 is a trademark or registered trademark of Lenovo in the United States, other
countries, or both.
PostScript is either a registered trademark or a trademark of Adobe Systems Incorporated
in the United States, and/or other countries.
Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the
United States, other countries, or both.
Microsoft and Windows are trademarks of Microsoft Corporation in the United States, other
countries, or both.
Intel is a trademark or registered trademark of Intel Corporation or its subsidiaries in the
United States and other countries.
UNIX is a registered trademark of The Open Group in the United States and other
countries.
Linux is a registered trademark of Linus Torvalds in the United States, other countries, or
both.
Other company, product, or service names may be trademarks or service marks of others.
Copyright IBM Corp. 1997, 2008 Trademarks ix

x AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
pref Instructor exercises overview

The objective of the AIX V6.1 System Administration exercises is to
have students understand and successfully perform the types of
activities a new AIX system administrator may encounter on a
day-to-day basis.
There are several connected exercises of which one or more must be
completed in order to complete a following exercise. The following is a
list of these exercise dependencies:
Exercise 11 - Managing File Systems, as well as Exercise 13 -
Backup and Restore, depend on the completion of Exercises 9 and
10, Working with LVM and Working with File Systems respectively.
Before teaching this course, read the Overview Exercise
Description, included in this guide, as well as the Student Exercises
book. This section explains the format of the different exercise
sections. This must be covered with the students as an overall
exercise introduction before the first exercise is started.
It is also imperative that you read and understand the Common
Student Problems and Known Hardware/Software Problems
sections provided with each exercise. These sections alert you to
known problems or common mistakes that are inherent to the
exercise.
Be sure you understand the equipment configuration for your teach
site before introducing the exercises. The password for root should be
set to ibmaix. During the first exercise students will create five users
taking the default information for their creation. IT IS MANDATORY
THAT STUDENTS HAVE ACCESS TO THE root PASSWORD as all
exercises require at some point for students to be the root user.
Supplemental exercise scripts have been created and placed in the
/home/workshop directory. Check for the existence of this directory
and its files. If /home/workshop does not exist or is empty, restore the
scripts from the instructor exercise diskette that is provided with the
instructor materials. Without these scripts, the following exercises
cannot be done:
Exercise 11 sections: Optional exercises, Using Fragments For
Disk Usage Efficiency, Using JFS Compression
The following is a synopsis of what these scripts do:
Exercise 18, lab6 - Creates problems with the printing subsystem.
Exercise 11, mkfile - Creates a file called 512-byte file whose size
is 512 bytes so it occupies only one 512-byte fragment.
Copyright IBM Corp. 1997, 2008 Instructor exercises overview xi

fragcopy - Makes copies of the 512-byte file to test the number of

512-byte files that can be stored in each file system. As it
processes, a count will be displayed showing the number of files
that have been successfully copied.
bigfile - It is a 1 MB file that is copied to each file system used to
assist in measuring performance.
Due to the variety of classrooms where this class may be offered, the
equipment available in any one classroom may be different from
another. It may be that the equipment is remote, and only terminals
that can remotely access the systems are available. There should be
two students per system. To alleviate the potential problems that can
be caused by terminal differences at various sites, directions are
provided for both ASCII interfaces and graphics interfaces. If you have
loaded the systems using mksysb images, the same printer support
that is found on a distribution tape or CD has been loaded onto the
mksysb image. Thus, you should have no problems providing the
necessary printer driver support for whatever printers are attached to
the AIX systems (assuming you are not working in a remote
environment).
There should be an extra unconfigured Ethernet adapter (either virtual
or physical) that will be used during Exercise 7, Devices, to give
students an opportunity to remove and discover a device. It is not
intended that the interface for this adapter be configured with an IP
address and so there is no need for it to be cabled to a hub or switch.
Local environments with physical non-partitioned lab systems can use
a CD-ROM as a substitute, but the lab instructions are written to use
an Ethernet adapter.
The Exercise Review/Wrap-up section at the end of each exercise in
this guide is meant to be conducted by you as an exercise review at
the end of each exercise session and before the beginning of the next
unit. The students do not have this section in their exercise notebook.
These are suggested review questions. To add to the exercise review,
keep track of questions and common problems students had during
the exercise. Use the board to cover these common problems. Simply
state that these are a few of the common problems seen in all classes.
Do not single out any one team or student. Keep it a group discussion.
xii AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
pref Exercises configuration

The following are the required software and hardware necessary to
teach AU14 (note that the official document on the lab setup is the Lab
Setup Guide for the course; what follows is a guide to what you should
already find on the platform after the instructions in that document are
complete):
1. Software: Each AIX system must be installed with AIX 5L V5.3 with
the following filesets and AIX 5L V5.3 TL6 for POWER6 servers:
- Java 1.4.1 applet filesets (java141.ext on Volume 3 of AIX
Installation CDs)
The AIX 5L V5.3 Web-based services, such as remote, wsm are
installed during the lab setup. The students do not install it, but
configure it. See the Lab Setup Guide for this course for more
information.
Also, check to be sure that Mozilla Foxfire is installed and functional on
the system (in an Xwindow environment execute: mozilla &).
TCP/IP needs to be configured in support of remote access and the
AIX 6.1 Migration installation from a NIM server, unless using
non-partitioned local lab systems.
Use the provided AU140 lab files CD to create the /home/workshop
directory. This directory contains numerous files used throughout the
machine exercises.
The following software packages should have been placed in the
/usr/sys/inst.images directory during setup. The filesets in this
package are used in Exercise 17, Printers and Queues.
printers
The following filesets should also be stored in the
/usr/sys/inst.images directory. These filesets are used in Exercise 4,
AIX Software Installation.
bos.games
bos.dosutil
2. Hardware:
A. Minimum:
43P-Model 140 and MCA machines are not supported by AIX
V6.1. You need at a POWER5 or POWER6 server (a POWER4
server would work, except that the lab HMC instructions
assume using HMCv7 which do not support POWER4).
Copyright IBM Corp. 1997, 2008 Exercises configuration xiii

For a class size of 24, there should be at least 12 AIX systems.

These system could be physical, but it is more likely that they will
be logically partitioned. Each system (whether physical or logical)
should have the following:
256 MB memory
Two hard drives; one with a minimum of 2.2 GB
CD-ROM (for physical systems)
Graphics terminal (for physical systems)
Have all systems connected via Ethernet
Note: If you have systems with disks less than 2.2 GB, the AIX
5L V5.3 and AIX 6.1 systems have to be loaded using both
drives. Thus, students are not able to do exercises which
require two disks. The preferred configuration is two disks at
least 2.2 GB in size.
More than two students per system is not recommended. Check to be
sure the /home/workshop directory has been set up on the student
systems and that the following files are in that directory:
mkfile
fragcopy
lab5 (only used for the serial devices exercise in the appendix)
lab5.1 (only used for the serial devices exercise in the appendix)
If the files are not there, you need to load them from the diskette
that was provided with your instructor materials.
For detailed lab set up instructions, refer to the Lab Setup Guide for
this course. These instructions also include how to set up X
Terminals and enable the AIX V6.1 online documentation.
xiv AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
pref Exercises description

Exercise Instructions This section contains what it is you are to
accomplish. There are no definitive details on how to perform the
tasks. You are given the opportunity to work through the exercise
given what you learned in the unit presentation, utilizing the unit
Student Notebook, your past experience and maybe a little intuition.
Exercise Instructions With Hints This section is an exact
duplicate of the Exercise Instructions section except that in addition,
specific details and/or hints are provided to help step you through the
exercise. A combination of using the Instructions section along with
Instructions With Hints section can make for a rewarding combination
providing you with no hints when you don't want them and hints when
you need them.
Optional Exercises This section gives you additional exercises to
perform relating to the unit of discussion. It is strictly optional and
should be performed when you have completed the required
exercises. The required exercises pertain to the most pertinent
information provided in the unit. This section may help round out the
hands-on experience for a related unit.
Copyright IBM Corp. 1997, 2008 Exercises description xv

Text highlighting
The following text highlighting conventions are used throughout this book:
Bold Identifies file names, file paths, directories, user names and
principals.
Italics Identifies links to Web sites, publication titles, and is used
where the word or phrase is meant to stand out from the
surrounding text.
Monospace Identifies attributes, variables, file listings, SMIT menus, code
examples of text similar to what you might see displayed,
examples of portions of program code similar to what you might
write as a programmer, and messages from the system.
Monospace bold Identifies commands, daemons, menu paths and what the user
would enter in examples of commands and SMIT menus.
<text> The text between the < and > symbols identifies information the
user must supply. The text may be normal highlighting, bold or
monospace, or monospace bold depending on the context.
xvi AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
EXempty Exercise 1. root login methods

(with hints)
Estimated time
00:10
What this exercise is about

In this exercise, you look at several ways to invoke the privileges of the
root user.
What you should be able to do

After completing this exercise, students should be able to:
Directly log in as root
Use the su command to log in as root
Introduction
Typically the system administrator should not use root as their normal
login account. Instead, they should use the su command to become
root for a short period of time to do administrative tasks.
Common Student Problems

None.
Known Hardware/Software Problems

None.
Copyright IBM Corp. 1997, 2008 Exercise 1. root login methods 1-1
Exercise Instructions with Hints
Preface
Two versions of these instructions are available; one with hints and one without. You can
use either version to complete this exercise. Also, please dont hesitate to ask the instructor
if you have questions.
All exercises of this chapter depend on the availability of specific equipment in your
classroom.
If your lab environment uses remote System p platforms, the instructor provide you with
instructions for accessing the remote environment. For example, this may involve using a
Web browser to access a Web server which provides access to a virtual private network
(VPN) or it may involve using a toll that extends a remote VPN to your desktop machine.
Once you have access to the remote environment you can choose a tool to connect to your
assigned machine. Once again, your instructor provides you with the addressing (IP
address or hostname) and instructions for accessing the remote machine. Your network
utility may be a telnet client, an X-Windows server, or a VNC client. Whatever tool you are
using, at various points in the labs, you may need to re-launch your remote access tool to
obtain a new login prompt.
If you are using an ascii based network client, such as PuTTY, be sure to specify the telnet
protocol in addition to providing the IP address of your assigned lab system. You may also
want to customize the terminal emulation, such as configuring the keyboard mapping to
generate <ctrl-H> when the backspace key is pressed. You might also wish to adjust the
window appearance characteristics such as the font size. If you did not customize before
launching the terminal emulation, you can make the changes later by right clicking on the
title bar and selecting change settings.
If you are using a VNC viewer (to access CDE environment), you need to enter (when
prompted) the IP address of your assigned lab system, immediately followed by :#, where
# is the port number for the vncserver. For example: 192.168.22.5:1.
The output shown in the answers is an example. Your output and answers based on the
output may be different.
All hints are marked with a >> sign.
Part 1 - Direct logins to root

__ 1. Establish connectivity to the remote lab environment. (If you are using local
non-partitioned machines, you only need to turn on the local attached terminal and
skip to step 3).
Your instructor should provide details on how to do this.
__ 2. Select and launch a network utility to access your assigned lab system.
1-2 AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
EXempty See the discussion in the above preface. Your instructor provides you with more
specific instructions such as the IP address of your assigned system.
__ 3. Log in as the user root. The password is ibmaix.
At the username prompt, enter: root
At the password prompt, enter root's password: ibmaix (You do not see the
password appear on the screen)
(you may only be prompted for a password - assume it is roots password)
__ 4. Open a terminal. (if necessary)
If you are using a CDE graphic interface:
On the tool bar at the bottom, click the small triangle above the pencil and paper
icon (fourth icon from the left). This will open a drawer of icons.
Click ONCE on the Terminal icon. This should bring up a terminal window. All of
the commands for this exercise should be typed at the command line in this
window. You may wish to resize the terminal window or adjust the font size being
used.
__ 5. Determine your current directory and your home directory.

Current directory: ________/__
Home directory: ________/__
# pwd
/
# echo $HOME
/
__ 6. To verify that you do, in fact, have root privileges, cat out the file
/etc/security/passwd. This file holds the encrypted passwords for the users on your
system. Only root can look at this file. More details on this file are mentioned later in
the course. Did you see the contents of the file? __________ Yes.
# cat /etc/security/passwd
__ 7. Log out of the system.

If you are using a CDE graphical interface:
On the blank area of the screen (not in a window), press and hold down the right
mouse button. A drop-down menu should appear. Continue holding the right
mouse button and drag the cursor to the bottom of the menu to the Log out...
selection. Then, release the mouse button.
Click Continue logout. A new log in box should appear (if using a local
terminal).
Otherwise, type exit to logout of the telnet window.

V4.1
EXempty Part 2 - Using the su command

__ 8. Log into the system using the login name team01 and the password team01.
At the username prompt, enter: team01
At the password prompt, enter password: team01 (You do not see the password
appear on the screen)

If you are using a CDE graphic interface:
icon (fourth icon from the left). This opens a drawer of icons.
Click ONCE on the Terminal icon. This should bring up a terminal window. All of
window.

Current directory: __________ /home/team01
Home directory: __________ /home/team01
$ pwd
/home/team01
$ echo $HOME
/home/team01
__ 11. Check to see if you have the authority to view /etc/security/passwd.

$ cat /etc/security/passwd
cat: cannot open /etc/security/passwd
Did you see the contents of the file? __________ No
__ 12. Change to the /tmp directory.

$ cd /tmp
$ pwd (to confirm)
/tmp
__ 13. Now, switch user to root without using the - (dash) option.
$ su or $ su root
root's Password: ibmaix

Current directory: __________ /tmp
Home directory: __________ home/team01
# pwd
/tmp
# echo $HOME
/home/team01
Has anything changed? _____________________________________________
The current directory has not changed, it is still /tmp.
__ 15. Do you have root privileges? Check /etc/security/passwd. _________ Yes.

<Output is the contents of the /etc/security/passwd file>
__ 16. Change to the /etc directory.

# cd /etc
# pwd
/etc
__ 17. Exit back to team01.

# exit
$
__ 18. Did the su and the cd to /etc (in the resulting subshell) affect the current directory
after exiting the subshell? ______ No
$ pwd
/tmp
__ 19. Now, try using the su command with the - (dash) option. What is your home
directory? If you're not sure, check before performing the su.
/home/team01.
$ echo $HOME
/home/team01
$ su - or su - root
#

V4.1
EXempty
Note
Make sure you include a space before and after the - (dash) root's Password:
ibmaix

# pwd
/
# echo $HOME
/
Current directory: _____/_____
Home directory: _____/_____
Is there any thing different this time? Yes, the current directory is now roots directory.
__ 21. Check /etc/security/passwd to make sure you have root privileges.

<Output is the contents of the /etc/security/passwd file>.
__ 22. Exit back to team01 and log out.

# exit
If you are using a CDE graphical environment:
In the blank area of the screen (not in a window), press and hold down the right
Click Continue logout. A new log in box should appear (if using a local
terminal).
End of exercise
Exercise Review/Wrap-up
Consider the following questions:
What is the difference between the su command used with and without the dash?
What situations would it be helpful to use the dash and not use the dash?
Are there disadvantages of logging in directly as root?

V4.1
EXempty Exercise 2. Using SMIT

(with hints)
Estimated time
00:20

AIX provides several different interfaces; ASCII (command-line),
Common Desktop Environment (CDE) and X-Window (Motif). This
exercise provides an opportunity to use the ASCII interface to the
System Management Interface Tool (SMIT). Optionally, you can also
use CDE and Motif interface to SMIT.

Log in from the ASCII interface
Start SMIT (ASCII and optionally the graphical version)
Add users with default settings
Navigate through the SMIT menus
Use SMIT function keys
Display the SMIT history files, smit.log, and smit.script
Start and use graphical SMIT
Introduction
This exercise is designed to give you experience working with the
system management tools supplied with AIX. In most classrooms,
there are two students per AIX system. In many cases, only one
person is able to perform the task at a time. You need to work together
with your teammate. Be sure to take turns at the keyboard.

For classes where the students are sitting at a locally attached graphic
AIX system console, if a student cannot start an X-Windows session
with xinit, make sure they haven't shelled out (F9) in SMIT to get to
Copyright IBM Corp. 1997, 2008 Exercise 2. Using SMIT 2-1

the command line. Also, <ctrl-c> may not work if students are
running a command when shelled out in SMIT.
When the students get to the place when they are presented with the
login screen for CDE, they may at first have difficulty logging in with
the command line. If they delay too long in logging in, the CDE login is
again displayed.
Students get confused over the different interfaces. You may need to
help them determine the benefits of each. Try to hold the discussion
until after the exercise. The exercise contains detailed instruction for
each interface.
There is a difference in function key activity between the ASCII
interface and the Motif interface. For example, in Motif, there is NO
function key to shell out to a command line (F9 in ASCII). It is assumed
that you can simply open another window and run the command.
If the class is being held with the System p platforms in a remote
location and you are using a tool (such as VNC) which is only
configured to support a specific graphic interface (such as CDE), warn
the students in Step 11 to leave CDE as the default interface.
Changing to a command line interface may prevent that tools from
working. If that happens, they will need to use telnet to get back into
the machine and change the user interface back.

Please tell the students that AIX 5L V5.3 is being used in this lab, but
that AIX 6.1 behaves the same. SMIT is being introduce, before you
migrate to AIX 6, because SMIT is needed for the Software Installation
unit.
While using SMIT in Graphic Mode, the system hangs when changing
a user password. You need to open a console session and restart the
system.

V4.1
EXempty Exercise Instructions with Hints

Preface
Two versions of these instructions are available; one with hints and one without. You can
use either version to complete this exercise. Also, please dont hesitate to ask the instructor
if you have questions.
Note that on some terminal emulations, the function keys are not operative and you may
need to substitute escape sequences. For example, instead of pressing F3, you might
need to press <ESC><3> for the same function.
classroom.
Part 1 - Using SMIT (ASCII interface)

__ 1. Log in as the user root. The password is ibmaix.
At the username prompt, enter: root
At the password prompt, enter root's password: ibmaix

If you are using a CDE graphical interface:
icon (fourth icon from the left). This opens a drawer of icons.
Click once on the Terminal icon. This should bring up a terminal window. All of
window.
__ 3. Start SMIT in ASCII mode.

# smitty
__ 4. Add five users to the system. They should be named team100 through team104.
Take all of the defaults. Customizing the attributes is covered later in the course.
Assign a password for each user that is the same as the user name. When the five
users and their passwords have been entered, exit from SMIT.
In SMIT, select:

Security & Users ->

Users ->
Add a User
* User NAME [teamxxx] A
Accept all other defaults.
Press Enter to run.
Press F3 and repeat the process until all five users are added.
When all users are added, press F3 until you reach the Security & Users menu.
Select Change a User's Password
User Name [team###]
Changing password for team###
teamxx's New password: team### (The password will not appear on the screen)
Enter the new password again: team###
Repeat that process until all new accounts have passwords. Then, press F10 and
exit from SMIT.
__ 5. Start SMIT using the options to redirect the smit.script and smit.log files to the
/home/team100 directory. From the main SMIT menu, select Processes &
Subsystems. Display all the current process running on your system. Examine the
output.
Which command do you think was executed? _______.
ps -ef
How can you find out? ___________________________________________
Pressing the F6 Command key from the output screen, displays a pop-up box
that shows the command that was executed.
Return to the main menu.
# smitty -s /home/team100/smit.script -l /home/team100/smit.log
Processes & Subsystems
Processes
Show All Current Processes
A Show THREADS information? pop-up box is displayed. Select no.
F6 (to view the command).
F3 - Cancel until you are back at the System Management screen.
__ 6. In a similar manner, select each of the main SMIT menu entries, examine the
submenu, then use the F3 key to back up to the previous menu. Do not execute

V4.1
EXempty any commands that may cause changes to the system. Because you are logged
in as root, you must be very careful.
Select each entry.
Press F3.
Repeat as appropriate.
__ 7. Shell out (F9) and execute the date command. Return to SMIT and do the following:
1) Stop syslogd subsystem.
2) View the command that was executed.
3) List ALL subsystems to see that syslogd is inoperative.
4) Restart syslogd.
5) Continue to the next step. Do not press F3 or exit SMIT.
F9; Enter
# date
<ctrl-d> to return to SMIT.
From the Main menu, select:
Processes and Subsystems ->
Subsystems ->
Stop Subsystem ->
Stop a Single Subsystem
F4 (to see the list of Subsystem Process IDs).
Select syslogd.
Enter (to stop the subsystem).
F6 (to view the command).
F3 (repeatedly) to return to the Subsystems menu.
Select List All Subsystems (syslogd should be inoperative).
F3 back to Subsystems menu.
Select Start a Subsystem.
F4 (to see the list of Subsystem Name).
Select syslogd.
Enter (to start the subsystem).

__ 8. You should have the COMMAND STATUS screen displayed. Capture an image of
this screen. Exit SMIT.
Press F8=(Image) and Enter.
F10
__ 9. Look at the SMIT history files using pg or more. Look for the image you captured in
the previous step.
What information is in each of these files? ____________________________
_____________________________________________________________
The smit.log file contains a reference to each menu screen selected, the
commands executed, output from the commands, and all screen images. The
smit.script file contains the high-level commands executed, preceded by a
comment showing the date and time when the command was executed.
Why should these files be monitored?
_____________________________________________________________
Both files should be monitored regularly and purged periodically so disk space is
not wasted.
# pg /home/team100/smit.log
# pg /home/team100/smit.script
__ 10. Look at the ownership of the SMIT history files of user team100 with ls -al.
Was it a good idea to create these files as root? ____ No
What will happen if user team100 will use SMIT?
_____________________________________________________________
User team100 is not the owner of the SMIT history files and has no write access
these files. If user team100 starts using SMIT, user team100 gets an error
message saying Could not open SMIT log file: ./smit.log.
# ls -al /home/team100/smit.log
# ls -al /home/team100/smit.script
__ 11. As the system administrator, you can select the default interface that the users can
use. Depending on your classroom environment, you may be able to set a different
interface. Ask your instructor if you would like to change the interface. For the new
interface to take effect, you need to reboot your system. Log out when you are done.

V4.1
EXempty
Note
If you are using VNC for remote graphics, do not change the system user interface
in this step; your VNC server is configured to only work with a CDE user interface.
# smitty
Select:
System Environments ->
Change System User Interface
At Select System Login User Interface, generate a list (F4) to see your choices.
Typically, your interface is set to AIX CDE 1.0. If you prefer the Command Line
interface, select it and press Enter. If you don't want to change the interface, press
F3 until you are out of SMIT and then logout (this is not an option if you are using a
remote LPAR system).
Press F3 twice to get back to the System Management screen.
If you did not make a change to the interface, select F3 - Cancel until you have
exited SMIT and log out. You are finished with Part 1 of the exercise.
If you changed the interface, you need to reboot the machine before it takes effect.
From this SMIT screen, select Stop the System.
* MESSAGE to all users on the system [No Message]
INTERACTIVE shutdown [no]
RESTART the system after shutdown [yes]
* TIME the system goes down [0]
Press Enter.
-OR-
This step could be achieved from the command line with:
# /usr/dt/bin/dtconfig -d (to disable CDE)
# shutdown -Fr
Upon reboot, your new interface should appear.
__ 12. If you did not change your System User Interface (and thus did not need to reboot),
terminate smit and log off of your system.
press F10
# exit

Part 2 - (Optional) Using SMIT (Motif interface)

__ 13. If using CDE, log into CDE desktop as root and the password ibmaix.
If using an X-Windows environment, log in to the AIX system from the ASCII
interface as root and the password ibmaix.
CDE environment (you may need to use a VNC client to remotely acquire a CDE
interface):
In the box, enter: root
In the box, enter root's password: ibmaix (You do not see the password appear
on the screen)
X-Windows environment:
Start an X-Window from the ASCII interface command prompt using
# xinit
-OR-
# startx
Note
If you are not sure how to login in an X-Windows environment, ask your instructor.
__ 14. If necessary, open a terminal window. If one already exists, skip to the next step.
CDE environment:
icon (fourth icon from the left). This will open a drawer of icons.
Click once on the Terminal icon. This should bring up a terminal window. All of
window.
X-Windows environment:
Move the cursor to an empty area on the screen and press and HOLD the right
mouse button.
While holding the mouse button down, drag the cursor to the first item on the
menu, New Window, and release the button. A new window should appear.
__ 15. Add five users to the system. They should be named team200 through team204.
Take all the defaults. Do not assign passwords for these users. When the five
users have been entered, exit from SMIT.

V4.1
EXempty # smit
Click Security & Users.
Click Users.
Click Add a User.
At the User NAME field, type in the user name team200 and click OK.
When you see the running man (at the top) stop and stand with his arms up in
victory, click Done to return to the Add a User menu.
Add users team201 through team204 in the same way.
When the five users are added, click Cancel to return to the Users window.
Once all users are added, click Cancel and then Exit to exit SMIT and return to the
command line.
__ 16. In a similar manner, select each of the main SMIT menu entries, examine the
submenu, and then use the Cancel button to go back to the previous menu. Do not
execute any commands that may cause changes to the system. Because you
are logged in as root, you must be very careful.
Select each entry.
Cancel.
Repeat as appropriate.
__ 17. Exit from SMIT. Look at the SMIT history files with pg or more.
What information is in each of these files?
_____________________________________________________________
The smit.log file contains a reference to each menu screen selected, the
commands executed, output from the commands, and all screen images. The
smit.script file contains the high-level commands executed, preceded by a
comment showing the date and time when the command was executed.
Why should these files be monitored?
_____________________________________________________________
Both files should be monitored regularly and purged periodically so disk space is
not wasted. Unlike the SMIT ASCII interface, the Motif interface automatically
records all screen images. The F8 key in the Motif interface will show you the
SMIT fastpath command instead.
Point and click Exit.
Select Exit SMIT.

# pg /smit.log
# pg /smit.script
__ 18. Log out of CDE or X-Windows.

If using CDE:
On the blank area of the screen (not in a window), press and hold down the right
Click Continue logout.
A new log in box should appear.
If using X-Windows:
On a blank area of the screen, press and HOLD the right mouse button. While
holding the right mouse button, drag the cursor to End Session and continue to
HOLD the mouse button. A new menu appears. While continuing to HOLD the
right mouse button, drag the cursor to Quit... and release the mouse button.
Click the OK button to Quit Mwm (Motif Window Manager).
You should now be back at the ASCII interface.
__ 19. Log out of the system.

# logout
End of exercise

V4.1
EXempty Exercise Review/Wrap-up

In an X-Window session, if I want to use the ASCII interface with SMIT, what command do I
type to invoke SMIT?
Answer: smitty


V4.1
EXempty Exercise 3. System startup and shutdown

(with hints)
Estimated time
00:30

This exercise familiarizes you with the system startup and shutdown
process, and using the System Resource Controller (SRC).

Start a System p in multi-user
Switch to the root user and shutdown the system
Execute the lssrc, stopsrc, and startsrc commands
Introduction
If there are two or more students per system, you must work as a
single team throughout this exercise. All instructions involve using the
system unit for starting and stopping the system in various modes.

Cover the user name and passwords the students are to use when
they log in. For the purpose of this exercise, it is assumed that the
login user names are team01 - team05. The passwords should be the
same as the user name. Ensure that if there is more than one student
on a system, that no two users use the same user name.
When using a local System p lab machine with attached console, if the
lab machines provide the CDE environment, and students did not
disable the CDE default, they will be prompted by the CDE interface
upon each reboot during this exercise. Have them select
command-line login.
Copyright IBM Corp. 1997, 2008 Exercise 3. System startup and shutdown 3-1

For classroom configurations where students are sharing one AIX
system, they must act as a single team during this exercise. This is
indicated in the Introduction section. Be sure to cover this as part of
an exercise introduction. Encourage them to take turns at the
keyboard. Suggest that everyone work as a team and go through the
exercise in its entirety. Any students wanting additional practice should
repeat this exercise before or after class or during free time when the
other students are not working on the system.
If the AIX systems being used to complete the exercise are remote
partitioned systems, the local workstations need access to the HMC
particularly to complete Part 2 - Boot Using System Management
Services.

V4.1

Preface
Two versions of these instructions are available; one with hints and one without. You
can use either version to complete this exercise. Also, please dont hesitate to ask the
instructor if you have questions.
classroom.
Part 1 - Multi-User mode

__ 1. Log in using the user name and password provided by your instructor. The login
name should be a user name such as teamxx where xx is a double digit number like
01, 02, and so forth. Switch to the root user. Display the alog file for the bootlog.
Page through the output to see what devices were configured and the other tasks
that were performed.
Note
If this is the first time you log in using the user name, you are prompted to change
your password. Keep it the same as the user name.
Log in as teamxx and enter the password.

$ su root
Enter root's password
# alog -o -t boot | pg
__ 2. Page out the /etc/inittab file. Take a moment to become familiar with its contents.
What is the run level of shdaemon? _______ 2
What does it mean? The run level for rc is 2. It represents AIX multi-user mode.
___________________________________
# pg /etc/inittab
__ 3. Log out of this session.
# exit
Part 2 - Boot using System Management Services

__ 4. If your remote AIX system is an LPAR on a remote managed system, then you can
do this SMS exercise by remotely accessing the HMC using your Web browser and
starting a virtual console session with your AIX system (the Web browser must have
connectivity to the HMC which may be inside a virtual private network (VPN).
__ 5. In order to boot to an SMS mode, you must first shutdown the AIX system. From the
command line, request a shutdown with no delay and no reboot.
# shutdown -F
__ 6. Access your systems HMC by starting a Web browser and using the HMCs IP
address as the URL. If you do not have this IP address, ask your instructor. (If you
are on a stand-alone system, skip to Step 4 below for booting in SMS mode.)
Start a Web browser.
For example, the standard method for IBM provided lab environments is to find the
Mozilla Firefox application icon in the Applications box of the e-lab Web interface.
Use of Internet Explorer, if available, should also work correctly.
Double-click the Firefox icon.
(click No, to make this your default browser )
# https://##.##.##.##
(click OK, to Accept the certificate temporarily for the session)

__ 7. Start the Hardware Management Console application by clicking the Launch link.
Log into your system HMC as hscroot. and password abc1234. Locate your
machine and the logical partition which your are assigned.
Click the link to launch the HMC application:
Log on and launch the Hardware Management Console Web application
Userid: hscroot
Password: abc1234
Click logon or press < Enter >

Click Not Now in response to the prompt: Do you want Firefox to
remember this password
You may also get an error about pop-ups being disabled. Click OK on the
error box. In Mozilla, pull down the Tools menu, select the Options
item, click on the Content icon, and uncheck the Block Popup Windows
option. Click OK. Then repeat the above sequence.
If prompted to accept security certificates, respond by clicking Yes.

V4.1
EXempty
You may also get a screen entitled Choose a Disconnected Session. Do
not use someone elses disconnected hscroot session. Click on the New
Session button and obtain a brand new hscroot session.
Expand the Systems Management selection in the Navigation window.
Expand the Servers selection in the Navigation window.
Click the server for your assigned LPAR. You may need to ask your instructor. All the
configured lpars appear in the Contents window.

__ 8. Locate the icon for your partition. The state of the partition should be not active.
Activate the partition, specifying that you want to open a terminal window and that
you want a boot mode of SMS.
To activate a logical partition from the HMC:
a) If not selected, click the Select box to the left of your lpar
b) Click the arrows which appeared to the right of the lpar
c) Drag the mouse to Operations, Activate and then click Activate
d) Check Open a terminal window or console session, and then click the
Advanced button
e) Click the drop-down menu for the Boot mode, select SMS, and then
click OK
(Do not change anything with the Keylock position)
f) This should return you to the Activate screen again. Click OK to
start the activation.
g) A virtual terminal window should start up on your workstation.
Since it is acting as the system console you should be able to watch
the boot messages and see when the SMS menu appears.
__ 9. If you are using a local non-partitioned system (otherwise skip to the next step): As
the root user, shut down and reboot the system to the System Management
Services mode. As soon as you see the keyboard and hear a beep, you should
press the numeric 1 key or <F1> key to access SMS.
# shutdown -Fr
When the first screen appears (PowerPC screen), after the keyboard icon but
before the last icon appears, press F1 (or 1, depending upon your environment). The
System Management Services screen then displays.
__ 10. System Management Services (SMS) menus and contents vary from one System p
machine/model to another. Take a few moments to explore the menu options which
display on your system. Do not make any changes or run any tests. Also, dont click
Exit on the main SMS screen until you are finished exploring the other options.
The following menu items may be different than what you see on your system.
Navigating the menu also may vary from system to system. On some systems, you
have a graphic interface and click the EXIT icon to return to a previous menu or
eventually exit from SMS. On other systems with an ASCII interface, you press the
<ESC> key to return to a previous menu or use the x response to exit from SMS.
Select Language - Type 1 and press <Enter>.
This shows a list of various languages to use while in the SMS menus.
Return to the previous menu <Esc>.
Set up Remote IPL - Type 2 and press <Enter>.
This prompts you for which interface you want to use for remote IPL and then
would prompt for networking information needed for a network boot.
Change SCSI Settings - Type 3 and press <Enter>.
This lists two options to modify: Hard Drive Spin Up Delay and Change SCSI
ID.
Return to the previous menu < Esc>.
Select Console - Type 4 and press <Enter>.
This will display a message on all eligible terminals to select as console. Press 0
(zero).
Select Boot Options - Type 5 and press <Enter>.
This provides boot options to select an install/boot device, configure the boot
device order, or enable/disable Multi-boot Startup Mode.
__ 11. When you are done exploring, exit the SMS program and allow the system to reboot
in multi-user mode.
Enter X and press <Enter> at the SMS screen. This will cause the system to
reboot in multi-user mode.

V4.1
EXempty Part 3 - System Resource Controller (SRC)

__ 12. Log in as root. List the active and inoperative subsystems on your system. Display a
screen at a time. Locate the syslogd subsystem. It should be marked as active.
login: root
root's Password: <enter root's password>
# lssrc -a | pg
__ 13. Using the appropriate SRC commands, stop the syslogd subsystem. Check that it
is inoperative. Restart it. Check that it is active.
# stopsrc -s syslogd
# lssrc -s syslogd
# startsrc -s syslogd
# lssrc -s syslogd
Part 4 - Resetting the run level (INIT)

__ 14. Shutting a system down to Single User mode removes any TCP/IP network access
to the system. Due to this, you need to use the system console while doing this lab
exercise.
If you are using a non-partitioned local system, then login at the attached console
and skip to step 18.
If you are using a partitioned system, you need to start a virtual terminal to your
partition, as you did earlier when booting to SMS mode. Do the following steps:
__ 15. Access your systems HMC by starting a Web browser and using the HMCs IP
address as the URL. If you do not have this IP address, ask your instructor. (If you
are on a stand-alone system, skip to Step 4 below for booting in SMS mode.)
For example, the standard method for IBM provided lab environments is to find the
Mozilla Firefox application icon in the Applications box of the e-lab Web interface.
Use of Internet Explorer, if available, should also work correctly.
# https://##.##.##.##

machine and the logical partition which your are assigned.
Userid: hscroot
Password: abc1234


V4.1
EXempty You may also get a screen entitled Choose a Disconnected Session. Do

__ 17. Start a virtual console window for your assigned partition. This acts as your system
console during the exercise.
In the contents window locate your partition and then:
a) Select your lpar by clicking the Select box to the left of the lpar
name.
b) Click on the arrows which appeared to the right of the lpar
c) Drag the mouse to Console Window, and then click Open Terminal
Connection
d) After a short delay a window should appear with an AIX login prompt
e) Login as root
__ 18. Reset the run level to Single User Mode without rebooting the system.
# shutdown -m
You will see several messages appear. Enter the root password when you
see the following:
INIT: Single-User Mode
PASSWORD:
__ 19. Verify that several subsystems have been stopped.

# lssrc -a | pg (This command will fail as the SRC daemon is stopped.)
__ 20. Verify that there are few processes running on the system.
# ps -ef | more
__ 21. Run /etc/rc.net to start the network interfaces and then reset the Run Level to
Multi-User Mode.
#/etc/rc.net
# telinit 2
__ 22. Ensure that all normal subsystems have been started and are working properly. If
you have been using CDE, and CDE fails/hangs now, you forgot to run /etc/rc.net
in the previous step.
login: root
root's Password: <Enter root's password.>
# lssrc -a | pg
__ 23. Shut down and reboot the system one more time.
# shutdown -Fr
End of exercise

V4.1

1. What do you need on a PCI RS/6000 to invoke Maintenance mode?
Answer: Access to the System Management Services firmware.

V4.1
EXempty Exercise 4. AIX software installation

(with hints)
Estimated time
00:25

In this exercise, you list the software currently installed on the AIX
system. You then install two software packages.

List the software installed on an AIX system
Install software
Install software fixes (optional)
View the history of installed software (optional)
Introduction
installation facilities supplied with AIX. If there is more than one
student sharing a system, then only one person can actually perform
these steps. All other students on that system should observe while
the steps are being done.
In this exercise, you install two software packages:
bos.dosutil
bos.games
In this exercise, you may also install a fix to already installed software
packages.
The instructions are written assuming you are using SMIT in ASCII
mode. However, it is easy to interpret the instructions so that they can
be used with SMIT in the AIXWindows environment. Feel free to use
SMIT in either mode.
When you select an item while using the ASCII user interface, use the
cursor keys on the keyboard to highlight a menu item, then press
Copyright IBM Corp. 1997, 2008 Exercise 4. AIX software installation 4-1
Enter. On a graphics display, choose (that is, click) the selected item
with the left mouse button.
The fix installation and display of the fix history is optional. It requires
that the lab setup for the exercises used an AIX 5L V5.3 TL6 base
installation. If the image used in your lab environment has additional
maintenance already installed and committed, you may not be able to
successfully complete the optional steps.
Instructor exercise overview

The fix installation and display of the fix history is optional. It requires
that the lab setup for the exercises used a base level AIX 5L V5.3
installation and that there is a fix that needs to be done. If the image
used in the lab environment already has the fix installed and
committed, the students are not able to successfully complete the
optional steps.

Be sure to explain the information in the Introduction of this Exercise.
In the classroom, students install software that has been preloaded to
the /usr/sys/inst.images directory. Explain that the installation from
this directory is the same as an installation from CD or tape.
However, if you are in an environment where you plan to provide the
students with the actual installation media, be sure to tell the students
which input media is being used.
Important: Since students instal certain UNIX games onto their
systems, it may be important to set some ground rules. They might
enjoy playing with the games during break, but stress that they are not
to play with them during lecture or when they should be working on
machine exercises. Most students will oblige. However, if a student
becomes disruptive with the games, the instructor always has the
option of using SMIT to remove the fileset.

V4.1
EXempty Exercise Instructions With Hints

Preface
classroom.
Part 1 - List and install AIX software

__ 1. Access your system. If remote, we recommend an interactive ascii prompt via telnet,
but CDE graphical session can also work.
__ 1. Log in to AIX as teamxx and su to root.
Log in as teamxx
If using CDE, after logging in start up a dtterm window to obtain an interactive
command prompt.
$ su root
__ 2. Use SMIT to list the software on your system.

What command is SMIT executing to provide this information?___________
lslpp -L all
Examine the output and verify that bos.games and bos.dosutil are not installed
on your system.
# smitty
Software Installation and Maintenance ->
List Software and Related Information ->
List Installed Software and Related Information ->
List Installed Software
* SOFTWARE name [all]
SHOW superseded levels? no
OK or Enter.
Press F6 to see the generated command and then F3 to return to the displayed
output.
While viewing the output, use Find ( / ) to verify that the bos.games and
bos.dosutil packages are not installed on your system.
Use F3 (repeatedly) to return to the Software Installation and Maintenance
menu.
__ 3. Now it is time to install the two software packages, bos.dosutil and bos.games:
These packages have been preloaded (but not installed) in a directory called
/usr/sys/inst.images. The procedure to install these packages from this directory
are the same as those to install software from CD or tape.
Use SMIT to install the bos.games and bos.dosutil packages. When this task has
successfully completed, return to the AIX command line.
In the SMIT Software Installation and Maintenance screen, select:
Install and Update Software ->
Install Software ->
* INPUT device / directory for software [/usr/sys/inst.images]
(Use F4 to select this standard repository location off the list)
SOFTWARE to install [bos.dosutil,bos.games]
(Use F4 to list the available packages from which to choose)
When selecting the packages, be sure to install the entire package. Use F7 to
indicate which packages to install.
Press Enter to perform the installation.
Verify that the output is correct and then press F10 to return to the command line.
__ 4. Verify that the bos.games fileset installed correctly and list the games which are
available.
# lslpp -L bos.games
# cd /usr/games
# ls
__ 5. Use SMIT to remove the bos.dosutil fileset.

# smitty install
Software Maintenance Utilities ->
* SOFTWARE name [bos.dosutil] +
PREVIEW only? [no] +
Press Enter to perform the fileset removal.

Verify that the output is correct and then press F10 to return to the command line.

V4.1
EXempty __ 6. If not doing the optional Part 2 - Working with AIX Fixes, log off the system at this
point.
Part 2 - Working with AIX Fixes (Optional)
__ 7. We have already downloaded AIX 5.3 TL6 Service Pack 1 and stored the fix files in
the /usr/sys/inst.images/fixes directory. Use SMIT to run a Comparison Report
between this fix repository and what is currently installed on your system. Only
report on installed software that is at a lower level than what is in the fix repository.
Briefly examine the generated report. Can you tell what problems or new capabilities
are included in which fileset update? ________________________________
# smit install
Software Service Management ->
Comparison Reports ->
Compare Installed Software to Fix Repository
FIX REPOSITORY location [/usr/sys/inst.images/fixes]
FILE location of lslpp -Lc output []
(Leave blank to compare against this system.)
Select which reports to run.
Installed Software that is at a LOWER level yes +
(lowerlevel.rpt)
Installed Software that is at a HIGHER level no +
(higherlevel.rpt)
Updates for filesets that are NOT INSTALLED no +
(notinstalled.rpt)
Installed Software with NO UPDATES found no +
(no_update_found.rpt)
DIRECTORY location for reports. [/tmp]
(Leave blank to omit.)
The report is only a listing of filesets and their version, release,
mod, and fix level.
__ 8. Return to the Software Installation and Maintenance menu and choose the menu
sequence which allows you to list software that is on installation media (which could
be a directory). List the APARs that are addressed by the filesets in the
/usr/sys/inst.images/fixes directory. Briefly examine the output.
Press F3 repeatedly to return to the Software Installation and Maintenance menu
List Software and Related Information ->
List Software on Media and Related Information
List Software Fixes (APAR) on the Installation Media
INPUT device / directory for software [/usr/sys/inst.images/fixes] +

V4.1
EXempty __ 9. Return to the Software Installation and Maintenance menu and follow a menu
sequence that presents a dialogue panel which allows you to install a fix by
specifying the related APAR number. Request to use our /usr/sys/inst.images/fixes
directory as the source of the fixes.
Press F3 repeatedly to return to the Software Installation and Maintenance menu
Install and Update Software ->
Update Software by Fix (APAR)
INPUT device / directory for software [/usr/sys/inst.images/fixes] +
__ 10. There is a bug in AIX 5L V5.3 TL6 where the usrck command can (incorrectly) lock
all user accounts. You do not want that to happen. In the dialogue panel, get a listing
of APARs which are available to be installed. Find the APAR which was just
described. Select and install that fix. Examine the output to verify that the fix was
installed. What filesets have been installed?
______________________________________________________________
______________________________________________________________
Use F4 on the FIXES to install line to get a list.
Use F7 to select the desired fix
FIXES to install [IY59384] +
PREVIEW only? (update operation will NOT occur) no +
Press <Enter> to install the fix.
The installed filesets were bos.rte.install and bos.rte.security
Press F10 to exit smit
__ 11. View the history of the installed filesets using the lslpp command.
# lslpp -h bos.rte.install
# lslpp -h bos.rte.security
__ 12. Log off from the system.
End of exercise

V4.1
EXempty Exercise 5. AIX 6.1 installation and configuration

(with hints)
Estimated time
01:00

In this exercise, you have the opportunity to migrate your AIX OS
image from AIX V5.3 to AIX 6.1, and work with the Configuration
Assistant. The Configuration Assistant is an application that can be
used to configure the initial system settings.

Install an Operation System by migrating to AIX 6.1
Use the Configuration Assistant
Introduction
The Hardware Management Console (HMC) and Configuration
Assistant are graphics-based applications that assist the system
administrator with various tasks. These applications are used to
perform the installation of the operating system and system
configuration respectively.

If the students are working in a remote lab environment, the DISPLAY
environment variable has to be exported for the configassist and
install_assist commands to work.

None.
Copyright IBM Corp. 1997, 2008 Exercise 5. AIX 6.1 installation and configuration 5-1

Preface
classroom.
The objective for the first part of this lab is to install an IBM AIX 6.1 mksysb image. The
instructions to accomplish this using an eLab environment and a stand-alone system is
listed. Ask your instructor to confirm that the /spot and lppsource (bos) have been
allocated for your system before starting.
__ 1. Log in to your assigned system.
__ 2. Before you start the migration/install, you need to identify the boot disk and the
adapter which has a configured interface. Use the bootlist -o command to
identify the boot disk. Use the netstat -in and the lsfg commands to obtain the
location of the configured interfaces adapter. If the configured interface is en4, then
the related adapter is ent4 (the number in both names will always match). You only
need the last two positions of the location code (either C#-T# or P#-T#)). You cover
location codes in more detail in the Device unit later in the course.
boot disk: __________________________________________________
configured interface and IP address ______________________________
location code of the network adapter _____________________________
To collect the information:
# bootlist -o -m normal
# netstat -in | grep en
# lscfg | grep -i ent
__ 3. If you do not already have a working HMC window, access your systems HMC by
starting a Web browser and using the HMCs IP address as the URL. If you do not
have this IP address, ask your instructor.
If you are not using a HMC, log on to your stand-alone system and skip to Step 3.
For example, standard for IBM provided lab environments is to find the Mozilla
Firefox application icon in the Applications box of the e-lab Web interface. Use of
Internet Explorer, if available, should also work correctly.

V4.1
EXempty Double-click the Firefox icon.

# https://##.##.##.##

__ 4. If you are not already logged into the HMC, start the Hardware Management
Console application by clicking the Launch link. Log into your system HMC as
hscroot. and password abc1234.
Userid: hscroot
Password: abc1234


__ 5. Locate your machine and the logical partition which your are assigned.
configured lpars will appear in the Contents window.
__ 6. Shutdown your assigned lpar, This can be performed using the HMC (Operating
System options only) or by using the shutdown command at an interactive root
user prompt.
If you performing this task at a root user interactive prompt, type:
# shutdown -F
-OR-
If you are performing this task using the HMC, follow the steps below:
a) Select your lpar by clicking the Select box to the left
c) Drag the mouse to Operations, Shutdown, and then click Shutdown
[DO NOT EXECUTE AN IMMEDIATE SHUTDOWN OF THE PARTITION FROM THE HMC.
THIS CAN CAUSE DISK CORRUPTION ON A RUNNING AIX SYSTEM]
d) Click Operating System Immedediate, and then OK
( After awhile, the Reference Code will show all zeros)
__ 7. If you currently have a virtual terminal open, close that window.
Click on the upper right corner of the window to close it.
(there were older versions of the HMC where this would not be sufficient and you
would also need to close the virtual session from the HMC menus)
__ 8. Boot your system in SMS mode. Open a console session if you are using a lpar
system and HMC configuration.
If you are performing this task using the HMC, follow the steps below:
a) If your lpar is not already selected, click the Select box to the
left of the lpar name.
b) Click the arrows which appeared to the right of the lpar
c) Drag the mouse to Operations, Activate and then click Activate
d) Check Open a terminal window or console session, and then click the
Advanced... button
e) Click the drop-down menu for the Boot mode, select SMS, and then
click OK
(Do not change anything with the Keylock position)
f) This should return you to the Activate screen again. Click OK to
start the activation.
If you have a local System p machine and locally attached console in front of you:
a) Power on your machine.
b) Soon after the LED reads E1F1, the keyboard is discovered and the
distinctive beeping will sound. Between that time and before the
system begins to search for the boot image, press either the F1 key
or numeric 1 key (depending on the type of machine) to invoke the
Service Mode boot list.
The console should eventually display the SMS menu.
__ 9. Configure your system Remote IPL parameters. You need the correct subnet mask,
plus the IP addresses of the NIM server and the gateway. These must be configured
to boot off the network adapter. The instructor provides you with the NIM and
gateway addresses.

V4.1
EXempty NIM server IP address: _______________________________________

Gateway IP address: _________________________________________
a) Select Setup Remote IPL (Initial Program Load) from the Main Menu,
by typing 2 at the prompt on the bottom of the screen and pressing
<Enter>.
b) Select the NIC Adapter Device one which matches the location that
you recorded ealier in step 2 by entering the number to the left of
that network adapter and pressing <Enter>.
c) select IP Parameters , by entering 1 and pressing <Enter>.
Make sure that all IP addresses are correct. These IP Addresses can
be changed by entering the number to the left, pressing <Enter>,
typing in the IP Address, and pressing <Enter> again.
d) When you finished entering all the IP Addresses, Press the <ESC>
until you return to the Main Menu.
DO NOT EXIT SMS AT THIS POINT
__ 10. Assign your network adapter as the first and only boot device. Be sure the location
of the adapter matches what you recorded earlier.
a) From the SMS Main menu, select Select Boot Options by entering
the number to the left <5>, and pressing <Enter>.
b) From the Multiboot menu, select Configure Boot Device Order, by
entering the number to the left <2>, pressing <Enter>.
c) From this menu, select Select 1st Boot Device by entering the
number to the left <1>, pressing <Enter>.
d) From this menu, select Network by entering the number to the left
<6>, pressing <Enter>.
e) From this menu, select the network adapter by entering the number to
the left <1>, pressing <Enter>.
f) From this menu, select Set Boot Sequence, by entering the number to
the left <2>, pressing <Enter>.
(You should see the Current Boot Sequence displayed, with the
network adapter as the first and only device.
g) Exit the SMS menu by press X. You will be prompted Are you sure.
Type the number for YES <1> and press <Enter>.
h) Your system should start booting and connecting to your NIM server.
__ 11. Your system should be downloading an AIX network boot image at this time. The
download and execution of the network boot image takes 2 to 3 minutes. When
done, you see a prompt; select the terminal console and the English during
installation.
Type 1 and press <Enter> for console, then type 1 and press <Enter> again to have
English during install.
__ 12. On the Main Installation and Maintenance menu, select Change/Show Setting and
Install.
Type 2 and press <Enter>.
__ 13. Verify that the Method of Installation is set to Migration and the Disk to Install is
the one you previously identified as the boot disk (probably hdisk0). When verified,
enter 0 to start the installation.
__ 14. You may get prompted once or twice for other verifications. For example, it may give
you the option of listing filesets and configuration files that are replaced. You may
choose to examine these; the lists are not very large. Eventually respond with a 0 to
continue with the migration.
When you see the Installing Base Operation System, you can take a break (though
the instructor may wish to use this time to discuss another topic).
This takes 30 to 40 minutes.
Type # and press <Enter> (the value of # depends on what you want to examine or
if you just want to continue.
__ 15. Once the install is complete, the system automatically reboots. Once the reboot is
completed, there will be a login prompt on the system console. At that point,
continue to the next step.
__ 16. Log in as root user and start the Configuration Assistant.
If you are using a graphical interface, type:
# install_assist
-OR-
# configassist
If you are using an ASCII interface, type:

# install_assist
__ 17. Ensure the date and time are set correctly.

Click Next on the opening screen.
Date and time are already selected. Click Next.
Make any changes if necessary to the date and time.
Click Next when finished.

V4.1
EXempty __ 18. Familiarize yourself with some of the other options. When youve finished, exit from
Configuration Assistant and select the option to Finish now, and do not restart
Configuration Assistant when restarting AIX.
Click Exit the Configuration Assistant, then click Next.
Click Finish now, and do not restart Configuration Assistant when restarting
AIX, then click Finish.
End of exercise
Have the students discuss the various things that can be configured with the Configuration
Assistant.

V4.1
EXempty Exercise 6. Web-based System Manager

(with hints)
Estimated time
00:20

This exercise provides an opportunity to use the Web-based System
Manager. If your classroom is networked, there is an optional part to
allow you to configure your server for Web-based System Manager
client access.

At the end of the lab, you should be able to:
Use the Web-based System Manager
Configure the Web-based System Manager for client machine
access
Use the Web-based System Manager from the client machine
using the WebSM client application
Introduction
Web-based System Manager in AIX. If you are sharing the machine,
only one person should perform each step. You must work from a
graphical interface for this exercise.
If your classroom is networked, you can configure and test the
Web-based System Manager for client machine access. The client
machine can be another AIX system or a PC. In both cases, the client
platform must have a browser that is Java-enabled.
Copyright IBM Corp. 1997, 2008 Exercise 6. Web-based System Manager 6-1

Preface
classroom.
If there are multiple interfaces on your system, your instructor should identify which
interface to use for this exercise. If you are in an environment where the AIX system is
accessed remotely, be sure to not deconfigure the interface used for that purpose.
Part 1: Use the Web-based System Manager

__ 1. For this exercise you need a graphic interface to your system.
If you are using a partitioned system, then you need to use a remote graphics utility
such as VNC or an Xwindows Server utility to access your system. You should be
presented with a CDE login prompt.
If you are using a local non-partitioned system you need to use the attached graphic
console.
__ 1. Log in as root.
__ 2. If you logged using a locally attached graphic console on a non-partitioned system,
then make sure you are in the AIX windows or CDE environment.
$ xinit (if using a local graphic console and not already in an AIXwindow or CDE
environment)
__ 3. If you are in a CDE environment and there is no terminal emulation window, open
the Personal Application subpanel (arrow over the pencil and paper icon) and click
on the Terminal icon.
Access the Web-based System Manager.
$ wsm &
__ 4. View system processes by both details and tree details. Viewing processes by tree
detail provides a pictorial view of the parent/child relationship of the processes.

V4.1
EXempty Return to the view of the task icons for Processes when you have completed this
task.
From the Web-based System Manager Contents area (on the right), double-click
Processes, then double-click All Processes. Once the Processes window
displays, scroll through the output.
There are two ways to display the processes as a tree:
Click View on the menu bar and then click Tree Details.
On the tool bar, click the Details icon, which is just to the left of the rightmost
one. Note: If you rest the mouse pointer over the icon, the icon name is
displayed.
Once you have reviewed the output, go back to the Processes plug-in by clicking
the back arrow on the title bar.
__ 5. View the Web-based System Manager window that displays the current system date
and time.
Single-click System Environment in the Navigation area.
In the System Environment window, double-click Settings.
Double-click Date/Time.
View the Date and Time and then click a Time Zone option. Since you are not the
root user, you will not be able to make any changes.
When you are done, click Cancel to close this task. window.
__ 6. Close the Web-based System Manager Launch Pad window and log out of the
system.
Close the Web-based System Manager by clicking the upper left of the title bar
(Console) and then clicking Exit on the pull-down menu.
Click the Exit Now button in the pop-up.
Click and hold the right mouse button in the root window area.
If in CDE environment, log-off by clicking the Log out menu item.
When prompted, click on Continue logout
If in a native Xwindows/Motif environment, click on the End Session menu item.
This brings up a sub-menu; click the Quit menu item.
If exiting the graphics environment results in a ASCII command prompt, type exit to
complete the log out of AIX.
Part 2 (Optional): Configure the Web-based System Manager

for AIX client access
Preface
For the optional sections, you need to have a functioning network environment in lab
room. The lab teams need to pair up and work together. Each team pair select which
machine is remotely managed (Web-based System Manager server) and which
machines are the remote system administrator platform (Web-based System Manager
client).
There is no need for additional configuration to use the Web-based System Manager to
manage an AIX platform from an AIX client with the Web-based System Manager
installed. Client-server mode does not require that the server run a Web server; the
connection is made directly between the two Web-based System Manager applications.
__ 7. Log into the WebSM Client system as root and using the correct root password.
__ 8. On the WebSM Client system, make sure you are in the AIX windows or CDE
environment. If you are in a CDE environment and there is no terminal emulation
window, open the Personal Application subpanel (arrow over the pencil and paper
icon) and click the Terminal icon.
Access the Web-Based System Manager on the client platform, but specify the IP
address or hostname of the Web-based System Manager server machine.
# xinit (if using a local non-partitioned system and not in a windows environment)
# wsm -host <server IP address>
-OR-
# wsm -host <server hostname>
__ 9. Do a couple things to validate that the interface is working correctly and then close
down the application.
__ 10. At the Web-based System Manager server system, disable client-server (and
remote client) access. This way you can use the Web-based System Manager
locally on a platform but not allow remote client-server access.
# /usr/websm/bin/wsmserver -disable

V4.1
EXempty __ 11. At the client platform, again start the Web-based System Manager specifying the IP
address or hostname of the Web-based System Manager server machine. Did it
work? _____
No
# wsm -host <server IP address>
-OR-
# wsm -host <server hostname>
__ 12. At the Web-based System Manager server platform, re-enable client-server (and
remote client) access.
# /usr/websm/bin/wsmserver -enable
Part 3 (Optional): Configure the Web-based System Manager

for remote client access
Preface
In order to do this optional exercise, you must have a qualified PC running a supported
version of Windows or Linux, a supported desktop environment, and a Java-capable
Web browser. The PC must have network connectivity to the AIX server platform.
When using a PC platform as the Web-based System Manager Remote Client, you
must first install the Web-based System Manager remote client application on that
platform. This is done using the Web server on the Web-based System Manager Server
platform.
One a PC has the WebSM Client, it can access any AIX system which has wsmserver
enabled (as in the previous exercise part). The AIX WebSM servers do not need the
following configuration completed, only the server that provides for Web browser-based
download of the WebSM client package.
The alternative to using a Web browser interface is to ftp download the install shield
executable file to the PC. The path to that file is:
/usr/websm/pc_client/setup.exe
__ 13. While a Web page server is not needed to use the Web-based System Manager
function, it is used to download the Web-based System Manager client to an remote
Linux or Windows platforms you might wish to work from. The HTTP Server Web
Server software (IHS2), which is delivered on the AIX 5L V5.3 Expansion Pack CD,
should already be pre-installed on your system.
Verify that it has been installed.
# smit list_software
List Installed Software and Related Information ->
List Installed Software
* SOFTWARE name [IHS2] +
SHOW superseded levels? no +
Press Enter
-OR-
# lslpp -L IHS2
__ 14. You need a graphic interface to configure the Web server support for providing the
WebSM PC client. In your graphic environment start a terminal emulation window
and execute configassist in the background.

V4.1
EXempty Configure the Web server support for Web-based System Manager using the
following values.
The root document directory is: /usr/HTTPServer/htdocs
The root CGI directory is: /usr/HTTPServer/cgi-bin
The HTTPServer start command is: /usr/HTTPServer/bin/apachectl restart
# configassist
At the Configuration Assist task list, select:
Configure the Web Server for Web-based System Manger remote management
Click on next
Confirm or change the displayed values
Click on Next until you return to the task list
Then exit out of the Configuration Assistant.
__ 15. On the client PC, start the Web browser and enter a URL of:
http://<wsmservername>/remote_client.html
__ 16. Click the InstallShield hyperlink on the new Web page. (The Java Web Start
method requires additional security software to be installed on both client and server
platforms).
__ 17. Identify the client as either a Linux or Windows platform on the resulting Web page.
__ 18. On the File Download pop-up, click the Save to Disk button. In the Save In: field,
navigate to store it on the desktop and again click the Save button. In the
Download complete pop-up, click the Open button. This runs the Install Shield
wizard. (There will be a command prompt window launch to execute java.exe just
before the Installer window appears; do not do anything with the command prompt
window.)
On the Installer window, click the Next button.
Use the default install Directory name and click the Next button.
Note the location and size of the install and click the Next button.
After the install is complete, click the Finish button.
__ 19. You should now see an icon on your desktop labeled: Web Based System
Manager Application.
Open this new icon.
In the Log On window, enter the Web-based System Manager server hostname
or IP address and click the Log On button.
After the handshake procedure is completed, enter root and the root password
and again click the Log On button.
__ 20. Once the Web System Manager window appears and has all the icons filled in, try a
few operations to prove it is operational and then exit the application.
End of exercise

V4.1
EXempty Exercise 7. Devices

(with hints)
Estimated time
00:30

This exercise familiarizes the students with listing device configuration
information on the system and configuring a defined device.

List device configuration on a system using SMIT or the
appropriate command
List and change system parameters
Configure a defined SCSI device
Introduction
This exercise is designed for you to become familiar with the methods
you can use to discover what devices are on your system and how
they are configured. It deals exclusively with adapters and SCSI types
of devices. Future optional exercises will deal with terminals and
printers.
Part 5 - Configuring a defined device, can only be done if you have an
externally attached tape drive or CD-ROM. Most lab facilities do not
have this.
Copyright IBM Corp. 1997, 2008 Exercise 7. Devices 7-1


Be sure to warn your students not to change any other system
parameters other than the one mentioned in the exercise, as other
changes could disable the system for other users.
In situations where you have many students to a single AIX system,
you need to coordinate the changing of system parameters section as
well as the section that deals with configuring a defined device.
In part 5, if a student fails to power the tape drive back on prior to
completing the last instruction, he/she will get the following message:
Method error (/etc/method/cfgsctape): 0514-047 Cannot
access a device.
Simply have the student power on the device and try the instruction
again.

None.

V4.1

Preface
classroom. If your lab environment uses remote System p/AIX platforms, the instructor
will provide you with the addressing (IP address or hostname) and instructions for
accessing the remote machine. For example, you might be using telnet, an X-windows
server, or a VNC client. Whatever tool you are using, at various points in the labs you
may need to re-launch your remote access tool to obtain a new login prompt.
Part 1 - List device configuration
__ 1. Log in as a regular user and obtain a command prompt. If using a graphical interface
(ex. CDE), then start a terminal emulation window.
__ 2. su to the root user. If more than one person is using the same system you need to
work as a team.
$ su root
__ 3. List all the supported devices on the system.

# lsdev -P -H | pg
- OR -
# smit
Devices ->
List Devices ->
List All Supported Devices
__ 4. List the configured devices on the system. Are your SCSI devices available?
____________________________________________________________
# lsdev -C -H | pg
- OR -

# smit
Devices ->
List Devices ->
List All Defined Devices
F3 to return to the menu.
__ 5. List information about the memory you have on the system. Record the amount of
memory you have.
# lsattr -E -l mem0
# lsattr -El sys0 -a realmem
-OR-
# smit
Devices ->
List Devices ->
Show Characteristics of a Defined Device
Cursor through the devices listed until you see the entries for memory. Select
one of the memory entries and press Enter.
__ 6. List the disks on your system. Record how many you have.
# lsdev -C -c disk
-OR-
# smit
Devices ->
Fixed Disk ->
List All Defined Disks

V4.1
EXempty Part 2 - List and change system parameters
__ 7. The kernel is represented by a special logical device, sys0. This can be

manipulated by commands or through SMIT.
List the attributes of the kernel. The only kernel attributes that can be changed are
those that are marked with True in the last column of output.
Which attributes cannot be changed? ________________________________
_________________________________________________________
# lsattr -E -l sys0 | more
-OR-
# smit
Change/Show Characteristics of Operating System
On our development system some of the values were:

keylock, realmem, conslogin.
Some PCI systems may also include:
fwversion, modelname, boottype, frequency, systemid.
LPAR systems may also have many additional unchangeable attributes which are
outside the scope of this class to discuss.
__ 8. Which attribute in the previous listing describes the number of processes a user can
run? _______________________________________
maxuproc
__ 9. By default, the system is configured so that each user can run no more than 128
processes at once. Change this value to 150 processes per user. The new value
takes effect immediately.
# chdev -l sys0 -a maxuproc=150
-OR-
# smit
Change/Show Characteristics of Operating System
Maximum number of PROCESSES allowed per user [150]
F10 to return to the command line.

Part 3 - Removing and redefining a network adapter device

Your machine must have an extra network adapter for which the corresponding
interface is not configured. Otherwise, skip to Part 5 - Changing the status of a disk
drive device.
__ 10. Identify the network interfaces that are configured (for example, available) on your
system by listing the if class of devices. Look for the names of the Standard Ethernet
Network Interfaces. Which of these interfaces are available and which ones are
defined?
______________________________________________________________
______________________________________________________________
# lsdev -Cc if
- OR-
# smit inet
List All Network Interfaces
On our development system, en0 was Available and en1 was Defined. An interface
is not made available until it is configured with an IP address.
__ 11. For the interfaces which are only defined, select one and list the adapter which the
interface is using. For example, if the interface is en4, then the corresponding
network adapter device name is ent4 (the numbers always match). What is the
status of the adapter?
_______________________________________________________________
In our hints throughout this part you refer to the Ethernet adapter that you
have identified as entx.
# lsdev -l entx
-OR-
# lsdev -Cc adapter | grep -i ethernet
- OR-
# smit ethernet
Adapter =>
List All Ethernet Adapters
On our development system ent1 was Available.

V4.1
EXempty __ 12. Put the Ethernet adapter device into a Defined state. Be careful to specify the
correct adapter and not the one with a configured interface. Display the adapter and
verify its status.
# rmdev -l entx
# lsdev -l entx
-OR -
# lsdev -Cc adapter | grep ent
(entx should be Defined)
there is no smit panel for removing an Ethernet adapter.
__ 13. Remove the Ethernet adapter from the customized database.

# rmdev -dl entx
# lsdev -l entx
-OR-
# lsdev -Cc adapter | grep ent
You should not see entx listed.
__ 14. Rediscover and configure the Ethernet adapter. Check its status.
# cfgmgr
# lsdev -l entx
(entx should be Available)
-OR-
# smit devices
Install/Configure Devices Added After IPL
Input device / directory for software [none]
OK or Enter.
F3 to go back to the Devices Menu.
List Devices ->
List All Defined Devices
entx should be Available.

Part 4 - Changing the status of a disk drive device
Your machine must have a an extra disk drive to continue with this section. Otherwise,
you are done with this exercise.
__ 15. Determine the logical names of your disk drives and the volume group they (if any)
that they are assigned to by using the lspv command. (We will be covering PVs and
VGs later in the course).
Which disks have a volume group of rootvg? ____________________________
Which disks have a volume group of none? _____________________________
# lspv
On our development system hdisk0 was in the rootvg and all other disks were listed
as none for the volume group column.
__ 16. Generate a device listing for the disk devices. Select one disk which was not in a
volume group and record the location of that disk. If it is SCSI attached, what is the
SCSI ID of the disk.
_______________________________________________________________
_______________________________________________________________
# lsdev -Cc disk
# lscfg | grep hdisk
-OR-
# smit
Devices ->
Fixed Disk ->
On our development system the disk hdisk1 was SCSI attached with an AIX location
code of 03-08-00-4,0. The physical location code ended in P1-T10-L4-L0.
__ 17. Using the disk selected above, put that disk device into a Defined state.
# rmdev -l hdisk1
# lsdev -Cc hdisk
(hdisk1 should be Defined)
__ 18. Remove the disk device from the customized database.

# rmdev -dl hdisk1

V4.1
EXempty # lsdev -Cc hdisk1

You should not see hdisk1 listed.
__ 19. Rediscover and configure that same disk for use. Does the device that appears
have the same location? _________________________________________
____________________________________________________________
# cfgmgr
# lsdev -Cc hdisk
# lscfg | grep hdisk
(hdisk1 should be Available)
-OR-
# smit devices
Install/Configure Devices Added After IPL
Input device / directory for software [none]
OK or Enter.
F3 to go back to the Devices Menu.
Devices ->
Fixed Disk ->
hdisk1 should be Available.
On our development system, the disk that reappeared with the same logical device
name also had the same location code. This is not always the case; never assume
that it is the same device just because a device name is the same.

Part 5 - Configure a tape device (optional)
If you have an externally attached tape drive, continue with this section. Otherwise, skip
to Part 4 - Configure a CD-ROM device.
__ 20. Issue the shutdown command to bring your system down. If you have more than one
user on your system, coordinate this task with the other users.
# shutdown
-OR-
# shutdown -F
__ 21. Turn off the tape drive. Reboot the system.

Push the tape drive on/off switch to off.
Power off the system unit, wait 15 seconds and then power it back up.
__ 22. Log in and su to root. Check the status of the tape drive.
Is it in a state where it can be used? ____ No
Why? _______________________________ Because the system was not able
to communicate with it during the reboot.
Record the name of the tape drive.
# lsdev -C -c tape
-OR-
# smit
Devices ->
Tape Drive ->
List All Defined Tape Drives
__ 23. Put the tape drive into the available state so it can be used. Use the name for the
tape drive that you recorded in the previous instruction.
Power on the tape drive.
# mkdev -l rmtx (where x stands for the drive recorded in the previous instruction)
-OR-
Power on the tape drive.
# smit

V4.1
EXempty Devices ->

Tape Drive ->
Configure a Defined Tape Drive
Select the appropriate drive and press OK or Enter.
End of exercise

1. Review the commands that list information on configured devices.
Answer: lsdev, lscfg, lsattr
2. Review the differences between an available device and a defined device.
3. Ask students what different ways could be used to make a defined device available.
Answer: You could turn on the device then do a shutdown and a reboot or you could
turn on the device and go into SMIT under devices and select Install/Configure
Devices Added After IPL.
4. Ask the students what could make a device become defined.
Answer: Either the device was removed from the system, it was turned off during the
last system reboot, it is broken, or another device was added with the same physical
address.

V4.1
EXempty Exercise 8. LVM commands

(with hints)
Estimated time
00:15

This exercise gives you a chance to look at the file system
configuration on your machine and introduce you to some LVM
commands.

Display information about your file systems
List the LVM information on your system
Copyright IBM Corp. 1997, 2008 Exercise 8. LVM commands 8-1


Preface
classroom.
__ 1. Log in as team01 and su to root.

$ su
password: ibmaix
__ 2. View the contents of /etc/filesystems and list the file systems on your machine.
# more /etc/filesystems
__ 3. Execute the command: lsfs.

What does it show? _____________________________________________
The list of all defined file systems on the machine.
# lsfs
__ 4. Execute the command: lsvg.

What does it show? _____________________________________________
The list of all volume groups defined on the machines.
# lsvg
__ 5. Execute the command: lspv.

What does it show? _____________________________________________
The list of all defined physical volumes on the machine.
# lspv

V4.1
EXempty __ 6. Execute the command: lsvg -l rootvg.

What does it show?
A list of all the logical volumes contained in rootvg.
# lsvg -l rootvg
End of exercise
Copyright IBM Corp. 1997, 2008 Exercise 8. LVM commands 8-3


V4.1
EXempty Exercise 9. Working with LVM

(with hints)
Estimated time
Parts 1 and 2 - 00:20
Parts 3, 4, and 5 - 00:40

The purpose of this exercise is to give students the opportunity to work
with the facility that manages physical and logical data storage, the
Logical Volume Manager (LVM). The Systems Management Interface
Tool (SMIT) is used extensively throughout this exercise.

Add and remove a disk from a volume group
Create and remove a volume group
List the characteristics of the system's volume groups, physical
volumes, and logical volumes
Add a logical volume
Extend a logical volume
Introduction
In this exercise you work extensively with SMIT to work with the
Logical Volume Manager. First, you investigate existing volume
groups, physical volumes and logical volumes. Once you are
comfortable with the layout of your storage, you have the opportunity
to add a physical volume to a volume group and then create a logical
volume. Finally, you make some changes to the logical volume
characteristics. In many cases, both the SMIT screens and the
command are provided in the Exercise Instruction With Hints
section. All tasks shown can also be done using the graphical version
of SMIT.
Copyright IBM Corp. 1997, 2008 Exercise 9. Working with LVM 9-1

At this point, students are still trying to understand the differences
between the various parts of the logical volume structure. They often
get physical volumes and volume groups mixed up. A large part of this
exercise is designed to give students an opportunity to view on a real
system the various pieces of the LVM. This is meant to give them a
greater understanding of the concepts and prepare them for working
with logical volumes. File systems are covered in the next exercise.
Some students prefer to use the commands. For those students, if
they use commands that don't have the appropriate option for the
resource, they will get error messages that imply the LVM is locked or
corrupted.
An error message like: Unable to find _____ in the Device
Configuration database is usually the result of entering a name for
a logical volume when the command was expecting a physical volume
or vice versa.
An error message like: Unable to find attribute hdisk0 in the
Device Configuration Database. Execute synclvodm to attempt
to correct the database. is usually the result of typing in
something like: lsvg -l hdisk0 when the correct form of the
command is: lsvg -l rootvg
It is common for the lab System p platforms to have disks that were
part of a non-rootvg volume group prior to the mksysb installation and
still have a VGDA on them. When trying to either add these disks to an
existing or new volume group, the operation will fail unless the -f
(force) option is used.

If you have lab systems with only one disk drive, your students won't
be able to do the section on creating their own volume group. In that
case, they won't be able to complete the Optional Exercise of the
backup of a volume group in the backup/recovery exercise. Warn them
about this section when introducing this exercise.
If you have more than one student per machine and they try to use the
LVM simultaneously, then students will get locking error messages.

V4.1

Preface
classroom.
Part 1 - Adding and removing a disk from a volume group

__ 1. Log in as teamxx and switch to the root user account.
$ su
__ 2. Ensure you have a hard disk available to create a new volume group.
What is the name of the disk that is currently available? _______________
# lspv
__ 3. From the disk information that you determined in the step above, add a free disk into
rootvg.
Note
You need to substitute your free disk name when hdiskn is referenced.
# smit vg
Set Characteristics of a Volume Group ->
Add a Physical Volume to a Volume Group ->
FORCE the creation of volume group? yes

VOLUME GROUP name [rootvg]
PHYSICAL VOLUME names [hdiskn]
Then, use F10 to exit from SMIT.
-OR-
# extendvg -f rootvg hdiskn
__ 4. Check to see if the disk is now associated with rootvg.

# lspv
__ 5. Take the same disk that you added to rootvg, out of rootvg. Make sure you remove
the correct disk.
# smit vg
Remove a Physical Volume from a Volume Group ->

FORCE deallocation of all partitions on no
this physical volume?
Then, use F10 to exit from SMIT.

-OR-
# reducevg rootvg hdiskn
__ 6. Verify that the disk is no longer associated with rootvg.

# lspv

V4.1
EXempty Part 2 - Creating and removing a volume group
__ 7. Using the free disk, create a new volume group called newvg.
# smit vg
Add a Volume Group ->
Add an Original Volume Group
VOLUME GROUP name [newvg]

Physical partition SIZE in megabytes 16
Force the creation of a volume group Yes
Then, use F3 to return to the Volume Groups menu.
-OR-
# mkvg -f -y "newvg" -s 16 hdiskn
__ 8. Verify that the new volume group was created.

What are the maximum number of physical volumes that can be added to this
volume group? _________________________________________________
Select List All Volume Groups
List only the ACTIVE volume groups? no
Use F3 to return to the Volume Groups menu.
Select List Contents of a Volume Group
List OPTIONS status
Use F3 to return to the Volume Groups menu.
-OR-
# lsvg
# lsvg newvg
Look at the Max PVs field in the output to get the maximum number of physical
volumes.
When creating a volume group, if a particular physical partition size is specified

and it is not large enough to address the all the storage in the largest disk in the
volume group using the default 1016 PPs/PV, then the mkvg command will
automatically make the -t factor large enough to allow a sufficient number of
PPs/PV. In this exercise, you have specified a PPsize of 16 MB. Lets say you
have a (nominally) 34 GB hard drive. 1016 x 16 MB only allows you to address

about 16.25 GB. If you increase the -t factor to 2, then you can address only 32.5
GB which is still not enough. In fact, the mkvg command would set the -t factor
equal to 3 which allows up to 3048 PPs/PV and can then address up to 48.75
GB. Remember, there is a trade off in the number of PVs/VG when increasing
the -t factor. if you had a -t factor of 1, you would be able to include 32 PPs in the
normal VG. By using a -t factor of 3, it has to divide that by the factor. This results
in a maximum of only 10 PPs. The numbers change depending on the size of
your disks, but the principle is the same.
__ 9. Convert newvg into a big volume group.

In SMIT, select:
Change a Volume Group
Change to big VG format yes
-OR-
# chvg -B newvg
Answer y to the prompt, if asked.
__ 10. What is the maximum number of physical volumes that can be added to the
converted newvg? _________________________________________________
List OPTIONS status
-OR-
# lsvg newvg
volumes.
The trade-off between PPs/PV and PVs/VG which you saw in step 8 is the same
here, except that in a big VG the default number of PVs/VG with a -t factor of 1 is
128. So a -t factor of 3 would result in 128/3 which would allow 42 PPs in our big
volume group, given the example 34 GB hard drive.

V4.1
EXempty __ 11. Convert newvg into a scalable volume group. In order to make this change the
volume group must first be varied offline. Vary it back online when the conversion is
completed.
varyoffvg newvg
In SMIT, select:
Change a Volume Group
Change to big scalable format yes
-OR-
# chvg -G newvg
Answer y to the prompt, if asked.
# varyonvg newvg
__ 12. What is the maximum number of physical volumes that can be added to the scalable
format newvg? _________________________________________________
List OPTIONS status
-OR-
# lsvg newvg
volumes.
The scalable VG has even more space in the VGDA to manage its constituent PVs.
And the management of the PPs is no longer on a PV by PV basis. Instead the PPs
are managed globally within the VG. You notice that there is no maximum number of
PPs per PV characteristic in a scalable VG.
__ 13. Remove the newvg volume group.
There are two Volume Groups menu items in SMIT that let you remove a volume
group. You can select Set Characteristics of a Volume Group and remove the
disk from the volume group. The volume group is automatically deleted when the
last disk is removed. Since there is only one disk in the newvg volume group,
removing that disk removes the volume group. The other menu choice of removing
the VG is obvious; yet, under the covers it is only doing a series of reducevg
requests until the last disk has been removed. Use this second choice.
Select Remove a Volume Group

Then, use F10 to exit SMIT.
-OR-
# reducevg newvg hdiskn
Note
There is no such command as rmvg.
__ 14. Verify that hdiskn is not associated with any volume group. If it is not, repeat the
steps above. You need to have this disk free for a later exercise.
# lspv

V4.1
EXempty Part 3 - Exploring your storage environment
__ 15. Log in as teamxx and su to root, if not already.

$ su
__ 16. List all of the system's volume groups. If you use SMIT, before pressing Enter for the
output, try using the F6 key to see the command that is being built.
# smit
System Storage Management (Physical & Logical Storage) ->
Logical Volume Manager ->
Volume Groups ->
List All Volume Groups
List only the ACTIVE volume groups? [no]

Pressing the F6 key shows: /usr/sbin/lsvg
Click OK or press Enter
-OR-
# lsvg
__ 17. List the contents of the rootvg volume group. If you are using SMIT, for List
OPTION use status.
Find the information that is necessary to answer the following questions about your
environment:
F3 to go back to the Volume Groups menu.
-> List Contents of a Volume Group
List OPTION status
-OR-
# lsvg rootvg
a) How many physical volumes are there? ______________________________

Check the field: TOTAL PVs.
b) How many logical volumes are there? _______________________________
Check the field: LVs.
c) How many total physical partitions are there for this volume group?_________
Check the field: TOTAL PPs.
d) How many of those physical partitions are free? ________________________
Check the field: FREE PPs.
e) How large are the physical partitions? _______________________________
Check the field: PP SIZE.
f) What is the Logical Track Group size? _______________________________
Check the field: LTG size.
g) What is the maximum number of PPs per physical volume? _______________
Check the field: Max PPs per PV.
__ 18. List the contents of the rootvg volume group once again this time selecting
physical volumes for the List OPTION. Note: the free distribution of physical
partitions is shown relative to the position on disk. These positions represent outer
edge, outer middle, center, inner middle, inner edge. Make a note of the free
partitions and their position on disk. ______ ______ ______ ______ ______
F3 to go back to the List Contents of a Volume Groups menu.

List OPTION physical volumes
-OR-
# lsvg -p rootvg
__ 19. List all of the physical volumes on the system.

F3 (three times) to go back to the Logical Volume Manager menu.
Physical Volumes ->
List all Physical Volumes in System
-OR-
# lspv
__ 20. List the contents of any physical volume on your system. If you are using SMIT, for
List OPTION use status.
environment:

V4.1
EXempty a) To which volume group does the physical volume belong? _______________
Check the field: VOLUME GROUP.
b) What is the physical partition size? _________________________________
Check the field: PP SIZE.
c) How many total physical partitions are there? __________________________
Check the field: TOTAL PPs.
d) How many of those physical partitions are used? _______________________
Check the field: USED PPs.
e) What is the maximum transfer size? _________________________________
Check the field: MAX REQUEST.
Try other options for listing information about the physical volume you select.
F3 to go back to the Physical Volumes menu.
List Contents of a Physical Volume
PHYSICAL VOLUME name [hdisk#]

List OPTION status
-OR-
# lspv hdisk#
(where # is the number for some disk on your system)
You find that we can only examine the contents of a physical volume if that PV is
assigned to a volume group. Thus, for our current lab situation, the disk which is
assigned to the rootvg is the only available choice.
__ 21. List all of the logical volumes in the rootvg volume group.
Logical Volumes ->
List all Logical Volumes by Volume Group
Note
This will list all volume groups. You will have to look through the list for rootvg, then
note the logical volumes for rootvg.
-OR-
# lsvg -l rootvg
__ 22. Show the characteristics for the logical volume hd4. If you are using SMIT, for List
OPTION use status.
environment:
a) To which volume group does it belong? ______________________________
Check the field: VOLUME GROUP.
b) How many copies? ______________________________________________
Check the field: COPIES.
c) How many physical partitions is it? __________________________________
Check the field: PPs.
d) How large is the logical volume in megabytes? _________________________
This is calculated: PP SIZE * PPs.
F3 to go back to the Logical Volumes menu.

-> Show Characteristics of a Logical Volume
LOGICAL VOLUME name [hd4]

List OPTION status
-OR-
# lslv hd4

V4.1
EXempty Part 4 - Adding a volume group

The completion of this section depends on having an available disk drive in your system
that is not assigned to a volume group. Check for a disk drive that has not been assigned to
a volume group (you can use the lspv command).
If all disk drives are assigned, continue with the next section Adding a logical
volume.
Important
Only one user per AIX system may complete this section. Please consult your
teammates before continuing.
__ 23. Only one physical volume has been defined to be a member of the rootvg volume
group. Using an available physical volume located on your system, create a new
volume group called datavg. Verify that the new volume group was successfully
created.
Volume Groups ->
Add a Volume Group ->
VOLUME GROUP name [datavg]

Physical partition SIZE in megabytes 16
PHYSICAL VOLUME names [hdisk1]
FORCE the creation of volume group? yes
Activate volume group AUTOMATICALLY yes
at system restart?
-OR-
# mkvg -f -y datavg -s 16 hdiskx (where x is number of the physical volume
that has not been assigned to a volume group)
To verify that your volume group has been created:

F3 (three times) to go back to the Volume Groups menu.
List All Volume Groups ->
List only the ACTIVE volume groups? no
-OR-
# lsvg
Part 5 - Adding a logical volume
__ 24. Add a logical volume, which is one partition in size, to the datavg. (Note: If you only
have one disk on your system, create the logical volume in rootvg.) Allocate the
partitions on the outer edge of the disk. Allow the system to default the logical
volume name and note the system generated name.
__________________________________
Logical Volumes ->
Add a Logical Volume
Logical volume NAME []
VOLUME GROUP name datavg
(or rootvg if you didn't add a new volume group)
Number of LOGICAL PARTITIONS [1]
PHYSICAL VOLUME names []
Logical volume TYPE []
POSITION on physical volume outer_edge
Note
Depending on your AIX version, this may show as edge.
...
-OR-
# mklv -a e datavg 1
__ 25. Rename the logical volume created above to yylv (where yy represents the initials
of your first and last name) instead of the system generated one, and verify that it
has been renamed.
F3 (two times) to go back to the Logical Volumes menu.
Set Characteristic of a Logical Volume ->
Rename a Logical Volume
CURRENT logical volume name [lv00]
NEW logical volume name [yylv]
-OR-
# chlv -n yylv lv00
To verify that your volume group has been created:

V4.1
EXempty F3 (three times) to go back to the Logical Volumes menu.

Select List All Logical Volumes by Volume Group
__ 26. Extend the yylv logical volume by one partition and specify that the added partition
be allocated on the outer edge of the disk. This logical volume is used in the next
exercise.
-> Set Characteristic of a Logical Volume
-> Increase the Size a Logical Volume
LOGICAL VOLUME name [yylv]
Number of ADDITIONAL logical partitions [1]
POSITION on physical volume outer_edge
Note
Depending on your AIX version, this may show as edge.
Notice that this is already selected because it is the current location of the existing
PPs in the LV.
-OR-
# extendlv yylv 1
End of exercise
1. Review each of the LVM terms with the students being sure they understand the
relationships between them.
2. Ask the students how they can tell whether a logical volume is being mirrored.
Answer: By the number of copies there are from the status listing of a logical volume.
3. Ask the students if they know the number of PPs in a logical volume and if the PP size
can you tell what the size (in MB) the logical volume is. Then ask how.
Answer: Yes, by multiplying the number of PPs times the PP size.

V4.1
EXempty Exercise 10. Working with file systems

(with hints)
Estimated time
00:20 Part 1
00:40 Parts 2 - 6

In this exercise the students are given the opportunity to work with file
systems.

Add and remove an AIX file system
Create an AIX file system specifying the number of bytes per inode
Create an enhanced JFS (JFS2) file system and illustrate the
AIX 6.1 file system shrink capability
Mount and unmount a file system
Extend a file system
Add and remove mirrors
Introduction
In this exercise, you are using SMIT extensively to work with file
systems. You add a file system to the logical volume you created in the
previous exercise and then mount it so users can access the storage
space. You copy files into this file system and then experiment with
unmounting and mounting the file system once again but in different
locations. You also get an opportunity to remove a file system and
create a file system. The last portion of the exercise provides exposure
to creating a mirrored file system and then changing some
characteristics of the mirrored file system.
Copyright IBM Corp. 1997, 2008 Exercise 10. Working with file systems 10-1

Often students try to unmount file systems and they have a process
running that is using that file system. This is particularly true when a
student uses the su command extensively to go from teamxx to root
back to teamxx again instead of using exit or <ctrl-d>.
WIth the df command, the default output is in 512 byte blocks which
some students have trouble translating into KB, MB, or GB. The
df command has flags to display the output in KB (with df -k),
MB (with df -m), and GB (with df -g). Because the SMIT panels
display only in blocks of 512 bytes, you may want to recommend the
command line hints for the displaying of information about the file
systems.
If you are in a situation where you have several students working on
terminals attached to a single system, you can run into locking error
messages with the LVM. This usually is not a problem, the student
simply needs to wait his/her turn. However, it could be a problem if
someone does a <ctrl-c> when in the middle of an LVM operation.
Locked files may not get released and the LVM could be locked
indefinitely. If this happens, try using the following command:
chvg -u VGname
This command is used to unlock a volume group. A volume group can
become locked when an LVM command terminates abnormally. This is
likely when the system crashes while an LVM operation is in process
or if an LVM command causes a core dump.
Some students prefer to use the commands. For those students, if
they use commands that don't have the appropriate option for the
resource, they will get error messages that imply the LVM is locked or
corrupted. An error message like:
Unable to find...
is usually the result of entering a name for an LV when the command
was expecting a physical volume or vice versa. An error message like:
Unable to find attribute hdisk0 in the Device
Configuration Database. Execute synclvodm to attempt to
correct the database.
is usually the result of typing in something like:
lsvg -l hdisk0
when the correct form of the command is:
lsvg -l rootvg

V4.1
EXempty Since, in the previous exercise, you used the second disk drive to
create a volume group, if students don't change the specification to
spread across multiple drives when they go to turn mirroring on, they
get a message that indicates there are not enough resources to satisfy
their request. They simply need to modify the appropriate field in SMIT
and try again.
In the first step in Part 6, if the student does not properly complete
reducevg for hdisk1 on the datavg, then the VGDA is still on the disk.
Attempting to specify that disk in the following extendvg would then
fail. First be sure the reducevg is properly completed. If necessary you
may use the -f (force) option on the extendvg command.

Preface
classroom.
Part 1 - Inodes and NBPI
__ 1. If you are creating a JFS file system intended to store files that are smaller than 512
bytes in size, what would you choose for the following values?
Fragment Size: _____ 512 NBPI: _____ 512
Regular or Large-file enabled: __________ Regular
__ 2. If you are creating a JFS file system intended to store files about 8 KB in size, what
would you choose for the following values?
Fragment Size: 4096 NBPI: 4096
__ 3. If you are creating a JFS file system intended to store files about 2 MB in size, what
would you choose for the following values?
Fragment Size: 4096 NBPI: 4096
__ 4. When should you use large file-enable file systems? _______________________

________________________________________________________________
Large-file enabled file systems are intended for files larger than 4 MB. If a large-file
enable file system is used for files less than 4 MB in size, the file will be stored the
same way as in a regular file system.

V4.1
EXempty __ 5. When should you use compressed file systems? __________________________

________________________________________________________________
Compressed file systems are good for JFS online archives and files that are not
accessed very often. Performance is significantly lower on compressed file systems
versus non-compressed file systems.
Part 2 - Creating a journaled file system
__ 6. Log in to the system and su to root. In order to use the space we set aside when we
created the logical volume in the previous exercise, we need to put a file system in it.
Create a journaled file system on the logical volume you created in the previous
exercise. Make the default mount point /home/teamxx/dir1 where the xx represents
your team number. (Wherever you see yylv substitute your initials for yy.) Request
that the filesystem be mounts at system restart.
$ su
# smit
System Storage Management (Physical & Logical Storage)->
File Systems ->
Add / Change / Show / Delete File Systems ->
Journaled File Systems ->
Add a Journaled File System on a Previously Defined Logical Volume ->
Add a Standard Journaled File System
* LOGICAL VOLUME name yylv
(where yy represents your initials)
* MOUNT POINT [/home/teamxx/dir1]
Mount AUTOMATICALLY at system restart? yes
OK or Enter.
-OR-
# crfs -v jfs -d yylv -m /home/teamxx/dir1 -A yes -p rw
__ 7. At this moment, can the space be used to create an AIX file?

_____________________________________________________________
_____________________________________________________________
No, because the file system has not yet been mounted. A file system has to be
mounted for it to be available to users.
Mount the file system. Verify that it is mounted. Check the owner and group of the
mount point. Change the owner and group of the mount point to teamxx and staff.
Copy all files and directories that are in the /usr/samples directory to the
/home/teamxx/dir1 directory.
F3 to go back to the File Systems menu
List All Mounted File Systems
Mount a File System
FILE SYSTEM name [/dev/yylv]

V4.1
EXempty OK or Enter.
F9 to launch a sub-shell (or open a window if in AIXWindows).
# ls -ld /home/teamxx/dir1
# chown teamxx.staff /home/teamxx/dir1
# cp -r /usr/samples /home/teamxx/dir1
# exit (to return to SMIT)
-OR-
# mount
# mount /home/teamxx/dir1
# mount
# ls -ld /home/teamxx/dir1
__ 8. Create an enhanced journaled file system in the rootvg and let the system create
the logical volume for you. Have it mount automatically at system restart. Set its
default mount point to /home/teamxx/dir2. First try to create it with a size of 1 MB.
If that fails, create the file system with a size of 16 MB.
F3 to go back to the File Systems menu.
Add/Change/Show/Delete File Systems ->
Enhanced Journaled File Systems ->
Add a Enhanced Journaled File System
Volume group name [rootvg]
SIZE of file system
Unit Size [Megabytes]
Number of Units [1]
*MOUNT POINT [/home/teamxx/dir2]
OK or Enter.
-OR-
#crfs -v jfs2 -g rootvg -a size=1M -m /home/teamxx/dir2 -A yes
__ 9. Check to see if it is mounted. It shouldn't be. Mount it. Verify that it is mounted.
Change the owner and group of the mount point to teamxx and staff.

Mount a File System
FILE SYSTEM name [/dev/lvxx]
(where xx is the system generated LV number)
OK or Enter.
Exit SMIT.
-OR-
# mount
# mount
__ 10. Add several files to the /home/teamxx/dir2 directory. (Note: A quick way to do this
is through the use of the touch command.)
# cd /home/teamxx/dir2
# touch file1 file2 file3 file4 file5 file6 file7
__ 11. List the files in the /home/teamxx/dir1 directory. Unmount /home/teamxx/dir1 and
list the files again. (Note: If you aren't root then su to root)
# ls -a
# cd ..
# umount /home/teamxx/dir1
# ls -a dir1
__ 12. List the files in the /home/teamxx/dir2 directory. It should have all the files you
created in it from instruction 5. Then mount yylv on /home/teamxx/dir2 and list the
files in /home/teamxx/dir2.
# ls -a dir2
# mount /dev/yylv /home/teamxx/dir2

V4.1
EXempty # ls -a dir2
# mount
What files are there now? _________________________________________
They are the files which were in the yylv logical volume which can be mounted at
any directory on the system.
Where did they come from? _______________________________________
These are the files that were copied from the /usr/samples directory.
What happened to the files that were originally seen in /home/teamxx/dir2?
_____________________________________________________________
The files that were originally seen in /home/teamxx/dir2 are still on the system,
they just can't be accessed at this time because the file system pointers are
pointing to the files in the yylv logical volume.
__ 13. Unmount /dev/yylv. Mount /home/teamxx/dir1. List the files in both

/home/teamxx/dir1 and /home/teamxx/dir2 to be sure they are set up as originally
created.
# umount /dev/yylv
# ls -a dir2
# ls -a dir1
# ls -a dir2
Part 3 - Changing the file system size
__ 14. Find out the names of the logical volumes on which the file systems are created.
(Hint: Use the Mount Point column in the output from the logical volume listing to
find the name of the logical volume.) Find the logical volumes the file systems are on
and the physical partitions in each file system and write it down.
File system: ____________________ /home/teamxx/dir1

Logical volume: ________________ yylv (where yy represents your initials)
Size: ___________________ Two physical partitions
File system: _____________________ /home/teamxx/dir2

Logical volume: ________________ fslv00
Size: _________________ One physical partition
Note that the two file systems are in two different volume groups which might
have different PPsizes.
# smit
Logical Volumes ->
List All Logical Volumes by Volume Group
-OR-
# lsvg -o | lsvg -i -l
# df -m
__ 15. Record the current size of the filesystem.

____________________________________________________________
If you are using SMIT, this will be displayed on the panel you will use
to extend the size in the next step.
-OR-
# lsfs /home/teamxx/dir1
__ 16. Extend the size of /home/teamxx/dir1 file system by ten 512-byte blocks.
What is the new size of the file system? ___________________

V4.1
EXempty How much did the file system actually get extended? ____________________
The file system actually got extended by the PPsize of your volume group.
Why didn't the size increase by 5,120 bytes? __________________________
Because the system allocates disk space in increments of physical partitions.
Did this change the size of the logical volume with which it is associated?____
Yes, it changed the size of the logical volume because in order for the file system
to expand, the logical volume has to expand as well.
F3 to go back to the System Storage Management menu.
File Systems ->
Change / Show Characteristics of a Journaled File System
Select the correct mount point from the selector screen.
SIZE of file system
Unit Size [512 bytes]
Number of Units [+10]
OK or Enter.
Logical Volumes ->
List All Logical Volumes by Volume Group
-OR-
# chfs -a size=+10 /home/teamxx/dir1
#lsvg -o | lsvg -i -l
# df -m
__ 17. Extend the size of the yylv logical volume by one physical partition.
What is the size of the logical volume? _______________________
What is the size of the file system? __________________________
Did this change the size of the file system? _____ No.
Why not? ______________________________________________________
______________________________________________________________
Extending the logical volume does not automatically extend the size of the file
system. However, in order for users to get access to that space, the file system
needs to then be expanded to fill the rest of the expanded logical volume.
Make it so the file system takes up all the space allocated to the logical volume.
F3 to go back to the Logical Volume Manager menu.
Logical Volumes ->
Increase the Size of a Logical Volume
Number of ADDITIONAL logical partitions [1]
OK or Enter.
F3 to go back to the System Storage Management menu
File Systems ->
Journaled File System ->
Change/Show Characteristics of a Journaled File System
File System Name [/home/teamxx/dir1]
OK or Enter.
-OR-
# extendlv yylv 1
# lsvg -l datavg
# df -m
At this point, your logical volume should be one physical partition larger in size than
the size of the file system. Increase the size of the file system by one 512-byte block.
The system rounds up to the next physical partition boundary.
SIZE of file system
(This will expand the file system by 1 logical partition and thus to the size of the
logical volume.)
-OR-
# df

V4.1
EXempty Part 4 - Reducing the size of a file system
__ 18. Change the characteristics of the /home/teamxx/dir1 standard file system to reduce
the size down to a final size of 1 megabyte.
Was this successful? ____________________________________________
This attempt failed because traditional journal file systems can only be shrunk
though a backup and restore procedure.
File Systems ->
Change / Show Characteristics of a Journaled File System
SIZE of file system
Number of Units [1]
OK or Enter.
-OR-
# chfs -a size=1M /home/teamxx/dir1
__ 19. Change the characteristics of the /home/teamxx/dir2 enhanced journal file system
to increase the size by one 512 byte block.
What is the final size resulting from this? _____________________________
You will note that the file system increased by a full Physical Partition, since this
is the unit of allocation used for obtaining disk space from the volume group.
File Systems ->
Enhanced Journaled File System ->
Change / Show Characteristics of an Enhanced Journaled File System
SIZE of file system
OK or Enter.
-OR-
# df -m
__ 20. Change the characteristics of the /home/teamxx/dir2 enhanced journal file system
to reduce the size down to a final size of 16 megabytes. Was this successful?
________________________________________________________________
The dynamic reduction in size is supported by the enhanced journal file system, but
it will only reduce the size down to the nearest PPsize boundary and no less than
the 16 MB minimum allocation for JFS2.
File Systems
Add / Change / Show / Delete File Systems
Enhanced Journaled File System
SIZE of file system
Number of Units [16]
OK or Enter.
-OR-
# chfs -a size=16M /home/teamxx/dir2
# df -m

V4.1
EXempty Part 5 - Removing a file system
__ 21. Remove the yylv logical volume, the one with /home/teamxx/dir1 as the mount
point. (where xx is the number of the logical volume assigned by the system when
you created the file system in Step 3.
Can it be done? ________________________________________________
No. Because the file system related to the logical volume is mounted.
Logical Volumes ->
Remove a Logical Volume

OK or Enter.
-OR-
# rmlv -f yylv
__ 22. Unmount /home/teamxx/dir1 and remove the yylv logical volume.

Does this remove all references to the logical volume and its mount point? ___
No, because the /etc/filesystems file still has the file system stanza in it which
includes reference to the logical volume and the mount point.
File Systems ->
Unmount a File System
NAME of file system to unmount [/home/teamxx/dir1]

OK or Enter.
Logical Volumes ->
Remove a Logical Volume

F3 to go back to the System Storage Management menu
File Systems ->

File System Name [/home/teamxx/dir1]
OK or Enter.
-OR-
# unmount /home/teamxx/dir1
# rmlv -f yylv
# cat /etc/filesystems
__ 23. Ensure that all references to the yylv logical volume and its mount point have been
removed by removing the file system.
Do you see any potential problems with removing a logical volume and then
removing the file system? _________________________________________
_____________________________________________________________
Yes, because if you remove the logical volume and forget to either remove the
entry in the /etc/filesystems file or remove the file system through SMIT, you
could create another file system that uses the same logical volume name
(especially if you allow the system to create the logical volume names) and then
remember at a later date that you needed to remove the old file system. If you
then do the remove operation through SMIT it uses the high level command rmfs
which also removes the logical volume. The logical volume it would be removing
is the new one which is not what you really wanted to do. For most situations, it is
best to remove the file system and let the system remove the logical volume for
you rather than removing the logical volume first and relying on remembering to
remove the file system right afterwards.
File Systems ->
Remove a Journaled File System
FILE SYSTEM name /home/teamxx/dir1
Remove Mount Point yes
OK or Enter.
-OR-
# rmfs -r /home/teamxx/dir1
# cat /etc/filesystems

V4.1
EXempty Part 6 - Working with mirrors
__ 24. Normally you would mirror across multiple disks. If your system does not have two
disks in a volume group, this will not be possible, and, for this lab, you should place
both copies on the same disk. If your system has two disks but one of the disks is in
a second volume group then you may wish to remove that volume group and extend
the rootvg to include the available disk.
# reducevg datavg hdisk1
The physical volume being reduced from the volume group must not have any
logical volumes allocated on it. It is recommended that you identify the logical
volumes (lsvg -l datavg), and either delete any remaining logical volumes (rmlv)
or move them (migratepv) to a physical volume that remains in the volume group.
You may use the -d flag for reducevg to automatically delete any remaining logical
volumes.
# extendvg rootvg hdisk1
__ 25. Change the characteristics of the fslvxx logical volume so that mirroring is turned on
to have a single mirror (two copies). Be sure to allow creating a mirror on the same
disk if you do not have two drives in your volume group. Other than that, take all the
defaults.
F3 back to the Logical Volumes menu.
Add a Copy to a Logical Volume
*LOGICAL VOLUME name [fslvxx]
*NEW TOTAL number of logical partition copies [2]

PHYSICAL VOLUME names [hdisk1]
(If you do not have two drives, leave it at the default of hdisk0.)
Allocate each logical partition copy on a SEPARATE physical volume?
[yes]
(If there is only one physical volume in the volume group, you must choose no.)
OK or Enter.
-OR-
# mklvcopy -s y fslvxx 2 hdisk1
(if there is only 1 PV in the VG, code -s n instead)
__ 26. Show the characteristics of the fslvxx logical volume.
Can you tell that it is mirrored? ________________________ Yes

What information indicates it is mirrored? _____________________________
You can tell it is mirrored by looking at the number of copies. If there is only one
copy, it is not mirrored. A two or three indicates either single or double mirroring.
What is the LV STATE? __________________________________ Stale
Why do you think it is marked stale? _________________________________
The exercise did not have you synchronize your copies when you turned
mirroring on.
What is the scheduling policy for writes? _______________________ Parallel
F3 back to the Logical Volumes menu.
Show Characteristics of a Logical Volume
LOGICAL VOLUME name [fslvxx]
List OPTION [status]
OK or Enter.
-OR-
# lslv fslvxx
__ 27. Change the characteristics of your mirrored logical volume to do sequential writes.
Synchronize the copies. Fix the problem based on the message you receive and try
again. Be sure it is mounted before continuing to the next instruction.
Try changing the write policy:
F3 back to the Logical Volumes screen.
Change a Logical Volume

SCHEDULING POLICY for writing logical [sequential]
partition copies
OK or Enter.
To fix the problem:
F3 back to the System Storage Management menu.
File Systems ->
Unmount a File System

OK or Enter.

V4.1
EXempty Try changing the write policy again:

Logical Volumes ->
Change a Logical Volume

SCHEDULING POLICY for writing logical [sequential]
partition copies
OK or Enter.
To synchronize the copies:
Press F9 to obtain a sub-shell
# syncvg -l fslvxx
# exit (to return to SMIT)
-OR-
Try changing the write policy:
# chlv -d s fslvxx
To fix the problem:
Try changing the write policy again:
# chlv -d s fslvxx
To synchronize the copies:
# syncvg -l fslvxx
__ 28. Show the characteristics of the fslvxx logical volume.

What values have changed? _______________________________________
The STALE PPs field should now have a value of 0 and the SCHED POLICY
field should show sequential.
Show Characteristics of a Logical Volume
* LOGICAL VOLUME name [fslvxx]
List OPTION [status]
OK or Enter.
-OR-
# lslv fslvxx
__ 29. Turn off mirroring on the fslvxx logical volume.

Set Characteristics of a Logical Volume ->
Remove a Copy from a Logical Volume
*LOGICAL VOLUME name [fslvxx]

*NEW maximum number of logical partition [1]
copies
OK or Enter.
-OR-
# rmlvcopy fslvxx 1
__ 30. Remove the, now unused, second disk from the rootvg.
Logical Volumes ->
Remove a Physical Volume from a Volume Group->
*VOLUME GROUP name [rootvg]
*PHYSICAL VOLUME name [hdisk1]
-OR-
# reducevg rootvg hdisk1
__ 31. Define the available disk as the volume group datavg.
Volume Groups ->

Add A Volume Group->
*VOLUME GROUP name [datavg]
Physical partition SIZE in megabytes [16]
*PHYSICAL VOLUME names [hdiskx]
(where x is the number of the physical volume that has not been
assigned to a volume group)
Force the creation of a volume group? yes

V4.1
EXempty ACTIVATE volume group AUTOMATICALLY

at system restart? yes
OK or Enter.
-OR-
# mkvg -f -y datavg -s 16 hdiskx (where x is number of the physical volume
that has not been assigned to a volume group)
End of exercise
JFS2 requires a minimum allocation of 16 MB. If the volume group has a PPsize which is
equal to or greater than this minimum, then a 1 MB request would result in a large enough
allocation to satisfy the requirement. If the PPsize is less than 16 MB, then a request for 1
MB would result in an allocation that is less than the requirement and thus the allocation
would fail.
1. Ask students what the difference is between adding a file system to a previously
created logical volume and simply adding a file system letting the system create the
logical volume. Why would you want to do it one way versus the other?
Answer: When you create a logical volume you have the ability to specify exactly
where on the disk and which disk within a volume group you wish to use. Thus, you
have some control on positioning for performance or availability. If you let the system
create both the file system and logical volume at the same time, you lose any control
over where the file system gets placed on disk.
2. Ask students what happens to files in a directory if you mount another logical volume
over it.
Answer: The files are still there, you just can't get to them because the file system
pointers now take you to the contents of a different location on disk.
3. Ask students how they think mirroring affects file system performance.
Answer: On writes it would degrade file system performance because two write
operations have to be scheduled. If the scheduling policy is sequential, that would slow
things down as well. On reads, it could improve performance because the quickest copy
to get to is read first.
4. Ask students what the discovered in the Removing A File System section.
Answer: They should have discovered that just because the logical volume has been
removed it doesn't mean that all references to the logical volume and the file system
have been removed as well.
5. Ask students what happens to a file system and all its data when you remove the logical
volume.
Answer: The file system and all its data gets deleted too.
6. Ask then if all data gets deleted, when might you want to remove a logical volume.
Answer: If a file system gets broken, you might wish to remove it, recreate it and then
restore from backup. You would remove it when the contents of it are no longer needed
or you might remove it when you are trying to relocate it to another volume group after
you have already created a new one and copied the necessary data into it.

V4.1
EXempty Exercise 11. Managing file systems

(with hints)
Estimated time
00:25

The purpose of this exercise is to give students the opportunity to
understand more about managing file system space and dealing with
file system problems.

Determine which directories and files are using the greatest
amount of file system space
Defragment a file system
Verify a file system using fsck
Use disk fragments to improve disk usage efficiency
Use JFS compression
Introduction
In this exercise you perform many of the activities required of a system
administrator to manage disk space. Most of these activities require
use of commands on a command line. This exercise requires the
successful completion of the last two exercises; Working with LVM and
Working with File Systems.
Throughout this exercise you are asked to change several times from
your regular user name, teamxx, to root and then back again. It is a
good idea to use exit to return to the previous user after completion of
a task by the current user instead of continually executing the su
command.
Copyright IBM Corp. 1997, 2008 Exercise 11. Managing file systems 11-1
In order to do Part 2 - Using Fragments For Disk Usage Efficiency and

Part 3 - Using JFS Compression, it is assumed that you have two disk
drives in your system and that you have completed Exercise 10 -
Working with LVM, Part 4 - Adding A Volume Group.

The students are asked to su several times from their regular user
name to root and back again. The Exercise Instructions With Hints
version of the exercise encourages the use of the exit command to
return back to a previous user. However, students invariably get
several shells deep and then when they get to the step where a file
system unmount occurs, they can't unmount the file system because
somewhere along the line a process has the mount point directory in
its environment. Encourage students to use exit in tandem with su
functions.

There are three shell scripts used in this exercise, mkfile, fragcopy,
and bigfile. They should be located in the /home/workshop
directory. This directory and the files in it should have been loaded
prior to the start of class. If they were not, load them from the instructor
exercise diskette if such a diskette was included in your instructor kit.

V4.1

Preface
classroom.
Part 1 - Determining file system usage

__ 1. Check that the /home/team01/dir2 file system is mounted. If it is not, then mount it.
# mount
# mount /home/team01/dir2 (if not already mounted)
__ 2. If you aren't already logged in, log in to your system and then su to root.
Recursively, copy the files under /usr/samples under your /home/teamxx/dir2
directory. Assume you have already determined that the /home/teamxx/dir2 file
system is close to full. Try to determine which of the directories is the biggest user of
the space. Then, try finding out which is the biggest file in that directory. Use the
sort command to organize the output in order of highest to lowest use. Finally,
determine who is the owner of that file.
# du /home/teamxx/dir2/samples | sort -nr | pg
# du -a /home/teamxx/dir2/samples/snmpd | sort -nr | pg
# ls -l /home/teamxx/dir2/samples/snmpd/dpi2/dpi20ref.txt
(Note: your system may have a different directory and file as the big
users of disk space then what we have listed above.)
__ 3. Let's assume that you have concluded that more space is needed (as opposed to
removing files to free up space); therefore, the file system needs to be expanded.
Expand the file system by the size of 1 physical partition. Verify the usage of this file
system.
# smit

File Systems ->
Enhanced Journaled File Systems ->
You are prompted to provide the file system name.
Select /home/teamxx/dir2.
File system name /home/teamxx/dir2
SIZE of file system
Press F10 to exit SMIT.
# df -m
-OR-
# df -m

V4.1
EXempty Part 2 - Using fragments for disk usage efficiency
__ 4. For this exercise, you create two JFS file systems, one with a fragment size of 512
bytes and the other with the default fragment size of 4096. For both file systems, use
a value of 512 for NBPI. Allocate 50000 512-byte blocks for each file system and put
them in the datavg volume group. Use /tmp/frag512 as the mount point for the file
system with a fragment size of 512 and /tmp/frag4096 as the mount point for the
one with the default fragment size. If more than one student is performing this part of
the exercise separately on the same System p system, it is necessary to modify the
file system names in order to make them unique. Once the file systems are created,
mount them.
# smit crjfs
Add A Standard Journaled File System
Select datavg for the Volume group name.
Volume group name datavg
SIZE of file system
MOUNT POINT [/tmp/frag512]
Fragment Size (bytes) 512
Number of bytes per inode 512
OK or Enter.
F3 to go back to the Add a Journaled File System screen.
Add A Standard Journaled File System.
*SIZE of file system
*MOUNT POINT [/tmp/frag4096]
OK or Enter.
# mount /tmp/frag512
# mount /tmp/frag4096
__ 5. Check both file systems with df -v and note any used space.
# df -v /tmp/frag512 /tmp/frag4096
You test the efficiency of the two file systems you just created for storing a large number
of very small files.
__ 6. First, you need a source file to use as input in the next step. Use the
/home/workshop/mkfile script to create a file that is 512 bytes in size. This script
requires the file size parameter to be passed to it and uses the suffix of bytefile. For
example: /home/workshop/mkfile 512 creates a file in your current directory called
512bytefile. Run this command from the /tmp directory creating a 512-byte file.
# cd /tmp
# /home/workshop/mkfile 512
__ 7. To test the number of 512-byte files that can be stored in each file system, run the
shell script /home/workshop/fragcopy. The shell script has the usage:
fragcopy numfiles dir/sourcefilename dir/targetfilename.
The fragcopy script creates copies of the sourcefilename until either the file
system is full or the numfiles value is reached. The files are put in the directory
specified in dir/targetfilename with the targetfilename as the prefix followed by
a . and a number which represents the count of the file currently being created.
While fragcopy processes, a count is displayed showing the number of files that
have been successfully copied. Have fragcopy copy /tmp/512bytefile 500 times
and have as the dir/targetfilename value: /tmp/frag512/frag500.
# /home/workshop/fragcopy 500 512bytefile frag512/frag500
__ 8. Look at the contents of the /tmp/frag512 directory. Note how many files are in this
directory.
# ls -l /tmp/frag512 | wc -l
The ls -l command output results in an extra line in the report, thus giving the
wrong number of files. Replace ls -l with ls.
__ 9. Do the same as step 7 only this time create copies of the 512bytefile in
/tmp/frag4096
# /home/workshop/fragcopy 500 512bytefile frag4096/frag500

V4.1
EXempty __ 10. Look at the contents for /tmp/frag4096. Note how many files are in this directory.
# ls -l /tmp/frag4096 | wc -l
The ls -l command output results in an extra line in the report, thus giving the
wrong number of files. Replace ls -l with ls.
__ 11. Now that there are two file systems with the same number of files, see how much
disk space each has utilized.
How many 512-byte blocks have the files used in /tmp/frag512? _____ 17048
How many inodes have the files used in /tmp/frag512? _______ 517
How many 512-byte blocks have the files used in /tmp/frag4096? _____ 20528
How many inodes have the files used in /tmp/frag4096?_______ 517
What can you conclude from this? __________________________________
For small files, a disk fragment size of 512 provides a more efficient use of
blocks for disk space and to store inode information.
# df -v /tmp/frag512 /tmp/frag4096
__ 12. When you are finished reviewing the results, unmount and remove both file systems.
# umount /tmp/frag512
# umount /tmp/frag4096
# rmfs -r /tmp/frag512
# rmfs -r /tmp/frag4096
Part 3 - Using JFS compression

__ 13. This exercise shows how to use some simple commands to investigate the effects of
JFS compression on both performance and disk space usage.
Create and mount two compressed journaled file systems, one with a fragment size
of 512 and a mount point of /tmp/compress512, the other with a fragment size of
2048, and NBPI of 2048 and a mount point of /tmp/compress. Make them both
65536 512-byte blocks in size and have them mount automatically at system restart.
Put them in the datavg volume group.
# smit crfs
Add a Journaled File System ->
Add a Compressed Journaled File System
SIZE of file system
MOUNT POINT [/tmp/compress512]
OK or Enter.
F3 to go back to the Add a Journaled File System menu.
Add a Compressed File System.
SIZE of file system
MOUNT POINT [/tmp/compress]
OK or Enter.
__ 14. Create a third file system (the same size as the two file systems you just created and
also in the datavg volume group) that does not use compression, that also has a
fragment size of 2048 and an NBPI of 2048, that has a mount point of
/tmp/uncompress, and that is automatically mounted at system restart. This will be
used for purposes of comparison.
F3 to go back to the Add a Journaled File System menu.

V4.1
EXempty Volume group name datavg

SIZE of file system
MOUNT POINT [/tmp/uncompress]
OK or Enter.
__ 15. Check that the journaled file systems have been correctly created.
Exit SMIT.
# lsfs -q | pg
__ 16. Mount the journaled file systems.

# mount /tmp/compress512
# mount /tmp/compress
# mount /tmp/uncompress
__ 17. Next, check the performance of the compressed file system by creating a 4 MB file
named bigfile by using the script /home/workshop/mkfile and then move it to
/home/workshop. Then copy it to each file system, and measure the performance
using timex. First, check that the logical volume configuration is similar. Then,
record copy times.
Real time for compress512: ____________________________________
Real time for compress: _______________________________________
Real time for uncompress: _____________________________________
# /home/workshop/mkfile 4194304
# mv 41943046bytefile /home/workshop/bigfile
#timex cp /home/workshop/bigfile /tmp/compress512
#timex cp /home/workshop/bigfile /tmp/compress
#timex cp /home/workshop/bigfile /tmp/uncompress
__ 18. Use du to check how much space is really used and how efficient the compression
is.
Disk usage for compress512: _____________________________________
Disk usage for compress: _________________________________________
Disk usage for uncompress: ______________________________________
# du /tmp/compress512
# du /tmp/compress
# du /tmp/uncompress
__ 19. Use ls -l to verify the normal size of each file. What is the amount of data in the
bigsize file in each filesystem?
File size of bigfile in compress512: ____________________________________
File size of bigfile in compress: _______________________________________
File size of bigfile in uncompress: _____________________________________
# ls -l /tmp/compress512
# ls -l /tmp/compress
# ls -l /tmp/uncompress

V4.1
EXempty Part 4 - Fixing file system problems
__ 20. Using the defragfs command, find out the current state of the file system
/tmp/compress512.
# smit
File Systems ->
Defragment a File System
File System Name /tmp/compress512
Perform, Query, or Report? query
Click OK or press Enter.
-OR-
# defragfs -q /tmp/compress512
__ 21. Defragment the /tmp/compress512 file system.

# smit
File Systems ->
Defragment a Journaled File System ->
File System Name /tmp/compress512
Perform, Query, or Report? perform
OK or Enter.
-OR-
# defragfs /tmp/compress512
__ 22. Perform a check on the /home/teamxx/dir2 file system so that if it finds something
wrong you are prompted as to whether or not to take any action.
# fsck /home/teamxx/dir2
What kind of message did you get?
The current volume is /dev/fslv00
File System is currently mounted.
Primary superblock is valid

fsck: Performing read-only processing does not produce dependable
results ______________________________
Unmount the file system and check it again. Mount the file system again after the
check is complete.
# fsck /home/teamxx/dir2

V4.1
EXempty Part 5 - Optional exercise

__ 23. There is a shell script provided by the system called skulker that periodically cleans
out files on the system. Find this file and add to it a line that gets rid of teamxx's
smit.log and smit.script files if they haven't been used in over two days. (This
could be run periodically using cron. You may want to refer back to this exercise
after completing the unit on scheduling.)
# vi /usr/sbin/skulker
Notice that this script removes files that are in /tmp, /var/tmp, /var/news, and so
forth. Towards the bottom of the script add the following line:
find /home/teamxx -name "smit.*" -mtime +2 -exec rm -f {}\;
End of exercise
1. In Part 1 - step 2, Why does the du -a size and the ls -l size for dpi20ref.txt differ?
Answer: du -a gives the amount of disk space allocated to the file while ls -l gives
the size of the file from the application perspective. If you are using a fragment size of 4
KB and the application writes a 100 byte file, then what size would each of these
commands return?
2. Ask students when they think they would want to defragment a file system.
Answer: When there is a strong indication that file system performance is being
impacted by file system fragmentation. Some file systems are set up to be fragmented
across disks (that is, striped) to improve performance through multiple actuator and
possibly multiple scsi adapter access. The performance of these file systems would be
reduced, not improved, if defragmented.

V4.1
EXempty Exercise 12. Paging space

(with hints)
Estimated time
00:20

This exercise is designed to give students the opportunity to work with
paging space changes on an active system.

List paging space
Add another paging space
Change the characteristics of a paging space
Remove paging space
Introduction
In this exercise you work with paging logical volumes, adding, deleting,
and changing them. To do this exercise, you need to work in teams -
one team per AIX system. Only one team member is able to do the
exercise, the other team members provide support and assistance
where needed.
Copyright IBM Corp. 1997, 2008 Exercise 12. Paging space 12-1

Preface
classroom.
__ 1. If you are not already logged into your system, log in as teamxx and su to root.
Check the current memory size on the system and then check the paging space.
$ su
# lsattr -E -l mem0
# lsps -a
__ 2. If you have two disks on your system and successfully created another volume
group named datavg during the Working with LVM exercise, create two page
spaces: first one in the rootvg volume group with a size of one partition and then the
other in the datavg volume group with an equivalent amount of space. (Be careful,
the volume groups may have different ppsize values.) Otherwise, create both page
spaces in the rootvg volume group. If you are short of space in rootvg, put both
paging spaces in datavg. Make them one partition each.
Have them become active now and at system restart.
# smit
System Storage Management (Physical and Logical Storage) ->
Paging Space ->
Add Another Paging Space ->
Select rootvg for the volume group.
Volume group name rootvg
SIZE of paging space (in logical partitions [1]
Start using this paging space NOW? yes
Use this paging space each time the system is RESTARTED? yes
OK or Enter.

V4.1
EXempty F3 to return to the menu.

Go back to the Paging Space menu.
Select Add Another Paging Space
A selector screen will pop up for volume group. Select datavg (or
rootvg if you dont have a second VG).
SIZE of paging space (in logical partitions) [8]
Start using this paging space NOW? yes
Use this paging space each time the system is RESTARTED? yes
OK or Enter.
-OR-
# mkps -s 1 -n -a rootvg
# mkps -s 8 -n -a datavg (or rootvg if you don't have datavg)
What names did the system assign to the two new paging spaces?____________
paging00 and paging01 to the new paging spaces.
__ 3. Check to be sure the new paging spaces are active. Create some system activity
and monitor the % of paging space used.
Exit SMIT.
# lsps -a
The following example commands use up memory on your system. Depending on
the amount of memory, you still may not be able to force paging.
# wsm & (if you are in a CDE graphical environment)
# find / -type f | xargs grep teasure >/dev/null &
# ls -laR / > /dev/null &
# lsps -a
Check the % used on the paging spaces.
What conclusion can you come to based on these observations?
_____________________________________________________________
The system begins to use active page spaces immediately and tries to
spread its usage across them as much as possible. Items in memory at the
time new page spaces are created and activated are not relocated but as
new system activity occurs, the new space begins to be used.
If necessary, run these commands again and notice how the page space usage
becomes balanced over time. Note that your system may have so much memory
that this step may not use enough to trigger an noticeable increase in paging space
utilization.
__ 4. Change the size of paging00 to four partitions. Check to be sure that paging00 has
been increased.
# smit
Paging Space ->
Change / Show Characteristics of a Paging Space
At the selector screen pop-up select paging00
Paging space name paging00
NUMBER of additional logical partitions [3]
OK or Enter.
F3 back to the Paging Space menu.
List All Paging Spaces
-OR-
# chps -s 3 paging00
# lsps -a
__ 5. Reduce the size of paging00 by one partition. Check to be sure that paging00 has
been decreased.
# smit
Paging Space ->
Change / Show Characteristics of a Paging Space
NUMBER of logical partitions to remove [1]
OK or Enter.
List All Paging Spaces

V4.1
EXempty -OR-
# chps -d 1 paging00
# lsps -a
__ 6. Remove paging01. First, deactivate it.

Deactivate a Paging Space
OK or Enter.
Remove a Paging Space
OK or Enter.
-OR-
# swapoff /dev/paging01
# rmps paging01
End of exercise
1. Review the steps students need to take to remove a paging space.
Answer: To remove a paging space, first you need to deactivate it and then remove it.
In AIX 5L V5.2 and later versions, this can be done without rebooting the system.
2. Review the reasons why additional page space may be needed and where it is best to
locate additional page space.
Answer: The amount of paging space required depends on the type of activities on the
system. If paging space runs too low, processes may be lost and if it gets full, the
system may halt completely. Since the performance of processes depend on access to
items in memory, the best place to put paging space is where you can get the best
performance. Thus, it is recommended to allocate paging space in the middle (center)
of the disk to reduce disk arm travel and to use multiple paging spaces, each allocated
on a separate physical volume. If there are multiple SCSI adapters on the system,
putting the paging spaces on disks on separate SCSI adapters may also help
performance.

V4.1
EXempty Exercise 13. Using backup and restore

(with hints)
Estimated time
Part 1 - 00:20
Parts 2, 3 and 4 - 00:40

The purpose of this exercise is to familiarize the students with backup
and restore techniques.

After completing this unit, students should be able to:
Backup a non-rootvg volume group
Perform an incremental backup by inode
Perform a backup by file name
Restore a file system and an individual file
In this exercise, you back up and restore a file system as well as
individual files. Some exercises may cover some options and
techniques that may be in the student notes, but not necessarily
covered in the instructor's presentation. Either refer to the unit notes or
check the hints in the Exercise Instructions with Hints section.
Backup media: You do not use tapes or diskettes for this exercise. You
create a new file system and use a file in it as your backup media.
Everything in the exercise works the same whether you backup to tape
or some other media.
Scenario: Existing and new file systems and files to be used for this
exercise are:
/home/teamxx/dir2 - File system to be backed up. Created in the
Working with File Systems exercise.
/home/bkupxx - New file system you create to use as the backup
media.
/home/bkupxx/weekly - File used for level 0 full inode backup.
/home/bkupxx/daily - File used for level 1 incremental backup.
Copyright IBM Corp. 1997, 2008 Exercise 13. Using backup and restore 13-1
/home/bkupxx/tardata - File used for tar backups.

/home/teamxx/mybkup - Directory used to demonstrate relative
versus full path name backups and importance of mounting versus
unmounting before restoring a backup.
Warning: There may be times in the exercise when you will be using
the recursive remove command rm -r. Please ensure that you are in
your /home/teamxx directory and not the /(root) directory. Being in
/(root) can cause some Irrecoverable problems that might only be
fixed by reloading your system.
The Exercise Instructions with Hints section provides both the
command line and SMIT methods. Select the one that is most
appropriate for you.

There are some exercise hints that use the . (dot) for current directory.
The dot may not be noticed by some students doing the Hints section.
Watch for any problems relating to the instructions that make
reference to the . (dot).
When students perform the full inode backup, ensure they have used
a zero 0 and not an uppercase O.

When listing a directory, the students are directed in the Exercise
Instructions with Hints section to do an ls -lR. If the students do
not use the recursive -R, they do not see a listing of files. They only
see a listing of directories. The files they are asked to list are from
/usr/samples. This directory consists of sublevel directories with all
the files contained within these directories. Students doing the
Exercise Instructions section may not use the appropriate options.
Direct them to use ls -lR.
If your classroom is equipped with a tape drive per student and you
can supply two tapes per student, you can adjust this exercise so the
students back up to tape rather than the new file system they are
asked to create. To ensure that this exercise could be done under any
circumstance, it was designed to use a user-defined file system as the
backup media.

V4.1
Important
You are backing up the file system that you created in the Working with File Systems
exercise; therefore, that exercise must have been completed successfully before
continuing with this exercise. When referencing the file system you can use either the
logical volume name, /dev/fslvxx or the file system name, /home/teamxx/dir2.
Preface
classroom.
Part 1 - Use SMIT to backup a non-rootvg volume group

__ 1. Create a file system called /home/savevg in rootvg that holds the backup image
and then mount it. Make sure the file system is in rootvg.
# smit fs
Add/Change/Show/Delete File Systems ->
Select rootvg for the volume group name
Volume group name: rootvg
SIZE of file system
Unit Size Megabytes
* Number of units [16]
*MOUNT POINT [/home/savevg]
Use F3 until you get back to the File Systems menu.
Select Mount a File System.
FILE SYSTEM name [/home/savevg]
Use F10 to exit SMIT

OR
# crfs -v jfs -g rootvg -a size=16M -m /home/savevg
# mount /home/savevg
__ 2. Back up the datavg volume group to /home/savevg/bkup.

# smit savevg
* Backup Device or FILE [/home/savevg/bkup]
* VOLUME GROUP to back up [datavg]
OR
# savevg -f /home/savevg/bkup -i datavg
__ 3. Make sure the backup completes successfully before doing this step. Remove the
datavg volume group. The volume group must not have any physical partitions
allocated to logical volumes. In order to remove these logical volumes in the datavg,
you must first make sure that they are all closed. In others words, the file systems
must be unmounted and the paging spaces must be inactive.You need to unmount
all mounted file systems in datavg first.
# smit vg
Select Remove a Volume Group
*VOLUME GROUP name [datavg]
OR
# reducevg -df datavg hdiskn
__ 4. Recover the volume group from your backup.

# smit restvg
* Restore DEVICE or FILE [/home/savevg/bkup]
SHRINK the filesystems? no
OR
# restvg -q -f /home/savevg/bkup hdisk1

V4.1
EXempty __ 5. Verify that the volume group is back.

# smit lsvg
Select List Content of a Volume Group
VOLUME GROUP name [datavg] +
List OPTION status +
F10 to exit SMIT.
__ 6. Examine the contents of the file /tmp/vgdata/datavg/datavg.data. What is the file

used for?
# pg /tmp/vgdata/datavg/datavg.data
Part 2 - Prepare file system for backup/restore
__ 7. Log in as teamxx and su to root. Use su with out the dash so that you retain the
teamxx home directory.
$ su
__ 8. Create a new file system in datavg named /home/bkupxx, where xx is your login
team number. Make the file system one partition in size.
# smit
File Systems ->
Add / Change / Show / Deleted File Systems ->
From the Volume Group Name pop-up box, select datavg.
SIZE of file system
Unit Size Megabytes +
* Number of units [1] #
* MOUNT POINT [/home/bkupxx]
Mount AUTOMATICALLY at system restart? yes +
OK or Enter.
OR
# crfs -v jfs -g datavg -a size=1M -m /home/bkupxx -A yes -p rw
__ 9. Mount the new file system. Verify that it is mounted.

In SMIT, F3 back until you are In the SMIT File System menu.
Select Mount a File System
FILE SYSTEM name [/dev/lv0x]
(where x is the number of your logical volume)
DIRECTORY over which to mount [/home/bkupxx]
OK or Enter.
F3 to go back to the File System menu.
Select List all Mounted File Systems

V4.1
EXempty F10 to exit SMIT.

OR
# mount /home/bkupxx
# mount
__ 10. Make sure the file system you created in the Working with File Systems exercise,
/home/teamxx/dir2 in logical volume /dev/fslvxx, is mounted. If it isn't, mount it.
# mount (check mounted file systems)
# mount /home/teamxx/dir2 (if not mounted)
__ 11. Ensure that all files from the /usr/samples directory were copied into the
/home/teamxx/dir2 file system. You are backing up these files to the new file
system, /home/bkupxx, which you created in Step 8.
# ls /home/teamxx/dir2 (check if data is in the file system)
# cp -r /usr/samples /home/teamxx/dir2 (if necessary)
Part 3 - Backup and restore by inode (file system)
__ 12. The bos.rte.archive fileset contains the backup and restore utilities. Check that the
bos.rte.archive fileset has been installed.
# lslpp -l bos.rte.archive
__ 13. You should be ready to back it up:

Change directory to /home/teamxx.
# cd /home/teamxx
Unmount the /home/teamxx/dir2 file system.
Verify that the file system is unmounted. If you got an error unmounting the file
system, it may be that you were residing in that directory. You must be out of the
file system before you can unmount it. Remember, it is a good idea to unmount
the file system before you back it up to ensure that there is no activity happening
in the file system at the time of backup.
# mount
Ensure that the mount point is empty. Do a listing on the directory. If it is empty,
proceed to the next step. If not, ensure you are in your home directory,
/home/teamxx, and delete the files in the directory /home/teamxx/dir2.
# ls -l /home/teamxx/dir2 (should be total 0)
# cd /home/teamxx
# rm -r /home/teamxx/dir2/* (if not empty)
__ 14. Back up the entire file system in inode format. Make sure to update
/etc/dumpdates. Your output media will be a file named weekly in the
/home/bkupxx file system.
# backup -0 -u -f /home/bkupxx/weekly /home/teamxx/dir2
OR
# smit
System Storage Management ->
File Systems ->
Backup a File System
*FILESYSTEM to Backup [/home/teamxx/dir2]
*Backup DEVICE [/home/bkupxx/weekly]

V4.1
EXempty Backup LEVEL [0]

Record backup in /etc/dumpdates? [yes]
OK or Enter.
__ 15. Display the table of contents of the /home/bkupxx/weekly backup to ensure that all
files in the /home/teamxx/dir2 file system were backed up properly. How are the
files listed? __ ____________________________________________________
The inode and relative path name of the files are displayed.
In SMIT, F3 back to the File Systems menu.
Select List Contents of a Backup
OK or Enter.
F10 to exit SMIT.
OR
# restore -T -f /home/bkupxx/weekly
__ 16. View /etc/dumpdates and analyze the contents.
Note
If there is more than one team on your system, you will be writing to
/etc/dumpdates simultaneously. Analyze the information that relates to your
logical volume name. You see the other team's logical volumes listed as well.
# pg /etc/dumpdates
__ 17. Mount the /home/teamxx/dir2 file system and simulate a daily change to the file
named /home/teamxx/dir2/samples/tcpip/resolv.conf. Add your name in
uppercase to the first line of the file.
# smit
File Systems ->
Mount a File System
FILE SYSTEM name [/dev/fslvxx]
DIRECTORY over which to mount [/home/teamxx/dir2]
OK or Enter.

F9 to access the command line.
# vi samples/tcpip/resolv.conf (add your name in uppercase).
Return to SMIT.
OR
# vi samples/tcpip/resolv.conf
# cd
__ 18. Now unmount the file system. Verify it is unmounted. Perform an incremental inode
backup of /home/teamxx/dir2 to the file /home/bkupxx /daily. It should backup just
the changed file, resolv.conf. You are simulating the fact that you have a base
backup in /home/bkupxx/weekly (analogous to the monthly backup explained in
class) and an incremental backup in /home/bkupxx/daily (analogous to the daily
backup explained in class).
In SMIT, go back to the File System menu.
Select Unmount a File System
F3 back to the File System menu.
Select List all Mounted File Systems
F3 back to the File System menu.
Select Back up a File System
FILESYSTEM to backup [/home/teamxx/dir2]
Backup DEVICE [/home/bkupxx/daily]
Backup LEVEL [1]
RECORD backup in /etc/dumpdates [yes]
OK or Enter.
OR
# cd /home/teamxx
# mount
# backup -u -1 -f /home/bkupxx/daily /home/teamxx/dir2

V4.1
EXempty
__ 19. Display the table of contents of the /home/bkupxx/daily backup to ensure that just
the changed file was backed up. Record the EXACT name of the file.
In SMIT, F3 back to the File System menu.
Backup DEVICE [/home/bkupxx/daily]
OR
# restore -T -f /home/bkupxx/daily
__ 20. View the contents of /etc/dumpdates. Analyze the contents.

# pg /etc/dumpdates
Note
The next few steps show command line hints rather than SMIT screens because of
the extra steps of removing files and destroying file systems, which is necessary to
perform simulated recovery functions. You may continue to use the SMIT F9
shellout feature to execute the crash procedures.
__ 21. Simulate a user accident and recover the lost file using the above backup:
Mount the file system. Change directory to /home/teamxx/dir2. Remove the
samples/tcpip/resolv.conf file. Check that it no longer exists.
# rm samples/tcpip/resolv.conf
# ls samples/tcpip/resolv.conf
__ 22. Before you start the restore, be sure you are in the directory /home /teamxx/dir2.
Why is this important? ______________________________________________
The files were backed up with relative path names. To restore them to their original
locations, you must be in the original directory.
You will see messages displayed before the restore takes place. When prompted to
specify next volume #, type in 1. When prompted to set owner/mode, type in no.
# pwd (if not in correct directory, cd to it)
# restore -xvf /home/bkupxx/daily

Extract requested files.
You have not read any media yet.
Unless you know which volume your file or files are on, you should start with the last
volume and work towards the first.
Specify next volume number 1.
Extract file./samples/tcpip/resolv.conf
Add links.
Set directory mode, owner, and times.
Set owner/mode for '.' ? [yes or no] no
# pg samples/tcpip/resolv.conf (look for your name in the file).
__ 23. Now, you perform a crash of the entire /home/teamxx/dir2 file system and then
restore it using your backups. Ensure you are in the /home/teamxx/dir2 directory.
Remove all files and directories in this file system. Check to see that they were all
removed. Restore the file system and check to ensure all the files were restored.
Which backup should you use first? ____________________________________
Use the weekly backup first, then the daily backup.
# rm -r /home/teamxx/dir2/*
# ls -lR (total 0)
# restore -rvf /home/bkupxx/weekly (Notice the messages on a level 0 backup)
# restore -rvf /home/bkupxx/daily (Notice the messages on a level 1 backup)
# ls -lR (You should see all your files)
__ 24. The restoresymtable file is created or updated upon each restore when working
with incremental inode backups. Do a long listing on this file to see that it exists.
Remove the restoresymtable.
Note
The restoresymtable file keeps track of links and deleted files that were created
between the backups. This file is only created when working with the -r restore
option. The system does not know when you are finished restoring all your
incremental inode backups, so it leaves you with the chore of cleaning up this
control file. This file did not appear during the single file restore. This is because

V4.1
EXempty the -r option is not appropriate for a single file restore. The -x option is the
option of choice when restoring a single file and the unusual problems of restoring
links and deleted files correctly is not an issue.
# ls -l restoresymtable
# rm restoresymtable
The next three steps demonstrate the advantage of backing up by relative path
name, by restoring into a directory that is different from where the files originally
resided.
__ 25. Create a new directory named /home/teamxx/mybkup. Change and verify that you
are in the new directory.
# mkdir /home/teamxx/mybkup
# cd /home/teamxx/mybkup
# pwd
__ 26. Restore the contents of /home/bkupxx/weekly into your new directory

/home/teamxx/mybkup. Don't forget to use the verbose option so you can see
what the restore command is doing.
# smit
File Systems ->
Restore a File System
*Restore DEVICE [/home/bkupxx/weekly]

*TARGET Directory [.]
VERBOSE output? yes
Number of BLOCKS to read in a single output operation [ ]
OK or Enter.
OR
# restore -rvf /home/bkupxx/weekly
__ 27. Check the directory /home/teamxx/mybkup. It should contain the contents of the
weekly backup. You can see that backing up files by relative path name allows you
to restore them back to ANY directory.
# ls -lR

V4.1
EXempty Part 4 - Backup and restore by name
__ 28. Make sure the /home/teamxx/dir2 file system is mounted.

Why would you want to have the file system mounted in order to back it up by
name? ________________________________________________________
To have the full path names backed up and ultimately restored by full path name.
# mount (to see if already mounted)
# mount /home/teamxx/dir2 (to mount if necessary)
__ 29. Back up the file system by name using the backup command. Also, use the full path
name to reference the file system. Once again, you use /home/bkupxx/weekly as
your output media instead of a tape.
# smit
System Storage Management
Files & Directories
Backup a File or Directory

*FILE or DIRECTORY to backup [/home/teamxx/dir2]
VERBOSE output? yes
OK or Enter.
OR
# find /home/teamxx/dir2 | backup -i -v -f /home/bkupxx/weekly
__ 30. List the table of contents to verify the backup was successful.
How were the files listed? _________________________________________
The files are listed with full path names.
# smit
File Systems ->
List Contents of a Backup

OK or Enter.

OR
# restore -T -f /home/bkupxx/weekly
__ 31. Let's take a look at the difference between relative and full path backups by
restoring by relative path name. First, cd to /home/teamxx/mybkup directory and
delete all the files. Ensure the directory is empty.
# rm -r /home/teamxx/mybkup/* (if necessary)
# ls -lR (total 0)
__ 32. The goal is to restore the files from /home/bkupxx/weekly into the
/home/teamxx/mybkup directory. Restore the files and note the way the files are
listed in the output of the restore command. Take a look at the contents of the
/home /teamxx/mybkup directory.
During restore where the filenames full or relative path? _________________
The file names are listed by their full path names.
Does it contain any files? _________________________________________
No. The files were restored by their full path name, which put them into the
/home/teamxx/dir2 file system.
# smit
Files and Directories ->
Restore a File or Directory

*Target Directory [/home/teamxx/mybkup]
VERBOSE output yes
OK or Enter.
F10 to exit SMIT.
# ls -lR (total 0)
OR
# restore -xvf /home/bkupxx/weekly
# ls -lR (total 0)

V4.1
EXempty __ 33. Back up the file system by name, but use a relative path name. Make sure you are in
/home/teamxx/dir2. List the contents of weekly. The file names should be in
a ./filename (relative path name) format.
# smit
Files & Directories ->
Backup a File or Directory

*File or Directory to Backup [.]
VERBOSE output? yes
OK or Enter.
F3 back to the Files & Directories menu.
*Backup DEVICE /home/bkupxx/weekly
F10 to exit SMIT.
OR
# find . | backup -ivf /home/bkupxx/weekly
# restore -Tf /home/bkupxx/weekly
__ 34. Now, you should be able to restore these files into the /home/teamxx/mybkup
directory. Change to the /home/teamxx/mybkup directory. Perform the restore.
# smit
Files & Directories ->
Restore a File or Directory

*Target Directory [.]
VERBOSE output? yes
OK or Enter.
F10 to exit SMIT.
OR
# restore -xvf /home/bkupxx/weekly
__ 35. List the contents of the /home/teamxx/mybkup directory. All of the files should be
listed.
# ls -lR /home/teamxx/mybkup

V4.1
EXempty Part 5 - (Optional)

The next eight steps are included to show what happens if you don't remember to
mount a file system before performing a restore. This is a very common error that
can cause a lot of headaches. Hopefully, by intentionally making this mistake in
class, you can avoid it back at work or be able to identify the problem immediately
and resolve it.
__ 36. The /home/teamxx/dir2 file system should still be mounted from the previous
restore. Once again, delete all the files and directories in the file system. (If you
clean out the file system now, it is much easier to see if your restore is working
properly later.) Double check to make sure the files are removed.
# mount
# ls -lR /home/teamxx/dir2 (should see total 0)
__ 37. Unmount the file system and verify that it is unmounted.

# cd /home/teamxx
# mount
__ 38. Change to the directory mount point and perform the restore.
# smit
File Systems ->
Restore a File system

*Target DIRECTORY [.]
VERBOSE output? [yes]
Number of BLOCKS to read in a single input operation [ ]
OK or Enter.
OR
# pwd
__ 39. Mount the file system and look at the contents to make sure that the restore worked.
Does the file system contain any files?_______ No
Why? ________________________________________________________
The files were restored to the /home/teamxx/dir DIRECTORY. When you
mounted the file system to the MOUNT POINT /home/teamxx/dir, you overlaid
the directory with the file system contents.
# mount
# ls -lR
__ 40. Unmount the file system and list the files in the mount point directory. You see that
this is where your files went when you performed the restore.
# cd /home/teamxx
# pwd
# ls -lR /home/teamxx/dir2
__ 41. Clean up the mount point by deleting all the files in the /home/teamxx/dir2
directory.
# ls -lR /home/teamxx/dir2 (total 0)
__ 42. Mount the file system. Verify it was mounted.

# mount
__ 43. Now that you have the file system mounted, try the restore again. List the files after
you are through with the restore.
# smit

V4.1
EXempty System Storage Management ->

File Systems ->
Restore a File system

*Target DIRECTORY [.]
VERBOSE output yes
Number of BLOCKS to read in a single input operation [ ]
OK or Enter.
F10 to exit SMIT.
# ls -lR
OR
# ls -lR
End of exercise

V4.1
EXempty Exercise 14. (Optional) Using tar and cpio

(with hints)
Estimated time
00:30

The purpose of this exercise is to familiarize the students with the
UNIX commands, tar and cpio.

Use tar to perform a backup and a restore
Use cpio to perform a backup and a restore
In this exercise you back up files and directories using tar and then
cpio. The backup media is the hard drive.
The Exercise Instructions with Hints section provides the command
line commands for this activity. SMIT does not support backups using
tar or cpio. You must perform these steps from the command line.
Copyright IBM Corp. 1997, 2008 Exercise 14. (Optional) Using tar and cpio 14-1

Preface
classroom.
Part 1 - Using tar

__ 1. Before getting started, perform some housekeeping. Delete all of the files in the
/home/teamxx/mybkup directory.
# rm -r /home/teamxx/mybkup/*
__ 2. Make sure that the /home/teamxx/dir2 file system is mounted. tar does not back
up file systems unless they are mounted. If it is not mounted, mount it.
# mount
# mount /home/teamxx/dir2 (if not mounted)
__ 3. Change directory to the file system and create a directory called myjunk. In that
directory, create four zero-length files named file1, file2, file3, and file4.
# mkdir myjunk
# cd myjunk
# touch file1 file2 file3 file4
__ 4. Back up the file system using the tar command to the file /home/bkupxx/tardata
using relative path names. List the contents of tardata to ensure all files are backed
up.
# tar -cvf /home/bkupxx/tardata . (note the dot as an argument)

V4.1
EXempty # tar -tvf /home/bkupxx/tardata
__ 5. Since the backup was done using relative path names, restore them to a different
directory. Use tar to restore the contents of the tardata directory to the
/home/teamxx/mybkup directory. Verify that all files were restored.
# tar -xpvf /home/bkupxx/tardata
__ 6. Change your directory to /home/teamxx/mybkup. Remove all the files in

/home/teamxx/mybkup. Verify that it is empty. Use tar to extract only the files
contained in the myjunk directory in tardata and load these files to the
/home/teamxx/mybkup directory. List the files in the /home/teamxx/mybkup
directory.
# ls -lR
# tar -xpvf /home/bkupxx/tardata ./myjunk
# ls -lR
__ 7. You have been using relative path names up to this point with tar. Specify a full path
name and tar backs it up in full path name format. Backup the contents of the file
system to tardata directory using full path names.
# tar -cvf/home/bkupxx/tardata /home/teamxx/dir2
__ 8. Check the contents of tardata. All file names should be referenced by full path
name. If you restored the files, they would restore back to the same directory that
you copied them from.
# tar -tvf /home/bkupxx/tardata
Part 2 - Using cpio (Optional)

In this section, you will back up and restore the same files again but this time using cpio.
__ 9. Change directory into /home/teamxx/dir2/myjunk and make sure your files are
there.
# cd /home/teamxx/dir2/myjunk
# ls
__ 10. Back up the file system using the cpio command to the file
/home/bkupxx/cpiodata using a relative path name. List the contents of cpiodata
to ensure all files were backed up.
# find . | cpio -ov > /home/bkupxx/cpiodata
# cpio -itv < /home/bkupxx/cpiodata
__ 11. Since this backup was done using a relative path names, restore them, using cpio,
to the directory /home/teamxx/mybkup. FIRST remove all the files from the
/home/teamxx/mkbkup directory.
# cpio -idv < /home/bkupxx/cpiodata
__ 12. Change your directory to /home/teamxx/mybkup. Remove all the files in

/home/teamxx/mybkup. Verify that it is empty. Use cpio to extract only the files
contained in the myjunk directory in cpiodata and load these files to the
/home/teamxx/mybkup directory. List the files in the /home/teamxx/mybkup
directory.
# ls -lR
# cpio -idv < /home/bkupxx/cpiodata myjunk/*
# ls -lR

V4.1
EXempty __ 13. You have been using relative path names up to this point with cpio. Specify a full
path name with the find command and cpio backs it up with a full path name
format. Back up the contents of the file system to cpiodata using full path names.
# find /home/teamxx/dir2/ | cpio -ov > /home/bkupxx/cpiodata
__ 14. Check the contents of cpiodata. All the file names should be referenced by full path
names. If you restored the files, they would restore back to the same directory that
you copied them from.
# cpio -itv < /home/bkupxx/cpiodata
End of exercise

V4.1
EXempty Exercise 15. Security files

(with hints)
Estimated time
00:30

The purpose of this exercise is to gain familiarity with security-related
files.

Execute commands that identify group membership
Identify default system groups
Display contents of security logs
Work with the SUID and sticky bits
Introduction
During this exercise, you are given an opportunity to look at some of
the files discussed in the unit. You also have a chance to work with the
SUID and sticky bits.

None.

None.
Copyright IBM Corp. 1997, 2008 Exercise 15. Security files 15-1

Preface
classroom.
Part 1 - Security control files

1. Log in as teamxx and su to root. Display a listing of the system-defined groups. Take a
moment to review the members of each group.
# cat /etc/group
-OR-
# smit
Security & Users
Groups
List All Groups
F3 to return to the menu
__ 2. Exit SMIT if you are in SMIT, and display the groups of which root is a member.
Next, without logging out, display the groups of which your teamxx login is a
member.
F10 to exit SMIT (if you are in SMIT).
# groups -OR- # setgroups
# groups teamxx
__ 3. Display the contents of the file that keeps track of the successful and failed su
attempts.
How can you determine the failed attempts? _________

V4.1
EXempty A failed attempt is noted by a- (minus). A successful attempt is noted by a +

(plus).
# pg /var/adm/sulog
__ 4. When a user logs in successfully, the login program writes the associated
information in the two files /var/adm/wtmp and /etc/utmp. Using the who command,
examine the contents of these two files.
# who /var/adm/wtmp
# who /etc/utmp
__ 5. Display in reverse chronological order, all previous logins and logouts recorded in
/var/adm/wtmp for your login user name, teamxx, and root.
# last teamxx
# last root
__ 6. Check to see if there were any invalid login attempts on your system due to an
incorrect login name or password.
How can you tell? ______________________________________________
Login names that are not recognized will be logged as UNKNOWN.
# who /etc/security/failedlogin
Part 2 - SUID and sticky bit
__ 7. Create a listing of all SUID programs on the system in a file called allsuids.
Page through allsuids to see all the programs that are SUID programs. Look for
some of the commands that you have already learned and executed.
# find / -perm -4000 > /tmp/allsuids
# pg /tmp/allsuids
__ 8. As an ordinary user, you have been executing the su command to switch user to
root to perform privileged tasks. The su program has the SUID bit turned on. Do a
long listing on /usr/bin/su to see the SUID bit.
# ls -l /usr/bin/su
__ 9. Turn the SUID bit off. Verify that the bit is turned off.
# chmod u-s /usr/bin/su
# ls -l /usr/bin/su (An x bit should replace the s bit.)
__ 10. Exit back to your teamxx login shell and try to su to root.
Can you? _____________________________________________________
No, because the SUID has been taken off the su program. Sometimes it is
necessary for unprivileged users to be able to accomplish tasks that require
privileges. When a SUID program is run, its effective UID becomes that of the
owner of the program rather than the user that is running it. When you did a long
listing on /usr/bin/su you saw that the owner was security.
$ su
Cannot su to root: Authentication is denied
__ 11. Since you cannot su to root to turn the SUID bit back on, login over yourself as root
and set SUID on the program /usr/bin/su. Verify that the SUID bit is turned on.
$ login root
# chmod u+s /usr/bin/su
# ls -l /usr/bin/su
__ 12. Log in over yourself as teamxx and su to root.

V4.1
EXempty # login teamxx

$ su
__ 13. Do a long listing on the /tmp directory. Locate the sticky bit that is set on the
directory.
# ls -ld /tmp
__ 14. Exit back to your teamxx login shell. Ensure you are in your teamxx home directory.
Create a subdirectory in your home directory named stickydir.
Change the bits on the stickydir directory for full access by user, group, and others
with the sticky bit turned on. Verify that the change was made.
# exit
$ pwd (You should be in /home/teamxx.)
$ mkdir stickydir
$ chmod 1777 stickydir
$ ls -ld stickydir
__ 15. Create a zero-length file in the stickydir directory named sticky1.

$ touch stickydir/sticky1
__ 16. Now, su to a teamyy user name that is not being used on your system.
Check with any other teams that may be sharing your system. The existing user
names should be team01 - team05 with passwords the same as the user name.
As this other teamyy user, try to first add and then remove a file named sticky2.
Can you do it? __________________________________________________
Yes. As a member of the staff group, teamyy has write authority to the stickydir
directory.

As this other teamyy user, try and remove the teamxx file named sticky1.
Can you do it? __________________________________________________
No. As the creator, teamxx, of the directory, stickydir, you set the mode of
/home/teamxx/stickydir so any user can create or delete their own files in this
directory, but cannot delete another user's files. The /tmp directory is the most
widely used directory with this feature.
$ su teamyy (where teamyy is a different user name from your login name).
$ touch sticky2
$ rm stickey2
$ rm /home/teamxx/stickydir/sticky1
Because you are not the owner the system should properly respond with Not
owner.
__ 17. Exit back to your teamxx login user name.

$ exit
End of exercise

V4.1

1. Who else can view the files in the /etc/security directory?
Answer: Members of the security group. If you do an ls -ld on the /etc/security
directory, you see that root has read/write permissions and those in the security group
have read-only permissions. Others has no permissions. You must determine who is a
member of the security group because of the read-only permissions available to that
group.
2. As a system administrator, how should you monitor the security control files like
/var/adm/sulog?
Answer: How big the file grows. If space is a premium on your system, files like these
should be analyzed, printed, and then reduced in size. You also want to monitor the
contents to see who is su-ing to root in case someone gets root's password who
shouldn't be using it.

V4.1
EXempty Exercise 16. User administration

(with hints)
Estimated time
00:40 Parts 1-5
00:10 Parts 6-7

This exercise introduces you to the management and administration of
users and groups.

Create a user with password restrictions
Implement and test user login controls
Remove a user from the system
Create a group
Execute pwdadm to change another user's password
Invoke user communication commands
Introduction
This exercise has two parts. Be sure to do only the currently assigned
part of the exercise.
This exercise concentrates on the SMIT interface for creating and
maintaining users and groups.
The exercise instructions direct you to create users usera, userb and
userc. If two students are sharing an AIX system, and performing this
exercise separately, create users useraxx, userbxx, usercxx where
xx represents the student's initials. This will ensure unique user IDs.
Copyright IBM Corp. 1997, 2008 Exercise 16. User administration 16-1

In the Introduction, it is pointed out that only the options the students
are to fill in are listed. All other options use the system default value as
presented on the dialog screens. Cover this with students before the
start of the exercise.
Also, in the Introduction, the students are asked to log in as root
rather than switching to root. Cover this with the students before the
start of the exercise; otherwise, the students might automatically log in
as teamxx and switch to root as they have done throughout the week.
This causes problems with some of the instructions.
The exercise instructions direct the students to create users usera,
userb and userc. If two students are sharing an AIX system, and
performing this exercise separately, they should create users useraxx,
userbxx, usercxx where xx represents the student's initials. This
ensures unique user IDs.
If any students are using the CDE interface they have to edit
the .dtprofile in their login directory and uncomment the last line:
DTSOURCEPROFILE=true
The files /etc/profile and $HOME/.profile are not read by default by
CDE. By editing the .dtprofile, CDE will use $HOME/.profile.
Step 30 has the students work with the news command. Although this
command was not covered in lecture, it is self explanatory. Information
on this facility is included in the Exercise Review/Wrap-up and the
instructor may wish to cover this with the students.
The following are common mistakes that can be made while editing
/etc/security/login.cfg in Part 2 of this exercise:
The new line and return codes must be preceded by a \ backward
slash and not a / forward slash.
There can be no return at the end of the customized login line
under default. The instructions tell the student to let vi do the line
wrap. To many students who are proficient typists, it is natural for
them to want to press the return key.
There must be double quotes around the login prompt information.
There must be spaces around the equal sign after the word
herald.
Should the student make any one of the above errors, they are all
recoverable and the students are able to login and edit the file.

V4.1

Preface
classroom.
Part 1: User administration

__ 1. Log in to the system as root. Using SMIT, list the current users defined on your
system.
# smit
Security & Users ->
Users ->
List All Users
__ 2. Create a new user named usera with the following restrictions:

Three-day password expiration warning
Two passwords before reuse
One week between expiration and lockout
Prevent other users from switching to usera
F3 back to the SMIT Users menu
Select Add a User
User NAME [usera]
Another user can SU to USER? false
Days to WARN USER before password expires [3]
NUMBER OF PASSWORDS before a reuse [2]
Weeks between password EXPIRATION & LOCKOUT [1]
OK or Enter.
__ 3. List the user attributes. Exit SMIT and look at the /etc/security/user file. Compare
the defaults in this file with the attributes listed for usera.
Go back to the Users menu.
Change/Show Characteristics of a User
User Name [usera]
OK or Enter.
F10 to exit SMIT.
# pg /etc/security/user
__ 4. To test the new user, log off and log in as usera.

Could you log in? ______ No
Why? ________________________________________________________
usera does not have a password and therefore, is considered an invalid user.
# exit
login: usera
usera's Password: <Enter password>)
You entered an invalid name or password
login:
__ 5. Log in as root and page through the /etc/passwd file.

What is in the password field for usera? ________
The * character.
What does this mean? ___________________________________________
That usera is not a valid user as no password was set; therefore, there is no
entry in /etc/security/passwd for usera.
# pg /etc/passwd
__ 6. Assign a password to usera. Display the /etc/passwd file.

What is in the password field for usera? _______________________
What does this mean? ___________________________________________
usera is a valid user with a password; therefore, there is an entry in
/etc/security/passwd for usera.
# pwdadm usera

V4.1
EXempty Changing password for "usera"

usera's New password: <enter-new-password>
Enter the new password again: <enter-new-password-again>
# pg /etc/passwd
__ 7. Do a long listing of all files including hidden files in usera's home directory.
What files were created? ______________________________ .profile
# ls -al /home/usera
__ 8. Test the new user ID. Log off and log in as usera. When prompted to change the
password, type in the same password assigned in step 6 to test the reuse restriction
set on the user ID from step 2. After you get the reuse error message, proceed to
change usera's password to something different. After successfully logging in as
usera, log out.
# exit
login: usera
usera's Password: <enter original password>
You are required to change your password.
Please choose a new one.
usera's New password: <enter the original password again>
Password was recently used and is not valid for reuse.
usera's New password: <enter new password>
Enter the new password again: <enter password>
$ exit
__ 9. Suppose usera is going on an extended vacation. Log in as root and using SMIT,
lock usera's login account.
# smit
Security & Users ->
Login Controls ->
Change/Show Login Attributes for a User
User NAME [usera]
Is this user ACCOUNT LOCKED [true]
OK or Enter.
__ 10. Exit SMIT and test the account lock by attempting to log in as usera. You should not
be able to. If you could, redo step 9 and test it again.
F10 to exit SMIT.

login: usera
usera's Password: <enter password>
Your account has been locked, please see the system administrator.
login:
__ 11. Log in as root and try to su to usera. Change to useras environment at the same
time.
Could you? _______ Yes
Why? _____ ___________________________________________________
root is not locked out of any restrictions placed on a user. Along with the account
locked feature, when you added usera you also put the restriction that no other
user could su to this account. As with the account locked feature, root is exempt
from this restriction as well.
Kill the subshell and return to root. Unlock usera's account.
# su - usera
$ <ctrl-d>
# smit login_user
User NAME [usera]
Is this user ACCOUNT LOCKED [false]
OK or Enter.
__ 12. Some users may try to guess other user's passwords. To help prevent illegal logins,
set the number of failed login attempts to two for usera. Exit SMIT and log out.
Use F3 to go back to the Change/Show Login Attributes for a User dialogue
panel
User NAME [usera]
Number of FAILED LOGINS before [2]
user account is locked
OK or Enter.
F10 to exit SMIT.
# exit
__ 13. Test the change by trying to log in as usera and entering an incorrect password
three times. On the fourth attempt, enter the correct password. If you are in a
Windows session, exit and execute this instruction from an ASCII command line.

V4.1
EXempty login: usera

usera's Password: <enter incorrect password>
You entered an invalid login name or password
login: usera
login: usera
The screen should scroll to refresh login prompt
login: usera
usera's Password: <enter correct password>
There have been too many unsuccessful login attempts; please see the system
administrator
login:
__ 14. Log in as root and reset the failed login count for usera.
login root
# smit user
Reset Users Failed Login Count
* User NAME [usera] +
Press <Enter>
F10 to exit SMIT.
__ 15. Attempt to login at usera with the correct password. Was it successful?
# exit
login usera
__ 16. Log in as root and change the failed login attempts limit to 0 for usera to disable the
failed login check.
# smit login_user
User Name [usera]
Number of FAILED LOGINS [0]
OK or Enter.
F10 to exit SMIT.
Part 2 - Group administration
__ 17. As your company grows you may need to add additional system administrators to
perform certain administrative tasks. As root, create a user named helper that
assists you in user and group administration only, by being a member of the
security group. Add security to helper's group set, not the primary group. Make
the password the same as the user name.
# smit
Security & Users ->
Users ->
Add a User
User NAME [helper]
Group SET [security]
OK or Enter.
Go back to Security & Users menu.
Passwords ->
Change a Users Password
User NAME [helper]
Changing password for "helper"
helper's New password: helper
Re-enter helper's New password: helper
OK or Enter.
__ 18. Exit SMIT. Page through the /etc/group file. Locate security and ensure that helper
is listed there.
F10 to return to the command line
pg /etc/group
__ 19. Log out and back in as helper. Execute the groups command to ensure that you
belong to two groups, staff and security. As a new system administrator, add a new
group called budget. List all groups to verify the addition of the new group.
# exit
login: helper
helper's Password: <enter password>

V4.1
EXempty $ groups or setgroups or id

$ smit
Security & Users ->
Groups ->
Add a Group
Group NAME [budget]
OK or Enter.
Go back to the Groups menu.
Select List All Groups
__ 20. Now add two new users, userb and userc making both members of budget as their
group set.
Go back to Security & Users menu.
Users ->
Add a User
User NAME [userb]
Group SET [budget]
OK or Enter.
Go back to Add a User menu.
User NAME [userc]
Group Set [budget]
OK or Enter.
__ 21. Exit SMIT. Assign the passwords for userb and userc. Make the password for each
user the same as the user name.
Exit SMIT
$ pwdadm userb
Changing password for "userb"
helper's Password: <type helper>
userb's New password: <type userb>
Re-enter userb's new password: <type userb>
$ pwdadm userc
Changing password for "userc"
helper's Password: <type helper>
userc's New password: <type userc>

Re-enter userc's new password: <type userc>
__ 22. Log off and back in as userb. (Because this is userb's first login, you are prompted
to change the password. Simply keep it the same. You should be familiar with what
to do; therefore, the prompts are not shown). Change to the budget group and
create a file named report. Add group write permission to report. Do a long listing
on the file to verify that the group access is the budget group. Return to the default
group.
Other than the command you used to change groups, what other command
could be used? _________________________________________________
The command setgroups budget is the same as newgrp budget. The
command setgroup -r is the same as newgrp.
$ exit
login: userb
userb's Password: <enter password>
$ id
$newgrp budget
$ id
$vi report
(type a few lines)
$ chmod g+w report
$ ls -l report
$ newgrp
$ id
__ 23. Let's test to see if a member of the budget group can edit the file report. Log out
and back in as userc. As in the step above, this is userc's first time to log in.
Change the password accordingly. Edit report and add your name. Log out.
Did userc have to issue any special command before editing the file? _____ No
Why? _________________________________________________________
When userc tries to edit the file in userb's directory, the group budget is
checked in /etc/group. Since userc's name appears in the budget group,
access is permitted.
$ exit
login: userc
userc's Password: <enter password>

V4.1
EXempty $ vi /home/userb/report
(type your name)
$ exit
Part 3 - Customizing the default .profile file
__ 24. When a user is created, the default .profile file is copied into the user's home
directory. As a system administrator, login, su to root, and edit the default .profile
file so that all newly created users have by default the EDITOR variable automatically
set up to /usr/bin/vi and their primary prompt string reflects the user's current
directory. (Hint: remember that the PS1 variable is for the primary prompt string and
the PWD variable contains the value of the current directory.)
Note
You may want to go back and remove the PS1 variable assignment from the
/etc/security/.profile file so the $ becomes the primary prompt string.
$ su - root
# vi /etc/security/.profile
EDITOR=/usr/bin/vi
PS1='$PWD=> '
export PATH EDITOR PS1
__ 25. Create a user named userd. Assign a password to userd. Log out and back in as
userd. You should see the current directory listed as your prompt. Next display the
value of the EDITOR variable. If the results are not what you expected, repeat the
previous step and try again.
# smit mkuser
User NAME [userd]
OK or Enter.
Exit SMIT
# pwdadm userd
# exit
login: userd
userd's Password: <enter password>
/home/userd=> echo $EDITOR
/usr/bin/vi

V4.1
EXempty Part 4 - Removing users
__ 26. Log in as root. Before you remove userb from the system, locate the stanza for
userb in /etc/security/passwd.
# pg /etc/security/passwd
__ 27. Remove userb from the system. Take the default to the option Remove
AUTHENTICATION which is yes.
What is this doing? ______________________________________________
When removing authentication, the stanza for userb is removed from the
/etc/security/passwd file. If you were to select no, userb's stanza would remain
in the file. The reason the option is given is to provide continued support to an
additional authentication method, should one be customized for your system.
Authentication methods are not covered in this course.
# smit rmuser
User NAME [userb]
OK or Enter.
F10 to exit SMIT
__ 28. Page through the /etc/security/passwd file to ensure that the stanza for userb was
removed.
# pg /etc/security/passwd
__ 29. When a user is removed from the system, the user's home directory and files owned
by the user are NOT removed. Verify that userb's home directory and files still exist.
Remove all files and directories associated with userb.
# ls -ld /home/userb
# ls -al /home/userb
# rm -r /home/userb
Part 5 - Communicating with users
__ 30. As root, edit /etc/motd and add to the file in all uppercase "TODAY IS USERA'S
BIRTHDAY. SURPRISE CAKE AT 11:00."First, check that the user access
permissions on the file have write permission. If not, set write for user.
# ls -l /etc/motd
# chmod u+w /etc/motd (if necessary)
# vi /etc/motd
TODAY IS USERA'S BIRTHDAY. SURPRISE CAKE AT 11:00
__ 31. Display the message by logging out and back in as usera.

# exit
login: usera
root's Password: <enter password>
__ 32. Now create a zero-length hidden file in usera's home directory named .hushlogin.
You want everyone that logs in to see the output of /etc/motd except usera. Log out
and back in as usera. The /etc/motd file should not have been displayed.
$ touch /home/usera/.hushlogin
$ exit
login: usera
usera's Password: <enter password>
__ 33. As root, create a file named newsitem in the /var/news directory. Type in a few
lines. Log out and back in as teamxx and read the news.
$ su
# cd /var/news
# vi newsitem1
(Type a few lines)
# exit
# exit
login: teamxx
teamxx's Password: <enter password>
$ news

V4.1
EXempty __ 34. To do this step you must have more than one user on your local system. If you are
the only person on your system and you have two terminals, log in on both terminals
using two different teamxx IDs. Perform the exercise from one terminal and look for
the message on the other terminal. As root, send a note to all users on the system
that the system shuts down in 30 minutes.
# wall The system will be shutdown in 30 minutes
Part 6 - Examine the security set up
__ 35. As root, examine any of the security files you have not yet seen.
$ su root
# more /etc/passwd
# more /etc/security/passwd
# more /etc/security/user
# more /etc/group
# more /etc/security/group
# more /etc/profile
# more /etc/environment
# more /etc/security/environ
# who /etc/security/failedlogin
# more /etc/security/login.cfg

V4.1
EXempty Part 7 - Customizing the login herald
__ 36. If you are sharing your machine with other students, you need to work together to
configure the new login prompt. ONLY ONE PERSON PER MACHINE CAN
PERFORM THIS STEP.
You can customize the login prompt that appears on an individual terminal or
customize it so that all terminals use the same login herald depending on what
stanza you change in /etc/security/login.cfg. In this step, you set up a global
default herald.
The herald should read:
. Your Company Name
Unauthorized users will be prosecuted.
login:
First, make a backup of the /etc/security/login.cfg file. If you corrupt this file, you
are not able to log in again.
# cp /etc/security/login.cfg /etc/security/login.cfg.bk
__ 37. Edit the default stanza in /etc/security/login.cfg and add the herald line.
Tip! - Do not use <Enter> at what looks like the end of the line. Let vi wrap the line.
The \r and \n are escape sequences that will signal to the <RETURN> and <NEW
LINE> for you. Note that for some terminal emulations, the \r <carriage return> is
displayed as a literal. On those emulations, the \n <new line> handles both the
carriage return and new line feed.
When finished, save the file and log out.
# vi /etc/security/login.cfg
default:
sak_enabled = false
logintimes =
logindisable = 0
logininterval = 0
loginreenable = 0
logindelay = 0
herald = "\n\n YOUR COMPANY NAME\r\nUnauthorized users will be
prosecuted..\r\nlogin: "
# exit
$ exit
__ 38. Check to see if the login prompt is what you expect. You must be using an ascii
interface to see the effect of the change.
For students using a local console: If you are using an ASCII screen, your new
herald should be displayed. If you are using CDE, you cannot see your changes
unless you select the Option button and selection Command Line. If you don't see
your new herald, then check your entry in /etc/security/login.cfg.
For student using remote access, attempt a new remote connection. For example
you might use a telnet protocol either via a telnet command or via PuTTY.
End of exercise

V4.1

1. There are many ways to prevent other users from accessing a particular user account
on the system. Two ways demonstrated in this exercise are not allowing others to su to
the user account and setting the number of failed logins. Did you notice any other
attributes that would enforce the same kind of action?
Answer: The following are similar or more restrictive:
a. User can log in remotely - If marked false, the user account cannot be remotely
accessed with the telnet or rlogin commands.
b. Allowed login times - Specifies time of day and days of week the user is allowed to
log in. Enter a comma separated list of day,time.
c. Valid ttys - Key in a list of terminals that can access this user account.
d. User can log in - This is more like the account locked feature. This option indicates
whether the user can log in using the login command.
2. When adding a user to a group, does the group have to already exist?
Answer: Yes. As you add users to a group, the user name is automatically placed in the
/etc/group file after the associated group name. If that group name is not in this file, the
task fails.
3. In Step 11, root could su to usera even though usera has the option Another user
can SU TO USER? set to no. Take a moment and log in as teamxx. Now try to su to
usera. You should receive the messages:
You are not allowed to su to this account. Cannot su to usera : account is
not accessible.
4. In Step 33, students work with the news command. The news command writes system
news items to standard output. This keeps you informed of news concerning the
system. Each news item is contained in a separate file in the /var/news directory. Users
should include the news -n command in their /etc/profile to display news each time
they log in. Any user having write permission to the /var/news directory can create a
news item.

V4.1
EXempty Exercise 17. Scheduling

(with hints)
Estimated time
00:30

The purpose of this exercise is to give you the opportunity to use the
scheduling facilities of at, batch, and crontab. These tools are very
important for balancing a system for off-peak usage as well as
scheduling periodic tasks.

Use at to schedule one-time-run jobs at specific times
Use batch to submit jobs into a queueing facility
Use crontab files to schedule repetitive jobs at specific times
Introduction
You can issue a command by entering it on the command line. It can
run in the foreground or the background. A command can also be
issued with nohup allowing you to log off and go home while the
command is still running. The scheduling utilities covered in this
exercise allow you to schedule commands to run after you have left for
the day.

The most common problem is the slow typist or the students that issue
commands to run five or ten minutes from the time they issued them
and then get impatient waiting. Help both types of students cope with
these issues during this exercise.
If students execute the steps in instruction 23 too quickly in
succession, they could lock the screen because the cron update may
not have completed the first edit before trying to do the second edit.
Copyright IBM Corp. 1997, 2008 Exercise 17. Scheduling 17-1

AIX V4.2 and later uses a version of the BSD mail daemon which may
cause problems when performing some of the steps in this exercise.
The BSD mail daemon assumes the use of DNS rather than
/etc/hosts. Thus, if /etc/hosts is used instead, it takes quite a while
for mail to appear. To fix this situation, it is necessary for the
/etc/hosts file to contain entries which look like the following:
10.0.0.1 sys1. sys1
To improve performance further, do the following:
vi /etc/netsvc.conf
Add the line: hosts=local and then save the file
chmod 400 /etc/netsvc.conf
In the June 2005 update of these materials, a series of steps that
depended on the students ability to view the LEDs of the lab system
being used were rewritten to facilitate offering of this course using a
wider range of hardware platforms and in a wider range of
environments, particularly with remote lab systems.

V4.1

Preface
classroom.
Part 1 - Using at
__ 1. Log in as teamxx. Submit a single at job that immediately echoes the message
Running an at command. and then echoes another message This is it! Check
your mail. Notice the shell script name assigned to your job.
$ at now
echo Running an at command
echo This is it!
<ctrl-d>
$ mail
Hint
Wait for a minute or so for the mail daemon to process and forward the mail. Keep
checking for new mail.
__ 2. Repeat the above instruction, but this time force your output to your screen.
(Suggestion: Use the tty command to get your device name.)
Hint
Enter tty at the command line to find out what your /dev/xxxx device name is.
$ at now
echo Running an at command > /dev/xxxx (xxxx is replaced by your device
name)
echo This is it! >/dev/xxxx
<ctrl-d>

__ 3. Experiment with the time specifications of next week and next month. After
entering jobs to be executed next week and next month, list the at jobs waiting to
run and check to see whether the jobs you just submitted appear in the list.
$ at now next week
echo $PWD
<ctrl-d>
$ at now next month
echo $PWD
<ctrl-d>
$ at -l
__ 4. Let's take a look at the environment when an at job executes. Change directory to
/etc. Submit an at job that returns the current directory to your screen and have it
execute in one minute. Now change directory to /var/adm and repeat the at
request. Change directory to the /(root) directory, list the at queued jobs, then wait
for the execution of your two jobs.
$ cd /etc
$ at now +1 min
echo $PWD >/dev/xxxx (xxxx is device you are using)
<ctrl-d>
$ cd /var/adm
$ at now +1 min
echo $PWD >/dev/xxxx (xxxx is device you are using)
<ctrl-d>
$ cd /
$ at -l
Wait for your jobs to run.
What can you deduce from the results you obtained in this exercise step?
_____________________________________________________________
The commands all execute relative to the directory you were in when you
executed the command. The first execution of echo $PWD was associated with
the /etc directory, where you were when you issued the command. The second
execution of echo $PWD was associated with the /var/adm directory, where you
were when you issued the command. When you switched to the /root directory
and waited for the commands to execute, it did not matter. The script was set up
when you did a <ctrl-d> to submit the job.

V4.1
EXempty __ 5. You should now have several at jobs scheduled. su to root to give yourself proper
permission for viewing these jobs. Change to the directory /var/spool/cron/atjobs.
Examine some of the files in this directory and notice the additional statements that
are created for the at job besides the ones that the user entered.
Note
pg won't work very well because lines get truncated. For one thing, you should see
that the directory for the at job does indeed get set specifically within the at job
itself. If the content of a file looks disjointed, it is because nulls are used as field
delimiters within the file. When you have finished, return to your teamxx ID.
$ su root
# cd /var/spool/cron/atjobs
# ls -l
# cat <filename>
# <ctrl-d>
__ 6. As teamxx, list your at jobs. Use the appropriate option to cancel several of them.
Note
You need to specify the whole job name.
$ at -l
$ at -r <jobname>
__ 7. Switch to root to gain proper permission to view the security files.

Page the /var/adm/cron/at.deny file. Notice that there are no entries.
What security implications does this have? ____________________________
Since this file exists, and it has no entries, anybody can use at.
For this exercise we won't modify this file, but if we were to, we would put one userid
on each line to specify each user for whom we wanted to deny access. Notice, also,
that there is no at.allow file. Exit the root subshell and return to teamxx.
$ su
root's Password: (enter password)
# cd /var/adm/cron

# pg at.deny
# ls -l /var/adm/cron
(Notice there is no at.allow file.)
# exit

V4.1
EXempty Part 2 - Using batch

__ 8. As teamxx, submit a job to batch that outputs the result of the date command to
your screen after waiting 60 seconds (Hint: Use the command sleep 60).
Remember, it is submitted exactly like an at job but without a date and time because
the time at which the job will be run is determined by system load. Unless your
system is very busy, you should see the output from your batch job quite soon. Wait
for your job to run.
$ batch
sleep 60
date > /dev/xxxx
<ctrl-d>
Part 3 - Using crontab Files

__ 9. Switch to root. Switch to the /var/adm/cron directory. Display the contents of the
cron.deny file. Also, look in the directory to see if a cron.allow file exists. The
expected result is that cron.deny exists, but contains no user names, and that
cron.allow does not exist.
$ su
# cd /var/adm/cron
# pg cron.deny
# ls -l
__ 10. Return to teamxx and change to your home directory.

# exit
$ cd
__ 11. Create a zero length file called datefile in the /tmp directory.
$ touch /tmp/datefile
__ 12. In your home directory, create a file called cron00 (which will be used as a crontab
file) which specifies that the output of the date command will be appended (using
the >> operator) to the file /tmp/datefile every minute from now on. Remember to
use full path names for commands used in crontab files. Remember, there can be

NO blank lines in the file including lines created by pressing Enter after the last
entry.
$ vi cron00
* * * * * /usr/bin/date >> /tmp/datefile
__ 13. Place your crontab file in the crontabs directory via the crontab command. Then,
use the appropriate option with crontab to look at the crontab entry that you
created.
$ crontab cron00
$ crontab -l
__ 14. Use the -f option of the tail command to check for additions to the file
/tmp/datefile. If you have done the last few steps correctly, you should see an
addition to the file in 1 minute, another in 2 minutes, and so on. Wait to see 3 or 4
additions to the file, then terminate the tail -f session.
$ tail -f /tmp/datefile
^c
$
__ 15. As teamxx, use the appropriate option of the crontab command to remove your
current crontab file. Verify that it has been removed.
$crontab -r
$crontab -l

V4.1
EXempty Part 4 - Optional

The remaining exercise steps are optional. Complete them if there is time available in
this lab session and if you would like more practice working with AIX scheduling
facilities.
__ 16. Continuing as teamxx, cd back to your home directory and create a shell script
called sizechk to:
__ a. Echo the message Active files over 2 blocks in /home.
__ b. Display the date and time.
__ c. Display the names of all files that are larger than 2 blocks in /home and that have
been modified within the last 2 days.
Hint
Use the find command, redirecting error messages to /dev/null).
__ d. Echo Done.
__ e. Make sure all output goes to the screen.
__ f. Make the script executable.
$ cd /home/teamxx
$ vi sizechk
echo Active files over 2 blocks in /home > /dev/xxxx
date > /dev/xxxx
find /home -size +2 -mtime -2 -print > /dev/xxxx 2>/dev/null
echo Done > /dev/xxxx
$ chmod +x sizechk
__ 17. Test your script. If you don't get any output, change the blocksize to 1.
$ sizechk
__ 18. Run sizechk as an at job without redirecting output here (let the script do it). Make
sure that your at job runs and that your output goes to your screen.
$ at now < sizechk

__ 19. Verify the system date and time. Create a file called cron01 that causes sizechk to
be invoked every minute for the next 5 minutes for today only. Remember to use full
path names for commands used in crontab files. Remember, there can be NO blank
lines in the file including lines created by pressing Enter after the last entry.
$ date
$ vi cron01
25-29 3 12 5 * /home/teamxx/sizechk
(This is just an example to be tailored for the current date and time. 25-29
represents each of the next 5 minutes, 3 represents the current hour using
military time, 12 represents the current day of the month, 5 represents the
current month of the year, * is in the position of the current day of the week, and
the last field contains the actual command. Make this execute for the current
date and time.)
__ 20. Place your crontab file in the crontabs directory via the crontab command. Wait
for the next 1 minute interval and check that your sizechk script did indeed get
invoked by cron and that it executed successfully. While you are waiting, use the
appropriate option with crontab to look at the crontab entry that you created.
$ crontab cron01
$ crontab -l
__ 21. Switch user to root. Change directory to /var/spool/cron/crontabs and look at the
files that are in this directory. Your cron01 should be there with the name teamxx.
Examine the contents of several of the files. They may contain statements that are
commented and could be uncommented and/or modified based on specific
installation requirements. They provide a starting point for the execution of some
administrative commands in different areas.
$ su
root's Password:(enter password)
# cd /var/spool/cron/crontabs
# ls -l
# pg teamxx
__ 22. Return to teamxx. Edit the current crontab file by adding to the existing crontab
data a line that executes the sizechk script on the hour, every hour of every day
except Sunday. Wait a moment, and then verify that this line has been added to your
crontab. Once you are done, remove this particular entry so it does not drive you
crazy the rest of the week. Verify that it was removed.

V4.1
EXempty # <ctrl-d>
$ crontab -e (add the following line and close the file)
0 * * * 1-6 /home/teamxx/sizechk
$ crontab -l
$ crontab -e (delete the lines you just added)
$ crontab -l
__ 23. Switch to root. Examine the file /etc/inittab and notice how the cron daemon
actually gets started.
$ su
# pg /etc/inittab
cron will respawn if it is not running. Why is this important?
_____________________________________________________________
You never have to worry about the cron daemon not running on your system as
it is respawned anytime it dies. If you are depending on cron to run jobs after
hours, it must remain up without operator intervention.
Return to your teamxx ID.
# exit
End of exercise

1. How many crontab files can a user have?
Answer: Each user can only have one crontab file
2. Why are crontab files given the name of the user id rather than letting a user name
their own file?
Answer: To ensure that no one overwrites someone else's crontab file accidentally by
giving it the same name that another user has used.
3. What is the major difference between at and cron?
Answer: at should be used for one-time jobs. cron is used for jobs that must be run
multiple times.
4. What happens if you submitted an at job to run at midnight and the system goes down
at 23:00 and doesn't come back up until 8:00 the next morning?
Answer: Luckily, because the at job is stored in the /var/spool/cron/atjobs directory, it
should run as soon as the system comes up.
5. What happens if you submitted a cron job to run at midnight and the system goes down
at 23:00 and doesn't come back up until 8:00 the next morning?
Answer: Unlike at and batch, cron jobs do NOT execute when the system comes back
up. You need to know if your system crashed and check to see if there were any
crontab files that should have run that did not.
6. Using at, where does standard out go by default?
Answer: The screen. You have to redirect standard out with the commands you specify
to at.

V4.1
EXempty Exercise 18. Printers and queues

(with hints)
Estimated time
00:40

This exercise is designed to provide the student with the opportunity to
configure and manage printers and queues using AIX Print Spooling.

Add a printer using SMIT
Add a queue using SMIT
Make changes to existing queues using SMIT
Install printer support software if needed
Introduction
This exercise is designed for you to become familiar with adding a
printer and a queue. In the area of printing, there are many commands
that can be used to do the same thing. In many of these cases, we
have provided only one command method and the other method takes
you through SMIT. If you are already comfortable with certain UNIX
commands for handling print activities, like submitting a print job,
please feel free to use that command. Some of the functions provided
by AIX can only be accessed through the command shown or through
SMIT.
Printer configurations vary widely in the classrooms. Check with your
instructor for specific instruction for your site.
If more than one student is sharing an AIX system, but performing this
exercise separately, be sure that the queue names are unique. For
example, name one of the queues xxdraft where xx is the student's
initials. Wherever you see draft in the exercise instructions, use your
unique name. Wherever you see the queue name quality, use a
queue name such as xxquality where the xx is the student's initials.
Copyright IBM Corp. 1997, 2008 Exercise 18. Printers and queues 18-1

This lab is designed to run without any physical printers, either local or
remote. To support working with print queues, the students need to
define a remote printer with the client also being the print server. There
are steps in the exercise which cannot be done in that situation. In
others, specifically the print status displays, they get error messages
about the print server not having the local printer to serve to the client.
The students are still able to complete most of the lab but they are not
able to see the job print out. The queues can still be created, jobs can
still be sent to the queues and the queues can still be managed.
Some classrooms with local systems have physical printers, either
local or remote. For those environments, if you wish to see printouts,
you need to customize Part 1 of the lab to use the local environment
information. For example you need to know the type of printer, or the
address of the network attached printer or the address of the server to
which the printer is attached. If the printer is server attached, you need
to configure that server to support the student systems acting as print
clients. If using physical printers, before starting this exercise it would
be a good idea to check all printers for a reasonable supply of paper.
Occasionally, the paper jams and causes delays. If this occurs and the
jam can't be quickly resolved, suggest that the people on the system
with the jammed printer team up with another group who are not
having paper problems.
There is a shell script that is used in the Optional Exercises. It should
be located in the /home/workshop directory. The shell script is named
lab6. This directory and the files in it should have been loaded prior to
the start of class. If they are not, you can easily create this script.
Below are the contents of the script. Be sure to make it executable.
Since it contains a subsystem command, only root can execute it.
#!/usr/bin/ksh
echo "Working, please wait .\c"
stopsrc -s qdaemon 2>/dev/null 1>/dev/null
echo ".\c"
echo "\n\n:\n" >>/etc/qconfig 2>/dev/null
echo ".\c"
If students request this script, refer them to Appendix D in the
Student Guide.

V4.1

Preface
classroom.
Part 1 - Add a local printer and a queue

__ 1. If your AIX system has printer port(s) available, then define a local printer (this Part
1) and then proceed to Part 3.
If your AIX system does not have printer port(s) available (possibly in a remote
LPAR environment), then create a file device queue instead (Part 2) and then
proceed to Part 3.
__ 2. Log in and su to root. Check that the AIX Print Spooling system is activated. Using
SMIT, configure your printer with a local queue called quality.
Note
If you don't have the correct software support you will be prompted to install the
software first before adding the printer and the queue. If this is the case, go to the
Part 2 - Installing Printer Support Software.
login: teamxx
Password: <password-for-teamxx>
$ su
# smit
Print Spooling ->
AIX Print Spooling ->
Add a Print Queue
You then get a series of pop-up menus that prompt you for selections of where the
printer is attached, who the manufacturer is, the model of the printer, the type of
printer interface and the parent adapter. The following shows the selections that
should be made for each pop-up menu:
Select local for the ATTACHMENT TYPE
Select the manufacturer of your printer, like IBM
Select the printer model you have, like 4201-2
At this point, you may get a menu titled Printer Support Not Installed if the
software support for the printer you selected was not installed.This notification
window takes you through the SMIT software installation screens. Use these to
install the printer support from the installation media or in the course environment
from /usr/sys/inst.images. When prompted for the input device, press F4 for a
listing of valid devices. Select the installation device, then press Enter.
You are prompted for the SOFTWARE to install:
Press F4 to get a listing of software that is on the installation media.
Cursor through the selections until you find the listing for your printer. When the
printer drivers you need are selected, press Enter.
Ensure that Preview Only is set to NO. Press Enter to install the software. This
takes a couple of minutes.
Now, continue with the following steps:
Select parallel for the Printer Interface
Select ppa0 Standard I/O Parallel Port Adapter for the Parent Adapter
This brings you to the dialog screen:
*Name of new PRINT QUEUE to add [quality]
__ 3. Using SMIT, add a second queue to the printer and name it draft.
# smit
Print Spooling ->
Add a Print Queue
Once again, you are presented with a series of pop-up menus (similar to the ones in
the previous step) that prompt you for selections of where the printer is attached,
who the manufacturer is, the model of the printer, the type of printer interface and
the parent adapter. The following shows the selections that should be made for each
pop-up menu:
Select local for the ATTACHMENT TYPE
Select the manufacturer of your printer, like IBM
Select the printer model you have, like 4201-2
(Select the same printer that you selected in the previous step).

V4.1
EXempty The next pop-up prompts you to either Add a new printer or select the existing
printer below. Select the existing printer.
This brings you to the dialog screen:
*Name of new PRINT QUEUE to add [draft]
__ 4. Exit SMIT and display the status of your print queues.

# lpstat
Part 2 - Add a file device queue

This part provides an alternative to working with a printer device in partitioned machine
environments where there are no serial or parallel port that can be used for a printer
definition.
__ 5. Log in and su to root.
__ 6. Using SMIT, configure an ASCII queue (called quality) with an attachment type of
file and a printer type of other (generic printer). The make the requested /dev device
name a value of null.
Any print jobs sent to this queue (if in a READY status) is sent to /dev/null. You
disable the queues, thus simulating a printer problem. In that state you are able to
practice the same print queue management procedures as would be used with any
print queue.

smit
Print Spooling ->
Add a Print Queue ->
Select file for the ATTACHMENT TYPE
Select other for the Type of Remote Printing
Confirm that your want a printer type of generic
* Name of existing FILE in /dev directory [null]
Press Enter
Names of NEW print queues to add
ASCII [quality]
Press Enter.
__ 7. Using SMIT, add a second queue name it draft.
Use the exact same procedure as described in the previous step.
__ 8. Exit SMIT and display the status of your print queues.
__ 9. Bring down both the quality and draft queues. Check the status of your queues to
be sure that both queues are down.
# disable quality
# disable draft
# lpstat

V4.1
EXempty Part 3 - Check the queue
__ 10. Examine the /etc/qconfig file to confirm the addition of the new queue and device
stanzas. Notice the location of the quality queue relative to the draft queue.
Which queue is listed first? ______ quality
# pg /etc/qconfig
__ 11. Using the qchk command, check the status of your newly created queues. They
should be marked READY.
Which queue is listed first? _______ quality
Which queue do you think is the default queue? ____________
The one that is listed first in the /etc/qconfig file.
# qchk -A
__ 12. Using SMIT, make the draft queue the system default queue.
# smit
Print Spooling ->
Manage Print Queues ->
Set the System's Default Print Queue
*System DEFAULT QUEUE name [draft]
__ 13. Exit out of SMIT and check the status and order of your queues.
Press F10 to exit SMIT
# qchk -A
__ 14. Create a file that can be used as a print test file. Put the file in the /tmp directory and
name the file printxx, where xx represents your initials. Put about 10 lines of data in
it.
# vi /tmp/printxx
Press i to begin inserting text.
Put 10 lines of anything you want in it.
<esc> to change to command mode < : >.
At the command entry prompt, type wq which will save and exit.
__ 15. Test your queues and printer by sending the printxx file you created in the step
above to both the draft queue and to the quality queue.
# qprt /tmp/printxx
# qprt -Pquality /tmp/printxx
-OR-
# lp /tmp/printxx
# lp -dquality /tmp/printxx
-OR-
# lpr /tmp/printxx
# lpr -Pquality /tmp/printxx
-AND-
# qchk -A

V4.1
EXempty Part 4 - Change the characteristics of a queue

If you are working with a remote queue, you are not able to set default print jobs
characteristics or the print queue discipline which are for local print queue management. If
that is your situation, then skip the next two steps.
__ 16. If you do not have a physical printer in the lab, you are not able to see the effects of
your changed job attributes, but it is still useful to know where to set these defaults.
Using SMIT, change the attributes of the quality queue by adjusting some printer
characteristics. These may be different depending on the printer used. Try making
some changes. Basically, change the look of the printed output. Here is a list of
some changes you might try (depending on your printer):
Use double spacing
Set a top margin of 20 lines
Specify 50 lines per page
Use double high printing
Provide a separator page at the head of the job
# smit
Print Spooling ->
Change/Show Print Queue Characteristics
*PRINT QUEUE name [quality]

At the pop-up menu you are provided with 4 choices. Select choice 2, Default Print
Job Attributes. That gives you a panel of attributes to change. Some possibilities
are:
DOUBLE SPACING? [yes]
TOP MARGIN (lines) [20]
LINES PER PAGE [50]
SEPARATOT PAGES [header for job; no trailer]
DOUBLE-HIGH print? [yes]
__ 17. If you have physical printers, send the printxx file to the quality queue to ensure
that the changes were successfully made to the queue.
# qprt -Pquality /tmp/printxx
-OR-
# lp -dquality /tmp/printxx
-AND-
# qchk -A
__ 18. Take the draft queue down, if not already down.

# disable draft
# lpstat
-OR-
# smit
Print Spooling ->

Stop a Print Queue
*PRINT QUEUE name [draft]
Press Enter to stop the queue.
Press F3 to go back to Manage Print Queues menu.
Show Status of Print Queues.
Include status of print queues on remote servers? [no]
Press Enter.
__ 19. Change the queue characteristics of the draft queue so that its discipline is set to
sjn, shortest job next. Check the status of the queues to verify that the change was
made successfully.
# chque -q draft -a 'discipline = sjn'
-OR-
# smit
Print Spooling ->

Change/Show Print Queue Characteristics
At the pop-up menu select: 4 Queuing Discipline
*Queuing DISCIPLINE [shortest job next]
Press Enter.

V4.1
EXempty
__ 20. Send the file /etc/security/user to the draft queue and check to see that it is in the
queue.
Next, send the /tmp/printxx file you created in the previous exercise to the draft
queue and again display the job queues.
Which job is scheduled to run first after the queue is brought back up?
________ /tmp/printxx
Why? ________________________________________________________
It is the smaller of the two files and the queuing discipline has been set up as
shortest job next.
# qprt /etc/security/user
# lpstat
# qprt /tmp/printxx
# lpstat
Part 5 - Manage jobs in queues
__ 21. Bring down the quality queue, if not already down. Check the status of your queues
to be sure that both queues are down.
# disable quality
# lpstat
__ 22. Send the files /etc/qconfig and /etc/filesystems to the quality queue and send the
file /etc/motd to the draft queue.
# qprt -Pquality /etc/qconfig
# qprt -Pquality /etc/filesystems
# qprt /etc/motd
__ 23. Check the order of the print jobs waiting in the two queues. They should all have the
same priority value (Pri). In the quality queue, the /etc/filesystems file should be
behind the /etc/qconfig file, because it was submitted later.
Change the print priority of the /etc/filesystems job to 25 so that it becomes the
first print job scheduled for the quality queue.
# lpstat -t
# qpri -#<job number> -a 25
# qchk -L -P quality
-OR-
# lpstat -t
# smit
Print Spooling ->
Manage Print Jobs ->
Prioritize a Print Job
*Print JOB NUMBER [] (Use F4 to list)
*Priority LEVEL [25]
Press Enter.
# qchk -L -P quality
__ 24. Let us assume that you really wanted the /etc/filesystems printed with the
characteristics associated with the draft queue. Move the /etc/filesystems print job
to the draft queue.

V4.1
EXempty # qmov -mdraft -#<job number>

# lpstat
-OR-
# smit
Print Spooling ->

Move a Job Between Print Queues
*DESTINATION PRINT QUEUE name [draft]
SOURCE (Select ONE of the following)
Print JOB NUMBER [job number]
Press Enter.
# lpstat
__ 25. You realize that you already printed /etc/security/user and you don't need to do it
again. Cancel the job.
# cancel <job number>
# lpstat
-OR-
# qcan -x<job number>
# lpstat
-OR-
# smit
Print Spooling ->

Cancel a Print Job
PRINT QUEUE containing job [draft]
*Print JOB NUMBER [job number]
Press Enter.
Press F10 exit SMIT.
# lpstat
__ 26. You want to print all the jobs currently in your queues except for the
/etc/filesystems job, which you want to print at a later time. Put a hold on the
/etc/filesystems job, verify the job is being held, and then bring up your two queues.
# qhld -#<job number>
# lpstat
# enable quality draft
# lpstat
-OR-
# smit
Print Spooling ->

Hold / Release a Print Job
Hold or Release jobs [hold]
Press Enter.
F3 back to the Print Spooling menu.
Show Status of Print Queues
Include status of print queues on remote servers? [no]
Press Enter.
F3 back to the Manage Print Queues menu.
Start a Print Queue
*PRINT QUEUE name [quality]
Press Enter.
F3 back to the Manage Print Queues menu.
Start a Print Queue
F10 to exit SMIT.
# lpstat
__ 27. Check to see all jobs have printed successfully. Then, release the /etc/filesystems
job so it too can print.
# qhld -r -#<job number>
-OR-

V4.1
EXempty # smit
Print Spooling ->

Hold / Release a Print Job
Hold or Release jobs [release]
Press Enter.
Part 6 - Add a remote print queue (optional)
__ 28. Work with another team to do remote printing. One team acts as the print server and
the other will act as the print client. Record, below, the IP address of the other
teams AIX system (not your own).
______________________________________________________________
# host $(hostname)
__ 29. Each team log in and su to root
__ 30. On both platforms, add a name resolution entry to you local hosts file for the other
teams machine if not already resolvable.
# host <hostname of other teams system>
if there is no resolution, then continue with the following steps:
# smit hosts
Add a Host
* INTERNET ADDRESS (dotted decimal) [other team IP address]
* HOST NAME [other team hostname]
press Enter
F10 to exit SMIT
__ 31. On the server platform, using SMIT, add print access for the client system and start
the print server subsystem (lpd daemon).
smit
Print Spooling ->
Manage Print Server ->
Add Print Access for a Remote Client
Name of the REMOTE CLIENT [client-hostname]
Press Enter.
F3 back to the Manage Print Server menu.
Select Start the Print Server Subsystem.
Take the defaults.
Press Enter.
F10 to exit SMIT.

V4.1
EXempty __ 32. On the client system, using SMIT, configure a remote print queue (called remote)
with your server systems hostname as the server with and a server print queue
name of quality. Define it with a timeout value of 1 minute.
smit
Print Spooling ->
Add a Print Queue ->
Select remote for the ATTACHMENT TYPE
Select standard processing for the Type of Remote Printing
Name of the QUEUE to add [remote]
HOSTNAME of the remote server [server-hostname]
Name of QUEUE on the remote server [quality]
Backend TIMEOUT period [1]
Press Enter.
__ 33. On the client system, send the files /etc/motd and /etc/filesystems to the remote
queue.
# qprt -Premote /etc/motd
# qprt -Premote /etc/filesystems
__ 34. On the client system, check to see the status of the print jobs.
# qchk -A
__ 35. On the server, see if the remote print jobs are in the quality queue.
# qchk -A
__ 36. Bring up your two local queues.

# enable quality draft
End of exercise
Optional Exercise
__ 37. As the root user, run the script /home/workshop/lab6. This script creates problems
with the printing subsystem. Try printing a file. Look at the error message and check
the file that is being referenced. Try to fix the problem and try printing a file again. If
the problem originally presented was fixed, a new error message is generated.
Make sure the resource mentioned in the error message is working and then try
printing a file again.
# ksh /home/workshop/lab6
# qprt /etc/motd
The error message you get looks something like this:
digest: error in config file /etc/qconfig, line 94.
digest: (FATAL ERROR): No device line in queue stanza.
enq: (FATAL ERROR): Error from digester /usr/lib/lpd/digest,
status=1024, rv = 11924.
The error numbers may be different depending on what is in /etc/qconfig.
# pg /etc/qconfig
# vi /etc/qconfig
<ESC>: set nu <shift G> (To get to the end of the file).
Look around line 94 (Or whatever line number your message indicated).
dd (Delete the line with the stray character).
<shift zz> (To save and quit the file).
# qprt /etc/motd
The error message you get looks something like this:
enq: (WARNING): Qdaemon appears to be dead.
enq: errno = 2 No such file or directory
enq: WARNING):Cannot awaken qdaemon.(request accepted anyway)
(The last 2 messages you will get several times)
# lpstat (Check to see if your job is in the queue).
# lssrc -s qdaemon
# startsrc -s qdaemon
End of exercise

V4.1

1. Ask if everyone has the AIX Print Spooling active and not the System V Print Spooling.
2. Ask if anyone had to install the printer software.
3. Ask the students what they thought would happen if they tried to change the print
characteristics of a queue to something that is not physically supported by the printer.
Answer: It basically ignores the change.
4. Ask how can you tell what is the default queue is.
Answer: You could look in the /etc/qconfig file. The first queue listed is the system
default queue.
5. Ask how do you send something to the default queue.
Answer: Simply use one of the print commands followed by the name of the file.
6. Ask what happened to the order of the jobs in the draft queue when the queuing
discipline was changed to shortest job next.
Answer: They moved around so that the shortest job was next in line to print.
7. Ask the students what they observed relative to the order for the jobs in the draft queue
when the longest job, /etc/security/user, was given a higher priority.
Answer: It became the first in line to print even though it was the longest job in the
shortest job next queue. Thus, priorities take precedence over type of queue.

V4.1
EXempty Exercise 19. Networking

(with hints)
Estimated time
00:40

This exercise provides an opportunity to configure and use TCP/IP.

Configure TCP/IP
Test TCP/IP connectivity
Use telnet and ftp to communicate on the network
Introduction
TCP/IP has been configured on your systems prior to class in order to
support the NIM-based installation/migration exercise and to provide
remote access to the lab systems. In this exercise, you record the
existing network configuration and then you remove that configuration.
At that point you only have access to the system from the system
console (either a local terminal or via an HMC virtual console). From
the console, you configure TCP/IP using the same values as it had
before.

During the initial machine setup, TCP/IP was configured. In this
exercise, students first de-configure TCP/IP and then reconfigure it.

If no one can telnet or ftp to any other system in the room or if the
two commands are extremely slow in execution, check for the
existence of the /etc/resolv.conf file on the system.
Copyright IBM Corp. 1997, 2008 Exercise 19. Networking 19-1

For Ethernet, be sure to tell the students which cable type to use in the
lab system. In most cases they can select n/a (the adapter connection
type can be auto-detected) or tp. Some local system environments
might be using BNC, though that would be rare these days.

V4.1

Preface
classroom.
If there are multiple interfaces on your system, your instructor identifies which interface
to use for this exercise. If you are in an environment where the AIX system is accessed
remotely, be sure to not deconfigure the interface used for that purpose.
Part 1 - Deconfigure TCP/IP

__ 1. If you are using a local non-partitioned system, login to the locally attached system
console as teamxx and su to root. Then go to step 6.
$ su
#
__ 2. If you are using a remote partitioned system, access your systems HMC by starting
a Web browser and using the HMCs IP address as the URL. If you do not have this
IP address, ask your instructor.
Start a Web browser, the Firefox Web browser should be available on the eLab Web
Interface. The Firefox application icon should appear in the Applications box.
# https://##.##.##.##

machine and the logical partition which you are assigned.

Userid: hscroot
Password: abc1234


configured lpars will appear in the Contents window.
__ 4. Start a virtual console window for your assigned partition. This acts as your system
In the contents window locate your partition and then:
a) Select your lpar by clicking the Select box to the left of the lpar
name.
c) Drag the mouse to Console Window, and then click Open Terminal
Connection
d) After a short delay a window should appear with an AIX login prompt
__ 5. Log in to your assigned partition as teamxx and su to root. This acts as your system
$ su root
#
__ 6. Collect the current network configuration information.

IP address: ____________________________________________________
IP address name resolution: _______________________________________

V4.1
EXempty network mask: _________________________________________________

default gateway: ________________________________________________
To collect the information run the following commands:
# netstat -in | grep en0
# hostname
# host $(hostname)
# lsattr -El en0
# netstat -r | grep default
__ 7. Identify a partner team with whom you validate network connectivity (this may have
been assigned by your instructor or this could be left to you to work out with other
students). Exchange IP address and hostname information with that partner team.
Record their configuration information below:
IP address: ____________________________________________________
IP address name resolution: _______________________________________
__ 8. ping your partner teams system to verify this currently works.
# ping <IP address or hostname of other system>
__ 9. Detach both the en0 and et0 interfaces.
# ifconfig en0 detach
# ifconfig et0 detach
__ 10. Remove your network interface (en0) and network adapter (ent0) from the ODM.
# rmdev -dl en0 -d
# rmdev -dl ent0
__ 11. Set your hostname to a null string and rename /etc/hosts to /etc/hosts.bak
# hostname ""
# mv /etc/hosts /etc/hosts.bak
__ 12. Run the configuration manager so that it rebuilds the adapter and interface you just
deleted. Once this is complete, you are ready to proceed with the TCP/IP
configuration.
# cfgmgr
__ 13. ping your partner teams system to verify that you cannot communicate.
# ping <IP address or hostname of other system>
__ 14. Display the network interfaces to show that the interface is not configured.
# netstat -in (you should only see the loopback interface)

Part 2 - Configure TCP/IP

__ 15. Configure the network interface for your network using SMIT, TCP/IP, Minimum
Configuration and Startup.
# smit
Communications Applications and Services ->
TCP/IP ->
Minimum Configuration & Startup
Select appropriate interface (ex.en0) from the Available Network Interface pop-up
menu.
*HOSTNAME [your hostname]
*Internet Address [your IP address]
*Network MASK (dotted decimal) [your netmask]
F10 to exit SMIT.
__ 16. Verify that the above action automatically set the hostname and the name
resolution for your own system.
# host $(hostname)
__ 17. Add an entry for your partner system in the /etc/hosts file. You may use SMIT or vi.
# vi /etc/hosts
-OR-
# smit
Communications Applications and Services ->

TCP/IP ->
Further Configuration ->
Name Resolution ->
Hosts Table (/etc/hosts) ->
Add a Host
*Internet Address (dotted decimal) [partner IP address]

*HOST NAME [partner hostname]
F3 to return to the Hosts Table (/etc/hosts) menu.

V4.1
EXempty Part 3 - Testing the configuration

__ 18. Use the network command, ping, to test connectivity to your default gateway by IP
address. If you can not get a response, check your en0 address and mask for
correctness.
# ping <gateway IP address>
<ctrl-c> (to stop the output)
__ 19. Use the network command, ping, to test connectivity to your partner system. ping
by IP address and then by host name. If you can not get a response back display the
contents of your /etc/hosts file to ensure you have an entry for the system and it is
correct. Ensure that the system you are trying to reach has completed the
configuration steps.
# ping <partner hostname>
# ping <partner IP address
__ 20. Execute the host command for the hostname of your partner system and, again, for
the IP addresses of your partner system.
# host <hostname of partner system>
# host <ip address of partner system>
__ 21. Execute the netstat -i command to display the status of the communication
adapter(s) on your system.
# netstat -i

Part 4 - Using telnet

__ 22. Select a system on the network to be your partner. telnet to your partner's system
and log in as root. Execute the hostname command to show that you are logged in to
another system.
# telnet <hostname of partner system>
login: root
# hostname
__ 23. While remotely logged in to your partner's system, create a file named sysxfile
where x is the number of your system. You use this file later in this exercise. List the
contents of root's directory to verify that your file exists.
# vi sysxfile
(Type a few lines that will identify this file as yours)
# ls
__ 24. Close the connection and verify that you are back on your local system by executing
hostname. You should see your host name.
# <ctrl-t>
telnet> close
# hostname
-OR-
# <ctrl-d>
# hostname

V4.1
EXempty Part 5 - Using ftp

__ 25. Transfer the file, sysxfile, from your partner's system to your system renaming it to
ftpdoc. When prompted to log in, log in as root.
# ftp sysy (where sysy is where your partner's system hostname would be)
Connected to sysy
220 sysx FTP server
Name (sysy:teamxx): root
331 Password required for root
Password:<enter root's password>
230 User root logged in.
ftp> get sysxfile ftpdoc
__ 26. Change to your teamxx's home directory ON YOUR system and list its contents.
Pick out a small file that you can transfer to your partner system's /home/teamxx
directory. Since you are root, you should not have any problems with permissions.
You may want to ask your partner which teamxx home directory to use.
ftp> lcd /home/teamxx
ftp> !ls -l
ftp> put <filename>
__ 27. End the file transfer session and verify that you are back on your local system.
ftp> quit
# hostname
End of exercise


V4.1
EXempty Appendix A. Optional exercise: AIX 5L V5.3

installation
(with hints)
Estimated time
Unknown

This exercise provides an opportunity to do an installation of the base
operating system of AIX 5L V5.3. The type of installation is a complete
overwrite installation with all the default settings.

Install the AIX 5L V5.3 operating system
Introduction
This exercise is designed to give you experience in installing a system
as if you had just received a new machine without anything on it. You
should be using the same type of distribution media you would get
from IBM.
If you have more than one student sharing a system (all systems have
a minimum of two terminals attached to them), then only one person
can actually perform these steps. All other students on that system
should observe while the steps are being done.
Requirements
A local system capable of running AIX 5L V5.3 with a CD-ROM
drive attached to it and at least 128 MB of memory and 2.2 GB of
physical disk space
Installation media (CD)
Copyright IBM Corp. 1997, 2008 Appendix A. Optional exercise: AIX 5L V5.3 installation A-1
An ASCII terminal connected to one of the built-in serial ports on

the system or a graphics display station directly attached to the
system that has a CD-ROM drive attached to it.

If the exercise equipment includes graphics displays directly attached
to the pSeries/AIX platform, then encourage the students to select that
as the console during installation. This makes the optional appendix
exercise relating to serial devices go a little bit smoother. If they don't
select the graphics display as the console, you need to help them get it
configured after the installation is complete.
Sometimes students have problems with the menus and figuring out
what to type where. The entry will be shown next to the Choice: line. If
a default is shown, it looks like this:
Choice [0]:
The student's choice should still be placed after the colon. If the enter
key is pressed with no choice made, what is in the brackets will be the
selection. That which is shown in the brackets is the default.

The target disk should not be smaller than 2.2 GB. The classroom
systems are likely have disk drives larger than this. So, the students
should be able to load AIX on a single disk.
A-2 AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
EXempty Exercise instructions

Preface
There is only one version of these instructions available; there is not a separate section
with Hints as the other exercises have. All instructions are included.
classroom.
__ 1. If your system is already on, log in as root. Then, shut down the system.
Console login: root's Password: ibmaix
# shutdown -F
When you see the Halt Completed message on the console, shutdown is
complete. The system then shuts itself off.
__ 2. Turn on all attached external devices, such as terminals, CD-ROM drives, tape
drives, monitors, and external disk drives. DO NOT TURN THE SYSTEM UNIT ON
YET.
__ 3. Insert the installation media into the CD-ROM drive.

Place the first Installation CD in the CD-ROM drawer and push the drawer in.
__ 4. If you are using a graphics display, skip to the next step. If you are using an ASCII
terminal, be sure the communications options on the terminal are set as follows:
Line Speed (baud rate) = 9600
Word Length (bits per character) = 8
Parity = no (none)
Number of Stop Bits = 1
Interface = RS-232C (or RS-422A)
Line Control = IPRTS
Be sure the keyboard and display options are set as follows:
Screen = normal
Row and Column = 24x80
Scroll = jump
Auto LF (line feed) = off
Line Warp = on
Forcing Insert = line (or both)
Tab = field
Operating Mode = echo

Turnaround Character = CR
Enter = return
Return = new line
New Line = CR
Send = page
Insert Character = space
On your IBM 3151 terminal, press the Ctrl+Setup keys to display the Setup Menu
and follow the on-screen instructions to check the settings. If you need to make
changes, follow the on-screen instructions for making changes.
__ 5. Turn the system unit power on. Press F5 after you see the keyboard icon on the
screen. The system will begin booting from the installation media.
__ 6. After about five minutes, a screen appears on all directly attached terminal or
display devices asking you to define the system console. Notice that you are given
these instructions in many different languages. Follow the instructions as they
appear on the device you wish to use as the system console.
__ 7. Next, a screen is displayed that prompts you to select an installation language.

Follow the directions on this screen to select the language in which the installation
instructions will be displayed.
__ 8. The Welcome to Base Operating System - Installation and Maintenance menu

will be displayed. Select item 2, Change/Show Installation Settings and Install.
Welcome to Base Operating System

Installation and Maintenance
Type the number of your choice and press Enter. Choice is indicated by >>>.
1 Start Install now with Default Setting

>>> 2 Change/Show Installation Settings and Install
3 Start Maintenance Mode for System Recovery
88 Help ?
99 Previous Menu
>>> Choice [1]: 2

V4.1
EXempty __ 9. The Installation and Settings menu is displayed. Select item 1, System Settings.
Installation and Settings
Either type 0 and press Enter to install with current settings or type
the number of the setting you want to change and press Enter.
1 System Settings
Method of Installation............. Preservation
Disk Where You Want to Install......hdisk0
2 Primary Language Environment Settings (AFTER Install)

Cultural Convention.......... C(POSIX)
Language..................... C(POSIX)
Keyboard..................... C(POSIX)
Keyboard Type ............... Default
3 More Options (Desktop, Security, Kernel, Software, ...)
0 Install AIX with the current settings listed above.

88 Help ?
99 Previous Menu
>>> Choice [0]: 1
__ 10. The Change Method of Installation menu is displayed. You are now given the
opportunity to change the method of installation. Select item 1, New and Complete
Overwrite.
Change Method of Installation
Type the number of the installation method and press Enter.
1 New and Complete Overwrite

Overwrites EVERYTHING on the disk selected for installation.
Warning: Only use this method if the disk is totally empty or if
there is nothing on the disk you want to preserve.
2 Preservation Install
Preserves SOME of the existing data on the disk selected for
installation.
Warning: This method overwrites the usr (/usr), variable (/var),
temporary (/tmp), and root (/) file systems. Other product
(application) files and configuration data will be destroyed.
3 Migration Install
Upgrades the Base Operating System to current release. Other product
(application) files and configuration data are saved.
88 Help ?
99 Previous Menu
>>> Choice [2]: 1
__ 11. The Change Disks Where You Want To Install menu is displayed. You are now
given the opportunity to select the disk drives where you want the system to be
loaded. Select the disk drive with the greatest amount of space on it, minimum of 2.2
GB, as long as it is marked bootable. If not, select both drives or if there isn't a single
drive with greater than 2.2 GB on it, select both drives.
Change Disks Where You Want to Install
Type one or more numbers for the disk(s) to be used for

installation and press Enter. To cancel a choice, type the
corresponding number and press Enter. At least one bootable
disk must be selected. The current choice is indicated by >>>.
Name Location Code Size (MB) VG Status Bootable
>>> 1 hdisk0 04-C0-00-4,0 2063 rootvg yes

2 hdisk1 04-C0-00-5,0 2063 rootvg yes
>>> 0 Continue with choices indicated above
55 More Disk Options

66 Disks not known to Base Operating System Installation
77 Display More Disk Information
88 Help ?
99 Previous Menu
>>> Choice [0]: <Enter>
__ 12. You are taken back to the Installation and Settings menu. Select item 2, Primary
Language Environment Settings (AFTER install).

V4.1
EXempty Installation and Settings
1 System Settings
Method of Installation............. Preservation

Cultural Convention.......... C(POSIX)
Language..................... C(POSIX)
Keyboard..................... C(POSIX)

88 Help ?
99 Previous Menu
>>> Choice [0]: 2

__ 13. Change the language environment and code set for the language you wish to have
after installation. The menu that is displayed looks like this:
Type the number for the Cultural Convention (such as date, time, and
money), Language and Keyboard for this system and press Enter, or type
106 and press Enter to create your own combination.
Cultural Convention Language Keyboard
>> 1. C (POSIX) C (POSIX) C (POSIX)

2. Albanian English (United States) Albanian
3. Arabic Arabic (Bahrain) Arabic (Bahrain)
... several screens later ...
106. Create your own combination of Cultural Convention, Language
and Keyboards.
88 Help ?
99 Previous menu
Choice [1]:
__ 14. You should now be back at the Installation and Settings menu. If you are happy
with your selections, select 0 to continue with the new and complete overwrite
installation.
Installation and Settings
1 System Settings
Method of Installation............. New and Complete Overwrite

Cultural Convention.......... English (United States)
Language..................... English (United States)
Keyboard..................... English (United States)

88 Help ?
99 Previous Menu
>>> Choice [0]: 0
__ 15. Once the installation begins, a status screen appears. Note the progress indicator at
the bottom of the screen. The installation process will take approximately one hour.

V4.1

1. Ask students if they noticed the different message displayed on the graphics display
versus the one displayed on the ASCII terminal when prompted to select a console.
Answer: The graphics display prompts to press the F1 key and the ASCII display
prompts to press 2.
2. Ask students what would happen if the ASCII display was being used as the console
but was set up differently than the settings the system was expecting.
Answer: Garbage is displayed on the screen.
3. Ask students how they would recover from the above described situation.
Answer: If you fix the terminal setup, and press the correct key, you will be able to
continue with the installation.
4. Ask the students at what point in the process do they think the disk will be corrupted by
the install. This is important if there already is a system loaded out there and they
decide to abort the installation.
Answer: The disk does not get overwritten by the install process until after the last
warning. Up until the last warning is issued, you can still abort the installation without
affecting what is on the hard disk.

V4.1
EXempty Appendix B. Serial devices

(with hints)
Estimated time
Unknown

This exercise gives you an opportunity to add and delete a serial
device and make changes to a serial device that is attached to a
RS/6000. It also gives you the opportunity to see what effect certain
terminal settings have on the behavior of your terminal.

Add a terminal
Change terminal characteristics
Change the settings on an ibm3151 ASCII terminal
Delete a terminal
In this exercise, you are adding a serial device, changing its
characteristics both on the RS/6000 and on the device itself, and then
delete the device. The serial device you use is an ibm3151 ASCII
terminal. The setup screen information may differ depending on
whether or not you are using a cartridge to define your language
environment. The example information provided relative to setup
screens is based on the ibm3151 terminal without a cartridge.
If you have another type of ASCII terminal, your instructor should
provide directions on how it should be setup.
You would use many of the same steps for configuring a modem
attached to a serial port. Your terminal should already be physically
cabled to the first built-in serial port on your system. For this exercise
you will need to work as a team with the other student who is working
on your system.
Copyright IBM Corp. 1997, 2008 Appendix B. Serial devices B-1


Sometimes students with some experience go ahead and configure the ASCII terminal
prior to this exercise. To handle this potential problem we have provided a way for these
students to delete the terminal first, then go back and add it in. For students who have
some knowledge of adding terminals, encourage them to do the delete section anyway
because they might not be aware of the fact that the terminal has to be disabled before
removing it.
Some students have difficulty understanding the built-in serial adapter and port labeling
scheme. The systems should be set up with the ASCII terminal attached to the s1 port.
An optional exercise (Part 6 - Fixing a hung terminal) is included which includes use of
the stty command which is not covered in class. Explain this to the students. The optional
exercise involves using the stty command to work with hung terminals.
For students who chose to do the Optional Exercises if, after executing the
stty -a < /dev/ttyx command, they get the error message /bin/ksh: /dev/ttyx:
cannot open they did not substitute their tty device number for the x in /dev/ttyx. Have
them try the operation again with the appropriate substitution.
In the Optional Exercises, if the lab5 script is run from an Low Function Terminal instead
of the ASCII terminal, the tty command does not work and the terminal may quit functioning
completely. Recovery is to reboot the machine.

Before start of class on Monday, be sure to check that all cables and interposers are in
place and that all terminals work. Occasionally, a cable, or other connector, may have been
borrowed for some other purpose at a site and not yet returned. When this occurs there is
no guarantee that what is left behind will work, particularly if the interposer was borrowed,
but the cable was left behind and plugged in. The only true test is to get a login prompt.
The exercise was written for an ibm3151 ASCII terminal without a cartridge. In some
countries language cartridges are used thus the example setup screens may not map
exactly to the information a student may see when in setup mode. Other ASCII terminals
can be used, however, some modifications will have to be made in the hints section where
the terminal type is provided. For example, some classrooms use ibm3101 ASCII
terminals. The setup menu on these terminals is accessed via <Ctrl>+ <break>. These
terminals must be configured as a Wyse 60. In addition, not all ASCII terminals function
exactly the same. Thus, some of the results may be slightly different, especially from the
optional exercise (Part 6 - Fixing a hung terminal).
There are two shell scripts that are used in the optional exercise (Part 6 - Fixing a hung
terminal). They should be located in the /home/workshop directory. The shell scripts are
named lab5 and lab5.1. This directory and the files in it should have been loaded prior to
the start of class. If they are not, you can easily create these scripts. Be sure the students
B-2 AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1
EXempty only execute these scripts against an ASCII terminal. Below are the contents of each script.
Be sure to make them executable for all users.
lab5
this_tty=`tty`
cat /bin > $this_tty
lab5.1
this_tty=`tty`
stty 38400
If students request these scripts, refer them to Appendix B of the Student Guide.


Preface
classroom.
Part 1 - Verify a terminal

__ 1. At the console, log in as your user and su to root. Check to see if your ibm3151
ASCII terminal is currently configured. If it is, continue to the section titled Part 2 -
Remove A Terminal. Otherwise, if your terminal has not been configured, skip to
the section titled Part 3 - Add a Terminal.
login: teamxx
Enter your password
# su root
# penable
If the output from the penable command looks something like this:
console
tty0
then your ASCII terminal is currently configured. Go to the section titled
Part 2 - Remove A Terminal. If your tty does not appear to be configured, skip to
the section titled Part 3 - Add a Terminal.

V4.1
EXempty Part 2 - Remove a terminal

__ 2. Using SMIT, remove the terminal you just finished setting up. Be sure to take the
definition out of the database. Make sure you are logged off. Does the system let
you?
________________________________________________________________
No, because the port is in use, even if there is only a login message displayed, the
getty program is using the port.
What is the error message it gave you? __________________________________
Cannot perform the requested function because the specified device is
busy.

# smit
Devices ->
TTY ->
Remove a TTY
You get a pop-up screen which lists the available tty devices. Select the terminal
from the list which corresponds to your terminal.
Keep definition in database [no]
OK or Enter.
The system should not have allowed you to remove an active terminal. You must
disable the terminal before you attempt to remove it.
Cannot perform the requested function because the specified device is busy.
__ 3. Disable the terminal.

# pdisable tty0
__ 4. Return to SMIT and complete the process of removing the ASCII terminal.
After you complete this step, be sure to do the next section, Part 2 - Add a terminal
because you will need this terminal for later exercises.

# smit
Devices ->
TTY ->
Remove a TTY

You get a pop-up screen which lists the available tty devices. Select the terminal
from the list which corresponds to your terminal.
Keep definition in database [no]
OK or Enter to perform the action.

V4.1
EXempty Part 3 - Add a terminal

__ 5. Using SMIT, add a terminal on the first built-in adapter. Take all the defaults except
the baud rate which you should set to 19200. Be sure your ibm3151 terminal is
configured to match the communications settings on the RS/6000.

# smit
Devices ->
TTY ->
Add a TTY
Select the type of terminal connection
tty rs232 AsynchronousTerminal
Select the port to which your terminal is attached:
sa0
The next screen is the dialog screen. The only value you need to insert is the
port number. Press F4 to view your choices. Select:
PORT number: s1
Enable LOGIN: enable
BAUD rate: 19200
TERMINAL type: ibm3151
OK or Enter.
Press F10 to return to the command line.

Part 4 - Check terminal communications settings

__ 6. On the ASCII terminal keyboard, press and hold the <ctrl> key and press the
<Setup> key. This puts you in setup mode. The cursor keys move you from one field
to another. The space bar shows you the options for whatever field is highlighted.
The <Send> key takes you to the next screen.

Press the setup key that is at the top right hand side of the numeric keypad of
the ASCII terminal.
This is what the settings may look like on the General menu.
Machine Mode = IBM 3151
Screen = NORMAL
Row and Column = 24x80
Scroll = JUMP
Auto LF (line feed) = OFF
CRT Saver = ON
Line Wrap = ON
Forcing Insert = OFF
Tab = FIELD
Press the <Send> key.
You should now be at the Communication menu. Check to be sure the line
speed (baud rate), word length, parity and stop bit match the system default
selections of 19200, 8, NO, and 1. The operating mode should be ECHO.
This is what the settings may look like on the Communications menu:
Operating Mode = ECHO
Line Speed (bps) = 19200
Word Length (bits) = 8
Parity = NO
Stop Bit = 1
Turnaround Character = CR
Line Control = IPRTS or PRTS
Break Signal (ms) = 500
Send Null Suppress = ON
If you have to make any changes, use the cursor key to take you to the
appropriate field, then use the space bar to cycle through the selections. Once
you have made all the changes, press the <Send> key.
Check the values for the Keyboard menu. Entries should be RETURN, NEW LINE,
CR, PAGE, SPACE.
This is what the settings may look like on the Keyboard menu:
Enter = RETURN

V4.1
EXempty Return = NEW LINE

New Line = CR
Send = PAGE
Insert Character = SPACE
Press the <Send> key again.
You should now be on the Function menu that allows you to save your changes.
Press the right arrow key once to Save.
Press the space bar to save the changes. The word Completed should be
flashing.
Press and hold the <ctrl> key and press the <setup> key to exit setup mode.
Check your terminal. Did you get a login prompt?
__ 7. On the Server console run the penable command that sends a login prompt to your
ASCII terminal.
# penable tty0
__ 8. Log in to the ASCII terminal as root. Try to go into SMIT.

What kind of message do you get? Message:
_____________________________________________________________
_____________________________________________________________
The environment variable TERM is currently set to a terminal type
that does not support the full screen display capabilities required
for SMIT. Set TERM to an appropriate terminal type or use
another terminal. login: root
# smit
__ 9. If SMIT did not run change the terminal type so you can use SMIT and then try
starting SMIT again.
# export TERM=ibm3151
# smit

Part 5 - Change terminal characteristics

__ 10. From another terminal and using SMIT, change the baud rate of your terminal to
9600. You will need to do this from your console. Be sure to disable the ibm3151
before making the changes and enable again once the changes are complete.

# pdisable tty0
# smit
Devices ->
TTY ->
Change/Show Characteristics of a TTY
You are presented with a pop-up screen listing the available tty devices. There
should be only one and it is highlighted.
Select your terminal from the list. You are now presented with the dialog screen.
Set the following characteristics:
Enable LOGIN [enable]
BAUD rate [9600]
TERMINAL type [ibm3151]
OK or Enter.
F10 to exit SMIT.
# penable tty0
__ 11. Once this change is complete on the system, can you work with the terminal? __ No
Why not? _________________________________________________________
Because the terminal settings don't match the port settings on the system.
Change the settings on your terminal so it can communicate with your system.
To change the settings on your terminal, follow the steps shown in Step 6 in Part 4
-Check Terminal Communications Settings. Look for the portion of these
instructions that deal with the Communication menu.

V4.1
EXempty Part 6 - Fixing a hung terminal (optional)

__ 12. DO NOT DO THIS FROM AN LFT. As root, from the ASCII terminal, run the shell
script /home/workshop/lab5. This shell script causes problems with the ibm3151
terminal. Determine what the problem symptoms are, such as a hung terminal
(doesn't show any character when you try to type), or strange characters displayed
when you try to type something. As part of troubleshooting, check the port attributes,
like the baud rate, and whether or not there are any processes running associated
with the terminal. Try to fix the problem using the suggestions in the student
notebook before taking the last resort of turning the terminal off then on.

# /home/workshop/lab5
From another terminal attached to the same system:
- # stty -a < /dev/ttyx
(where x is the number of the tty device. In our case most likely 0)
- # ps -ef | pg
From the problem terminal:
- Try the key sequence <ctrl-q>
- Try the key sequence <ctrl-j> stty sane <ctrl-j>
- # pdisable ttyx
- # penable ttyx
- Wait a couple of minutes. If this doesn't work try to turn off the terminal and
turn it back on again.
Note
The lab5 shell script caused a problem because it attempted to cat several binary
files to your ASCII terminal.
__ 13. DO NOT DO THIS FROM AN LFT. As root, from the ASCII terminal, run the shell
script /home/workshop/lab5.1. This shell script causes display problems with the
ibm3151 terminal. Determine what the problem symptoms are, such as a hung
terminal (doesn't show any character when you try to type), or strange characters
displayed when you try to type something. As part of troubleshooting, check the port
attributes, like the baud rate, and whether or not there are any processes running

associated with the terminal. Try to fix the problem using the suggestions in the
student notebook before taking the last resort of turning the terminal off then on.

# /home/workshop/lab5.1
- # stty -a < /dev/ttyx
(Notice that the baud rate is different than the terminal setting.)
- # stty 9600 < /dev/ttyx (if this doesn't work then try ...)
# pdisable ttyx
# penable ttyx
Turn off the terminal and turn it back on again.
Note
The lab5.1 shell script caused a problem by using the stty command to reset the
baud rate on the ASCII terminal to something that did not match the terminal setup.
End of exercise

V4.1

1. Ask the students what they had to do to get SMIT to work the first time they tried to get
into SMIT.
Answer: They had to set the TERM variable for their terminal type.
2. Ask the students what they had to do to get the terminal to communicate with the
system after they had changed the terminal characteristics.
Answer: They had to change the terminal setup to match the port characteristics.
3. Ask the students what they had to do to be able to remove a terminal.
Answer: First they had to disable the port through the use of the pdisable command.


V4.1
EXempty Appendix C. Optional exercise: The System V

print subsystem
(with hints)
Estimated time
Unknown

This lab demonstrates configuring a printer and printing using the
System V print subsystem.

Add a System V printer
Submit jobs to the printer
Display status of the printer and print jobs
Remove a System V printer
Introduction
This exercise reinforces the material covered in the System V Print
Subsystem appendix. It consists of questions and hands-on activities.
There are two levels of information:
Exercise Instructions
Use any level that facilitates your learning process. Use all available
resources: team members, documentation, your instructor.
Requirements
System running AIX 5L with connected local printer.
Copyright IBM Corp. 1997, 2008 Appendix C. Optional exercise: The System V print subsystem C-1
Instructor exercise overview

These exercises require a local printer for each team of students.
You need to provide students with the following information about their printer:
Printer device (which device driver to use)
How the printer is connected (serial or parallel and which adapter and port)
If the printer is serially connected: baud rate, parity, bits per character, stop bits, and
flow control.
Printer type (terminfo entry)
Content type for the printer (usually this will be simple)
Interface script to use (use standard for non-PostScript printers)
(This lab was written for non-PostScript printers.)
You should TEST these exercises BEFORE you teach this class to verify that your values
will work!!!
The exercises are currently configured to use the command line.
If printers are not available, most of the exercise can be done by printing to a file (specify a
regular file name for the -v flag to lpadmin).
C-2 AIX System Administration I Copyright IBM Corp. 1997, 2008

V4.1

Preface
classroom. If your lab environment uses remote pSeries/AIX platforms, the instructor
will provide you with the addressing (IP address or hostname) and instructions for
accessing the remote machine. For example, you might be using telnet, an X-windows
server, or a VNC client. Whatever tool you are using, at various points in the labs you
may need to re-launch your remote access tool to obtain a new login prompt.
If your classroom does not have printers, you can perform most of these exercises by
printing to a file. Notes are provided in the instructions for this option.
Part 1 - Switch print subsystems

__ 1. Use switch.prt -d to determine which print subsystem is active.
Unless someone has changed it, the AIX print subsystem should be active.
__ 2. If the System V print subsystem is active, change to the AIX print subsystem.
What command did you use? ____________________ switch.prt -s AIX
__ 3. What file is /usr/bin/lpstat linked to? ___________________ /usr/aix/bin/lpstat
__ 4. Switch to the System V print subsystem.

What command did you use? _____________________ switch.prt -s SystemV
__ 5. What file is /usr/bin/lpstat linked to? _________________ /usr/sysv/bin/lpstat
Part 2 - Add a System V printer

__ 6. Collect the following information from the instructor:
Printer device _____________________
(If you will be printing to a file, the device will be the name of the target file.)
How the printer is connected:
- (serial or parallel) _____________________,
- which adapter _____________________ and
- which port it is connected to _____________________.
(If you will be printing to a file, you do not need this information.)
If the printer is connect via a serial connection, details of the connection (baud
rate, parity, bits per character, stop bits, and flow control):
____________________________________________________________
_____________________________________________________________
(If you will be printing to a file, you do not need this information.)
Printer type _____________________
(If you will be printing to a file, use printer type unknown.)
Printer content type _____________________
(If you will be printing to a file, use content type simple.)
Interface script _____________________
(If you will be printing to a file, use the standard interface script.)
__ 7. Choose a name for your printer: _____________________
__ 8. Log in as root.
__ 9. Configure the printer device:

(If you are printing to a file, skip the steps in the table and simply create the target
file using the touch command, for example:
# touch /tmp/dummy_printer)

V4.1
EXempty
Step Action
Use SMIT fastpath smit pdp to open the Printer/Plotter Devices
1
menu.
2 Move the cursor to select Add a Printer/Plotter; press <Enter>.
3 Scroll through the device list and select the printer device.
4 Select the type of interface (connection).
5 Select the parent adapter.
For parallel connected printers, enter <p> in the PORT number field and
6
press <Enter>.
For serial connected printers, enter the port number in the PORT
7
number field, fill in the baud rate and so forth and press <Enter>.
__ 10. List the newly created printer device.

What command did you use? ______________________
If using a printer:
# lsdev -Cc printer
If printing to a file:
# ls -l directory
(where directory is the directory containing your target file)
__ 11. Use lpadmin to define a new System V printer using the printer device you just
created. Use the information you gathered in steps 1 and 2 to configure the printer.
What command did you use? ___________________________________
# lpadmin -p printer_name -v device_name -T printer_type \
-I content_type -m interface_script
# lpadmin -p myprinter -v /dev/lp0 -T ibmproprinter \

-I simple -m standard
# lpadmin -p myprinter -v /dev/lp0 -T bj-300
__ 12. Use lpstat to verify queueing status.

What command did you use? _____________________
Is the printer accepting requests? _____________________ No
What reason is given? _____________________ New printer

# lpstat -a (lists status for all printers)
-OR-
# lpstat -a printer_name (lists status for printer_name only)
__ 13. Use lpstat to verify printing status.

What command did you use? _____________________ lpstat -p
Is the printer enabled? _____________________ No
__ 14. Use lpstat to get a long listing of your printers characteristics.

What command did you use? ________________ lpstat -p -l printer_name
__ 15. Using the output from the previous step, answer these questions:
What users are allowed to access this printer? ____________________ all
What action will be taken On fault? (This is the default Alert action.) _________
mail to root once
__ 16. View the printer configuration file.

What command did you use? ___________________________________________
cat /etc/lp/printers/printer_name/configuration
__ 17. Navigate to the spool directory.

What command did you use? _____________ cd /var/spool/lp/tmp/host_name
__ 18. List the files in this directory: __________________________________________

Unless there are other printers defined on this host, the only file should be .SEQF.

V4.1
EXempty Part 3 - Submit print jobs

__ 19. Attempt to print /etc/motd (the message of the day file) on your new printer.
What command did you use? _____________________
lp -d printer_name [-L C] /etc/motd
What happened? __________________________________________
Print request was refused with this error message:
UX:lp: ERROR: 0920-067 Requests for destination canon aren't being
accepted.
TO FIX: Use the lpstat -a command to see why this destination is not
accepting requests.
__ 20. Enable queueing using the accept command. (Dont enable printing yet.)
What command did you use? _____________________ accept printer_name
__ 21. Resubmit your print job.

Note the request-ID here: ________ printer_name-X (where X is the job number)
__ 22. Use lpstat to list outstanding job status.

What command did you use? ____________________ lpstat -o
__ 23. List the files in the spool directory: ________ ls /var/spool/lp/tmp/host_name
__ 24. Examine the request file (X-0) and answer these questions:
What file will be printed? ____________________ File to be printed is the F line
What is the files size? ______________ size is in the O line, the flist attribute
__ 25. Submit another request to print /etc/motd, but this time, request that the print
service copy the file to the spool directory.
What command did you use? _____________ lp -d printer_name -c /etc/motd
__ 26. List the new files in the spool directory: __________________________________

Y-0 and Y-1 (where Y is the request number)
__ 27. Examine the two new files.

Which file is the request file? _____________________ request file: Y-0
Which file is the copy of the file to be printed? _______________ file to print: Y-1
__ 28. Enable printing on your printer.

What command did you use? ___________________ enable printer_name
__ 29. Your print jobs should begin printing as soon as you enabled the printer.
Use lpstat -p while the printer is printing to show what job is being printed.
(If your jobs do not print, ask the instructor for help.)
__ 30. Examine the request log.

What command did you use? _______________ pg /var/lp/logs/requests
Look for request IDs in the file.
What character at the beginning of the line indicates a new entry (new print
request) in this file? _____________________________________________
Each new entry starts with a line starting with =
__ 31. Disable printing on your printer using the disable command.

What command did you use? __________________ disable printer_name
__ 32. Disable queueing using the reject command.

What command did you use? ___________________ reject printer_name
__ 33. Use lpadmin to remove a System V printer.

What command did you use? ________________ lpadmin -x printer_name

V4.1
EXempty Part 4 - (Optional) Configure a remote print server and client

__ 34. Use Printing for Fun and Profit under AIX 5L (Redbook), Chapter 3. System V
network printing as a reference for this exercise.
__ 35. You need to work with another team.
__ 36. One team configures their system as a print server machine (see Section 3.3 in the
Redbook). The other team configures their system to use the other machines
remote printer (see Section 3.1 in the Redbook).
__ 37. Test your configuration by attempting to print from the client machine to the printer
on the server machine.
__ 38. Make notes below of the steps you took and your results.
End of exercise
Exercise Review
In the exercise the following material was covered:
switch.prt was used to display the active print subsystem and to switch print
subsystems
SMIT was used to configure a printer device
lpadmin was used to define a System V printer
lpadmin -x was used to remove a System V printer
lpstat -a and lpstat -p were used to display queueing status and printing status
(respectively)
The printer configuration file (/etc/lp/printers/printer_name/configuration) was
examined
lp was used to submit print jobs
accept was used to enable queueing
reject was used to disable queueing
lpstat -o was used to display job status
enable was used to enable printing
disable was used to disable printing
Spool files (/var/spool/lp/tmp/host_name/*)were examined
The request log (/var/lp/logs/requests) was examined

V4.1
backpg
Back page

PK8YF A40 Instructorhints PDF

Uploaded by

Copyright:

Available Formats

PK8YF A40 Instructorhints PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PK8YF A40 Instructorhints PDF

Uploaded by

Copyright:

Available Formats

What is the purpose of the document?

What is the purpose of the document?

What printing subsystems are discussed in the document?

What printing subsystems are discussed in the document?

V4.

Front cover

Instructor Exercises Guide

IBM certified course material

February 2008 edition

Intel is a trademark or registered trademark of Intel Corporation or its subsidiaries in the

iii AIX System Administration I Copyright IBM Corp. 1997, 2008

iv AIX System Administration I Copyright IBM Corp. 1997, 2008

Instructor exercises overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi

Exercises configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

Exercise 1. root login methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

Exercise 2. Using SMIT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

Exercise 3. System startup and shutdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1

Exercise 4. AIX software installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

Exercise 5. AIX 6.1 installation and configuration . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

Exercise 6. Web-based System Manager . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

Exercise 7. Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-1

Exercise 8. LVM commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1

Copyright IBM Corp. 1997, 2008 Contents v

Exercise 9. Working with LVM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9-1

Exercise 10. Working with file systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10-1

Exercise 11. Managing file systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11-1

Exercise 12. Paging space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12-1

Exercise 13. Using backup and restore . . . . . . . . . . . . . . . . . 13-1

Exercise 14. (Optional) Using tar and cpio . . . . . . . . . . . . . . . . 14-1

Exercise 15. Security files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15-1

Exercise 16. User administration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16-1

vi AIX System Administration I Copyright IBM Corp. 1997, 2008

TOC Part 6 - Examine the security set up . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16-16

Exercise 17. Scheduling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-1

Exercise 18. Printers and queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18-1

Exercise 19. Networking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19-1

Appendix A. Optional exercise: AIX 5L V5.3 installation . . . . . . . . . . . . . . . . . . . A-1

Appendix B. Serial devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B-1

Appendix C. Optional exercise: The System V print subsystem. . . . . . . . . . . . . C-1

Copyright IBM Corp. 1997, 2008 Contents vii

viii AIX System Administration I Copyright IBM Corp. 1997, 2008

Copyright IBM Corp. 1997, 2008 Trademarks ix

x AIX System Administration I Copyright IBM Corp. 1997, 2008

pref Instructor exercises overview

Copyright IBM Corp. 1997, 2008 Instructor exercises overview xi

fragcopy - Makes copies of the 512-byte file to test the number of

xii AIX System Administration I Copyright IBM Corp. 1997, 2008

pref Exercises configuration

Copyright IBM Corp. 1997, 2008 Exercises configuration xiii

For a class size of 24, there should be at least 12 AIX systems.

xiv AIX System Administration I Copyright IBM Corp. 1997, 2008

pref Exercises description

Copyright IBM Corp. 1997, 2008 Exercises description xv

xvi AIX System Administration I Copyright IBM Corp. 1997, 2008

EXempty Exercise 1. root login methods

What this exercise is about

What you should be able to do

Common Student Problems

Known Hardware/Software Problems

15. Do you have root privileges? Check /etc/security/passwd. _______ Yes.