V11.

cover

Front cover
Course Exercises Guide
with hints
Power Systems for AIX - PowerVM I:
Implementing Virtualization
Course code AN30G ERC 8.0
June 2019 edition
Notices
This information was developed for products and services offered in the US.
IBM may not offer the products, services, or features discussed in this document in other countries. Consult your local IBM representative
for information on the products and services currently available in your area. Any reference to an IBM product, program, or service is not
intended to state or imply that only that IBM product, program, or service may be used. Any functionally equivalent product, program, or
service that does not infringe any IBM intellectual property right may be used instead. However, it is the user's responsibility to evaluate
and verify the operation of any non-IBM product, program, or service.
IBM may have patents or pending patent applications covering subject matter described in this document. The furnishing of this
document does not grant you any license to these patents. You can send license inquiries, in writing, to:
IBM Director of Licensing
IBM Corporation
North Castle Drive, MD-NC119
Armonk, NY 10504-1785
United States of America
INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY
KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some jurisdictions do not allow disclaimer
of express or implied warranties in certain transactions, therefore, this statement may not apply to you.
This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein;
these changes will be incorporated in new editions of the publication. IBM may make improvements and/or changes in the product(s)
and/or the program(s) described in this publication at any time without notice.
Any references in this information to non-IBM websites are provided for convenience only and do not in any manner serve as an
endorsement of those websites. The materials at those websites are not part of the materials for this IBM product and use of those
websites is at your own risk.
IBM may use or distribute any of the information you provide in any way it believes appropriate without incurring any obligation to you.
Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other
publicly available sources. IBM has not tested those products and cannot confirm the accuracy of performance, compatibility or any other
claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those
products.
This information contains examples of data and reports used in daily business operations. To illustrate them as completely as possible,
the examples include the names of individuals, companies, brands, and products. All of these names are fictitious and any similarity to
actual people or business enterprises is entirely coincidental.
Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines Corp., registered in many
jurisdictions worldwide. Other product and service names might be trademarks of IBM or other companies. A current list of IBM
trademarks is available on the web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml.
© Copyright International Business Machines Corporation 2009, 2019.
This document may not be reproduced in whole or in part without the prior written permission of IBM.
US Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.
V11.3
Contents

TOC

Contents
Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii

Exercises description . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv

Exercise 1. Virtual I/O Server Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-1

Part 1: Configure a new Virtual I/O Server installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2
Part 2: Activate the Virtual I/O Server . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-13

Exercise 2. Dual VIOS Virtual SCSI Device Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-1

Part 1: Configure storage access on VIOS1 and VIOS2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-4
Part 2: Configure a virtual SCSI disk for LPAR1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-7
Part 3: Configure path and disk attributes on LPAR1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-21
Part 4: Explore path priorities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-23
Part 5: Configure a virtual SCSI disk for LPAR2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-24
Part 6: Check virtual adapter configuration from HMC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-31
Part 7: Test the client MPIO configuration - OPTIONAL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-35

Exercise 3. Configuring File-Backed Optical Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-1

Part 1: Create the Virtual Media Repository and store an ISO file . . . . . . . . . . . . . . . . . . . . . . . . . . 3-2
Part 2: Install AIX Image From Virtual Optical Media Device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-9

Exercise 4. Dual VIOS Virtual Fibre Channel Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

Part 1: Configure virtual Fibre Channel adapters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-2
Part 2: View the status of the N_Ports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-6
Part 3: Configure MPIO in the client partition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-7
Part 4: Check the virtual Fibre Channel tunables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
Part 5: Test the paths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-12

Exercise 5. Virtual Ethernet Adapter Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

Part 1: Configure an internal Virtual Ethernet Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-2
Part 2: Configure internal Virtual Ethernet Adapters on client . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-4
Part 3: View virtual Ethernet statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-8
Part 4: View virtual Ethernet configuration in partition profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5-9

Exercise 6. Dual VIOS Shared Ethernet Adapter configurations . . . . . . . . . . . . . . . . . . . . . . . . . 6-1

Part 1: Determine current IP address configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-2
Part 2: Create a Dual VIOS SEA configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-3
Part 3: View and test the dual VIOS SEA configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-10
Part 4: Configure a virtual Ethernet interface on the client in the bridged network . . . . . . . . . . . . 6-12
Part 5: View activity on the adapters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-16
Part 6: Using network analysis tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-19

Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover . . . . . . . . . . . . . . . 7-1
Part 1: Create an SR-IOV vNIC with failover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-2
Part 2: View vNIC configuration in VIOS, HMC, and partition profiles . . . . . . . . . . . . . . . . . . . . . . . 7-9
Part 3: Test vNIC failover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-17
Part 4: (Optional) Test vNIC Failover (VIOS failure) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7-22

Exercise 8. PowerVM system maintenance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-1

© Copyright IBM Corp. 2009, 2019 i

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Contents

TOC Part 1: Update the Virtual I/O Server software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-2

Part 2: Using viosbr to backup the virtual and logical configuration . . . . . . . . . . . . . . . . . . . . . . . . . 8-4
Part 3: Using cron to schedule tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8-8

© Copyright IBM Corp. 2009, 2019 ii

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Trademarks

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, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business
Machines Corp., registered in many jurisdictions worldwide.
The following are trademarks of International Business Machines Corporation, registered in many
jurisdictions worldwide:
AIX 6™ AIX® C3®
developerWorks® Express® OS/400®
Passport Advantage® Power® PowerVM®
POWER6® POWER7® POWER8®
POWER9™ Redbooks® System Storage®
Tivoli®
Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both.
Java™ and all Java-based trademarks and logos are trademarks or registered trademarks of
Oracle and/or its affiliates.
Other product and service names might be trademarks of IBM or other companies.

© Copyright IBM Corp. 2009, 2019 iii

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercises description

pref

Exercises description
In the exercise instructions, you can check off the line before each step as you complete it to track
your progress.
Most exercises include required sections, which should always be completed. It might be
necessary to complete these sections before you can start later exercises. If you have sufficient
time and want an extra challenge, some exercises might also include optional sections that you can
complete.
This course includes two versions of the course exercises: "with hints" and "without hints".
The standard "Exercise instructions" section provides high-level instructions for the tasks you
should perform. You need to apply the knowledge you gained in the unit presentation to perform the
exercise.
The "Exercise instructions with hints" provide more detailed instructions and hints to help you
perform the exercise steps.

© Copyright IBM Corp. 2009, 2019 iv

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

Exercise 1. Virtual I/O Server

Configuration
(with hints)

Overview
This lab exercise covers the creation and configuration of a new Virtual I/O Server partition.

Objectives
At the end of the lab, you should be able to:
• Configure a new Virtual I/O Server installation
• Use the cfgassist command

Introduction
In this exercise, students will configure a new Virtual I/O Server installation by creating a partition,
configuring its resources, and configuring a network interface using the cfgassist tool.

Requirements
• This workbook
• A workstation with a web browser connected to a network
• A POWER8 or later processor-based server with a dedicated HMC
• A system from which to start an SSH session to the HMC

© Copyright IBM Corp. 2009, 2019 1-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
Exercise Instructions with hints

Preface
• All exercises for this unit depend on the availability of specific equipment in your lab and at your
desk. You’ll need:
▪ A managed system
▪ A system from which to execute remote commands to the HMC (via SSH)
▪ All lab systems need to be accessible to each other on a network
▪ One disk with the Virtual I/O Server software image installed
• Hints are provided for lab exercises in case you need them and provide solutions for the lab
exercise steps. All hints are marked by a » sign.

Introduction
The managed system in the IBM lab is a:
▪ POWER8 processor-based server with 20 physical processors and 64 GB of memory
There are four Fibre Channel controllers. One Fibre Channel controller will be assigned to each of
four Virtual I/O Servers (VIOS). The Fibre Channel controller provides the VIOS with access to SAN
LUNs to be used as vSCSI-based rootvg for all the LPARs, and to provide access to NPIV-based
disks to be used in a datavg in LPAR2. A VIOS software image has been installed on the first disk
to be used on each VIOS created by a student. The server has four Ethernet adapters for each
VIOS to use in Shared Ethernet Adapters. Each server has a Single Root Input/Output
Virtualization (SR-IOV) Ethernet adapter. The SR-IOV adapter is used on each the LPAR1 and
LPAR2 partitions for Ethernet network access. The server has an integrated disk controller
connected to physical internal disks. The internal disks will not be used.
In this exercise, you will create the VIOS partition and complete the configuration of the existing
newly installed image.
When a task is performed in the HMC GUI in this exercise, the Enhanced+ mode will be used. HMC
v8 R8.6.0 Service Pack (SP) 1 is installed. This release (specifically SP1) introduces the login
screen where Enhanced+ mode is the default login.

Part 1: Configure a new Virtual I/O Server installation

__ 1. Start a web browser and connect to your assigned HMC. Logon with your assigned HMC
userid.
Note: Just before you log in, there are likely to be security questions. Read the windows and
answer positively or to continue in spite of certificate issues.
Below is a sample of the new login screen showing login by a student with the login ID of
studentx in Enhanced+ mode. Use the login ID and password provided by the instructor:

© Copyright IBM Corp. 2009, 2019 1-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

» Start the web browser.

» Using the https protocol, enter your assigned HMC’s hostname or IP address as the web
address. Example: https://HMC_IP_address
» Answer appropriately to any security questions. Accept security risks associated with
certificates.
» Use your assigned HMC login userid and password provided by your instructor.

© Copyright IBM Corp. 2009, 2019 1-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

Note

In the exercise instructions, when using the HMC Enhanced+ GUI, the Toolbar will be used to
access and select resources, and the Menu Pod will be used to perform tasks on those resources.
The Toolbar is located in the upper left corner of the HMC GUI. The Menu Pod is along the left edge
of a selected resource’s window. The Toolbar and Menu Pod look like this:

Many of the instructions will be based on operating from the Menu Pod of the managed
system. From the Menu Pod of the managed system, many of the tasks and resources on
the managed system are accessed. In the graphic above, notice the Menu Pod for a
managed system is divided into sections. The sections are Capacity, System Actions,

© Copyright IBM Corp. 2009, 2019 1-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
Partitions, Properties, PowerVM, Capacity On Demand, Serviceability, and Topology.
Selecting the Partitions link will display the AIX/Linux or IBM i partitions on the managed
system (this is the default view of the managed system). Selecting the Virtual I/O Servers
link will display the VIOS partitions on the managed system. From either of the displays, the
individual partitions are selected and tasks are performed on those partitions. When the
exercise requires navigating back to a previous view or accessing other partitions, the
instructions will indicate to use the “bread crumbs” located above the view that is being
displayed. Here is an example of navigating back to the managed system called bud to
select a new resource or task to perform:

__ 2. In the Toolbar, navigate to the All Systems view, if the login did not take you to that view.
__ 3. Click on the name of your assigned managed system. Here is an example showing the All
Systems view, selecting the managed system named bud.

__ 4. The Partitions view is displayed for the selected managed system. This shows the partitions
that are defined on the managed system. 
You will be creating your VIOS partition. To do that, select the Virtual I/O Servers link under
the PowerVM section in the Menu Pod.

© Copyright IBM Corp. 2009, 2019 1-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

__ 5. The Virtual I/O Servers view is displayed. If there are VIOS partitions defined by other
students in the class, they will be displayed. There might not be any VIOS partitions defined.

Note

The display is in one of three views, Gallery, Table or Relationship. The default is Gallery. Select the
other views, and settle on one that is most comfortable for you. The instructions will show the Table
view when lists of partitions are displayed.

__ 6. Select Create Virtual I/O Server button. This will start the Add VIOS Wizard. Navigate the
wizard as indicated in the steps below to create your VIOS partition. Use the values your
instructor has provided for your assigned VIOS partition, rather than what is shown in the
example steps below. The table in this step provides the VIOS and LPAR assignments that
the instructor should have given you. The examples in the steps below are for team 1,
student 2 on the managed system.

Table 1: Student VIOS and LPAR assignments

Assigned Second LPAR1 LPAR2 LPAR3
VIOS VIOS
Student 1 - Team 1 T1_VIOS1 T1_VIOS2 S1_LPAR1 S1_LPAR2 S1_LPAR3
Student 2 - Team 1 T1_VIOS2 T1_VIOS1 S2_LPAR1 S2_LPAR2 S2_LPAR3
Student 3 - Team 2 T2_VIOS1 T2_VIOS2 S3_LPAR1 S3_LPAR2 S3_LPAR3
Student 4 - Team 2 T2_VIOS2 T2_VIOS1 S4_LPAR1 S4_LPAR2 S4_LPAR3
__ a. On the General tab
VIOS Name = T#_VIOSN, where:
T# is T1 for team1, T2 for team2 (your assigned team)
N is 1 for VIOS1 and 2 for VIOS2 (your assigned VIOS)
Partition ID = Your student number (1, 2, 3, or 4)
If you get the following error, change it to your student number plus 50 (for example,
student 2’s alternate partition ID would be 52).

© Copyright IBM Corp. 2009, 2019 1-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

Click Next.
__ b. On the Processor tab
Processor Mode = Shared
Weight = 255
Virtual Processors: Maximum = 4, Allocated = 2, Minimum = 1
Processing Units: Maximum = 2, Allocated = 1, Minimum = 0.5
Click Next.

__ c. On the Memory tab

Choose GB as the units (click the slidebar)
Memory Allocation: Maximum = 4, Allocated = 2, Minimum = 1
Click Next.

© Copyright IBM Corp. 2009, 2019 1-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

__ d. Skip Physical I/O assignments...click Next

The physical I/O will be assigned after the VIOS is created to provide practice with
modifying an existing LPAR’s configuration.

© Copyright IBM Corp. 2009, 2019 1-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

__ e. On the Configuration Summary tab...take the defaults, click Finish.

__ f. Monitor the progress of the wizard. There should be no errors as the wizard displays
progress as shown in the example below.

© Copyright IBM Corp. 2009, 2019 1-9

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
If you get errors, try to figure out the cause. If you are uncertain how to proceed, contact
the instructor.
__ g. When you get the message that the VIOS Partition was created successfully, close the
Add VIOS Wizard window by clicking the x in the upper right corner of the window.
__ 7. Verify that your newly created VIOS is in the Virtual I/O Servers view. The example shows
the table view where T1_VIOS1 is already active, and T1_VIOS2 has been created and is
starting. Your VIOS name might not contain the system-name prefix (sys8702-)as shown in
the example. 
If you do not see your newly created VIOS, navigate to the All Systems view, and then to
your assigned managed system view, then click on the Virtual I/O Servers link to refresh
the table.

Important

Recall that you didn’t add the physical I/O adapters to your VIOS when you created it. That is the
next task to perform, and it will provide more exposure to and practice with the Enhanced+ GUI.
You must add a Fibre Channel Adapter, and a 2-Port PCI Ethernet adapter to your VIOS partition.

__ 8. Select your VIOS by clicking on the name. This will display the details of the VIOS. Notice
the Menu Pod changed to display sections and tasks related to your VIOS.
__ 9. Select Physical I/O Adapters from the Menu Pod.

© Copyright IBM Corp. 2009, 2019 1-10

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

__ 10. In the Physical I/O Adapter display, select the Add Adapter button.
__ 11. The Add Physical I/O Adapter window is displayed. 
Be careful at this step! Choose the correct adapters for your VIOS.
The instructor will provide you with the adapter slot numbers for a Fibre Channel adapter,
and a 2-port Ethernet adapter to assign to your VIOS. Find those adapter slot numbers. If
you do not see those slot numbers, or can not find the information provided by the
instructor, contact the instructor. Do not randomly choose adapters.
In the list, choose the Fibre Channel and 2-port Ethernet adapters that are assigned to you.
Choose only those adapters, and double check that they are the correct ones.
The example below shows team2 selecting the C5 Fibre Channel adapter and C9 2-Port
Ethernet adapter:

© Copyright IBM Corp. 2009, 2019 1-11

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty

__ 12. Click OK
__ 13. You are returned to the Physical I/O Adapter window. The adapters have not been added
until you save the change. Notice at the top of the window, the Save button is in focus. You
must click on it to save the new adapter into the VIOS configuration.
Click the Save button.
__ 14. Verify a few more settings before activating your VIOS. 
Click on General Properties in the Menu Pod.
Notice that the Save button at the top of the display is not active.
__ a. Select the Advanced button to reveal additional settings at the bottom of the display.
You will see a blue checkmark in focus on the button after you click it.
__ b. Select the option to Allow Performance Information Collection
__ c. If you plan to use this VIOS in Live Partition Mobility operations, select the Mover
Service Partition option, and optionally select the Enable Time Reference option.

© Copyright IBM Corp. 2009, 2019 1-12

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
__ d. Lastly, verify that the Save Configuration changes to profile option is set to Enabled.
It should be enabled as the default when creating the partitions from the wizard.
__ e. If you changed anything, click Save to save the configuration changes.
__ 15. This completes the creation of the VIOS partition.

Important

The VIOS is now ready to be installed. During lab setup, the lab manager installed a VIOS 2.2.6.21
image for you on the first SAN LUN on the Fibre Channel adapter that you assigned to your VIOS.
Instead of installing the VIOS, you will boot that pre-installed image, and then configure the HMC
and VIOS to enable your AIX partitions to use your VIOS for virtualized storage and network
resources.
In the next part of the lab, you activate the VIOS into SMS mode, and select the first SAN LUN as
the boot device. You will then assign TCP/IP network settings on your VIOS.

Part 2: Activate the Virtual I/O Server

__ 16. You will need a console when activating the VIOS partition. The console can be started from
the GUI, but this can be unreliable because of various Java-related complications. Instead,
you will open the console using the HMC CLI and the vtmenu command.
__ a. Open an SSH session to your assigned HMC using Putty or a terminal session on your
workstation. If you are having problems connecting to the HMC using SSH, contact the
instructor.
__ b. At the login prompt, login with your assigned HMC userid and password.
__ c. Enter the command, vtmenu
__ d. From the list of partitions, choose the number associated with your VIOS. In the
example below, option 13 is chosen to open the console for T1_VIOS1. The console is
open and waiting for input.
----------------------------------------------------------
Partitions On Managed System: sys870
OS/400 Partitions not listed
----------------------------------------------------------
1) S1_LPAR1 Not Activated
2) S1_LPAR2 Not Activated
3) S1_LPAR3 Not Activated
4) S2_LPAR1 Not Activated
5) S2_LPAR2 Not Activated
6) S2_LPAR3 Not Activated
7) S3_LPAR1 Not Activated

© Copyright IBM Corp. 2009, 2019 1-13

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
8) S3_LPAR2 Not Activated
9) S3_LPAR3 Not Activated
10) S4_LPAR1 Not Activated
11) S4_LPAR2 Not Activated
12) S4_LPAR3 Not Activated
13) T1_VIOS1 Not Activated
14) T1_VIOS2 Not Activated
15) T2_VIOS1 Not Activated
16) T2_VIOS2 Not Activated

Enter Number of Running Partition (q to quit): 13

13
Opening Virtual Terminal On Partition T1_VIOS1 . . .

Open in progress

Open Completed.
__ 17. Switch back the HMC GUI window. It should still be on the General Settings window for your
newly created VIOS.
__ 18. To activate the VIOS logical partition, select the VIOS Actions link in the Menu Pod.
__ a. Select the Operations link
__ b. Select the Activate task as shown below:

© Copyright IBM Corp. 2009, 2019 1-14

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
__ c. In the Install VIOS window, leave all the fields set at the default, except the following:
Activation Options = Activate (Normal)
Select VIOS Configuration = default_profile
Boot Mode = System Management Services
__ d. Click Finish
__ e. Close the Install VIOS window when it displays the message Power ON VIOS
successful.
__ 19. Switch to the VIOS console window (the HMC CLI SSH session)
__ 20. Interact with the SMS menus to set the boot device. The disks that contain operating
systems will be listed. Look for the disk that is listed as containing VIOS 2.2.6 or greater.
Choose this disk and boot the operating system. If there is more than one (most likely there
will be because of multi-pathing), pick the first one.
If you choose the wrong disk, the boot will fail. If this happens, shut down the partition from
the HMC, go back to SMS and try again.
Here are exact instructions to help with the SMS menu options:
__ a. You may need to type a 0 (zero) to choose your window as the console.
__ b. To boot from the correct disk, from the main SMS menu choose option 5: Select Boot
Options.
__ c. Choose 1: Select Install/Boot Device.
__ d. Choose 3: Hard Drive.
__ e. Choose 2: SAN.
__ f. There will be two Fibre adapter ports. Choose the first one by typing a 1 followed by the
ENTER key.
__ g. The firmware must scan the bus to build the list of assigned SAN LUNs that are
bootable. 
Choose option 2 to Scan for available boot devices. 
Then choose option 3 to initiate the scan (this will take a minute or two)

© Copyright IBM Corp. 2009, 2019 1-15

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
__ h. Choose the disk that has the Virtual I/O Server boot image. In the example below, you
can tell disk #1 has the VIOS operating system by the (VIOS 2.2.6.00) at the end of its
description. Note that for the other disks listed on your system there will be two or more
disks with AIX installed.
Version FW860.30 (SV860_103)
SMS (c) Copyright IBM Corp. 2000,2016 All rights reserved.
-------------------------------------------------------------------------------
Select Device
Device Current Device
Number Position Name
1. 1 42 GB FC Harddisk, part=2 (VIOS 2.2.6.00)
( loc=U78CB.001.WZS06PK-P1-C5-T1-W500507680c12332b-L0000000000000000 )
2. - 10 GB FC Harddisk, part=2 (AIX 7.2.0)
( loc=U78CB.001.WZS06PK-P1-C5-T1-W500507680c12332b-L0001000000000000 )
3. - 10 GB FC Harddisk, part=2 (AIX 7.2.0)
( loc=U78CB.001.WZS06PK-P1-C5-T1-W500507680c12332b-L0002000000000000 )
4. - 42 GB FC Harddisk, part=2 (VIOS 2.2.6.00)
( loc=U78CB.001.WZS06PK-P1-C5-T1-W500507680c12334e-L0000000000000000 )
5. - 10 GB FC Harddisk, part=2 (AIX 7.2.0)
( loc=U78CB.001.WZS06PK-P1-C5-T1-W500507680c12334e-L0001000000000000 )
6. - 10 GB FC Harddisk, part=2 (AIX 7.2.0)
( loc=U78CB.001.WZS06PK-P1-C5-T1-W500507680c12334e-L0002000000000000 )

-------------------------------------------------------------------------------
Navigation keys:
M = return to Main Menu N = Next page of list
ESC key = return to previous screen X = eXit System Management Services

-------------------------------------------------------------------------------
Type menu item number and press Enter or select Navigation key:1

__ i. Choose 1 and press enter

__ j. Choose 2: Normal Boot.
__ k. Choose 1: Yes. (You are sure you want to exit SMS.)
__ 21. The boot process might take 5-10 minutes. It’s a good time for a break :)
__ 22. When the Virtual I/O Server has finished booting, login using the padmin login. You will
NOT be prompted to change the password because this is not the first login attempt after
installation. Use the password provided by your instructor.
The licenses have already been accepted. If this were the first login after installation, you
would have to run the license -accept command before performing any other tasks on the
Virtual I/O Server command line.

© Copyright IBM Corp. 2009, 2019 1-16

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
__ 23. Execute the hostname command. Verify that the hostname of the VIOS is your assigned
VIOS (T#_VIOSN). If the hostname is not your assigned VIOS name, you chose the wrong
FC adapter. You must shutdown the VIOS, delete the incorrectly assigned FC adapter from
the VIOS, assign the correct FC adapter, and reactivate the VIOS. If you continue to have
problems, contact the instructor. Do not proceed if the hostname of your VIOS is not
your assigned VIOS name.
__ 24. Set the bootlist for the VIOS so that it can be rebooted if required without having to interact
with SMS again. Example command:
$ bootlist -mode normal hdisk0
__ 25. Set the time zone for the partition but leave the date and time as is currently configured. You
can use either the cfgassist utility or use the chdate command. The new time zone will
take effect after the next operating system boot. Use the time zone EST5EDT for the
Eastern time zone in the USA.
» Run cfgassist. Use the Set Date and TimeZone menu option. Answer yes to the question:
“Does this time zone go on DAYLIGHT SAVINGS TIME?” Set the time zone to EST5EDT. The
date and time should be fine as is. Do not change these to your own local date and time. Simply
press Enter to use the current configuration.
» If you wish to try the chdate command, first determine the current date and time with the date
command. Then when you enter the chdate command to change it, be sure to use the same
date and time. Enter the following command in the CLI where mmddHHMMYYYY is the actual
month, date, hour, minutes, and year and TZ is the actual time zone:
chdate mmddHHMMYYYY -timezone TZ
For example, this command changes the date to January 25th, 2008 at 2:58 in the
afternoon on United States Pacific time:
$ chdate 012514582008 -timezone PST8PDT
__ 26. Type the help command to see what CLI commands are available.
__ 27. Run the ioslevel command to display the version of the Virtual I/O Server software that is
installed.
__ 28. Type in oem_setup_env to get an AIX root shell. From here you can invoke AIX commands
and access SMIT if desired. We won’t need to do this in this lab exercise, so type exit to
get out of the oem_setup_env shell and go back to the Virtual I/O Server CLI.
__ 29. The TCP/IP networking for the VIOS partition should already be set. If you assigned a
different (wrong) Ethernet adapter to the VIOS, ent0 and ent1 devices will be in the Defined
state. The new Ethernet adapter that you assigned will be ent2 and in the Available state
(with no IP address). You must check to see if there is an ent2, and if so, remove the wrong
Ethernet adapter and add the correct one.

© Copyright IBM Corp. 2009, 2019 1-17

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
__ 30. Display the Ethernet adapters assigned to the VIOS. If you see more than one 2-Port
Ethernet adapter (that is, if the output shows ent2 and ent3 devices), shutdown the VIOS,
delete the incorrectly assigned Ethernet adapter from the VIOS, assign the correct Ethernet
adapter, and reactivate the VIOS. If you continue to have problems, contact the instructor.
$ lsdev -type adapter | grep ^ent
ent0 Available PCIe2 2-Port 10GbE Base-T Adapter
(e4148e1614109204)
ent1 Available PCIe2 2-Port 10GbE Base-T Adapter
(e4148e1614109204)
__ 31. Test network connectivity with a ping to your HMC IP address or hostname. If it is
successful, you have verified the minimum configuration of the Virtual I/O Server.
__ 32. Use the lstcpip command to check the network configuration. View the man page for more
information on options.
» You can use the lstcpip -stored command to view the configuration. You can also use the
following command flags to view different parts of the configuration (note that the nameserver
might not be configured in the lab):
lstcpip -hostname
lstcpip -interfaces
lstcpip -namesrv
lstcpip -num
lstcpip -num -routinfo
__ 33. Now you will set the recommended default device attributes using the rules command. The
following steps will show that changing the default device attributes to the recommended
values will change the reserve_policy attribute from single_path to no_reserve. You will
use hdisk5 as the resource to check and verify.
__ a. Invoke oem_setup_env to enter a root shell to check the ODM before setting the default
attributes.
__ b. Query the ODM to see the predefined device entry for hdisk5. The PdDvLn field the key.
# odmget -q "name = hdisk5" CuDv

CuDv:
name = "hdisk5"
status = 1
chgstatus = 2
ddins = "scsidisk"
location = "00-01-02"
parent = "fscsi1"
connwhere = "W_2"

© Copyright IBM Corp. 2009, 2019 1-18

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
PdDvLn = "disk/fcp/mpioosdisk"
__ c. Query the predefined ODM class see the default value for the reserve_policy attribute
# odmget -q "uniquetype = disk/fcp/mpioosdisk and attribute =
reserve_policy" PdAt

PdAt:
uniquetype = "disk/fcp/mpioosdisk"
attribute = "reserve_policy"
deflt = "single_path"
values = "no_reserve,single_path,PR_exclusive,PR_shared"
width = ""
type = "R"
generic = "DUC"
rep = "sl"
nls_index = 96
__ d. Notice that the default (deflt field) value is single_path. After running the rules
command, this will change to no_reserve.
__ e. Exit the root shell, returning to the padmin restricted shell.
__ f. Run the rules command to show that there is a difference between the recommended
setting and the system setting for reserve_policy for the example disk type.
# rules -o diff -s -d | grep mpioosdisk
devParam.disk.fcp.mpioosdisk:reserve_policy device=disk/fcp/mpioosdisk
single_path | no_reserve
__ g. Run the rules command again, to update the system settings to the recommended
device attribute settings.
$ rules -o deploy -d
A manual post-operation is required for the changes to take effect, please
reboot the system.
__ h. Run the rules command again to show that there are no differences between the
recommended settings and the system settings for any device attribute. Note that there
will be no output, because there are not differences.
$ rules -o diff -s -d
$
__ i. Return to the root shell as you did in the earlier steps and check the value for the
reserve_policy attribute. It should now show no_reserve as the default.
__ j. Exit the root shell.
__ k. Shutdown and reboot your VIOS to make the rules changes take effect.

© Copyright IBM Corp. 2009, 2019 1-19

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 1. Virtual I/O Server Configuration

EXempty
$ shutdown -restart -force
__ 34. That is the end of this lab exercise. Notify your instructor that you are finished.

END OF LAB (with hints)

© Copyright IBM Corp. 2009, 2019 1-20

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

Exercise 2. Dual VIOS Virtual SCSI

Device Configuration
(with hints)

Overview
This lab covers the configuration of virtual SCSI devices in a Dual VIOS configuration.

Objectives
At the end of the lab, you should be able to:
• Configure attributes for paths and disks on the VIOS and client to support proper MPIO failover
actions
• Configure a partition to use virtual SCSI disks hosted in a Dual Virtual I/O Server configuration,
using physical volumes as backing storage
▪ Use the HMC Enhanced+ GUI to configure one LPAR
▪ Use the VIOS CLI and the HMC CLI to configure another LPAR
• Use the VIOS CLI and HMC to view virtual SCSI configuration information
• Test setting different path priorities for a MPIO virtual SCSI disk
• (Optional) Test the client MPIO failover configuration

Introduction
In this exercise, students will configure two Virtual I/O Servers to host virtual SCSI devices for AIX
partitions. Students will properly configure the VIOS Fibre Channel adapters and disk access, then
configure storage resources and virtual target devices to provide clients with access to virtual SCSI
disks using physical volumes as backing storage. The backing storage devices (vSCSI disks) will
be used in LPAR1 and LPAR2 as rootvg disks as shown in the diagram below. Students will also
examine virtual adapter configurations from the HMC and the Virtual I/O Server. Students will verify
the configuration is correct by testing failover from the primary VIOS path to the secondary.

© Copyright IBM Corp. 2009, 2019 2-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

Requirements
• This workbook
• A workstation with a web browser connected to a network
• A POWER8 or later processor-based server with a dedicated HMC
• A system from which to start an SSH session to the HMC

© Copyright IBM Corp. 2009, 2019 2-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
Exercise Instructions with hints

Preface
• All exercises for this unit depend on the availability of specific equipment in your lab and at your
desk. You’ll need:
▪ A managed system configured with 3 assigned AIX partitions per student and two assigned
VIOS partitions per team. Two students will be on a team. One student on the team is
assigned one VIOS as primary, the other student on the team is assigned the other VIOS as
primary. You will use the AIX partitions assigned to you and the VIOS partitions assigned to
your team. AIX partitions will have names that start with S1, S2, S3, or S4 for student 1,
student 2, student 3 and student 4, respectively. VIOS partitions will have names that start
with T1 or T2 for team1 and team 2, respectively.
▪ A system from which to execute remote commands to the HMC (via SSH)
▪ All lab systems need to be accessible to each other on a network
▪ Two disks with AIX V7.2 pre-installed (all disks are attached to both Virtual I/O Server
partitions).
▪ The lab instructions refer to the LPAR names generically, without the S#_ or T#_
prefix. When you see LPAR1 or LPAR2, you should work with your assigned
S#_LPAR1 and S#_LPAR2. When you see VIOS1 and VIOS2, you should work with
your assigned T#_VIOS1 and T#_VIOS2
See the table below for specific assignments:

Table 2: Student VIOS and LPAR assignments

Assigned VIOS Second VIOS LPAR1 LPAR2 LPAR3
(VIOS1) (VIOS2)
Student 1 T1_VIOS1 T1_VIOS2 S1_LPAR1 S1_LPAR2 S1_LPAR3
Student 2 T1_VIOS2 T1_VIOS1 S2_LPAR1 S2_LPAR2 S2_LPAR3
Student 3 T2_VIOS1 T2_VIOS2 S3_LPAR1 S3_LPAR2 S3_LPAR3
Student 4 T2_VIOS2 T2_VIOS1 S4_LPAR1 S4_LPAR2 S4_LPAR3
• Hints are provided for lab exercises in case you need them and provide solutions for the lab
exercise steps. All hints are marked by a » sign.

Introduction
This exercise is presented in 7 parts:
• Part 1 - Configure storage access on VIOS1 and VIOS2
• Part 2 - Configure a physical volume as a virtual SCSI disk for LPAR1 to use as its boot device
using the HMC GUI
• Part 3 - Configure path and disk attributes on LPAR1
• Part 4 - Explore path priorities on LPAR1
• Part 5 - Configure a physical volume as a virtual SCSI disk for LPAR2 to use as its boot device
using the HMC CLI and the VIOS CLI
• Part 6 - Check virtual device configuration from HMC
• Part 7 - (Optional) Test the client MPIO configuration

© Copyright IBM Corp. 2009, 2019 2-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
Part 1: Configure storage access on VIOS1 and VIOS2
__ 1. The table below shows the hdisk assignments on your assigned VIOS partitions for your
AIX partitions. Find the row that matches your assigned student number, and make note of
the associated VIOS partitions and hdisk names. The hdisk name will be the same on both
of your VIOS partitions.

Table 3: AIX LPAR rootvg disk assignments per student

Assigned Second LPAR1 LPAR2 LPAR3
VIOS VIOS rootvg rootvg rootvg
Student 1 T1_VIOS1 T1_VIOS2 hdisk2 hdisk3 hdisk4
Student 2 T1_VIOS2 T1_VIOS1 hdisk5 hdisk6 hdisk7
Student 3 T2_VIOS1 T2_VIOS2 hdisk2 hdisk3 hdisk4
Student 4 T2_VIOS2 T2_VIOS1 hdisk5 hdisk6 hdisk7
__ 2. IMPORTANT: This lab depends on the other student in your team having completed
Exercise 1. That is, your second VIOS as outlined in the table above must be Running and
fully booted. Check that your second VIOS is Running in the HMC. If it is not active, contact
the instructor to check on the status. You can not proceed in this lab without both of the
VIOS partitions on your team in the Running state.
__ 3. Log into your assigned VIOS and your second VIOS as padmin using the credentials
supplied by your instructor. 
If you have trouble logging into your second VIOS, contact the instructor.
__ 4. Determine if the PVIDs for all of your LPARs SAN disks in both VIOS partitions match. If so,
then the same disk is available in both VIOS partitions. This is key for a client MPIO
configuration.
Run lspv in both VIOS partitions for all three of your AIX partition hdisks. 

Here is example partial output showing that hdisk2 on VIOS1 is the same disk as hdisk2
on VIOS2:
On VIOS1:
$ lspv | grep hdisk2
hdisk2 00f9e0966d37f9d8
On VIOS2:
$ lspv | grep hdisk2
hdisk2 00f9e0966d37f9d8
If the PVID value shows “none”, then run the following command, replacing hdiskX with the
hdisk name:
$ chdev -dev hdiskX -attr pv=yes
hdiskX changed
In example above, the PVIDs match. If they don’t, you’ll need to find the disk that does
match on both VIOS partitions. Look for any additional SAN disks that might match. In a
typical training lab environment, they will match. Contact your instructor if you can’t find
disks that match.

© Copyright IBM Corp. 2009, 2019 2-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 5. View the paths to all three disks on both VIOS partitions. You can use the lspath command
to do this. Notice that the path configuration on VIOS1 is similar to the configuration on
VIOS2. There are four paths through FC port fscsi0. Both VIOS partitions have
implemented MPIO access to the SAN device.
$ lspath -dev hdisk2
status name parent connection

Enabled hdisk2 fscsi0 500507680c52332b,1000000000000

Enabled hdisk2 fscsi0 500507680c52334e,1000000000000
Enabled hdisk2 fscsi0 500507680c32332b,1000000000000
Enabled hdisk2 fscsi0 500507680c32334e,1000000000000
__ 6. Check the reserve_policy for all three disks on both VIOS partitions.
$ lsdev -dev hdiskX -attr reserve_policy
value

single_path
In this case, the reserve_policy is single_path. Your disks might be set to no_reserve.
__ 7. If the reserve_policy attribute for any of your disks on either VIOS partition is
single_path, change it to no_reserve.
This command should have the following output:
$ chdev -dev hdiskX -attr reserve_policy=no_reserve
hdiskX changed
__ 8. Check the algorithm attribute for all three disks on both VIOS partitions. If it is not set to
round_robin already, set it to round_robin. This enables load balancing over the two
paths to the LUN from both VIOS partitions.
Example command and output showing that the attribute is not set to round_robin.
$ lsdev -dev hdiskX -attr algorithm
value

fail_over
In the above example, it is currently set to fail_over. Use chdev to change it to
round_robin with the following command.
$ chdev -dev hdiskX -attr algorithm=round_robin
Example output:
hdiskX changed
__ 9. Recall the rule for the hcheck_interval attribute. It should not be set lower than the
read/write timeout (rw_timeout) value on the disk. Check the rw_timeout value on all three
disks on both VIOS partitions.

© Copyright IBM Corp. 2009, 2019 2-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
Example command and output that shows a 30 second read/write timeout value:
$ lsdev -dev hdiskX -attr rw_timeout
value

30
__ 10. The hcheck_interval value is likely already set to the recommended value on all three
disks on both VIOS partitions. If it is not, then change it.
Example command and output that shows a 60 second interval:
$ chdev -dev hdiskX -attr hcheck_interval=60
__ 11. The hcheck_mode attribute value for all three disks on both VIOS partitions is likely already
set to nonactive. If it is not, then change it.
Example command and output:
$ chdev -dev hdiskX -attr hcheck_mode=nonactive
__ 12. On your assigned VIOS partition, check the physical path attributes for both paths to the
three disks. Are the attributes configured with the recommended values?
List the path attributes for both paths. Example command and output:
$ lsdev -dev fscsi0 -attr
attribute value description user_settable

attach switch How this adapter is CONNECTED False

dyntrk no Dynamic Tracking of FC Devices True+
fc_err_recov delayed_fail FC Fabric Event Error RECOVERY Policy True+
scsi_id 0x10f00 Adapter SCSI ID False
sw_fc_class 3 FC Class for Fabric True

The recommended values are that dyntrk should be set to yes and fc_err_recov should
be set to fast_fail. You should find that your system has these set to no and
delayed_fail as shown in the example above.
__ 13. If any of the device attributes need to be changed to the recommended values on your
assigned VIOS, change them now. If you need to change the attributes, this time, you will
make the change using the concurrent update flag (-U) so as to avoid the need to reboot the
VIOS. To use the concurrent update flag, you must be in the root shell.
Change the device attributes. Example commands and their outputs:
$ oem_setup_env
# chdev –l fscsi0 –a dyntrk=yes -a fc_err_recov=fast_fail –U
fscsi0 changed
# exit
$

© Copyright IBM Corp. 2009, 2019 2-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
Part 2: Configure a virtual SCSI disk for LPAR1

Note

Now that your Virtual I/O Server is installed and minimally configured, and the disks have been
prepared for Dual VIOS access, you can configure virtual devices.
In the last exercise, when you created the VIOS partition, there was no place to create virtual SCSI
server adapters for connection to client LPARs. In this part of this exercise you will see that when
using the Enhanced+ GUI, the creation of the virtual SCSI adapters and the definition of the virtual
target devices in the VIOS is performed without any need to login to the VIOS.
You will configure a virtual SCSI disk to your VIOS using the Enhanced+ GUI that uses a disk
(physical volume) as backing storage.

__ 1. Open an SSH session to your assigned VIOS partition and login as padmin.
__ 2. Recall from the previous part of the exercise the hdisks that you are to use for your AIX
LPARs. Make note of the rootvg disk to use for your LPAR1 partition. 

Assigned Second LPAR1 LPAR2 LPAR3
VIOS VIOS rootvg rootvg rootvg
Student 1 T1_VIOS1 T1_VIOS2 hdisk2 hdisk3 hdisk4
Student 2 T1_VIOS2 T1_VIOS1 hdisk5 hdisk6 hdisk7
Student 3 T2_VIOS1 T2_VIOS2 hdisk2 hdisk3 hdisk4
Student 4 T2_VIOS2 T2_VIOS1 hdisk5 hdisk6 hdisk7
You will use the LPAR2 disk in part 2 of this exercise as the rootvg for your LPAR2 partition. 
You will use the LPAR3 in the next exercise as the rootvg for your LPAR3 partition.
__ 3. List out all the virtual devices in the available state. Note that there are no virtual SCSI
server adapters. They will be created as part of the assignment of a virtual SCSI disk to
LPAR1. Your output might be different. If it includes virtual SCSI adapters, other the other
student has created a virtual device to their LPAR1. Make note of any virtual SCSI adapter
names (vhost0, vhost1, etc) that are present before you assign a virtual device to LPAR1.
Here is an example command and its output where no virtual devices have been assigned:
$ lsdev -virtual -state available
name status description
pkcs11 Available PKCS#11 Device
vbsd0 Available Virtual Block Storage Device (VBSD)
vsa0 Available LPAR Virtual Serial Adapter
__ 4. Start a web browser and connect to your assigned HMC in Enhanced+ mode. Logon with
your assigned HMC userid and password. See Exercise 1, Step 1 for details on login
procedures.
__ 5. Navigate to the Partitions table view for your assigned system by clicking on the name of
your assigned system in the All Systems view. See Exercise 1, Steps 2 and 3 for details on
accessing the Partitions table view for your assigned system.

© Copyright IBM Corp. 2009, 2019 2-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 6. In the Filter field of the Partitions table view for your assigned system, type your student
prefix to display only your assigned LPARs. You will type S1, S2, S3, or S4 depending on
your assignment. The example below shows the S2 student filtering on their LPARs:

__ 7. You should see three of your assigned AIX LPARs. Your AIX LPARs (S#_LPAR1,
S#_LPAR2, and S#_LPAR3) should show a status of Not Activated.
__ 8. Select your LPAR1 partition by clicking on the name. This will display the General
Properties of the partition. The example shows student #2 selecting LPAR1. Be sure you
are choosing your assigned LPAR1.

© Copyright IBM Corp. 2009, 2019 2-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 9. The Enhanced+ GUI tasks depend on the synchronization of the operations that are
performed into the LPARs profile. When the Save configuration changes to profile option
is Enabled, the operations will be synchronized. Enable that feature as follows:
__ a. Click the Advanced button in the upper right corner of the General Properties screen

__ b. Scroll to the bottom of the General Properties screen (in the Advanced Settings
section), and change the Save configuration changes to profile: option to Enabled (it
should be Disabled)

__ c. Click the Save button to complete the enablement.

__ 10. In the Menu Pod for your LPAR1 partition, select Virtual Storage.

© Copyright IBM Corp. 2009, 2019 2-9

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

__ 11. In the Virtual Storage display, under the Virtual SCSI tab, notice that there are No items to
display in all of the tables. This indicates that there are no virtual SCSI devices assigned to
this partition. 
To add a virtual SCSI disk, select the Add Virtual SCSI Storage button.

© Copyright IBM Corp. 2009, 2019 2-10

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

__ a. Ensure that the Physical Volume radio button is selected (it is the default).
__ b. In the User Defined Name field, type S#_LPAR1-rootvg, where S# is replaced with
your student number.
__ c. Click the Show assigned physical volumes option. The list should now contain Team
#1’s hdisk2 and hdisk3 disks and Team #2’s hdisk5 and hdisk6 disks. They are
preloaded with an AIX operating system image, so they will not be in the list of available
disks.

© Copyright IBM Corp. 2009, 2019 2-11

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ d. Important: Select the right disk in the next step (and do not do anything in the VIOS to
overwrite the volume group that is on hdisk2, hdisk3, hdisk5, or hdisk6). Be sure to
choose your assigned LPAR hdisk that is associated with your assigned and second
VIOS partitions, not another student’s. 

Once again, here are the hdisk assignments by student and by VIOS:

Assigned Second LPAR1 LPAR2 LPAR3

VIOS VIOS rootvg rootvg rootvg
Student 1 T1_VIOS1 T1_VIOS2 hdisk2 hdisk3 hdisk4
Student 2 T1_VIOS2 T1_VIOS1 hdisk5 hdisk6 hdisk7
Student 3 T2_VIOS1 T2_VIOS2 hdisk2 hdisk3 hdisk4
Student 4 T2_VIOS2 T2_VIOS1 hdisk5 hdisk6 hdisk7
__ e. Select the hdiskX disk (hdisk2 if your student number is odd, or hdisk5 if your student
number is even) that is on your assigned VIOS partitions from the list of available disks.
The other students’ disks might be in the list too so you might have to scroll the list of
disks to find your assigned LPAR1 disk on your assigned VIOS partitions. 

Notice in the example that hdisk5 on T1_VIOS1 and T1_VIOS2 is selected because this
example is for student #2 using the Team #1 VIOS partitions.

__ f. Click OK to execute the task. This will take some time because many subtasks are
being performed, from DLPAR operations on the VIOS partitions, to creation of the
adapters on the LPAR1 partition (and applying the change to the current configuration),
to creating the virtual target device on the VIOS partitions.
__ 12. When the task completes, verify that the Virtual Storage view has been updated to show a
new entry in the Physical Volume table. Observe that the Connections column shows both
VIO Servers.

© Copyright IBM Corp. 2009, 2019 2-12

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

__ 13. In later lab steps, you will need to know which virtual SCSI adapters in LPAR1 map to
VIOS1 and VIOS2. You can see that in the HMC GUI. Choose the newly added disk by
clicking the radio button next to it.
__ 14. Click the Action menu, and choose the View Device Mappings option.

__ 15. In the display, make note of the client adapter ID that maps to each VIOS. In the example
below, client adapter ID 2 maps to VIOS1, and client adapter ID 3 maps to VIOS2 and the
server adapter name in both VIOS partitions is vhost0.

© Copyright IBM Corp. 2009, 2019 2-13

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 16. Close the View Device Mapping dialog by clicking Close.
__ 17. Return to your SSH session on your assigned VIOS. Check for virtual SCSI resources
again, using the command that you ran earlier. Notice the two new resources; the virtual
SCSI server adapter (vhost0 or vhost1) and the virtual target device (S#_LPAR1-rootvg):
$ lsdev -virtual -state available
name status description
vasi0 Available Virtual Asynchronous Services Interface (VASI)
vbsd0 Available Virtual Block Storage Device (VBSD)
vhost0 Available Virtual SCSI Server Adapter
vsa0 Available LPAR Virtual Serial Adapter
S2_LPAR1-rootvg Available Virtual Target Device - Disk
IMPORTANT: Your output might show another virtual SCSI server adapter and virtual target
device. The new virtual SCSI server adapter in this output (compared to the output from
Step 3 on page 2-7) is the one you just created. Make note of the name of the new virtual
SCSI server adapter. If there is more than one new virtual SCSI server adapter, make note
of both and run the next step using both of the vhost devices names, looking for the hdisk
that you assigned to your LPAR1.
__ 18. Use lsmap -vadapter vhostX to verify the mapping of the virtual target, where X is
replaced by the number of the new virtual SCSI server adapter discovered in the previous
step.
The output of lsmap -vadapter vhostX should be similar to the following. This example
shows hdisk5 as the backing device with a virtual target device (VTD) name of
S2_LPAR1-rootvg.
$ lsmap -vadapter vhostX
SVSA Physloc Client Partition ID
--------------- -------------------------------------------- ------------------
vhostX U8286.41A.2182AFV-V1-C2 0x00000006

VTD S2_LPAR1-rootvg
Status Available
LUN 0x8100000000000000
Backing device hdisk5
Physloc U78C9.001.WZS03PT-P1-C12-T2-W500507680235E1AB-L1000000000000
Mirrored false
__ 19. Back in the HMC Enhanced+ GUI, view the LPAR1 partition profile to see the client virtual
SCSI adapter that was created and to which virtual SCSI server adapter in each VIOS the
client adapter points. Make note that the mapping is to both VIOS partitions.
Open the Partition Actions section of the Menu Pod of your LPAR1, select Profiles, then
Manage Profiles from the expanded section. The example shows student #2’s selection:

© Copyright IBM Corp. 2009, 2019 2-14

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

__ 20. This opens the list of profiles for this partition. Click on the name of the only profile in the
table.
__ a. This opens the Logical Partition Profile Properties window (familiar to anyone who has
used the Classic GUI.)
__ b. Click on the Virtual Adapters tab. Notice the following:
- Client SCSI adapter entry, and the Adapter ID column for that adapter,
- The Server/Client Partition column which points to both VIOS partition names,
- The Partner Adapter column which displays the virtual SCSI server adapter ID on both
VIOS partitions.
- Cancel the Properties window.
- Close the Managed Profiles window.

© Copyright IBM Corp. 2009, 2019 2-15

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

__ 21. Select the breadcrumb that returns you to your assigned managed system view.

__ 22. From the Menu Pod of your assigned managed system, select the Virtual I/O Servers link
from the PowerVM section.
__ 23. The display will change to either a table or gallery list of all configured VIOS partitions. Open
your assigned VIOS partition by clicking on the name.
__ 24. Open the VIOS Actions section in the Menu Pod, then open the Profiles sub-section, and
then select Manage Profiles.
__ 25. This opens the list of profiles for this partition. Click on the name of the only profile in the
table.
__ a. This opens the Logical Partition Profile Properties window (familiar to anyone who has
used the Classic GUI.)
__ b. Click on the Virtual Adapters tab. Notice the following:
- Server SCSI adapter entry, and the Adapter ID column for that adapter matches the
one to which the client SCSI adapter points,

© Copyright IBM Corp. 2009, 2019 2-16

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
- The Server/Client Partition column which points to your LPAR1 partition name,
- The Partner Adapter column which matches the virtual SCSI client adapter ID on your
LPAR1.
- Cancel the Properties window.
- Close the Managed Profiles window.

__ 26. Repeat Step 23 on page 2-16 through Step 25 on page 2-16 for your second VIOS partition.
You will see that there is a similar mapping between your LPAR1 and your second VIOS
partition.
The next step is to boot your LPAR1 partition from the newly assigned virtual SCSI disk. As was
mentioned earlier, the AIX operating system was already installed in the SAN LUN that is assigned
as the virtual SCSI disk.
__ 27. You will need a console when activating the LPAR1 partition. The console can be started
from the GUI, but this can be unreliable because of various Java-related complications.
Instead, you will open the console using the HMC CLI and the vtmenu command.
__ a. Open an SSH session to your assigned HMC using Putty or a terminal session on your
workstation. If you are having problems connecting to the HMC using SSH, contact the
instructor.
__ b. At the login prompt, login with your assigned HMC userid and password.
__ c. Enter the command, vtmenu
__ d. From the list of partitions, choose the number associated with your LPAR1. The console
is open and waiting for input.

© Copyright IBM Corp. 2009, 2019 2-17

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 28. Switch back the HMC GUI window. Return to your assigned managed system view. This will
display the list of partitions on your assigned managed system. Enter your student prefix
(S1, S2, S3, or S4) in the Filter field.
__ 29. Select your LPAR1 partition by clicking on the name. This will display the General
Properties of the partition.
__ 30. To activate the LPAR1 logical partition, select the Partition Actions link in the Menu Pod.
__ a. Select the Operations link
__ b. Select the Activate task
__ c. In the Activate window, leave all the fields set at the default
__ d. Click Finish
__ e. Close the Activate window when it displays the message Partition activated.
__ 31. Switch to the LPAR1 console window (the HMC CLI SSH session)
__ 32. Did your LPAR1 boot to the login prompt? 
If it booted to the login prompt, skip to Step 36 on page 2-19.
If the boot failed, proceed with the next step.
__ 33. Did your LPAR1 boot to the SMS menu? 
If it booted to the SMS menu you need to set the boot list using the SMS menu, skip to
Step 35 on page 2-18
__ 34. Did your LPAR1 fail to boot to the SMS menu? 
If the boot failed to reach the AIX prompt and the SMS menu, you need to boot to SMS and
set the boot list using the SMS menu. Shutdown and reactivate your LPAR1 to the SMS
menu as follows:
__ a. Switch back to the HMC GUI and perform a shutdown immediate on the LPAR1 partition
from the Partition Actions -> Operations -> Shutdown task in the Menu Pod.
__ b. When the LPAR1 partition changes to the Not Activated state, activate it again following
the same steps as in Step 30 on page 2-18 above, but, before you click Finish at Step d,
be prepared to switch back to the console session window as soon as you click Finish.
__ c. Hit the number 1 key as soon as you see the word Keyboard in the console. This will
bring up the SMS menu.
__ d. Proceed with the next step.
__ 35. Interact with the SMS menus to set the boot device. There should be only one disk available
and will be listed as a SCSI disk. Choose this disk and boot the operating system. 
Here are exact instructions to help with the SMS menu options:
__ a. You may need to type a 0 (zero) to choose your window as the console.
__ b. To boot from the correct disk, from the main SMS menu choose option 5: Select Boot
Options.
__ c. Choose 1: Select Install/Boot Device.
__ d. Choose 3: Hard Drive.
__ e. Choose 1: SCSI.

© Copyright IBM Corp. 2009, 2019 2-18

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ f. Choose 1: To select the first SCSI adapter (either one will work because this is a Dual
VIOS configuration).
__ g. Choose the only disk displayed. It should indicate that it contains an AIX 7.2 image. If
not, try to figure out the problem. If you need help, contact the instructor.
Version FW860.10 (SV860_056)
SMS (c) Copyright IBM Corp. 2000,2016 All rights reserved.
-------------------------------------------------------------------------------
Select Device
Device Current Device
Number Position Name
1. 1 SCSI 9 GB Harddisk, part=2 (AIX 7.2.0)
( loc=U8286.41A.2182AFV-V6-C2-T1-L8100000000000000 )

-------------------------------------------------------------------------------
Navigation keys:
M = return to Main Menu
ESC key = return to previous screen X = eXit System Management Services

-------------------------------------------------------------------------------
Type menu item number and press Enter or select Navigation key:1

__ h. Choose 1 and press enter

__ i. Choose 2: Normal Boot.
__ j. Choose 1: Yes. (You are sure you want to exit SMS.)
__ 36. When the operating system in the LPAR1 partition has finished booting, login as root with
the password provided by your instructor.
__ 37. Execute the hostname command. Verify that the hostname matches your assigned LPAR1
(S#_LPAR1). If the hostname is not your assigned LPAR1 name, you chose the wrong disk.
You must shutdown LPAR1, delete the incorrectly assigned disk from LPAR1, assign the
correct disk to LPAR1, and reactivate LPAR1. If you continue to have problems, contact the
instructor. Do not proceed if the hostname of the LPAR is not your assigned LPAR1
name.
__ 38. List the disk devices. Determine the name of the new virtual SCSI disk drive. Note that there
may be additional hdisks listed in the Defined state on your lab system. Use the command
lsdev -c disk -S a to view just the available disks.

© Copyright IBM Corp. 2009, 2019 2-19

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
New virtual SCSI disk drive name: __________________
Here is an example output. In this example output, hdisk0 is the virtual disk you just
configured.
# lsdev -c disk -S a
hdisk0 Available Virtual SCSI Disk Drive
__ 39. Set the normal mode bootlist for the LPAR1 partition to the new virtual disk, if not set
already.
Here is an example command to set the bootlist to hdisk0.
# bootlist -m normal hdisk0
__ 40. Find the active Ethernet interface. You will configure the IP address for the LPAR on that
Ethernet interface.
» Here is an example command to find the active Ethernet interface. In this example, the en1
interface is active and will be used in the next step.
# lsdev -Cc adapter -S a | grep ^en
ent1 Available 01-00 PCIe2 100/1000 Base-TX 4-port Converged Network Adapter
VF (df1028e214103c04)
__ 41. Configure the interface found in the previous step with the IP address information provided
by the instructor for LPAR1 using smitty chinet
Change / Show a Standard Ethernet Interface

Type or select values in entry fields.

Press Enter AFTER making all desired changes.

[Entry Fields]
Network Interface Name en1
INTERNET ADDRESS (dotted decimal) [10.8.xxx.yyy]
Network MASK (hexadecimal or dotted decimal) [255.255.255.0]
Current STATE up +
Use Address Resolution Protocol (ARP)? yes +
BROADCAST ADDRESS (dotted decimal) []
Interface Specific Network Options
('NULL' will unset the option)
rfc1323 []
tcp_mssdflt []
tcp_nodelay []
tcp_recvspace []
tcp_sendspace []
Apply change to DATABASE only no +
__ 42. To ensure that the default gateway is associated with the correct interface, reconfigure the
inet0 device.
# mkdev -l inet0
__ 43. Verify that the address is configured correctly by pinging your HMC address.

© Copyright IBM Corp. 2009, 2019 2-20

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 44. At this point you can log out of the LPAR1 console session and close it in the HMC. Then
open an SSH session to LPAR1 to perform the rest of the steps in this part of the exercise.
__ 45. List the paths to the rootvg disk. There are two paths for hdisk0, each represented by a new
vscsi# device.
Example command and output:
# lspath -l hdisk0
Enabled hdisk0 vscsi1
Enabled hdisk0 vscsi2
You have successfully configured VIOS1 and VIOS2 to provide access to a SAN LUN for
LPAR1. Next you will configure the paths on LPAR1 to enable proper failover.
__ 46. List the slot numbers (adapter ID numbers) of the adapters. You can use this information to
determine which vscsi# maps to which VIOS.
# lsslot -c slot
# Slot Description Device(s)
U8284.22A.21C193V-V7-C0 Virtual I/O Slot vsa0
U8284.22A.21C193V-V7-C2 Virtual I/O Slot vscsi0
U8284.22A.21C193V-V7-C3 Virtual I/O Slot vscsi1
__ 47. Recall that at Step 15 on page 2-13 you determined which client adapter ID mapped each of
the VIO Servers. From the output in the previous step, make note of the vscsi adapter
device that maps to each of the VIO Servers. 
Given the examples in this guide, vscsi0 (ID 2) maps to VIOS1 and vscsi1 (ID 3) maps to
VIOS2.
__ 48. Another way to find the client-server virtual SCSI relationship on an AIX client LPAR is to
run the following command. Verify that this information matches the information you
discovered in the previous two steps:
# echo 'cvai' | kdb | grep vscsi
vscsi0 0x000007 0x0000000000 0x0 T1_VIOS1->vhost0
vscsi1 0x000007 0x0000000000 0x0 T1_VIOS2->vhost0

Part 3: Configure path and disk attributes on LPAR1

__ 49. Check the attributes of both paths to the new VSCSI disk in LPAR1. Are all the attributes set
correctly?
Example commands and outputs which show the attributes and their values for the paths to
the disk:
# lsattr -El vscsi0
rw_timeout 0 Virtual SCSI Read/Write Command Timeout True
vscsi_err_recov delayed_fail N/A True
vscsi_path_to 0 Virtual SCSI Path Timeout True

# lsattr -El vscsi1

rw_timeout 0 Virtual SCSI Read/Write Command Timeout True
vscsi_err_recov delayed_fail N/A True
vscsi_path_to 0 Virtual SCSI Path Timeout True

© Copyright IBM Corp. 2009, 2019 2-21

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
In the example command output above, the vscsi_err_recov needs to be changed to
fast_fail and the vscsi_path_to needs to be enabled by setting some number of
seconds, 30 or more, for the time out value
__ 50. In LPAR1, configure the path attributes to the recommended values. Use the -P option to
change the ODM. You’ll have to reboot in a later step for the changes to take affect.
Example commands and their outputs:
# chdev -l vscsi0 -a vscsi_err_recov=fast_fail -a vscsi_path_to=30 -P
vscsi0 changed
# chdev -l vscsi1 -a vscsi_err_recov=fast_fail -a vscsi_path_to=30 -P
vscsi1 changed
__ 51. Check the range of options for the algorithm attribute for the client’s disk.
In LPAR1, list the range of algorithms for the SAN disk. What are the options?
Example command to list the algorithm range for hdisk0 which shows only the fail_over
option:
# lsattr -Rl hdisk0 -a algorithm
fail_over
Recall that for virtual SCSI disks, the only algorithm option is fail_over.
__ 52. Check the rest of the disk attributes for the new disk on LPAR1. Are they set to the
recommended values?
Example command and its output to view the current settings on the disk:
# lsattr -El hdisk0
PCM PCM/friend/vscsi Path Control Module False
algorithm fail_over Algorithm True
hcheck_cmd test_unit_rdy Health Check Command True+
hcheck_interval 0 Health Check Interval True+
hcheck_mode nonactive Health Check Mode True+
max_transfer 0x40000 Maximum TRANSFER Size True
pvid 00f9e0966d37f9d80000000000000000 Physical volume identifier
False
queue_depth 3 Queue DEPTH True+
reserve_policy no_reserve Reserve Policy True+

Given the above output, only the hcheck_interval must be changed. It should be set to 60
seconds.
__ 53. In LPAR1, set the hcheck_interval to 60 on the disk. Use the -P flag because the hdisk is
in use.
Example command and its output:
# chdev -l hdisk0 -a hcheck_interval=60 -P
hdisk0 changed
__ 54. Reboot LPAR1.

© Copyright IBM Corp. 2009, 2019 2-22

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
Part 4: Explore path priorities
__ 55. Once LPAR1 has rebooted, restart the SSH session, and check the priority of the paths to
the rootvg virtual disk. You should see that both paths have the same priority.
Use the lspath command. For example:
# lspath -El hdisk0 -p vscsi0
# lspath -El hdisk0 -p vscsi1
These commands should each have this output showing that they have equivalent priorities:
priority 1 Priority True
__ 56. Check the path IDs associated with each path and virtual SCSI adapter.
Use the lsmpio command. For example:
# lsmpio -l hdisk0
name path_id status path_status parent connection

hdisk0 0 Enabled Sel vscsi0 810000000000

hdisk0 1 Enabled vscsi1 810000000000
The example output shows that vscsi0 is path ID 0.
__ 57. In Step 47 on page 2-21 you determined which VIOS was associated with which vscsi
adapter. Based on that determination, which VIOS is associated with path ID 0 on your
LPAR1? Using the examples presented so far in this guide, vscsi0 is associated with
VIOS1, therefore, the path to VIOS1 is the active path being used by the PCM. Determine
which VIOS is the active path on your LPAR1. In the steps below you will change the
primary path to the path that is not primary at this point. If your LPAR1 primary path is not
vscsi0 to VIOS1, substitute the appropriate values for the vscsi devices in the steps below
to switch the primary path on your LPAR1 to the other VIOS.
__ 58. Explicitly set the priority for the vscsi0 path using to the priority of 2. This will ensure that the
path through vscsi1 (and the associated VIOS) is used as the primary path. This can also
allow load balancing in situations when a VIOS pair provide storage access to multiple
clients (some clients can specify VIOS1 as the primary path and others can specify VIOS2
by changing the priorities on each client LPAR appropriately).
On LPAR1, use the chpath command to change the priority of the vscsi0 virtual path to 2.
For example:
# chpath -l hdisk0 -p vscsi0 -a priority=2
The command output should be:
path Changed
__ 59. Open an SSH session (or find an open SSH session) to both VIOS partitions.
__ 60. Position the two SSH session windows so you can see both windows. Using the examples
in the guide, the client vscsi0 goes through VIOS1 and the client vscsi1 goes through
VIOS2.

© Copyright IBM Corp. 2009, 2019 2-23

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 61. In both your VIOS sessions, start the viostat -adapter 2 command. Watch for a few
seconds to see if any data is flowing through vhost0 or vhost1 and hdisk2 or hdisk5. You
are going to write some data to a filesystem in LPAR1’s rootvg in the next step. You will
observe that the virtual SCSI server adapter that maps to the vscsi1 VIOS is used.
__ 62. Make sure that the VIOS session where you expect to see the traffic is visible. In the
examples in the guide, that would be VIOS2.
__ 63. In LPAR1, run cp /unix /tmp command. Immediately watch the viostat output in the
VIOS sessions. You should see one report with data for vhost0 or vhost1, depending on
the VIOS. If you miss it, try again.
__ 64. In which VIOS did you see activity in the viostat output? You will see activity in the VIOS
for which you did not set the path priority to 2 in LPAR1. Again, following the example in this
guide, you will see activity in the VIOS2 session.
Recall that the algorithm for the disk is set to fail_over so the I/O traffic will use one path
or the other. By using path priorities you configured the path to the other VIOS to be the
primary. You should see no data being written to the vhost adapter in non-primary VIOS
session.
__ 65. Stop both viostat commands with CTRL+C.
__ 66. Back in LPAR1, check the selected path ID now that traffic has been generated. Notice that
the selected path is now path ID 1, vscsi1. That is VIOS2 per the examples in this guide.
Use the lsmpio command. For example:
# lsmpio -l hdisk0
name path_id status path_status parent connection

hdisk0 0 Enabled vscsi0 810000000000

hdisk0 1 Enabled Sel vscsi1 810000000000

Part 5: Configure a virtual SCSI disk for LPAR2

You will configure a virtual SCSI disk for LPAR2, but this time instead of using only the HMC GUI,
you will use the HMC CLI and the VIOS CLI to map the device.
__ 67. If you do not have an HMC CLI session open, open an SSH session to your assigned HMC
and login using the userid and password supplied by the instructor.
__ 68. You should have VIOS CLI sessions open where you had verified the mappings for LPAR1’s
virtual SCSI disk. If so, proceed to the next step. If not, open an SSH session to both your
assigned VIOS and your second VIOS. Login using padmin and the password supplied by
your instructor.
__ 69. In the steps that follow, instead of referring to your assigned VIOS, the refers to your VIOS1
and VIOS2 partitions. Regardless of your assigned VIOS (VIOS1 or VIOS2) you will be
working with both VIOS partitions in your team. Pay close attention to the tables below that
contain your assigned slot numbers. These values are assigned to avoid conflicts.

© Copyright IBM Corp. 2009, 2019 2-24

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

Note

In the steps that follow, the commands are provided after the description of the task to be performed
in the step. Replace the string S# with S1, S2, S3, or S4 depending on your student number, and
replace T# with T1 or T2 depending on your team number. Replace the string m-sys-name with the
name of your assigned managed system.
The step will indicate whether to run the command in the HMC CLI session, or in the VIOS CLI
session. The command example shows the prompt that is appropriate for the session, but the
prompt string is not to be typed as part of the command.
student#@hmc:~> : is the HMC CLI
$ : is the VIOS CLI
# : is the LPAR2 CLI
The command to be typed is in bold type. The command should be typed as a continuous string.
No continuation characters are used in the example text to simplify the examples.

__ 70. In the HMC CLI session, verify that the DLPAR commands will be saved to the last activated
profile. Run the command command below, specifying your VIOS1, VIOS2 and your LPAR2
using the compound filter.
student#@hmc:~> lssyscfg -r lpar -m m-sys-name -F sync_curr_profile,name
--filter 'lpar_names="T#_VIOS1,T#_VIOS2,S#_LPAR2"'
0,S#_LPAR2
1,T#_VIOS1
1,T#_VIOS2
If the sync_curr_profile value for both VIOS partitions is not set to 1, contact the
instructor.
The sync_curr_profile must be set to 1 for your LPAR2. You must change that in the next
step.
__ 71. In the HMC CLI session, enable the sync_curr_profile option on LPAR2 so that HMC
Enhanced+ GUI commands will successfully update the profile when a DLPAR operation is
performed.
student#@hmc:~> chsyscfg -r lpar -m m-sys-name -p S#_LPAR2 -i
"sync_curr_profile=1"
__ 72. In the HMC CLI session, verify that the sync_curr_profile option on LPAR2 is enabled.
student#@hmc:~> lssyscfg -r lpar -m m-sys-name -F sync_curr_profile,name
--filter "lpar_names=S#_LPAR2"
1,S#_LPAR2
__ 73. In the HMC CLI, check for vSCSI adapters on your LPAR2 (should see no adapters).
student#@hmc:~> lshwres -r virtualio --rsubtype scsi -m m-sys-name --level
lpar --filter lpar_names=S#_LPAR2
No results were found.

© Copyright IBM Corp. 2009, 2019 2-25

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 74. In the HMC CLI, check for vSCSI adapters on your VIOS partitions (might see adapters for
both of your team’s LPAR1 partitions).
student#@hmc:~> lshwres -r virtualio --rsubtype scsi -m m-sys-name --level
lpar -F lpar_name,slot_num,remote_lpar_name,remote_slot_num --filter
'lpar_names="T#_VIOS1,T#_VIOS2"'
T#_VIOS1,2,S#_lpar1,2
T#_VIOS1,3,S#_lpar1,2
T#_VIOS2,2,S#_lpar1,3
T#_VIOS2,3,S#_lpar1,3
__ 75. To avoid adapter definition conflicts in the shared VIOS partitions, use the following slot
numbers when defining adapters. Find the row that indicates your student #. In that row you
see two adapters slot numbers for your LPAR2, and their corresponding adapter slot
numbers in the two VIOS partitions. For example, student 3 will define LPAR2 adapter slot 2
mapped to VIOS1 adapter slot 10, and adapter slot 3 mapped to VIOS2 adapter slot 11.

Table 4: LPAR2 virtual SCSI adapter slot number assignments

Student # LPAR2 slot numbers (L1# / L2#) VIOS slot numbers (V#)
2 T1_VIOS1: 10
Student 1
3 T1_VIOS2: 10
2 T1_VIOS1: 11
Student 2
3 T1_VIOS2: 11
2 T2_VIOS1: 10
Student 3
3 T2_VIOS2: 10
2 T2_VIOS1: 11
Student 4
3 T2_VIOS2: 11
__ 76. In the HMC CLI, add a virtual SCSI server adapter to your VIOS1 that is connected to your
LPAR2 using slot number values from the table above. 
For example, student #2 would assign LPAR2 slot 2 to T1_VIOS1 slot 11. 

In the example below (typed as one continuous command), replace the T# with your team
number, the S# with your student number, V# with your T#_VIOS1 slot number, and L1#
with your LPAR2 slot number for VIOS1:
student#@hmc:~> chhwres -r virtualio --rsubtype scsi -m m-sys-name -o a -p
T#_VIOS1 -s V# -a
adapter_type=server,remote_lpar_name=S#_LPAR2,remote_slot_num=L1#
__ 77. In the HMC CLI, add a virtual SCSI server adapter to your VIOS2 that is connected to your
LPAR2 using slot number values from the table above. 
For example, student #2 would assign LPAR2 slot 3 to T1_VIOS2 slot 11. 

In the example below (typed as one continuous command), replace the T# with your team
number, the S# with your student number, V# with your T#_VIOS2 slot number, and L2#
with your LPAR2 slot number for VIOS2:
student#@hmc:~> chhwres -r virtualio --rsubtype scsi -m m-sys-name -o a -p
T#_VIOS2 -s V# -a
adapter_type=server,remote_lpar_name=S#_LPAR2,remote_slot_num=L2#

© Copyright IBM Corp. 2009, 2019 2-26

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 78. In the HMC CLI, list the assigned virtual SCSI adapters for your VIOS1 and LPAR2, noting
that a new adapter is on VIOS at slot 10 or 11. The example output below is for student 1’s
LPAR2.
student#@hmc:~> lshwres -r virtualio --rsubtype slot -m m-sys-name --level
slot --filter 'lpar_names="T#_VIOS1,S#_LPAR2"' -F lpar_name,slot_num,config
T1_VIOS1,2,scsi
T1_VIOS1,3,scsi
T1_VIOS1,10,scsi
T1_VIOS1,1,serial
T1_VIOS1,0,serial
S1_LPAR2,1,serial
S1_LPAR2,0,serial
__ 79. In the HMC CLI, list the assigned virtual SCSI adapters for your VIOS2 and LPAR2, noting
that a new adapter is on VIOS at slot 10 or 11. The example output below is for student 1’s
LPAR2.
student#@hmc:~> lshwres -r virtualio --rsubtype slot -m m-sys-name --level
slot --filter 'lpar_names="T#_VIOS2,S#_LPAR2"' -F lpar_name,slot_num,config
T1_VIOS2,2,scsi
T1_VIOS2,3,scsi
T1_VIOS2,10,scsi
T1_VIOS2,1,serial
T1_VIOS2,0,serial
S1_LPAR2,1,serial
S1_LPAR2,0,serial
__ 80. In the HMC CLI, verify that this new adapter was written to the current VIOS1 profile (note
that your LPAR ID values may be different and there might be one more LPAR1 virtual SCSI
adapter in the list for the other student on your team):
student#@hmc:~> lssyscfg -m m-sys-name -r prof -F virtual_scsi_adapters
--filter lpar_names=T#_VIOS1
"2/server/6/S#_LPAR1/2/0,V#/server/5/S#_LPAR2/2/0"
__ 81. Repeat the previous step for your VIOS2 profile.

Note

At this point, you must add a virtual SCSI adapter to LPAR2 for VIOS1 and VIOS2. One adapter will
be added to your LPAR2 that maps to VIOS1, and another adapter will be added that maps to
VIOS2. The adapters will be added in the same command. Use the adapter slot information in
Table 4 on page 26 as a reference.
To add the virtual SCSI client adapters to LPAR2, a chhwres is not the command to use, because
your LPAR2 is not activated. Instead, the adapters will be added to LPAR2’s profile, using the
--force option because profile synchronization is enabled. Then the profile will be applied to the
current configuration to put the LPAR back in a synchronization enabled mode.

© Copyright IBM Corp. 2009, 2019 2-27

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 82. In the HMC CLI, add the virtual SCSI client adapter to LPAR2 that maps to both VIOS
partitions. The command assumes the profile is named default, but that could be
determined using lssyscfg or by looking at the HMC GUI. Replace the T# with your team
number, the S# with your student number, the V# with your T#_VIOS# slot number (it’s the
same for both VIOS), the L1# with your LPAR2 slot number for VIOS1, and the L2# with
your LPAR2 slot number for VIOS2.

Execute the following in the HMC CLI:
student#@hmc:~> chsyscfg -r prof -m m-sys-name -i
'lpar_name=S#_LPAR2,name=default,virtual_scsi_adapters="L1#/client//T#_VIOS1
/V#/0,L2#/client//T#_VIOS2/V#/0"' --force
__ 83. In the HMC CLI, verify that the profile was updated (note that the VIOS partition IDs will be
different in your output):
student#@hmc:~> lssyscfg -m m-sys-name -r prof -F virtual_scsi_adapters
--filter lpar_names=S#_LPAR2
"2/client/10/T#_VIOS1/V#/0,3/client/11/T#_VIOS2/V#/0"
__ 84. In the HMC CLI, check the profile synchronization state (0=disabled, 1=enabled, 2=out of
sync-requires reactivation from profile):
student#@hmc:~> lssyscfg -r lpar -m m-sys-name -F name,sync_curr_profile
--filter lpar_names=S#_LPAR2
S#_LPAR2,2
__ 85. In the HMC CLI, apply the updated profile to change the sync state back to enabled:
student#@hmc:~> chsyscfg -r lpar -m m-sys-name -p S#_LPAR2 -n default -o
apply
__ 86. In the HMC CLI, verify that sync is enabled again (can activate from current configuration):
student#@hmc:~> lssyscfg -r lpar -m m-sys-name -F name,sync_curr_profile
--filter lpar_names=S#_LPAR2
S#_LPAR2,1
__ 87. In the HMC CLI, verify that the virtual SCSI adapter is now in the current configuration:
student#@hmc:~> lshwres -r virtualio --rsubtype scsi -m m-sys-name --level
lpar --filter lpar_names=S#_LPAR2
lpar_name=S#_LPAR2,lpar_id=4,slot_num=2,state=0,is_required=0,adapter_type=c
lient,remote_lpar_id=2,remote_lpar_name=T#_VIOS1,remote_slot_num=10
lpar_name=S#_LPAR2,lpar_id=4,slot_num=3,state=0,is_required=0,adapter_type=c
lient,remote_lpar_id=3,remote_lpar_name=T#_VIOS2,remote_slot_num=10
__ 88. Now switch to the your VIOS CLI sessions to assign a LUN to the virtual SCSI server
adapter for your LPAR2 to use as a boot disk

© Copyright IBM Corp. 2009, 2019 2-28

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 89. In both VIOS CLI sessions, confirm that the virtual SCSI server adapter in slot 10 or 11 was
added and determine its device name (in the output below, the virtual SCSI adapter is
vhost2):
$ lsdev -slots
# Slot Description Device(s)
U8284.22A.21C193V-V11-C0 Virtual I/O Slot vsa0
U8284.22A.21C193V-V11-C2 Virtual I/O Slot vhost0
U8284.22A.21C193V-V11-C3 Virtual I/O Slot vhost1
U8284.22A.21C193V-V11-C10 Virtual I/O Slot vhost2
__ 90. In both VIOS CLI sessions, list the virtual SCSI mappings for the vhost discovered in the
previous step (there should be no devices mapped to the vhost adapter):
$ lsmap -vadapter vhost2
SVSA Physloc Client Partition ID
--------------- -------------------------------------------- ------------------
vhost2 U8284.22A.21C193V-V11-C10 0x00000005

VTD NO VIRTUAL TARGET DEVICE FOUND

__ 91. Find your assigned LPAR2 disk in the table below. That is, you will either be configuring
hdisk3 (student 1 and student 3) or hdisk6 (student 2 or student 4) to your LPAR2. And be
sure you are logged into your team’s VIOS pair (T1_ or T2_). Ask the instructor if you are
uncertain which disk to use.

Assigned Second LPAR1 LPAR2 LPAR3

VIOS VIOS rootvg rootvg rootvg
Student 1 T1_VIOS1 T1_VIOS2 hdisk2 hdisk3 hdisk4
Student 2 T1_VIOS2 T1_VIOS1 hdisk5 hdisk6 hdisk7
Student 3 T2_VIOS1 T2_VIOS2 hdisk2 hdisk3 hdisk4
Student 4 T2_VIOS2 T2_VIOS1 hdisk5 hdisk6 hdisk7
__ 92. Recall that since the disk will be used in a dual VIOS configuration, you changed the
reserve_policy to no_reserve earlier in this lab.
__ 93. In both VIOS CLI sessions, assign your LPAR2 disk to the vhost adapter displayed in
Step 89 on page 2-29 using the name shown in the example below replacing # with your
student number. If the vhost adapter names are different in your two VIOS CLI sessions,
use the appropriate vhost adapter names in the VIOS CLI.:
$ mkvdev -vdev hdisk# -vadapter vhost# -dev S#_LPAR2_rootvg
S#_LPAR2_rootvg Available
__ 94. In both VIOS CLI sessions, verify that the mapping was completed successfully
$ lsmap -vadapter vhost#
SVSA Physloc Client Partition ID
--------------- -------------------------------------------- ------------------
vhost# U8286.41A.2182AFV-V2-C3 0x00000004

VTD S#_LPAR2_rootvg

© Copyright IBM Corp. 2009, 2019 2-29

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
Status Available
LUN 0x8100000000000000
Backing device hdisk#
Physloc 
U78C9.001.WZS03PT-P1-C11-T2-W500507680235E1AB-L2000000000000
Mirrored false

__ 95. In the HMC CLI, activate your LPAR2:

student#@hmc:~> chsysstate -r lpar -m m-sys-name -o on -n S#_LPAR2
__ 96. Open a console on your LPAR2 using vtmenu on the HMC. If your LPAR2 does not boot to
the login prompt, shutdown your LPAR2 and repeat the steps you used in Step 32 on
page 2-18 to boot your LPAR1 partition.
__ 97. Login as root using the password provided by the instructor.
__ 98. List the disk devices. Determine the name of the new virtual SCSI disk drive. Note that there
may be additional hdisks listed in the Defined state on your lab system. Use the command
lsdev -c disk -S a to view just the available disks.
New virtual SCSI disk drive name: __________________
Here is an example output. In this example output, hdisk0 is the virtual disk you just
configured.
# lsdev -c disk -S a
hdisk0 Available Virtual SCSI Disk Drive
__ 99. Set the normal bootlist for the LPAR2 partition to the new virtual disk.
Here is an example command to set the bootlist to hdisk0.
# bootlist -m normal hdisk0
__ 100.Find the active Ethernet interface. You will configure the IP address for the LPAR on that
Ethernet interface.
» Here is an example command to find the active Ethernet interface. In this example, the en1
interface is active and will be used in the next step.
# lsdev -Cc adapter -S a | grep ^en
ent1 Available 01-00 PCIe2 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)

© Copyright IBM Corp. 2009, 2019 2-30

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 101.Configure the interface found in the previous step with the IP address information provided
by the instructor for LPAR2 using smitty chinet
Change / Show a Standard Ethernet Interface

Type or select values in entry fields.

Press Enter AFTER making all desired changes.

[Entry Fields]
Network Interface Name en1
INTERNET ADDRESS (dotted decimal) [10.8.xxx.yyy]
Network MASK (hexadecimal or dotted decimal) [255.255.255.0]
Current STATE up +
Use Address Resolution Protocol (ARP)? yes +
BROADCAST ADDRESS (dotted decimal) []
Interface Specific Network Options
('NULL' will unset the option)
rfc1323 []
tcp_mssdflt []
tcp_nodelay []
tcp_recvspace []
tcp_sendspace []
Apply change to DATABASE only no +
__ 102.To ensure that the default gateway is associated with the correct interface, reconfigure the
inet0 device.
# mkdev -l inet0
__ 103.Verify that the address is configured correctly by pinging your HMC address.
__ 104.In both VIOS CLI sessions, check the output of lsdev -virtual -type disk | grep S#
to verify that there are now two VTD entries for disks (one for your LPAR1 and one for your
LPAR2).
The output of the command should now show a new virtual target device. The example
shows output for student 2:
$ lsdev -virtual -type disk | grep S#
name status description
S2_LPAR1_rootvg Available Virtual Target Device - Disk
S2_LPAR2_rootvg Available Virtual Target Device - Disk

Part 6: Check virtual adapter configuration from HMC

You can check the configurations of virtual adapters on the HMC Enhanced+ GUI in the managed
system’s Virtual Storage view (VIOS-centric view) or in the partition’s Virtual Storage view
(LPAR-centric view).
__ 105.Login to your HMC GUI in Enhanced+ mode, using your assigned HMC userid and
password.

© Copyright IBM Corp. 2009, 2019 2-31

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 106.Navigate to your managed system view, and select the Virtual Storage link in the Menu
Pod.
In the Virtual Storage task, there are two views, the default view is the Storage View, The
Adapter View provides a visual mapping of the virtual SCSI or virtual Fibre Channel
relationships with a VIOS.
__ 107.In the Virtual Storage view, you see a table of VIO Servers that are managed by this HMC.
The default view is the Storage View. From this view, storage virtualization can be managed
on any of the VIO Servers in the table.
Select the Adapter View button in the upper right corner.

__ 108.A window opens displaying a table containing with the same VIO Servers. Each VIO Server
entry must be expanded to display their mappings, both virtual SCSI disks and virtual Fibre
Channel ports. Expand the section for your VIOS partitions, then for each of your LPARs,
and then for each of their vhost adapters to see the VTD name, backing device, type of
backing device, and size for each virtual SCSI disk. The example graphic shows the
expansion of Student #1’s LPAR2 on Team #1’s VIOS1, VIOS2. This output should look
similar to yours. If not, contact the instructor.

© Copyright IBM Corp. 2009, 2019 2-32

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty

In systems where there are many client partitions, the HMC view from the VIOS perspective might
not be the best method due to the volume of data in the table. Instead, one might view the storage
mappings from the client partition perspective.
__ 109.Click on the Partitions link in the managed system’s Menu Pod. This displays a list of
partitions on that managed system. Type your student prefix (S1, S2, S3, or S4) into the
filter box to display only your assigned LPARs.
__ 110.Select your LPAR2 partition in the table by clicking on the name of the partition.

© Copyright IBM Corp. 2009, 2019 2-33

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 111.The view changes to your LPAR2’s General Properties and Menu Pod. Select the Virtual
Storage link in the Menu Pod. The example graphic below shows the selection being done
for student #1’s LPAR2:

__ 112.In the Virtual Storage display, you will see three tabs and three tables. The first tab is Virtual
SCSI. In the Virtual SCSI tab, the first table is Physical Volume. In that tab and table you
should see your LPAR’s rootvg virtual SCSI disk. Notice that size and VIOS connection
information is displayed in the table.
Select that disk by clicking the radio button next to it.

__ 113.Choose View Device Mapping from the Action menu.

© Copyright IBM Corp. 2009, 2019 2-34

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 114.In the View Device Mapping window, notice the detailed information from both the client
and VIOS perspective. The example graphic below (and the view you will have) shows a
dual VIOS virtual SCSI disk relationship. This information is the same as was displayed
from the VIOS perspective under the Managed System’s Virtual Storage view. Close the
view when you are done observing the mappings.

Part 7: Test the client MPIO configuration - OPTIONAL

To execute this part of the lab, you must coordinate with the instructor and the other member of
your team before proceeding.
Next you will test the path failover capability. You will do this by shutting down VIOS2 (don’t do it
yet!), which currently provides the primary path to the LUN for LPAR1. You will verify that even
when the primary VIOS is not available, LPAR1 can still access the disk. When the failed VIOS
recovers, you will verify that the path through the recovered VIOS is re-enabled by the PCM.

Stop

You must work with the other student on your team, and the instructor before proceeding in this part
of the lab. You will be shutting down VIOS2 which is disruptive to the other student on your team.
Do not proceed until you know that the other student is working with you.

__ 115.IMPORTANT STEP: At this point, you must ensure that the primary path for your LPAR1
rootvg disk (hdisk2 or hdisk5) is the path to VIOS2. Is LPAR1’s rootvg disk primary path
currently set to the path to VIOS2? See the steps at the end of Part 3, "Configure path and
disk attributes on LPAR1," on page 2-21 and all of Part 4, "Explore path priorities," on
page 2-23 to determine if your LPAR1 disk’s primary path is through VIOS2.
If your LPAR1 disk’s primary path is through VIOS2, skip to Step 120 on page 2-36.

© Copyright IBM Corp. 2009, 2019 2-35

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 116.Only proceed to the next step if your primary path for LPAR1 rootvg disk is currently set to
VIOS1.
__ 117.In LPAR1, change the path priority back to the VIOS on the VIOS2 path. Do this by
changing the path priority for the vscsiX adapter that maps through VIOS1 to 3, making the
path to VIOS1 a lower priority.
For example:
# chpath -l hdisk0 -p vscsiX -a priority=3
The command output should be:
path Changed
__ 118.In LPAR1, rerun the cp /unix /tmp command. This generates traffic to force the PCM to
switch paths.
__ 119.In LPAR1, check the selected path ID now that traffic has been generated. Verify that the
path to VIOS2 is now the selected path.
__ 120.In LPAR1, create a script, using the contents below. Call the script “test_path.sh” and
create it in /tmp on LPAR1. The script writes the output of the date command to a file in
/tmp in a continuous while loop and also prints the output to the screen.
# vi /tmp/test_path.sh
#!/bin/ksh

>/tmp/datefile
while ((1))
do
date | tee -a /tmp/datefile
sleep 1
done
__ 121.In LPAR1, make the /tmp/test_path.sh script executable on LPAR1.
# chmod +x /tmp/test_path.sh
__ 122.In LPAR1, start the /tmp/test_path.sh script. The output of the date command should be
printing to standard out. Position the window so that you can see both the LPAR1 session
and an SSH session to VIOS2 on the vscsi1 path.
__ 123.From an SSH session to VIOS2, run the shutdown command. Answer y to the Continue?
question.
__ 124.Watch the /tmp/test_path.sh output on LPAR1 to confirm there is no disruption to the
date command output. Remember, the output of the command is also going to the
/tmp/datefile.
__ 125.On LPAR1, after about a minute, note the last timestamp in the output for the script, then
CTRL+C the script. Check the end of the /tmp/datefile file and compare the timestamp to
the one that last appeared in standard out to see that all of the date command output was
written to the file.

© Copyright IBM Corp. 2009, 2019 2-36

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 2. Dual VIOS Virtual SCSI Device Configuration

EXempty
__ 126.Confirm that the primary path (through vscsi1 in our examples) has failed. Use the lspath
command to view the status of the paths. If you don’t see a failure yet, wait a minute or two
more. Sometimes it takes time to register the failure.
Example command and output which shows that the vscsi1 path has failed.
# lspath -l hdisk0
Enabled hdisk0 vscsi0
Failed hdisk0 vscsi1
__ 127.Re-activate the shutdown VIOS to its current configuration from the HMC from the GUI or
CLI.
Alternatively, from the HMC CLI, the command would be, where <sys-name> is your
assigned managed system name:
hmc:~> chsysstate -o on -r lpar -m <sys-name> -n VIOS2
__ 128.Wait a couple of minutes after the VIOS2 operating system has booted, then verify that the
path from LPAR1 changes back from the Failed state to the Enabled state. Use the lspath
command to do this.
» Example command and its output:
# lspath -l hdisk0
Enabled hdisk0 vscsi0
Enabled hdisk0 vscsi1
That is the end of the lab exercise. Let your instructor know you have completed the exercise.

END OF LAB (with hints)

© Copyright IBM Corp. 2009, 2019 2-37

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty

Exercise 3. Configuring File-Backed

Optical Devices
(with hints)

Overview
This lab covers the configuration of file-backed virtual optical media SCSI devices.

Objectives
At the end of the lab, you should be able to:
• Configure a file-backed Virtual Media Repository
• Install an AIX partition using an install image loaded from a Virtual Media Repository using a
virtual optical media disk

Introduction
Students will configure storage resources and virtual target devices to provide clients with access to
file-backed optical media devices.

Requirements
• This workbook
• A workstation with a web browser connected to a network
• A POWER8 or later processor-based server with a dedicated HMC
• A system from which to start an SSH session to the HMC

© Copyright IBM Corp. 2009, 2019 3-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
Exercise Instructions with hints

Preface
• All exercises for this unit depend on the availability of specific equipment in your lab and at your
desk. You’ll need:
▪ A managed system configured with 3 assigned AIX partitions named LPAR1, LPAR2, and
LPAR3 and one assigned VIOS partition per student team. You will continue to work with
your assigned AIX partition (S#_LPAR3) and just your assigned VIOS partitions (T#_VIOS#)
▪ A system from which to execute remote commands to the HMC (via SSH)
▪ All lab systems need to be accessible to each other on a network
▪ An AIX V7.2 image file in ISO format in /home/padmin on the Virtual I/O Server.
• Hints are provided for lab exercises in case you need them and provide solutions for the lab
exercise steps. All hints are marked by a » sign.

Part 1: Create the Virtual Media Repository and store an ISO file
__ 1. Login to your assigned VIOS partition’s CLI. Refer to the table below to find your assigned
VIOS parition name. Be sure to login to your assigned VIOS.

Table 5: Student VIOS and LPAR assignments

Assigned Second LPAR1 LPAR2 LPAR3
VIOS VIOS
Student 1 T1_VIOS1 T1_VIOS2 S1_LPAR1 S1_LPAR2 S1_LPAR3
Student 2 T1_VIOS2 T1_VIOS1 S2_LPAR1 S2_LPAR2 S2_LPAR3
Student 3 T2_VIOS1 T2_VIOS2 S3_LPAR1 S3_LPAR2 S3_LPAR3
Student 4 T2_VIOS2 T2_VIOS1 S4_LPAR1 S4_LPAR2 S4_LPAR3
__ 2. In your assigned VIOS partition’s CLI, run the lspv -free command and note the physical
volumes that are available for use. You should find that your LPAR3 disk and hdisk1 are
listed. There might be other disks in the list.
The LPAR3 rootvg disk for students 1 and 3 is hdisk4 and the LPAR3 rootvg disk for
student 2 and 4 is hdisk7. Refer to the table below to validate your assigned LPAR3 rootvg
disk.
All students will use hdisk1 on their assigned VIOS for the VIOS Media Repository.

Table 6: AIX LPAR rootvg disk assignments per student

Assigned Second LPAR1 LPAR2 LPAR3
VIOS VIOS rootvg rootvg rootvg
Student 1 T1_VIOS1 T1_VIOS2 hdisk2 hdisk3 hdisk4
Student 2 T1_VIOS2 T1_VIOS1 hdisk5 hdisk6 hdisk7
Student 3 T2_VIOS1 T2_VIOS2 hdisk2 hdisk3 hdisk4
Student 4 T2_VIOS2 T2_VIOS1 hdisk5 hdisk6 hdisk7
__ 3. Create a new logical volume storage pool (or volume group) which includes hdisk1 in this
new storage pool. Call the storage pool datasp.

© Copyright IBM Corp. 2009, 2019 3-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
Example command to create a new volume group:
$ mkvg -f -vg datasp hdisk1
Alternatively, you could use the mksp command to create a storage pool with the following
command. (Do not run both commands.)
$ mksp -f datasp hdisk1
__ 4. View the output of lspv to verify that hdisk1 is in the datasp volume group. Also run the
lssp command to verify the datasp storage pool (LVPOOL). These two commands will
show you the new datasp volume group/storage pool.
$ lspv
NAME PVID VG STATUS
hdisk0 00f9c19175915859 rootvg active
hdisk1 00f9c1917653458d datasp active
hdisk2 00f9c19175aa626b None
hdisk3 00f9c19175abd40d None
hdisk4 none None
hdisk5 00f9c19175abfe25 None
hdisk6 00f9c19175ab3386 None
hdisk7 none None
$ lssp
Pool Size(mb) Free(mb) Alloc Size(mb) BDs Type
datasp 10168 10168 8 0 LVPOOL
rootvg 40704 13312 256 0 LVPOOL
__ 5. Create a 9 GB virtual media repository with the mkrep command. Create it in the datasp
volume group that you created previously.
$ mkrep -sp datasp -size 9G
Virtual Media Repository Created
Repository created within "VMLibrary" logical volume
__ 6. List out the repository and check its size.
$ lsrep
Size(mb) Free(mb) Parent Pool Parent Size Parent Free
9178 9178 datasp 10168 952
__ 7. Create a virtual optical media disk in the repository using the contents of the aix.iso file in
the /home/padmin directory. Create it with the name aix7211 and use the read-only (-ro)
flag (the command takes about a minute to complete).
Example listing of the /home/padmin directory showing the *.iso file.
$ ls
aix.iso ioscli.log smit.script tivoli
config smit.log smit.transaction
Example mkvopt command to create the virtual optical media disk and use the *.iso file as
its contents.
$ mkvopt -name aix7222 -file aix.iso -ro
__ 8. View the repository again and verify that the virtual optical media disk was created.

© Copyright IBM Corp. 2009, 2019 3-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
Example command and its output:
$ lsrep
Size(mb) Free(mb) Parent Pool Parent Size Parent Free
9179 7643 datasp 10168 952

Name File Size Optical

Access
aix7222 3107 None
ro

Note

Now the VIOS partition has a virtual media repository that contains an ISO file. In the next few steps
you make a virtual SCSI disk available to your LPAR3 partition to install AIX to, and assign the ISO
file to your LPAR3 partition to install AIX from. Use the HMC Enhanced+ GUI to add the virtual SCSI
disk and virtual optical drive to a new virtual SCSI adapter on your LPAR3 partition.

__ 9. Start or return to an existing HMC Enhanced+ GUI session. Navigate to your assigned
managed system’s default view (list of partitions on the managed system.) In the filter field,
enter your team prefix (S1, S2, S3, or S4) to display only your assigned LPARs.
__ 10. Click the name of your LPAR3 partition to display the General properties.
__ 11. Because LPAR3 has never been activated, the profile content must be synchronized. When
presented with the following dialog box, click OK.

© Copyright IBM Corp. 2009, 2019 3-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
__ 12. The Enhanced+ GUI tasks depend on the synchronization of the operations that are
performed into the LPARs profile. When the Save configuration changes to profile option
is Enabled, the operations will be synchronized. Enable that feature as follows:
__ a. Click the Advanced button in the upper right corner of the General Properties screen

__ b. Scroll to the bottom of the General Properties screen (in the Advanced Settings
section), and change the Save configuration changes to profile: option to Enabled (it
should be Disabled)

__ c. Click the Save button to complete the enablement.

__ 13. Click the Virtual Storage link under Virtual I/O section.
__ 14. In the Virtual Storage view, note that the Virtual SCSI tab is in focus. 
Click the Add Virtual SCSI Storage button to create a virtual SCSI disk mapping from both
VIOS partitions to your LPAR3.

© Copyright IBM Corp. 2009, 2019 3-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty

__ 15. Name the device S#_LPAR3-rootvg, where S# is your student number, and select your
LPAR3 rootvg hdisk# from your two VIOS partitions (T#_VIOS1, and T#_VIOS2). Be sure
to pick your assigned hdisk# (as shown in Table 6 on page 2) on your assigned VIOS
partitions (as shown in Table 5 on page 2). 
Click OK to execute the task.

© Copyright IBM Corp. 2009, 2019 3-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
__ 16. When the command completes you will be returned to the Virtual Storage display. Verify
that the new virtual SCSI disk is in the Physical Volume table.

__ 17. Select the Virtual Optical Device tab.

__ 18. Click Add Virtual Optical Device. This displays the Add Virtual Optical Drive dialog box.
__ 19. In the Add Virtual Optical Drive dialog box, name the device S#_LPAR3-cd, and select
your assigned VIOS from the list of Virtual I/O Servers. The example shows the selection
for student 1 on team1:

__ 20. Click Edit Connections to choose the existing virtual SCSI adapter for the virtual SCSI
disk. This displays the Edit Connection dialog box.
Choose the Server Adapter ID that is displayed in the drop-down menu. It might be 4 or 5.
Click OK to return to the Add Virtual Optical Device window.

© Copyright IBM Corp. 2009, 2019 3-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty

__ 21. Back in the Add Virtual Optical Device window, click OK to execute the task.
__ 22. Verify that the table in the Virtual Optical tab sows the S#_LPAR3-cd device. Notice that the
Media File field is empty. In the next step, you will load the AIX ISO file into the new virtual
optical device.
__ 23. Select the virtual optical device in the table (click the radio button next to it). The choose
Load from the Action menu. This will display the Load Media File dialog box.

__ 24. In the Load Media File dialog box, choose the aix7222 media file and then click OK to
execute the task.
__ 25. Back in your assigned VIOS CLI session, view the repository and see what is different in the
output from the last lsrep command that you ran in a previous step.

© Copyright IBM Corp. 2009, 2019 3-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
Notice that the “Optical” column now lists the VTD name of the virtual optical device.
$ lsrep
Size(mb) Free(mb) Parent Pool Parent Size Parent Free
9179 7643 datasp 10168 952

Name File Size Optical

Access
aix7222 3107 S1_LPAR3-cd
ro
Now you have a virtual optical media drive loaded with data ready to be used by a partition. In
the next part of this exercise you will configure LPAR3 to see this media, boot from it, and install
the AIX operating system.

Part 2: Install AIX Image From Virtual Optical Media Device

__ 26. You will need a console when activating the LPAR3 partition. The console can be started
from the GUI, but this can be unreliable because of various Java-related complications.
Instead, you will open the console using the HMC CLI and the vtmenu command.
__ a. Open an SSH session to your assigned HMC using Putty or a terminal session on your
workstation. If you are having problems connecting to the HMC using SSH, contact the
instructor.
__ b. At the login prompt, login with your assigned HMC userid and password.
__ c. Enter the command, vtmenu

© Copyright IBM Corp. 2009, 2019 3-9

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
__ d. From the list of partitions, choose the number associated with your LPAR3. In the
example below for student 1, option 3 is chosen to open the console for S1_LPAR3. The
console is open and waiting for input.
----------------------------------------------------------
Partitions On Managed System: sys870
OS/400 Partitions not listed
----------------------------------------------------------
1) S1_LPAR1 Running
2) S1_LPAR2 Running
3) S1_LPAR3 Not Activated
4) S2_LPAR1 Running
5) S2_LPAR2 Not Activated
6) S2_LPAR3 Not Activated
7) S3_LPAR1 Not Activated
8) S3_LPAR2 Not Activated
9) S3_LPAR3 Not Activated
10) S4_LPAR1 Not Activated
11) S4_LPAR2 Not Activated
12) S4_LPAR3 Not Activated
13) T1_VIOS1 Running
14) T1_VIOS2 Running
15) T2_VIOS1 Running
16) T2_VIOS2 Running

Enter Number of Running Partition (q to quit): 3

6
Opening Virtual Terminal On Partition S1_LPAR3 . . .

Open in progress

Open Completed.

__ 27. Switch back the HMC GUI window.

__ 28. To activate the LPAR3 logical partition, select the Partition Actions link in the Menu Pod.
__ a. Select the Operations link
__ b. Select the Activate task

© Copyright IBM Corp. 2009, 2019 3-10

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty

Note

If you encounter this error, click Cancel:

Click General Properties in the Menu Pod or refresh your browser window and attempt the
activation again. If the failure persists, contact the instructor.

__ c. In the Activate S#_LPAR3 window, leave all the fields set at the default, except the
following:
Under Advanced Settings:
Boot Mode = System Management Services
__ d. Click Finish
__ e. Close the Activate window when it displays the message Partition activated.
__ 29. Switch to the LPAR3 console window (the HMC CLI SSH session)
__ 30. Interact with the SMS menus to set the boot device. There should be only one disk available
and will be listed as a SCSI disk. Choose this disk and boot the operating system. 
Here are exact instructions to help with the SMS menu options:
__ a. You may need to type a 0 (zero) to choose your window as the console.
__ b. In the main SMS menu choose option 5: Select Boot Options.
__ c. Choose 1: Select Install/Boot Device.
__ d. Choose 2: CD/DVD.
__ e. Choose 1: SCSI.
__ f. Choose 1: If your server adapter ID was 4 in the Edit Connections dialog earlier, or
Choose 2: if your server adapter ID was 5 in the Edit Connections dialog earlier.
__ g. Choose 1: To select the SCSI CD-ROM drive (virtual optical drive). If you get an error,
return to the previous menu and choose the other vSCSI adapter.

© Copyright IBM Corp. 2009, 2019 3-11

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 3. Configuring File-Backed Optical Devices

EXempty
__ h. Choose 2: Normal Boot.
__ i. Choose 1: Yes. (You are sure you want to exit SMS.)
» Install AIX as you normally would using all the defaults for the questions. If you have questions
about the installation steps, contact the instructor.
__ 31. Once you see the Installing Base Operating System screen start to log progress by
having a percentage greater than zero in the ”% tasks completed” message, let your
instructor know you’re finished with the exercise. Your LPAR3 will finish its installation
(takes about 15 minutes) and will reboot on its own.
Note: If you happen to check the LPAR later and want to login, there is no root password.
No network adapters are configured for LPAR3 (not enough hardware in the systems).
Therefore, you can not configure networking.

END OF LAB (with hints)

© Copyright IBM Corp. 2009, 2019 3-12

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty

Exercise 4. Dual VIOS Virtual Fibre

Channel Configuration
(with hints)

Overview
This exercise covers a popular dual Virtual I/O Server storage configuration with virtual Fibre
Channel adapters. You will configure a client to access storage through two virtual Fibre Channel
paths, one through each VIOS partition. The client will use AIX MPIO to access the NPIV storage
over the dual paths. You will configure the VIOS partitions and the client, then test failover by
shutting down the VIOS partition providing the primary path.

Objectives
At the end of the lab, you should be able to:
• Configure virtual Fibre Channel server adapters and NPIV on two VIOS partitions to provide
dual-path access to the same storage for a virtual client
• Configure virtual Fibre Channel client adapters for the client partition
• Manage MPIO to the SAN storage on the client partition to support proper failover actions and
load balancing between VIOS partitions
• Test the client MPIO failover configuration

Introduction
In this lab exercise, you will configure both assigned VIOS partitions with virtual Fibre Channel
server adapters. You’ll map these to local physical Fibre Channel ports using NPIV. The client will
be configured with virtual Fibre Channel client adapters. The result will be two paths from the client
partition through the VIOS partitions to a SAN LUN. You will verify the configuration is correct by
testing failover from one VIOS path to the other.

© Copyright IBM Corp. 2009, 2019 4-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
Exercise instructions with hints

Preface
• Two versions of these instructions are available; one with hints and one without hints. You can
use either version to complete this exercise. Also, please don’t hesitate to ask the instructor if
you have questions. In some cases, the answer given may be just an example, and there may
be other correct answers.
• All exercises of this chapter depend on the availability of specific equipment in your classroom.
• All hints are marked by a » sign.

Part 1: Configure virtual Fibre Channel adapters

__ 1. Before adding the virtual Fibre Channel adapter, determine the pre-zoned WWPNs that
have a SAN LUN mapped for your LPAR’s use. 
On your LPAR2, run the script 
/wwpn_display.sh <student ID #>,
where <student ID #> is replaced with your student number.
The following example is for student 2 in sys869:
# /wwpn_display.sh 2

Virtual fibre channel adapter #1 WWPN values are ...

c869100000000021 and c869100000000022

Virtual fibre channel adapter #2 WWPN values are ...

c869100000000023 and c869100000000024
__ 2. Start a web browser and connect to your assigned HMC in Enhanced+ mode. Logon using
the userid and password supplied by your instructor.
» Start a web browser.
» Using the https protocol, enter your assigned HMC’s hostname as the web address.
» Answer appropriately to any security pop up windows.
» Use the login userid and password given to you by your instructor. Choose to login in
Enhanced+ Mode.
__ 3. On the HMC, navigate to your LPAR2 partition’s Virtual Storage view.
__ 4. In LPAR2’s Virtual Storage view select the Virtual Fiber Channel tab. Notice the table of
devices is empty.
__ 5. Click the Add Virtual Fiber Channel Device button. The Add Virtual Fiber Channel dialog
box is displayed.

© Copyright IBM Corp. 2009, 2019 4-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty

__ 6. In the Add Virtual Fiber Channel dialog box, select the fcs0 port for your assigned VIOS1
and fcs0 port for your assigned VIOS2. Be careful to choose the correct VIOS partitions.
Then click Edit Connections. In the Edit Connection dialog box:
__ a. In the WWPN fields, fill in the two WWPN boxes for each adapter with the two WWPNs
for each adapter that were displayed in Step 1 on page 4-2.
__ b. In the Server Adapter field, set the Virtual Fiber Channel server adapter ID for both
VIOS partitions based on the following. Hint: Highlight the text in the box, overwrite it
with the number noted below, then hit Enter to affix the value into the box.
student 1: 21
student 2: 22
student 3: 23
student 4: 24
__ c. Click OK to return to the Add Virtual Fiber Channel dialog box. The example below is
for student 1.

© Copyright IBM Corp. 2009, 2019 4-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty

__ 7. Click OK in the Add Virtual Fiber Channel dialog box to execute the task.
__ 8. Verify that the new Virtual Fiber Channel device was added and is mapped through both
VIOS partitions

__ 9. Log into your assigned VIOS1 partition.

__ 10. Log into your assigned VIOS2 partition.
__ 11. Log into the LPAR2 partition. Run the cfgmgr command.
__ 12. In VIOS1, run the lsdev -vpd | grep vfchost command. You should see the virtual FC
server adapter that was added. Look at the location code. Recall, the server adapter slot
number that you specified in the previous step for the virtual Fiber Channel server adapter
ID. Make note of the vfchost (virtual fibre channel server adapter) that maps to the server
adapter number you specified in the previous step
Example command and its output which shows student 2, with vfchost1 assigned for server
adapter ID number 22:
$ lsdev -vpd | grep vfchost
vfchost1 U8284.22A.21C193V-V10-C22 Virtual FC Server Adapter
__ 13. In VIOS1, use the lsnports command to see available NPIV ports. Verify that the fcs0
physical port shows less than 64 available ports (aports) because you assigned one to
LPAR2. The number of available ports should be 62 or 63.

© Copyright IBM Corp. 2009, 2019 4-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
Example command and its output showing two physical ports one with 63 available ports
and one with 64 available virtual ports:
$ lsnports
name physloc fabric tports aports swwpns awwpns
fcs0 U78C9.001.WZS0B44-P1-C11-T1 1 64 63 2048 2043
fcs1 U78C9.001.WZS0B44-P1-C11-T2 1 64 64 2048 2048
An available port has a ‘1’ in the fabric column and there is one less port available in the
aports column, that is, 63 vs. 64.
__ 14. In VIOS1, use lsmap to check the configuration of the vfchost adapter. Replace the
vfchost# in the example below with the vfchost you found in Step 12 on page 4-4. The
VIOS1 configuration is complete if you see in the output that the status is LOGGED_IN. If
you do not see the logged in status, double check your virtual adapter configuration.
Example command and its output:
$ lsmap -npiv -vadapter vfchost#
Name Physloc ClntID ClntName ClntOS
------------- ---------------------------------- ------ -------------- -------
vfchost# U8286.41A.21E096V-V1-C22 4 LPAR2 AIX

Status:LOGGED_IN
FC name:fcs0 FC loc code:U78C9.001.WZS0B44-P1-C11-T1
Ports logged in:3
Flags:a<LOGGED_IN,STRIP_MERGE>
VFC client name:fcs0 VFC client DRC:U8286.41A.21E096V-V4-C3
__ 15. In VIOS2, run the lsdev -vpd | grep vfchost command. You should see the virtual FC
server adapter that was added earlier in this exercise. The location code should show the
same virtual FC server adapter ID that you specified for this VIOS too.
Example command and its output which shows vfchost1 as the server FC adapter with the
adapter ID of 22:
$ lsdev -vpd | grep vfchost
vfchost1 U8284.22A.21C193V-V11-C22 Virtual FC Server Adapter
__ 16. In VIOS2, use the lsnports command to see available NPIV ports. Verify that the fcs0
physical port shows 63 available ports (aports) because you assigned one to LPAR2.
Example command and its output showing two physical ports:
$ lsnports
name physloc fabric tports aports swwpns awwpns
fcs0 U78C9.001.WZS0B44-P1-C11-T1 1 64 63 2048 2043
fcs1 U78C9.001.WZS0B44-P1-C11-T2 1 64 64 2048 2048
__ 17. In VIOS2, use lsmap to check the configuration of the vfchost adapter. Replace the
vfchost# in the example below with the vfchost you found in Step 12 on page 4-4. The
VIOS2 configuration is complete if you see the status is LOGGED_IN. If you do not see the
logged in status, double check your virtual adapter configuration.

© Copyright IBM Corp. 2009, 2019 4-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
Example command and its output:
$ lsmap -npiv -vadapter vfchost#
Name Physloc ClntID ClntName ClntOS
------------- ---------------------------------- ------ -------------- -------
vfchost# U8284.22A.21C193V-V11-C22 4 LPAR2 AIX

Status:LOGGED_IN
FC name:fcs0 FC loc code:U78C9.001.WZS0B44-P1-C12-T2
Ports logged in:1
Flags:a<LOGGED_IN,STRIP_MERGE>
VFC client name:fcs1 VFC client DRC:U8286.41A.21E096V-V4-C4

Part 2: View the status of the N_Ports

__ 18. Open an SSH session to the HMC.
__ 19. Let’s look at some HMC commands to view information about the virtual Fibre Channel
devices. Use the lshwres command to view the current Fibre Channel adapter
configuration for your LPAR2 partition. Notice there are two connections, one to each VIOS
in a dual VIOS configuration.
Example command and output for student #2 and system bud:
student2@bud-hmc:~> lshwres -r virtualio --rsubtype fc -m bud --level lpar
--filter 'lpar_names=S2_LPAR2'
lpar_name=S2_LPAR2,lpar_id=8,slot_num=4,adapter_type=client,state=1,is_requi
red=0,remote_lpar_id=10,remote_lpar_name=T1_VIOS1,remote_slot_num=22,"wwpns=
c869100000000023,c869100000000024"
lpar_name=S2_LPAR2,lpar_id=8,slot_num=5,adapter_type=client,state=1,is_requi
red=0,remote_lpar_id=11,remote_lpar_name=T1_VIOS2,remote_slot_num=22,"wwpns=
c869100000000021,c869100000000022"
__ 20. Use the lsnportlogin HMC command to view the status of the ports on LPAR2. There
should be status for four WWPNs, two mapped through VIOS1 and two mapped through
VIOS2. The wwpn_status of 2 means unknown which will be addressed in the upcoming
steps.
Example command and output for student #1 and system sys865:
student1@sys865-hmc:~> lsnportlogin -m sys865 --filter lpar_names=S1_LPAR2
lpar_name=S1_LPAR2,lpar_id=12,profile_name=default,slot_num=5,wwpn=c86510000
0000013,wwpn_status=2,logged_in=unknown,wwpn_status_reason=null
lpar_name=S1_LPAR2,lpar_id=12,profile_name=default,slot_num=5,wwpn=c86510000
0000014,wwpn_status=2,logged_in=unknown,wwpn_status_reason=null
lpar_name=S1_LPAR2,lpar_id=12,profile_name=default,slot_num=4,wwpn=c86510000
0000011,wwpn_status=2,logged_in=unknown,wwpn_status_reason=null
lpar_name=S1_LPAR2,lpar_id=12,profile_name=default,slot_num=4,wwpn=c86510000
0000012,wwpn_status=2,logged_in=unknown,wwpn_status_reason=null

© Copyright IBM Corp. 2009, 2019 4-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
__ 21. The wwpn_status value for the ports might initially show up as unknown (a value of 2). If this
is the case, use the chnportlogin command to log all of LPAR2's ports into the fabric, then
run the command again to log the inactive ports out. What do you notice about the output of
the lsnportlogin command after logging the ports in and out of the fabric?
» First, use the chnportlogin command to cause all of the ports of the LPAR to log in to the
fabric. Here is an example command for student #2 and system bud:
student2@bud-hmc:~> chnportlogin -m bud -p S2_LPAR2 -o login
» View the port status with the lsnportlogin command. This time specify in the command to
display only the WWPN values, and login status fields. Below is an example command and its
output for student #2 and system bud. Notice that the WWPN status is now active (1) for both
ports. Notice that one says it is logged in from the client (the active port) and one says it is
logged in from the VIOS. This is because the chnportlogin command logs into the fabric from
the VIOS using the non-active WWPN values, not from the client LPAR.
student2@bud-hmc:~> lsnportlogin -m bud --filter lpar_names=S2_LPAR2 -F
wwpn,wwpn_status,logged_in
c869100000000021,1,client
c869100000000022,1,vios
c869100000000023,1,client
c869100000000024,1,vios
» Use the chnportlogin command to log all non-active ports out of the fabric. Here is an
example command for student #2 and system bud:
student2@bud-hmc:~> chnportlogin -m bud -p S2_LPAR2 -o logout
» View the port status again by repeating the last lsnportlogin command. Here is an example
command and its output for student #2 and system bud. Notice that the WWPN status is now
active (1) for the port that is actually in use, and non-active (0) for the one that is not being used.
It also shows that the port (WWPN) was logged in to the fabric from the client LPAR.
student2@bud-hmc:~> lsnportlogin -m bud --filter lpar_names=S2_LPAR2 -F
wwpn,wwpn_status,logged_in
c869100000000021,1,client
c869100000000022,0,none
c869100000000023,1,client
c869100000000024,0,none

Part 3: Configure MPIO in the client partition

Next, we need to configure the client partition.

Note

A SAN LUN was mapped to the WWPNs for the virtual Fiber Channel adapters you added.
Ordinarily, this involves a call to the SAN administrator, requesting one or more LUNs to be mapped
to all four of those WWPNs. However, in our lab we maintain a set of pre-zoned WWPNs with LUNs
already mapped to them.

© Copyright IBM Corp. 2009, 2019 4-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
__ 22. Switch to your LPAR2 session or log into the LPAR2 partition.
__ 23. In the login session to LPAR2, list the available disks. In the lsdev -c disk output, notice
the description is not Virtual SCSI Disk Drive for the SAN disk. With NPIV, the client
operating system can see the type of device and will show an AIX location code.
Example command and expected output showing only one virtual SCSI disk drive:
# lsdev -c disk
hdisk0 Available Virtual SCSI Disk Drive
hdisk1 Available C2-T1-01 MPIO IBM 2145 FC Disk
__ 24. Do you see the new disk in LPAR2, coming from the SAN? If you do not see a new disk in
LPAR2, go back and check the WWPNs that you specified for the virtual Fiber Channel
adapters. Contact the instructor if you are unable to see a new disk on LPAR2 after you
investigate the problem.
__ 25. On LPAR2, view the Fibre Channel devices with the lscfg command. Notice the name of
the virtual Fibre Channel client adapter device.
Example command and its output:
# lscfg | grep fcs0
+ fcs0 U8284.22A.21C18FV-V7-C3-T1 Virtual
Fibre Channel Client Adapter
__ 26. On LPAR2, use the fcstat command to view the active WWPN for the virtual FC adapter.
Example command and its output which shows one WWPN (C865100000000021) for the
fcs0 adapter:
# fcstat fcs0 | grep World
World Wide Node Name: 0xC865100000000021
World Wide Port Name: 0xC865100000000021
__ 27. On LPAR2, use the lscfg command to view the only WWPN for the virtual FC adapter in
the ODM.

© Copyright IBM Corp. 2009, 2019 4-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
Example command and its output which shows one WWPN (C865100000000021) for the
fcs0 adapter:
# lscfg -vl fcs0
fcs0 U8284.22A.21C18FV-V7-C3-T1 Virtual Fibre Channel Client
Adapter

Network Address.............C865100000000021
ROS Level and ID............
Device Specific.(Z0)........
Device Specific.(Z1)........
Device Specific.(Z2)........
Device Specific.(Z3)........
Device Specific.(Z4)........
Device Specific.(Z5)........
Device Specific.(Z6)........
Device Specific.(Z7)........
Device Specific.(Z8)........C865100000000021
Device Specific.(Z9)........
Hardware Location Code......U8284.22A.21C18FV-V7-C3-T1
__ 28. List the paths for the newly discovered MPIO disk.
Example command and expected output:
# lspath -l hdisk1
Enabled hdisk1 fscsi0
Enabled hdisk1 fscsi0
Enabled hdisk1 fscsi0
Enabled hdisk1 fscsi0
Enabled hdisk1 fscsi1
Enabled hdisk1 fscsi1
Enabled hdisk1 fscsi1
Enabled hdisk1 fscsi1
__ 29. Make note of the relationship between the fiber channel adapters and the VIOS partitions. In
the examples below, fcs0 is adapter ID 2 and maps to VIOS1, while fcs1 is adapter ID 3
and maps to VIOS2.
# lsslot -c slot | grep fcs
U8286.41A.21E096V-V4-C2 Virtual I/O Slot fcs0
U8286.41A.21E096V-V4-C3 Virtual I/O Slot fcs1
# echo "vfcs fcs0" | kdb | grep vfchost
host_name: T1-VIOS1 host_device: vfchost0
# echo "vfcs fcs1" | kdb | grep vfchost
host_name: T1-VIOS2 host_device: vfchost0
__ 30. Check the algorithm attribute on the disk. It is probably set for fail_over which is okay for
this exercise.
# lsattr -El hdisk1 -a algorithm
algorithm fail_over Algorithm True

© Copyright IBM Corp. 2009, 2019 4-9

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
__ 31. Check the range of options for the algorithm attribute.
# lsattr -Rl hdisk1 -a algorithm
fail_over
round_robin
shortest_queue
__ 32. Check the path attributes for both paths. Are they set to the recommended values?
Example command for the first path and example output:
# lsattr -El fscsi0
attach switch How this adapter is CONNECTED False
dyntrk yes Dynamic Tracking of FC Devices True
fc_err_recov fast_fail FC Fabric Event Error RECOVERY Policy True
scsi_id 0x10e01 Adapter SCSI ID False
sw_fc_class 3 FC Class for Fabric True
__ 33. If any of the path attributes are not set to the recommended values then change them. You’ll
need to use the -P flag to change the ODM. You’ll reboot in a later step.
You should find that dyntrk is already set to yes and fc_err_recov is already set to
fast_fail, which are the recommended values.
If any of the values are not set to the recommended values, use the following command
syntax to help you set the values:
# chdev -l fscsi0 -a dyntrk=yes -P
fscsi0 changed
__ 34. Check the disk attributes. Are they set to the recommended values?
Example command and expected output:
# lsattr -El hdisk1 -a reserve_policy,algorithm,hcheck_mode,hcheck_interval
reserve_policy single_path Reserve Policy True+
algorithm fail_over Algorithm True+
hcheck_mode nonactive Health Check Mode True+
hcheck_interval 60 Health Check Interval True+
The output above shows the hcheck_interval could be changed to the read/write timeout
value of the disk, reserve policy could be changed to no_reserve, and we’ll leave the
fail_over value for algorithm.
__ 35. Check the read/write timeout value (rw_timeout) on the disk.
Example command and output that shows the timeout value is 30.
# lsattr -El hdisk1 -a rw_timeout
rw_timeout 30 READ/WRITE time out value True
__ 36. Configure the disk attributes for the recommended values. Again, leave the algorithm set
to fail_over for now. If in AIX and the disk was in use, you could use the -U option to
make the changes concurrently.) Use the lsattr command to check the updated settings.
# chdev -l hdisk1 -a hcheck_interval=30 -a reserve_policy=no_reserve
# lsattr -El hdisk1 -a hcheck_interval

© Copyright IBM Corp. 2009, 2019 4-10

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
__ 37. A reboot of LPAR2 is necessary only if dyn_trk or fc_err_recov were changed. Wait until
the end of the next section to perform the reboot, if necessary.

Part 4: Check the virtual Fibre Channel tunables

__ 38. On both VIOS partitions, document the max_xfer_size and the num_cmd_elems for the
physical Fibre Channel port that the LPAR2 partition is using.
max_xfer_size: _____________________
num_cmd_elems: ___________________
First, make sure you have the correct physical FC port on each VIOS. List the NPIV
mappings, looking for your S#-LPAR2 and then note the physical Fibre Channel adapter
port (fcs0 in the example) that was configured for the virtual Fibre Channel server adapter.
Then display the values for the two attributes for that physical adapter.
$ lsmap -all -npiv
Name Physloc ClntID ClntName ClntOS
------------- ----------------------- ------ -------------- -------
vfchost# U8202.E4B.06BCC8P-V1-C22 2 S#-LPAR2 AIX

Status:LOGGED_IN
FC name:fcs0 FC loc code:U78AA.001.WZSGHVK-P1-C4-T1
Ports logged in:14
Flags:a<LOGGED_IN,STRIP_MERGE>
VFC client name:fcs0 VFC client DRC:U8202.E4B.06BCC8P-V1-C5
Check the values of the two attributes:
$ lsdev -dev fcs0 -attr | grep -i maximum
max_xfer_size 0x400000 Maximum Transfer Size
True
num_cmd_elems 4014 Maximum number of COMMANDS to queue to the adapter
True
__ 39. Now document the values for the same attributes on the virtual FC adapter on LPAR2.
max_xfer_size: _____________________
num_cmd_elems: ___________________
# lsattr -El fcs0 | grep -i maximum
intr_priority 3 Interrupt priority False
lg_term_dma 0x800000 Long term DMA True
max_xfer_size 0x100000 Maximum Transfer Size True
num_cmd_elems 200 Maximum Number of COMMAND Elements True
sw_fc_class 2 FC Class for Fabric True
__ 40. If necessary, change both attributes in LPAR2 so that the values match those used on the
physical port in both VIOS partitions (but do not exceed 2048 for num_cmd_elems). A reboot
will be necessary to change the values on the running system.

© Copyright IBM Corp. 2009, 2019 4-11

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
Example command given the physical adapter values presented in a previous hint:
# chdev -l fcs0 -a max_xfer_size=0x400000 -a num_cmd_elems=2048 -P
fcs0 changed
Here’s an example command below and its output that shows the results. Note that you’ll
have to reboot the LPAR for the values to be used.
# lsattr -El fcs0
intr_priority 3 Interrupt priority False
lg_term_dma 0x800000 Long term DMA True
max_xfer_size 0x400000 Maximum Transfer Size True
num_cmd_elems 2048 Maximum Number of COMMAND Elements True
sw_fc_class 2 FC Class for Fabric True
Reboot LPAR2 using the shutdown -Fr command.

Part 5: Test the paths

In this part of the exercise we’ll test path failover to paths by disabling one of the paths.
__ 41. Return to your LPAR2 session (or open one if necessary)
__ 42. Add the new NPIV disk on LPAR2 to a volume group. Call the volume group npivvg.
Example command: # mkvg -y npivvg hdisk1
__ 43. In LPAR2, create a 1 GB jfs2 filesystem in the npivvg volume group with the /datafs mount
point.
Example command:
# crfs -v jfs2 -m /datafs -g npivvg -a size=1G
__ 44. Mount the /datafs filesystem on LPAR2 and verify it is mounted successfully.
Example commands and output:
# mount /datafs
# mount | grep datafs
/dev/fslv00 /datafs jfs2 Sept 19 18:24 rw,log=/dev/loglv00
__ 45. Have a login session open to both VIOS1 and VIOS2. Position the windows so you can see
all three windows to VIOS1, VIOS2, and LPAR2.
__ 46. In both VIOS partitions, run the nmon tool. Type the up-caret (^) character to get to the panel
that shows Fibre Channel adapter port statistics.
__ 47. In LPAR2, run the cp /unix /datafs command at the same time that you are watching
both nmon outputs. Can you determine which VIOS is in use for access to the SAN LUN on
LPAR2? Note the name of the primary VIOS: ____________________
You should see a short burst of activity on one of the VIOS partitions. Here is an example of
what you might see in one VIOS. The other VIOS will not show any statistics much above
zero.

© Copyright IBM Corp. 2009, 2019 4-12

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty

__ 48. Recall that you found the relationship between S#_LPAR2’s virtual fiber channel adapters
and the VIOS through which each adapter is mapped at Step 29 on page 4-9. If necessary,
return to that step to determine which virtual fiber channel adapter maps to the primary
VIOS.
__ 49. Force LPAR2’s NPIV disk traffic onto the secondary VIOS by disabling the path to the
primary VIOS. In the examples below, it is assumed VIOS1 through fcs0 is the primary path
and VIOS2 through fcs1 is the secondary path, that is, all the paths have a path priority of 1,
so path 0 is primary. If your configuration is different, run the appropriate commands to
change from primary to secondary paths.
Below are example command that list the paths and path ID information followed by
commands that disable the VIOS1 connections through fscsi0:
# lspath -tl hdisk1
Enabled hdisk1 fscsi0 0
Enabled hdisk1 fscsi0 1
Enabled hdisk1 fscsi0 2
Enabled hdisk1 fscsi0 3
Enabled hdisk1 fscsi1 4
Enabled hdisk1 fscsi1 5
Enabled hdisk1 fscsi1 6
Enabled hdisk1 fscsi1 7

# chpath -l hdisk1 -i 0 -s disable

path Changed
# chpath -l hdisk1 -i 1 -s disable
path Changed
# chpath -l hdisk1 -i 2 -s disable
path Changed
# chpath -l hdisk1 -i 3 -s disable
path Changed

© Copyright IBM Corp. 2009, 2019 4-13

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 4. Dual VIOS Virtual Fibre Channel Configuration

EXempty
__ d. Display the hdisk1 paths to verify that the disable operation was successful.
# lspath -tl hdisk1
Disabled hdisk1 fscsi0
Disabled hdisk1 fscsi0
Disabled hdisk1 fscsi0
Disabled hdisk1 fscsi0
Enabled hdisk1 fscsi1
Enabled hdisk1 fscsi1
Enabled hdisk1 fscsi1
Enabled hdisk1 fscsi1

__ e. Now that the traffic is pushed over the secondary VIOS path, use the cp /unix
/datafs operation with the nmon tools running in VIOS1 and VIOS2 to verify that there
is activity on the secondary VIOS.
__ 50. Type a 'q' in both of the nmon windows to stop the tool.
__ 51. You have shown that one path can be disabled and this will force traffic to the other path
when using MPIO with NPIV storage. This mimics a failover situation where one VIOS may
have an unplanned outage.
__ 52. Re-enable the paths that were disabled
Below is an example command that to enable the VIOS1 connections through fscsi0:
# chpath -l hdisk1 -i 0 -s enable
path Changed
# chpath -l hdisk1 -i 1 -s enable
path Changed
# chpath -l hdisk1 -i 2 -s enable
path Changed
# chpath -l hdisk1 -i 3 -s enable
path Changed

__ 53. Let your instructor know you have completed all of the lab steps.

End of exercise

© Copyright IBM Corp. 2009, 2019 4-14

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty

Exercise 5. Virtual Ethernet Adapter

Configuration
(with hints)

Overview
This exercise has students configure virtual Ethernet adapters. You will use both the HMC GUI and
CLI.

Objectives
At the end of the lab, you should be able to:
• Configure and test partition communications with virtual Ethernet adapters

Introduction
In this exercise, you will create a virtual Ethernet connection between partitions. You will use both
the HMC graphical interface and HMC commands to complete the tasks.

Requirements
• This workbook
• A workstation with a web browser connected to a network
• A POWER8 or later processor-based server with a dedicated HMC
• A system from which to start an SSH session to the HMC

© Copyright IBM Corp. 2009, 2019 5-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty
Exercise Instructions with hints

Preface
• All exercises of this chapter depend on the availability of specific equipment in your classroom.
You’ll need:
▪ A managed system configured with 2 running AIX partitions
▪ A system from which to execute remote commands to the HMC (via SSH)
▪ All lab systems need to be accessible to each other on a network

Part 1: Configure an internal Virtual Ethernet Network

In this part of the exercise, you will create an internal virtual Ethernet network using the Enhanced+
GUI, and then verify that it was created using the HMC CLI. All examples in the HMC CLI steps are
to be typed as one continuous command.
__ 1. Start a web browser and connect to your assigned HMC. Log on in Enhanced+ mode using
the userid and password provided by your instructor.
__ 2. Start an SSH session to your assigned HMC. Log on using the userid and password
provided by your instructor.
__ 3. In the HMC GUI, navigate to the Menu Pod for your assigned managed system. Choose
Resources -> All Systems, then click the name of your assigned managed system.
__ 4. Choose Virtual Networks from the PowerVM section of the Menu Pod.

© Copyright IBM Corp. 2009, 2019 5-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty
__ 5. The Virtual Networks panel is displayed. There might be virtual networks created by another
student using the same managed system. If so, they will be in the table under the Virtual
Networks twistie. In the example below, there is one virtual networks (internal), created by
the student #1:

__ 6. Click the Add Virtual Network button, to create your internal network. The Add Virtual
Network wizard is started.
__ a. In the first tab, Network Name, type the name of your internal network in the Virtual
network name field. Name the network S#-internal, where S# is your student number.
__ b. Click the Internal Network radio button
__ c. Type your student number in the VLAN ID field
__ d. Open the Advanced Settings twistie and observe the options. You will not use any of
these in this class, but notice that you choose the virtual switch, you can create a new
virtual switch, and you can make this network accessible on all VIOS partitions with a
single click, if desired.
__ e. Click Next
Here is an example for student #2:

© Copyright IBM Corp. 2009, 2019 5-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty

__ f. In the Summary tab, click Finish.

__ 7. You are returned to the Virtual Networks panel. Verify that your new internal network is now
in the table.
__ 8. Switch to the window where you opened the SSH session to the HMC, your HMC CLI
session.
__ 9. Use the lshwres command to display the list of virtual networks defined on your assigned
managed system. The example shows the results of student #2 adding the s2-internal
virtual network.
student2@hmc:~> lshwres -r virtualio -m bud --rsubtype vnetwork
vnetwork=s1-internal,is_tagged=0,vswitch=ETHERNET0,vlan_id=1
vnetwork=s2-internal,is_tagged=0,vswitch=ETHERNET0,vlan_id=2
__ 10. Using the lshwres command, list the virtual Ethernet adapters that are associated with the
VLAN you specified when creating the internal network. Expect the results to be that no
adapters are defined on your VLAN. 
The example below shows the results of student #2 after adding the S2-internal virtual
network. Adapters will be added in the next part of the exercise.
student2@hmc:~> lshwres -r virtualio -m bud --rsubtype eth --level lpar
--filter 'vlans="2"' -F lpar_name,port_vlan_id,slot_num,state --header
No results were found.

Part 2: Configure internal Virtual Ethernet Adapters on client

In this part of the exercise, you will associate two virtual Ethernet adapters with the new network,
one adapter in each of two partitions. These two virtual adapters will be configured for the same

© Copyright IBM Corp. 2009, 2019 5-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty
Virtual LAN ID. You will configure the virtual adapters and test that the partitions can ping each
other.
__ 11. Back in the HMC GUI session, click the Partitions link in the Menu Pod to display the list of
partitions on the managed system. Enter your student prefix (S1, S2, S3, or S4) in the filter
field to display only your assigned LPARs.

__ 12. Your LPAR1 and LPAR2 should be Running. If they are not, Activate them using their
current configuration. If you are unsure how to do this, contact the instructor.
__ 13. Add LPAR1 to the new internal network. Start the process by clicking on the name of your
LPAR1 in the table or gallery of LPARs that is being displayed. This will display the General
properties of LPAR1.
__ 14. In LPAR1’s Menu Pod, select the Virtual Networks task from the Virtual I/O section. The
example shows student #2 selecting Virtual Networks for their LPAR2.

© Copyright IBM Corp. 2009, 2019 5-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty

__ 15. In the Virtual Networks display, the Virtual Networks table will be empty, because you
have not created any virtual Ethernet adapters yet.
__ 16. Click the Manage Network Connections button to add a new connection for this LPAR.

__ 17. In the table that is displayed, choose the internal network that you just added. Click the
check box and then click OK.

© Copyright IBM Corp. 2009, 2019 5-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty

__ 18. Back in your HMC CLI session, using the lshwres command, list the virtual Ethernet
adapters that are associated with the VLAN you specified when creating the internal
network. This time you expect the results to show a virtual Ethernet adapter defined on your
VLAN. 
The example below shows the results of student #2 after adding the S2-internal virtual
network.
student2@hmc:~> lshwres -r virtualio -m bud --rsubtype eth --level lpar
--filter 'vlans="2"' -F lpar_name,port_vlan_id,slot_num,state --header
lpar_name,port_vlan_id,slot_num,state
S2_LPAR1,2,4,1
__ 19. Repeat steps from Step 13 on page 5-5 to Step 18 on page 5-7 substituting LPAR2 for
LPAR1 in each step.
__ 20. You added the internal virtual Ethernet adapter dynamically. What must you do to have AIX
recognize the new device?
__ a. Open an SSH session for both LPAR1 and LPAR2. Login as root using the password
provided by your instructor.
__ b. Run the cfgmgr command to discover the new virtual Ethernet adapters.
__ 21. Verify that there is one new virtual Ethernet adapter available in each partition (LPAR1 and
LPAR2). Note the adapter name in each LPAR. Use the -S a option to lsdev to see only
Available devices.

© Copyright IBM Corp. 2009, 2019 5-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty
Here are example commands and outputs. In this example, ent2 is the new virtual Ethernet
adapter.
# lsdev -c adapter -S a | grep ^en
ent1 Available 00-00 PCIe2 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)
# cfgmgr
# lsdev -c adapter -S a | grep ^en
ent1 Available 00-00 PCIe2 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)
ent2 Available Virtual I/O Ethernet Adapter (l-lan)
__ 22. Now you have two Ethernet adapters in each LPAR. Determine the adapter with the active
interface (IP address defined). In the example above, the IP address is defined on ent1, the
SR-IOV VF adapter. In the rest of the lab you configure the network interface on the new
internal virtual Ethernet adapter with an internal address on each LPAR and verify that they
can communicate with each other, but not externally. Perform the following steps for both
LPARs:
__ a. Use the ent device names displayed in the previous command if your device names are
different.
__ b. Run the ifconfig -a command and verify the SR-IOV interface that is configured with
the IP address, again in the example that is en1.
__ c. Run the smitty chinet command, choose the internal virtual Ethernet adapter (this is
en2 using the example output above), and fill in the appropriate information that was
provided by your instructor. The following information is likely to be true, however follow
your instructor’s information if it differs:
Internet Address = 10.0.0.1 (LPAR1) or 10.0.0.2 (LPAR2)
Netmask = 255.255.255.0
Current state = up
Use Address Resolution Protocol (ARP) = yes
Press ENTER to run the commands to set the IP address on en2.
__ d. Once the operation has completed, exit from SMIT and run ifconfig -a to check that
en2 is now configured.
__ 23. Now that both running AIX LPARs have internal virtual Ethernet adapters with an IP address
on the same VLAN, ping from one LPAR to another using the 10.0.0.x IP addresses. You
should find that LPAR1 and LPAR2 can ping successfully to each other.
Both LPARs are still able to ping external IP addresses (like the HMC) because there was
no change to the SR-IOV VF configuration that provides external access.

Part 3: View virtual Ethernet statistics

__ 24. Use netstat -i and entstat ent# to view configuration information and packet receive
and transmit statistics. In netstat -i, look for the Ipkts and Opkts fields. In the entstat
output, look for the Transmit and Receive Statistics near the top of the output.
Run these commands and compare the outputs before and after pinging the 10.0.0.x IP
addresses between LPAR1 and LPAR2.

© Copyright IBM Corp. 2009, 2019 5-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty
Helpful tips:
- You can use ping with the -c 10 flag to send 10 packets.
- When using entstat, you can pipe it to the head command to just see the top 10 lines
which will include the packet counts.
- You can reset the statistics in entstat with the -r flag so that it is easier to calculate
and compare the packet statistics. Example: entstat -r en2
- A consistent small amount of packet traffic on the network is normal even when there is
no user activity.

Part 4: View virtual Ethernet configuration in partition profiles

The commands executed in Part 2 that created the virtual Ethernet adapters updated the running
configuration of each partition, and also updated the last activated partition profile for each partition.
In this part of the exercise, you will review the partition profiles to see the new resources that were
added.
__ 25. Switch to your HMC Enhanced + GUI session.
__ 26. When you left the HMC GUI session, you were in the Virtual Network display for LPAR2. If
you are there, you can proceed. Otherwise, navigate to your LPAR2 partition’s Menu Pod.
__ 27. Open the Partition Actions section of the Menu Pod by clicking on it as shown in the
example above.
__ 28. Open the Profiles section. Then select the Manage Profiles task.

© Copyright IBM Corp. 2009, 2019 5-9

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty
__ 29. This opens a panel that is probably familiar if you have experience with the HMC prior to the
Enhanced+ interface. Open the only profile that is in the list of profiles for the partition by
clicking the name of the profile.

Important

In the steps that follow, only view the profile content, DO NOT change anything.
Also, when closing the windows, only use Close and Cancel, DO NOT use OK.
If you see this dialog box at any point, click No and then click Cancel, not OK:

If you click Yes on this panel, you will experience problems in the rest of the lab.

__ 30. This will open the profile for the partition. Click the Virtual Adapters tab.
__ 31. The list of virtual adapters is displayed. You should see an Ethernet adapter. This adapter
was added in Step 16 on page 5-6 and Step 17 on page 5-6 when a new network
connection was added to a virtual network.
__ 32. Click the adapter ID number (it is a link), to display the details of the virtual Ethernet adapter.
Note that the Port Virtual Ethernet (VLAN ID) field is the VLAN ID that is associated with
the virtual network that was chosen in Step 17 on page 5-6.

© Copyright IBM Corp. 2009, 2019 5-10

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty

__ 33. Click the View Virtual Network... option. This opens a panel that lists virtual switches and
VLANs. Choosing different combinations of these resources shows their relationships with
the partitions on the managed system.
Choose the ETHERNET0 virtual switch, and then select the VLAN ID you defined in the
virtual network above. In the details table you will see your two partitions on the internal
network VLAN. Notice too that the slot number for the virtual ethernet adapter is displayed
for each partition. 
The example below shows the selections made for student #2.

© Copyright IBM Corp. 2009, 2019 5-11

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 5. Virtual Ethernet Adapter Configuration

EXempty

__ 34. Change the View by: selection to Partitions. This table shows the relationships between all
the partitions and the defined VLANs. You might see bridged VLANs if another student has
completed the next exercise. If not, you will return to this view in the next exercise to see the
bridged VLANs.
__ 35. Close the Virtual Network Management window.
__ 36. Cancel out of the Virtual Adapter properties window.
__ 37. Cancel out of the Logical Partition Profile properties window.
__ 38. Close the Managed Profiles window.
__ 39. You can close all HMC and login windows and inform your instructor that you have
completed the lab exercise.

END OF LAB (with hints)

© Copyright IBM Corp. 2009, 2019 5-12

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty

Exercise 6. Dual VIOS Shared Ethernet

Adapter configurations
(with hints)

Overview
This exercise covers virtual Ethernet configurations when using a Dual Virtual I/O Server with a
Shared Ethernet adapter.

Objectives
At the end of the lab, you should be able to:
• Create and test a Shared Ethernet adapter
• Use the following VIOS CLI commands to view SEA statistics: seastat, topas, and entstat

Introduction
Students will configure a SEA failover configuration. Virtual I/O Server utilities for checking SEA
statistics will be explored.

Requirements
This guide and a server with two Virtual I/O Servers and one client partition. Both Virtual I/O Server
partitions must be connected to the same physical network so that they can participate in the Dual
VIOS SEA configuration.

© Copyright IBM Corp. 2009, 2019 6-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
Exercise instructions with hints

Preface
• Two versions of these instructions are available; one with hints and one without hints. You can
use either version to complete this exercise. Also, please don’t hesitate to ask the instructor if
you have questions. In some cases, the answer given may be just an example, and there may
be other correct answers.
• All exercises of this chapter depend on the availability of specific equipment in your classroom.
• All hints are marked by a » sign.

Important

Extremely important!
Parts 1, 2, 3, 5 and 6 of this lab exercise must be done as a team due to limited resources in the
lab.
In Part 4, you must work independently
There should be two students per team. The instructor will prepare you for this lab by putting you
into a session with your team members. Only one team member can execute the steps in this lab. It
is suggested that you share the time executing the steps and choose one student’s LPAR1 for
testing in Part 5. For example, one team member does Parts 1 and 3. The other team member does
Parts 2, 5 and 6.
If you are uncertain who your team members are, ask the instructor.

Part 1: Determine current IP address configuration

__ 1. Did you read the Important message above this Part of the lab? If not, read it now.
__ 2. Are you aware of your team members, and are you working as a team? If not, contact the
instructor for directions. Do not proceed in this exercise without coordination with your team
members.
__ 3. Log in to your assigned HMC Enhanced+ GUI. Navigate to the All Systems view. Click the
Display Relational View button to display all partitions on the host. All four (six when
working as a team) of your assigned partitions should be running from the last exercise. If
they are not running, activate VIOS1 and VIOS2, waiting for their login prompts, then
activate LPAR1 and LPAR2. Activate the partitions to their current configuration.
__ 4. Login to VIOS1 and VIOS2.

© Copyright IBM Corp. 2009, 2019 6-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 5. Display the Ethernet devices on both VIOS partitions. You should see two Ethernet devices
ent0 and ent1. They are two ports on a PCIe adapter.
$ lsdev | grep ^ent
ent0 Available PCIe2 2-Port 10GbE Base-T Adapter
(e4148e1614109204)
ent1 Available PCIe2 2-Port 10GbE Base-T Adapter
(e4148e1614109204)

__ 6. Check the interfaces and their IP addresses using lstcpip -stored. In the example below,
the IP address is on en0.
$ lstcpip -stored
T1_VIOS1
Network Interface Attributes

attributes: en0
IPv4 address = 10.8.XX.XX
Network Mask = 255.255.255.0
State = up

attributes: en1
State = down

attributes: et0
State = down

attributes: et1
State = down

Static Routes:
Route 1:
hopcount = 0
default gateway = 10.8.XX.254

Part 2: Create a Dual VIOS SEA configuration

Now we need to add the SEA using physical device where the IP address is configured. The virtual
Ethernet adapters that are required for an SEA are added as part of the process when using the
HMC Enhanced+ GUI.
__ 7. Did you read the Important message above Part 1 of the lab? If not, read it now. This Part of
the lab must be done as a team.
__ 8. Are you aware of your team members, and are you working as a team? If not, contact the
instructor for directions. Do not proceed in this exercise without coordination with your team
members.
__ 9. Log in to your assigned HMC Enhanced+ GUI. Navigate to the All Systems view.
__ 10. Click the name of your assigned managed system.

© Copyright IBM Corp. 2009, 2019 6-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 11. Open the Virtual Networks twistie and then click Add Virtual Network. This opens a
tabbed wizard. In the steps that follow, you will complete each tab to build the commands
necessary to create the SEA.

__ 12. In the Network Name tab, fill the Virtual network name box with vlan10 (for team 1) or
vlan20 (for team 2), ensure that the Bridged Network option is checked, and fill the VLAN
ID box with 10 (for team 1) or 20 (for team 2). No IEEE 802.1q Tagging will be done on this
bridge. Then click Next.

Note

All of the examples that follow are for team 2. Team 1’s screens would contain the team 1 values
and will look slightly different.

© Copyright IBM Corp. 2009, 2019 6-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty

__ 13. In the Network Bridge tab, change the Failover setting to Yes and leave everything else
default, because you are not creating a Load Sharing configuration. Click Next to proceed.

© Copyright IBM Corp. 2009, 2019 6-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 14. In the VIOS and Adapters tab, ensure your team’s VIOS1 is in the first row (Primary)
Virtual I/O Server box, and then choose the Ethernet device with the IP address assigned
that was discovered earlier as the Physical Adapter Location - Description using the
drop-down menu. Do not click Next yet.
__ 15. Still in the VIOS and Adapters tab, in the second row of the table, select your team’s VIOS2
in the Virtual I/O Server box, and then choose the Ethernet device with the IP address
assigned that was discovered earlier as the Physical Adapter Location - Description
using the drop-down menu. Click Next to proceed.
In the example below, Team 1 is choosing their VIOS1 and VIOS2 partitions. Your VIOS
partition names might not contain the system names as shown in the example.

__ 16. Review the settings on the Summary tab, then click Finish to create the SEA.

© Copyright IBM Corp. 2009, 2019 6-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty

__ 17. When the process finishes, the new SEA device will appear in the table. The internal
networks created earlier in the course, and the other team’s VLAN and SEA might already
be in the table.

__ 18. In VIOS1 list the virtual Ethernet adapters. In the example below, you see that ent2 and
ent3 were created. The ent2 device is the virtual Ethernet adapter for the SEA, and the
ent3 device is the SEA.
$ lsdev -virtual | grep ^ent
ent2 Available Virtual I/O Ethernet Adapter (l-lan)
ent3 Available Shared Ethernet Adapter

© Copyright IBM Corp. 2009, 2019 6-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 19. Check the newly created SEA. What devices are configured in the SEA? As expected, ent2
is the virtual Ethernet adapter, and ent0 is the physical adapter.
$ lsmap -all -net
SVEA Physloc
------ --------------------------------------------
ent2 U8284.22A.21C193V-V10-C7-T1

SEA ent3
Backing device ent0
Status Available
Physloc U78CB.001.WZS0638-P1-C9-T1
__ 20. Another way to check the adapters in the SEA is with entstat. The entstat command shows
statistics for all of the adapters in the SEA, including the individual ports in the link
aggregate.
$ entstat -all ent3 | grep ent[0-9]
ETHERNET STATISTICS (ent3) :
Statistics for adapters in the Shared Ethernet Adapter ent3
ent2: 20
Real Adapter: ent0
ETHERNET STATISTICS (ent0) :
Virtual Adapter: ent2
ETHERNET STATISTICS (ent2) :
Control Channel Adapter: ent2
ETHERNET STATISTICS (ent2) :
__ 21. Check the VLAN settings for the SEA. Verify that the Port VLAN ID is 10 or 20 depending
on which team you are on, and that there are no other VLAN IDs.
$ entstat -all ent3 | grep VLAN
VLAN Ids :
Enabled VLAN IDs: None
Invalid VLAN ID Packets: 0
Port VLAN ID: 20
VLAN Tag IDs: None
Management VLAN: Enabled
Invalid VLAN ID Packets: 0
Port VLAN ID: 20
VLAN Tag IDs: None
Management VLAN: Enabled

© Copyright IBM Corp. 2009, 2019 6-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 22. What happened to the IP address? The interface where the address was prior to the
creation of the SEA (the physical Ethernet adapter, en0 in the example below), is now
assigned to the SEA (en3 in the example below). 
Run the lstcpip -stored again to see where the VIOS IP address is. It should be on the
SEA device, because, again, the HMC Enhanced+ GUI moved the address.
$ lstcpip -stored
T1-VIOS1
Network Interface Attributes

attributes: en0
IPv4 address = 10.8.XX.XX
Network Mask = 255.255.255.0
State = detach

attributes: en1
State = down

attributes: et0
State = down

attributes: et1
State = down

attributes: en2
State = down

attributes: en3
IPv4 address = 10.8.XX.XX
Network Mask = 255.255.255.0
State = up

attributes: et2
State = down

attributes: et3
State = down

Static Routes:
Route 1:
hopcount = 0
default gateway = 10.8.12.254

DNS information:

Some error messages may contain invalid information

for the Virtual I/O Server environment.

cat: 0652-050 Cannot open /etc/resolv.conf.

© Copyright IBM Corp. 2009, 2019 6-9

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 23. Switch to your VIOS2 login session and repeat Step 18 on page 6-7 through Step 22 on
page 6-9 to verify that the SEA was created on both VIOS partitions.

Part 3: View and test the dual VIOS SEA configuration

__ 24. Did you read the Important message above Part 1 of the lab? If not, read it now. This Part of
the lab must be done as a team.
__ 25. Are you aware of your team members, and are you working as a team? If not, contact the
instructor for directions. Do not proceed in this exercise without coordination with your team
members.
__ 26. What about the control channel? There were no options for configuring one, but can you
see that it is configured? On VIOS2, check the control channel attribute of the SEA.
$ entstat -all ent3 | grep "Control Channel"
Control Channel PVID: 4095
Control Channel Adapter: ent2
Notice that a control channel VLAN was assigned, 4095. The control channel virtual
Ethernet adapter is ent2, the virtual Ethernet adapter in the SEA.
__ 27. On VIOS2, check the state of the SEA and the trunk priority setting for the virtual Ethernet
adapter in the SEA.
$ entstat -all ent3 | grep -E "Trunk|Priority"
Priority: 2
Trunk Adapter: True
Priority: 2 Active: False
Trunk Adapter: True
Priority: 2 Active: False
Notice that obviously this is a trunk adapter, and that the trunk priority was set to 2 (making
this the secondary SEA). Also note that the Active: False indicates that the primary SEA is
currently active.
__ 28. On VIOS1, check the same attributes of the original SEA and note their differences from the
backup.
$ entstat -all ent3 | grep "Control Channel"
Control Channel PVID: 4095
Control Channel Adapter: ent2
Notice that the primary VIOS sees the control channel on VLAN 4095 too. The control
channel virtual Ethernet adapter is ent2, again, the virtual Ethernet adapter in the SEA.

© Copyright IBM Corp. 2009, 2019 6-10

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 29. On VIOS1, check the state of the SEA and the trunk priority setting for the virtual Ethernet
adapter in the SEA.
$ entstat -all ent3 | grep -E "Trunk|Priority"
Priority: 1
Trunk Adapter: True
Priority: 1 Active: True
Trunk Adapter: True
Priority: 1 Active: True
The trunk priority has not changed, it was originally set to 1 (making this the primary SEA).
Also note that the Active: True indicates that the SEA on this VIOS is currently active.
__ 30. On VIOS1, cause a fallover of the SEA from VIOS1 to VIOS2 by changing the SEA from
primary to backup (standby).
$ chdev -dev ent3 -attr ha_mode=standby
ent3 changed
__ 31. On VIOS1, observe the change in the state of the SEA.
$ entstat -all ent3 | grep -E "Trunk|Priority"
Priority: 1
Trunk Adapter: True
Priority: 1 Active: False
Trunk Adapter: True
Priority: 1 Active: False
The trunk priority has not changed, it was originally set to 1 (making this the primary SEA).
Also note that the Active: False indicates that the SEA on this VIOS is currently the inactive
or backup.
__ 32. Likewise, on VIOS2, observe the change in the state of the SEA.
$ entstat -all ent3 | grep -E "Trunk|Priority"
Priority: 2
Trunk Adapter: True
Priority: 2 Active: True
Trunk Adapter: True
Priority: 2 Active: True
The trunk priority has not changed here either. But now the state of the SEA is Active: True
which indicates that the SEA on this VIOS is currently active.
__ 33. Return to the VIOS1 window and cause the SEA to return to VIOS1 from VIOS2 by
changing the SEA from backup to primary (auto).
$ chdev -dev ent3 -attr ha_mode=auto
ent3 changed

© Copyright IBM Corp. 2009, 2019 6-11

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 34. On VIOS1, observe the change in the state of the SEA.
$ entstat -all ent3 | grep -E "Trunk|Priority"
Priority: 1
Trunk Adapter: True
Priority: 1 Active: True
Trunk Adapter: True
Priority: 1 Active: True
Again, the trunk priority has not changed. It is still set to 1. The state of the SEA is now
Active: True which indicates that the SEA on this VIOS is currently active.
__ 35. Likewise, on VIOS2, observe the change in the state of the SEA.
$ entstat -all ent3 | grep -E "Trunk|Priority"
Priority: 2
Trunk Adapter: True
Priority: 2 Active: False
Trunk Adapter: True
Priority: 2 Active: False
Again, the trunk priority has not changed here either. The Active: False indicates that the
SEA on this VIOS is currently the inactive or backup.

Part 4: Configure a virtual Ethernet interface on the client in the bridged

network
In this part of the exercise, you will configure a virtual Ethernet adapter on LPAR1 in the same
VLAN as the virtual network of your Shared Ethernet adapter. You will then move the IP address of
LPAR1 to the bridged virtual Ethernet adapter and verify that it can access the external network.

Important

You do not perform this Part of the lab as a team. You must work independently for this Part of the
exercise.

__ 36. Back in the HMC Enhanced+ GUI session, click the Partitions link in the Menu Pod to
display the list of partitions on the managed system.

© Copyright IBM Corp. 2009, 2019 6-12

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty

__ 37. Start the process of adding LPAR1 to the new bridged network by clicking on the name of
your LPAR1 in the table or gallery of LPARs that is being displayed. This will display the
General properties of LPAR1.
__ 38. In LPAR1’s Menu Pod, select the Virtual Networks task from the Virtual I/O section.
__ 39. In the Virtual Networks display, the Virtual Networks table contains only internal the
virtual Ethernet adapter that you defined for earlier LPAR1.
__ 40. Click the Manage Network Connections button to add a new connection for this LPAR.

__ 41. In the table that is displayed, choose the bridged network that you added earlier. Notice that
it shows the VLAN ID of 10 or 20 that you defined. Click the check box next to the VLAN you
created (vlan10 for team 1, vlan20 for team 2) and then click OK.

© Copyright IBM Corp. 2009, 2019 6-13

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty

__ 42. You will be returned to the Virtual Networks window, where you will now see a virtual
Ethernet network connection that is associated with the bridged network that was just
defined in your VIOS1 and VIOS2 partitions.
__ 43. You added the bridged virtual Ethernet adapter dynamically. What must you do to have AIX
recognize the new device? Perform this step in the console window for LPAR1.
The best way to open a console to a partition is to SSH to the HMC command line and run
the vtmenu command. Choose the LPAR from the menu and log in. 
Note: To cancel out of the console connection and return to vtmenu, type a tilde followed by
a period (~.) and answer y to the Terminate session? question.
» Run the cfgmgr command to discover the new virtual Ethernet adapters.
__ 44. Verify that there is a new virtual Ethernet adapter available in LPAR1. This is the bridged
virtual Ethernet adapter that you just added. Note the adapter name. If you do not see the
new device, run the cfgmgr command and check again.

© Copyright IBM Corp. 2009, 2019 6-14

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
Here are example commands and outputs. In this example, ent3 is the new bridged virtual
Ethernet adapter. Your new bridged virtual Ethernet adapter might be different.
# lsdev -c adapter | grep ^en
ent0 Defined 00-00 PCIe3 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)
ent1 Available 01-00 PCIe3 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)
ent2 Available Virtual I/O Ethernet Adapter (l-lan)
# cfgmgr
# lsdev -c adapter | grep ^en
ent0 Defined 00-00 PCIe3 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)
ent1 Available 01-00 PCIe3 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)
ent2 Available Virtual I/O Ethernet Adapter (l-lan)
ent3 Available Virtual I/O Ethernet Adapter (l-lan)
__ 45. Verify that you have the correct virtual Ethernet adapter using the entstat command,
looking for the VLAN ID of the bridged virtual network.
In the example below for team 2, the bridged VLAN 20 is on the ent3 virtual Ethernet
adapter.
# entstat -d ent3 | grep -i vlan
Invalid VLAN ID Packets: 0
Port VLAN ID: 20
VLAN Tag IDs: None
__ 46. Now you have an Ethernet adapter on a bridged network in LPAR1. In the example above,
the adapter with the active interface (IP address defined) is ent1, the active SR-IOV VF
adapter. If your Ethernet adapter with the IP address defined is different, make note of that
and use it in the steps below instead of ent1. It might be ent0.
__ 47. Reconfigure the network interface so that the virtual Ethernet adapter has the active
interface. Because network connectivity is lost when you perform these steps, they must be
run from the console window of the partition. Perform the following steps:
__ a. Run the lsdev -c adapter command and verify that ent1 is the SR-IOV VF and ent3 is
the bridged virtual Ethernet adapter. Make a note if your device names are different.
__ b. Run the ifconfig -a command and verify that which interface has the IP address
defined (might be en0 or en1. Use the interface that has the IP address configured in
the following steps.
__ c. Run the ifconfig enX down detach command to unconfigure the enX (SR-IOV)
interface.
__ d. Run the smitty mktcpip command, choose enY (the bridged virtual Ethernet adapter),
and fill in the appropriate information that was provided by your instructor. Be sure to
check that you’re using the correct IP address that is assigned to the LPAR. Your
instructor will provide you with the necessary TCP/IP information.
Press ENTER to run the screen.

© Copyright IBM Corp. 2009, 2019 6-15

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ e. Once the operation has completed, exit from SMIT and run ifconfig -a to check that
en3 is now configured.
__ 48. Now that LPAR1 has a virtual Ethernet adapter on the bridged virtual network, ping your
assigned HMC IP address. You should find that LPAR1 can successfully ping the HMC.
This is because the LPAR1 is using the VIOS Shared Ethernet Adapter bridge to
communicate externally.
__ 49. Remove the SR-IOV adapter and interfaces from AIX. Do this with by issuing the three
rmdev commands as shown here:
# rmdev -dl etX
etX deleted
# rmdev -dl enX
enX deleted
# rmdev -dl entX
entX deleted

Part 5: View activity on the adapters

__ 50. Did you read the Important message above Part 1 of the lab? If not, read it now. This Part of
the lab must be done as a team.
__ 51. Are you aware of your team members, and are you working as a team? If not, contact the
instructor for directions. Do not proceed in this exercise without coordination with your team
members.
__ 52. Next, you’ll see the adapters are used in the SEA. First, reset all the statistics for entstat
for the SEA device so we can view any changes more easily. Use the following command
on the VIOS1 CLI. Change the ent device name to match your SEA if ent3 is not correct.
$ entstat -reset ent3
__ 53. Now, check the packet statistics for the SEA device and all of the adapters in use for the
SEA. The following commands and output assume the SEA device is ent3, the physical
interface is ent0, and ent2 is the virtual Ethernet adapter in the SEA configuration. Use
more to look through the entstat output and document the transmit and receive statistics
for each of the five devices. You should find that all of the transmit and receive statistics are
very low. (Some packet traffic is caused by switch and partitions negotiating configurations.)
There will be many pages of output for the entstat command and the statistics for each
device are listed sequentially. Look for the beginning of each of the Ethernet device statistic
sections for the Transmit and Receive Statistics (example shown below).
Hint: If you pipe the output to more, you can page through the output and use the search
capabilities of more (/string then n to go to a next occurrence and b to go to a previous
occurrence).

© Copyright IBM Corp. 2009, 2019 6-16

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
Command: $ entstat -all ent3 | more
Example output showing SEA transmit and receive statistics:
-------------------------------------------------------------
ETHERNET STATISTICS (ent5) :
Device Type: Shared Ethernet Adapter
Hardware Address: 00:14:5e:51:74:20
Elapsed Time: 0 days 0 hours 10 minutes 37 seconds

Transmit Statistics: Receive Statistics:

-------------------- -------------------
Packets: 64 Packets: 64
(lots of output removed)
Look for the Transmit and Receive packet statistics for each of the following devices and
make note of the values similar to the layout of table below. Given the above output, you’d
put 64 in both the Transmit and Receive columns for the ent3 device:

Table 7.
Adapter Transmit packets Receive packets
SEA ent3
Phys adapter ent0
VEth adapter ent2

__ 54. Now, use the following ftp command from your LPAR1 partition to a partition that is not on
your server to generate some traffic over the SEA. You will need to use a partition that is on
one of the lab systems in your class. Use the lab assignments visual to determine the IP
address of another LPAR1 in your class. The other students in your class may be still be in
the process of changing the SEA configuration on their system, so if a ping to one LPAR1
does not work, try another until you find one. If you cannot find one, contact your instructor
for help.
Here is the ftp command to use where ftp> is the prompt:
ftp> put "| dd if=/dev/zero bs=1M count=100" /dev/null

© Copyright IBM Corp. 2009, 2019 6-17

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
Here is an example ftp session. Use an appropriate IP address.
ftp 10.40.#.#
Connected to 10.40.#.#.
220 hurston101.beaverton.ibm.com FTP server (Version 4.2 Wed Oct 1
09:40:05 CDT 2008) ready.
Name (9.47.87.101:root): root
331 Password required for root.
Password:
230-Last login: Sun Dec 14 17:52:34 CST 2008 on ftp from ::ffff:9.27.27.18
230 User root logged in.
ftp> put "| dd if=/dev/zero bs=1M count=100" /dev/null
200 PORT command successful.
150 Opening data connection for /dev/zero.
100+0 records in.
100+0 records out.
226 Transfer complete.
104857600 bytes sent in 8.843 seconds (1.158e+04 Kbytes/s)
local: | dd if=/dev/zero bs=1M count=100 remote: /dev/null
ftp> bye
221 Goodbye.
__ 55. Back in the VIOS1 session, use entstat to view the transmit and receive packet statistics
for the SEA adapters. You’ll need to use entstat -all for the SEA device and page
through the output as you did before. Fill in the table with the statistics you find.
Example entstat command: $ entstat -all ent3 | more

Table 8.
Adapter Transmit packets Receive packets
SEA ent__
Phys adapter ent__
VEth adapter ent__

» Example entstat statistics which show that the ent0 physical adapter is the one that was used.

Table 9.
Adapter Transmit packets Receive packets
SEA ent3 77336 77165
Phys adapter ent0 72075 3199
VEth adapter ent2 5055 72228

© Copyright IBM Corp. 2009, 2019 6-18

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
Part 6: Using network analysis tools
This section explores network analysis tools specifically designed to show Shared Ethernet adapter
activity.
__ 56. Did you read the Important message above Part 1 of the lab? If not, read it now. This Part of
the lab must be done as a team.
__ 57. Are you aware of your team members, and are you working as a team? If not, contact the
instructor for directions. Do not proceed in this exercise without coordination with your team
members.
__ 58. In your VIOS1 partition, run the topas command. Notice the Network section in the upper
left quadrant of the screen. Notice that the line shown is for the total activity. Press the letter
n and see what happens. The Network section disappears. Press the letter n again. Now,
you should see a line for the configured interface on your partition and a line for the
loopback interface.
__ 59. Open a new SSH session to your VIOS1 partition. Login as padmin. Then, while watching
the topas output in one screen with the two interfaces shown in the Network section, in the
second screen, ping an external host such as your HMC system. You should see activity in
the en# line in topas.
__ 60. Next, while watching the topas screen, ping this partition’s own IP address. You should see
some activity in the loopback interface line. Stop the ping. Note: You may have to watch for
a few moments to see a pattern. A ping does not generate much load.
__ 61. Optional steps to see more activity. Watch the topas Network area while performing these
steps:
__ a. In LPAR1, run this command: cp /unix /tmp/afile
__ b. In VIOS1, invoke ftp and login to LPAR1 as root. Set your local (lcd) directory to /tmp.
Set your remote directory (cd) to /tmp. Use the get subcommand to 
“get afile” while watching topas. You should see a short burst of activity in the en#
device.
__ c. You can repeat the ftp on VIOS1 to its own IP address and see what happens. For
example, you can ftp to your VIOS1 IP address, then do a cd /tmp and a get afile
afile1 command.
__ 62. Stop any topas, ping, or ftp sessions.
__ 63. Run seastat -d ent# on VIOS1 where ent# is your SEA device. What happens? What
must you do to enable seastat? Perform the step to enable the seastat command to run.
» Example commands:
$ seastat -d ent3
Device ent3 has accounting disabled
$ chdev -dev ent3 -attr accounting=enabled
__ 64. Run seastat -d ent# where ent# is your SEA device. If you receive an error, wait a
moment and run it again. You should see output for at least your assigned VIOS. If neither
LPAR1 or LPAR2 shows in the output, ping one of them and run the seastat command
again. The SEA client should appear in the seastat report.

© Copyright IBM Corp. 2009, 2019 6-19

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 6. Dual VIOS Shared Ethernet Adapter configurations

EXempty
__ 65. Close all sessions to your LPARs and tell your instructor that you’ve completed the lab.

End of exercise

© Copyright IBM Corp. 2009, 2019 6-20

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

Exercise 7. Virtual Network Interface

Controllers (vNICs) and
vNIC Failover
(with hints)

Overview
This lab covers the configuration and testing of vNICs and vNIC Failover on clients in a Dual VIOS
configurations.

Objectives
At the end of the lab, you should be able to:
• Configure SR-IOV vNICs in a dual VIOS setup.

Introduction
This lab exercise has three parts:
In the first part, students configure an SR-IOV vNIC in a failover configuration between the two VIO
servers.
In the second part, students view the vNIC that was created in the HMC and VIOS LPARs.
In the third part, students test the vNIC configuration. Students also explore the vNIC Auto Priority
Failover setting, modifying it and then testing the behavior.

Requirements
This guide and a server with an SR-IOV adapter, two Virtual I/O Server partitions and two client
partitions.

© Copyright IBM Corp. 2009, 2019 7-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
Exercise instructions with hints

Preface
• Two versions of these instructions are available; one with hints and one without hints. You can
use either version to complete this exercise. Also, please don’t hesitate to ask the instructor if
you have questions. In some cases, the answer given may be just an example, and there may
be other correct answers.
• All exercises of this chapter depend on the availability of specific equipment in your classroom.
• All hints are marked by a » sign.

Part 1: Create an SR-IOV vNIC with failover

The SR-IOV vNIC with failover is to be created on LPAR2. You need the SR-IOV ports from your
LPAR1 and LPAR2 partitions. In the previous lab LPAR1’s ent0 (the SR-IOV port device) was
removed from AIX.
In the beginning of this part you remove LPAR1’s SR-IOV logical port in the HMC. Then you
remove the Ethernet device (ent0) and the SR-IOV logical port from LPAR2’s. Since SR-IOV logical
port is the interface that has LPAR2’s IP address, you must open a console to remove the IP
address and then remove the devices. LPAR2 must then be shutdown to add the vNIC because if it
is running without an IP address, it will not have an RMC connection, causing the task in the HMC
Enhanced+ GUI to fail (it attempts an DLPAR operation if the LPAR is running). With LPAR2
shutdown, the task updates the profile and current configuration. Then when LPAR2 is activated,
from the current configuration, the vNIC will be present.
First remove the SR-IOV port from LPAR1. The Ethernet resources were removed from AIX on
LPAR1 in the previous lab. If you have problems removing the SR-IOV logical port from LPAR1,
return to the part of the previous lab where the Ethernet resources are removed from AIX and
repeat those steps.
__ 1. Navigate to the Hardware Virtualized I/O window for your LPAR1 partition. You can get
there from the Partitions link in the Menu Pod of your assigned managed system. Then click
on LPAR1. The Hardware Virtualized I/O option is in the LPAR1’s Menu Pod.

CAUTION

Make sure you have chosen the correct LPAR1 from the list of LPARs. There are four
students on the HMC. If you choose the wrong LPAR1, you might interrupt the other
student’s lab. You do not want to do that :)

__ 2. Select the SR-IOV Logical Port from the table (it should be the only entry).

© Copyright IBM Corp. 2009, 2019 7-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

__ 3. Select the Remove Logical Port option from the Action menu.

__ 4. Click OK in the dialog box that is presented (Do you want to...).
__ 5. The SR-IOV Logical Ports table should now be empty.

© Copyright IBM Corp. 2009, 2019 7-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

Next, remove the SR-IOV port from LPAR2. Start by removing the Ethernet resources from AIX on
LPAR2.
__ 6. Open a console session on your LPAR2 partition. You will be changing the IP address from
one interface to another, so you must be on the console, not an SSH session.
As was shown earlier in the class, open an SSH session to your HMC, then use the vtmenu
command to open the console.
__ 7. Determine the adapter on which the IP address is defined using netstat. In the example
below, en1 interface has the IP address. It should be the active SR-IOV VF device.
# netstat -in
Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll
en1 1500 link#2 c6.2e.b8.bb.95.0 110 0 86
0 0
en1 1500 10.8.17 10.8.17.22 110 0 86
0 0
lo0 16896 link#1 114 0 114
0 0
lo0 16896 127 127.0.0.1 114 0 114
0 0
lo0 16896 ::1%1 114 0 114
0 0

__ 8. Using the interface that was determined in the previous step, run the ifconfig enX down
detach command to unconfigure the enX (SR-IOV) interface.
__ 9. Remove the SR-IOV adapter and interfaces from AIX. Do this with by issuing the three
rmdev commands as shown here:
# rmdev -dl etX
etX deleted
# rmdev -dl enX
enX deleted
# rmdev -dl entX
entX deleted

© Copyright IBM Corp. 2009, 2019 7-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 10. Shutdown your LPAR2 partition with shutdown -F
__ 11. When you see Halt completed... on the console, switch back to the HMC GUI session.
Now, remove the SR-IOV port from LPAR2. This is done in the HMC GUI session.
__ 12. Navigate to the Hardware Virtualized I/O window for your LPAR2 partition. The Hardware
Virtualized I/O option is in the LPAR2’s Menu Pod.

CAUTION

Make sure you have chosen the correct LPAR2 from the list of LPARs. There are four
students on the HMC. If you choose the wrong LPAR2, you might interrupt the other
student’s lab. You do not want to do that :)

__ 13. Select the SR-IOV Logical Port from the table (it should be the only entry).

__ 14. Select the Remove Logical Port option from the Action menu.

__ 15. Click OK in the dialog box that is presented (Do you want to...).
__ 16. The SR-IOV Logical Ports table should now be empty.

© Copyright IBM Corp. 2009, 2019 7-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

Finally, add the SR-IOV vNIC adapter to LPAR2 specifying both VIOS partitions to create a failover
configuration.
__ 17. Select the Virtual NICs task in the Menu Pod

__ 18. Click the Add Virtual NIC button. This will display the Add Virtual NIC - Dedicated window.

© Copyright IBM Corp. 2009, 2019 7-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 19. In the Add Virtual NIC - Dedicated window, do the following:
__ a. Select your VIOS1 in the Hosting Partition column
__ b. Choose the T3 entry in the dropdown list in the Physical Port Location Code column
__ c. Click the Add Entry button to add the connection through VIOS2
__ d. In the new row, select your VIOS2 in the Hosting Partition column,
__ e. And choose the T4 entry in the dropdown list in the Physical Port Location Code
column
__ f. Allow the Capacity and Failover Priority values to remain at their default values
__ g. Click OK
The following shows the correct values:

__ 20. The operation will take a minute or two.

__ 21. Activate your LPAR2 partition. Navigate to Partition Actions -> Operations -> Activate.
Then select the Activate (Normal) and Current Configuration options, and click Finish.
__ 22. Close the activation window.
__ 23. Monitor the console window. Login as root when the login prompt appears.
__ 24. Back in the HMC, click the Refresh button in the Virtual NIC window. Ignore the internal
error that is displayed. This is because LPAR2 does not have an RMC (or network
connection) to the HMC. A new error will be displayed every time to click Refresh until the
RMC connection is established.

© Copyright IBM Corp. 2009, 2019 7-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 25. Observe that the table now shows the vNIC, with the Device Name field showing
unavailable. The Backing Device State should show Operational for both VIOS
partitions. The Hosting Partition column should show both VIOS partition names.
Ignore any errors on the refresh.

__ 26. Return to the LPAR2 console session.

__ 27. Determine the adapter that is the vNIC. It might be ent0 or ent1.
__ a. Run the lsdev command to see the Ethernet adapters. In the example, the vNIC
adapter is ent1.
# lsdev -c adapter | grep ^ent
ent0 Defined 00-00 PCIe3 100/1000 Base-TX 4-port Converged Network
Adapter VF (df1028e214103c04)
ent1 Available Virtual NIC Client Adapter (vnic)
__ 28. Reconfigure the network so that the vNIC adapter has the IP address. Perform the following
steps:
__ a. Run the smitty mktcpip command, choose enX (the vNIC adapter), and fill in the
appropriate information that was provided by your instructor. Be sure to check that you
are using the correct IP address that is assigned to the LPAR. Your instructor will
provide you with the necessary TCP/IP information.
Press ENTER to run the screen.
__ b. Once the operation has completed, exit from SMIT and run ifconfig -a to check that
enX is now configured.
__ 29. Now that LPAR2 has been switched to a vNIC, ping your assigned HMC to test access to
external systems. You should find that LPAR2 can successfully ping the HMC. This is
because the LPAR2 is using the vNIC and the SR-IOV VF through one of the VIOS to
communicate externally.
__ 30. Exit the console session.
__ 31. Use the ~. key sequence to close the vterm through vtmenu. Type a y to terminate the
console session.
__ 32. Type a q to exit vtmenu.
__ 33. Open an SSH session to LPAR2 as a final test of connectivity.

© Copyright IBM Corp. 2009, 2019 7-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
Part 2: View vNIC configuration in VIOS, HMC, and partition profiles
__ 34. In the HMC GUI, in your LPAR2 Menu Pod (where you left off above), open the profile for
LPAR2 and observe the new vNIC virtual adapter. 
To do that, you can navigate to Partition Actions -> Profiles -> Manage Profiles. 
Open the only profile in the list, and click the Virtual Adapters tab.

Click the Adapter ID for the vNIC adapter to review the details of the adapter.

Don’t change anything in the profile.

__ 35. Close the windows associated with LPAR2’s profiles using the Close or Cancel buttons.

© Copyright IBM Corp. 2009, 2019 7-9

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 36. Return to the managed system’s Menu Pod by clicking the managed system’s name in the
breadcrumbs. Here is an example for the managed system named bud:

__ 37. Select the Virtual NICs task from the Menu Pod. Notice that your vNIC is in the list and the
details that are provided. There might be as many as eight entries in the table. If the other
students have completed the lab to this point, there will be eight entries. Use the Connected
Partition column to see the relationship of the rows to an individual client LPAR (and your
LPAR2 in this case).
The active SR-IOV device is displayed. This is useful. This view is informational only.

__ 38. Next you will look at the SR-IOV assignments

__ a. Now select Hardware Virtualized I/O from the managed systems Menu Pod.

© Copyright IBM Corp. 2009, 2019 7-10

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

__ b. Select the only SR-IOV adapter in the list

__ c. Select the Logical Ports view. 

Notice in the table that appears, that there are some ports with a check mark in the
Virtual NIC Backing Device column. 
Looking at the rows where the vNIC backing device has a check mark, find the rows
listing your LPAR2 as the Connected Partition. 
Notice that the Partition column shows the your VIOS partitions as owners. Expectedly,
there are two rows, one for each VIOS, per the failover configuration that was done
earlier.
You can choose the ports and click the Actions menu. You will see that there is only an
option to view the port. Again, this is an informational view only.

© Copyright IBM Corp. 2009, 2019 7-11

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

__ 39. Finally, look at the VIOS information in the HMC GUI. For both of your VIOS partitions, look
at the Hardware Virtualized I/O task, and open the Profile to see what is represented there.
You can follow the steps below to review the VIOS information.
Remember, always cancel or close the profile windows.
__ a. Click the Virtual I/O Servers link in the managed systems Menu Pod, then click the
name of your VIOS partitions, one at a time. (If you want to do this for just one of the
VIOS, that is OK.)

© Copyright IBM Corp. 2009, 2019 7-12

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

__ b. Click the Hardware Virtualized I/O task in the VIOS Menu Pod.

© Copyright IBM Corp. 2009, 2019 7-13

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

__ c. Find the SR-IOV logical port assigned to your LPAR2. This is your vNIC backing device
on this VIOS. Notice that the port is identified as such with a check mark in the Virtual
NIC Backing Device column. In both VIOS partitions, there is a vNIC backing device
because the vNIC was added with two entries, one for each VIOS. Notice that you can
not change the port that is a vNIC backing device. Changing the vNIC is done at the
LPAR level, not in the VIOS. The example output is from VIOS1.

__ d. Now open the profile for either of your VIOS partitions. 

To do that, you can navigate to VIOS Actions -> Profiles -> Manage Profiles. Then
open the only profile in the list.

© Copyright IBM Corp. 2009, 2019 7-14

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
- Click the Virtual Adapters tab. Notice no vNIC adapters. This is only on the LPAR.
- Click the SR-IOV Logical Ports tab. On both VIOS partitions the SR-IOV table is empty.
The vNIC SR-IOV port is not represented in the VIOS profiles.

- Cancel or Close all profile windows.

__ 40. Let’s move on to looking at the SR-IOV and vNIC resources from the HMC CLI. Open an
SSH session to your HMC.
__ 41. Start by looking at the SR-IOV port assignments. Look for your assigned partitions only, as
the output will contain SR-IOV information for the other students on the same managed
system. 
In the classroom lab, you should see two SR-IOV ports assigned, one to each of your VIOS
partitions. These ports are part of the vNIC failover configuration. They will have an
associated LPAR, your LPAR2. You might see other student’s LPAR2 vNIC resources as
well.
studentN@hmc:~> lshwres -r sriov -m <sys-name> --rsubtype logport --level \
eth -F lpar_name logical_port_id is_vnic_backing_device mode \
client_lpar_names --header
lpar_name,logical_port_id,is_vnic_backing_device,mode,client_lpar_names
S1_LPAR1 27004007 0 null
S1_LPAR2 27004008 0 null
T1_VIOS1 2700400a 1 ded S2_LPAR2
T1_VIOS2 2700400b 1 ded S2_LPAR2
The example output above shows the results after student 2 has completed the steps in this
lab. The last two lines of output in the example are for S2’s vNIC failover definition for
LPAR2, one SR-IOV port on VIOS1 and one on VIOS2. This is clear from the
is_vnic_backing_device=1 designation. Note the logical port IDs. They will be useful later
when looking at the vNIC adapters.
Here is the output after student 1has completed the lab to this point:
lpar_name logical_port_id is_vnic_backing_device mode client_lpar_names
T1_VIOS1 2700400a 1 ded S1_LPAR2
T1_VIOS2 2700400b 1 ded S1_LPAR2
T2_VIOS1 27004007 1 ded S2_LPAR2
T2_VIOS2 27004008 1 ded S2_LPAR2

© Copyright IBM Corp. 2009, 2019 7-15

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 42. Display the information about the newly defined vNIC for LPAR2 substituting your LPAR2
name for the string LPAR2 in the example:
studentN@hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnic --level
lpar -F lpar_name,backing_device_states --filter lpar_names="LPAR2"
LPAR2,"sriov/27004011/1/Operational,sriov/27004012/0/Operational"
Only one vNIC is defined for each student. It is associated with their LPAR2. It is in a vNIC
failover configuration, with both VIOS SR-IOV ports operational. The logical port ID that you
saw earlier is displayed in the output. From the information displayed earlier, you see that
the VIOS1 SR-IOV port is active in the vNIC because of the number one in the state output
(sriov/27004011/1/Operational).
__ 43. Look closer at the vNIC backing devices to see their state and VIOS association (there
might be output for four pairs of VIOS if the other students have done the lab steps to this
point):
studentN@hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnicbkdev -F
lpar_name,logical_port_id,status,is_active --header
lpar_name,logical_port_id,status,is_active
VIOS1,27004011,Operational,1
VIOS2,27004012,Operational,0
Again, you see that VIOS1 is the active SR-IOV port from the last field of the output. You
must use the output from the two previous steps to map the logical port ID for your LPAR2’s
vNIC backing devices to the two VIOS partitions. In the classroom lab environment, you
might see as many as 8 lines of output for the above command.
__ 44. Now that you know an SR-IOV logical port is assigned to each of your VIOS partitions, open
an SSH session on VIOS1 to look at the resources that were defined there. Run the
following commands and observe the output.
__ a. List all the vNIC mappings on this VIOS and make note of the vnicserverX device
associated with your LPAR2 in the ClntName field: lsmap -all -vnic
Name Physloc ClntID ClntName ClntOS
------------ ------------------------------- ------ ------------- -------
vnicserver0 U8286.41A.21E096V-V1-C32897 4 LPAR2 AIX

Backing device:ent6
Status:Available
Physloc:U78CB.001.WZS06PK-P1-C10-T3-S10
Client device name:ent0
Client device physloc:U8284.22A.21C18FV-V12-C4
This command displays most of what you want to know about the VIOS vNIC resource
relationships. To understand more details, you can use lsdev -dev <dev> -attr, for
each of the devices in the vNIC relationship.
__ b. Check the adapter list for vNIC resources: lsdev -type adapter | grep -E
'VF|vnic'
ent4 Available PCIe3 4-Port 10GbE CU Adapter VF(df1028e21410e404)
vnicserver0 Available Virtual NIC Server Device (vnicserver)

© Copyright IBM Corp. 2009, 2019 7-16

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ c. Check the attributes of the vNIC server resource that you discovered in the earlier step:
lsdev -dev vnicserverX -attr
attribute value description
user_settable

eth_dev_loc U78CB.001.WZS0638-P1-C10-T3-S8 Backing device location code False

eth_dev_name ent4 Backing device name False
You see that the backing device is ent4.

Part 3: Test vNIC failover

In the previous part, you determined that the active port in LPAR2’s vNIC is on VIOS1. Both VIOS
have the same failover priority which makes it necessary to manually move the active port back to
the original VIOS after a failover. In this part, you change the active port to VIOS2 using the HMC
GUI. Still in the HMC GUI, you then change the failover priority so that VIOS1 and VIOS2 have
different priorities, with VIOS2 being the higher priority. You use both the HMC GUI and the HMC
CLI to cause failovers.
In an optional part of the exercise (that involves working with the other student on your team) you
can test a shutdown of VIOS2 to see that the automatic failover occurs. Upon reactivation of
VIOS2, you see that the vNIC was switched back to VIOS2.
__ 45. Ensure that you are logged into the HMC Enhanced+ GUI.
__ 46. You should have SSH sessions to LPAR2, VIOS1, and the HMC open from earlier steps in
the lab. If not, open any needed SSH sessions using the userid and password supplied by
the instructor.
__ 47. On LPAR2, start a ping to the HMC IP address.
__ 48. In the HMC GUI, navigate to LPAR2’s Virtual NIC view.
__ 49. Select the non-active backing device (it is associated with VIOS2)
__ 50. In the Action menu, choose the Make the Backing Device Active option.

© Copyright IBM Corp. 2009, 2019 7-17

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 51. Click OK to initiate the failover operation.
If possible, keep the LPAR2 session where the ping is active in view. If necessary, switch
back as soon as the HMC GUI task to switch to the backup VIOS completes.

__ 52. In the LPAR2 session, you should observe a loss of 3-8 packets.
__ 53. In the LPAR2 session, stop the ping and check the error log for the entries related to the
failover. The entries should show that the vNIC went down and then came back up.
# errpt | more
4FB9389C 0405123418 I S ent0 VNIC Link Up
F655DA07 0405123418 I S ent0 VNIC Link Down
3DEA4C5F 0405123418 T S ent0 VNIC Error CRQ
Open a detailed view of the VNIC Error CRQ entries, to see that the event that was received
was a PARTNER_FAILOVER.
__ 54. In the HMC GUI, click the Refresh button, to observe that VIOS2 is now the Active.

Questions

Did you notice that the vNIC Auto Priority Failover feature was disabled because you performed a
manual failover?
It indicated that in the confirmation message for the failover. If the VIOS partitions have different
priorities, this might not be a good state to stay in. This manual fallover operation disabled
auto-fallback when the higher priority VIOS comes back online.
You can check the vNIC Auto Priority Failover feature in the HMC GUI by running this command,
where 0 indicates disabled, and 1 indicates enabled (looking for your LPAR2 line in the output):

© Copyright IBM Corp. 2009, 2019 7-18

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
studentN@-hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnic --level lpar -F
lpar_name,auto_priority_failover
LPAR2,0

__ 55. In the HMC GUI, re-enable vNIC Auto Priority Failover for the vNIC. Select the active row
of the vNIC, and then choose Modify vNIC from the Action menu.

__ 56. Ensure that the vNIC Auto Priority Failover feature is Enabled, as shown below. It might
be marked as Enabled even though it is not. Click OK to perform the task.

Now let’s experiment with the auto priority failover feature.

__ 57. Change the failover priority for VIOS2 to 1, and for VIOS1 to 2. This sets up for an automatic
fallback for VIOS2 when it becomes available if it is not the active port.
In the HMC GUI, again, choose the first row of the vNIC, then in the Action menu, choose
the Modify Backing Devices task.

© Copyright IBM Corp. 2009, 2019 7-19

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

Change the Failover Priority for VIOS2 to 1, and for VIOS1 to 2.

Notice, you could have enabled the auto priority failover here too.
__ 58. In the LPAR2 session, check the error log again. Notice any new activity? There might be.
In some versions of the HMC, the change in failover priority was disruptive. That’s good to
know.
Now use the HMC CLI to perform the same testing steps.

© Copyright IBM Corp. 2009, 2019 7-20

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 59. To perform the manual failover from the HMC CLI, you need to know the slot number of the
vNIC adapter in LPAR2, and the logical_port_id of the SR-IOV port in the VIOS that you
intend to make active (VIOS1 in our case). The steps that follow determine those values
and then issue the command to failover the vNIC from VIOS2 to VIOS1. Type all commands
as continuous strings, without the “\” character.
__ a. To find the slot number and backing device logical port IDs of the vNIC adapter in
LPAR2: (replace <sys-name> with your assigned managed system name, and
<LPAR2-name> with your assigned LPAR2’s name)
studentN@hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnic \
--level lpar -F slot_num,backing_devices --filter lpar_names=LPAR2
5,"sriov/VIOS1/1/1/2/27004011/2.0/2.0/2,sriov/VIOS2/2/1/3/27004012/2.0/2.
0/1"
In the lab, only one slot will be displayed (slot 5 in the example above). From the
backing devices output for slot 5, note the two logical port IDs which are the fifth field
in the sriov entries. In the example, the vNIC backing device on VIOS1 has a logical
port ID of 27004011, and the backing device on VIOS2 has a logical port ID of
27004012. Make note of these values as they will be used in a future step.
__ b. To find the nonactive vNIC backing devices, use the following command. Make note of
the logical port ID that is associated with the row that has a value of 0 in the is_active
column.
studentN@hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnicbkdev \
-F lpar_name logical_port_id is_active --filter \
"\"lpar_names=T#_VIOS1,T#_VIOS2"\" --header
lpar_name logical_port_id is_active
T#_VIOS1 27004011 0
T#_VIOS2 27004012 1
__ c. From the previous two commands, you should conclude that the slot number is 5, and
the logical_port_id of the nonactive SR-IOV port is 27004011. Your values will be
different values, so use your values in the command on the next step.
__ d. Restart the ping in the LPAR2 session to monitor the failover.
__ e. Therefore, the command to failover from VIOS2 to VIOS1 is:
studentN@hmc:~> chhwres -m <sys-name> -r virtualio -o act --rsubtype \
vnicbkdev -p LPAR2 -s 5 --logport 27004011
__ f. As soon as you hit enter on the previous step, the ping should pause on LPAR2, and
then resume.
__ g. Check the status of the vNIC backing devices on the HMC. VIOS1 should now be active
(is_active = 1).
studentN@hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnicbkdev \
-F lpar_name logical_port_id status is_active --filter \
“\”lpar_names=T#_VIOS1,T#_VIOS2”\” --header
lpar_name logical_port_id status is_active
T#_VIOS1 27004011 Operational 1
T#_VIOS2 27004012 Operational 0

© Copyright IBM Corp. 2009, 2019 7-21

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 60. What happened to the auto priority failover setting for the vNIC? You should expect that it
got disabled. Check it on the HMC, recalling that 0=disabled (find your LPAR2 in the output
if there is more than one line returned):
studentN@hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnic --level \
lpar -F lpar_name,slot_num,auto_priority_failover --filter \
lpar_names=LPAR2
LPAR2,5,0
__ 61. Change the auto priority failover setting for the vNIC back to enabled:
studentN@hmc:~> chhwres -o s -r virtualio -m <sys-name> --rsubtype vnic -p \
LPAR2 -s 5 -a auto_priority_failover=1
__ 62. Check the ping that is active in the LPAR2 window, noticing again that there was an
interruption on the vNIC device.
__ 63. Check the auto priority failover setting again. It should now be enabled, recalling that
1=enabled:
studentN@hmc:~> lshwres -r virtualio -m <sys-name> --rsubtype vnic --level \
lpar -F lpar_name,slot_num,auto_priority_failover --filter \
lpar_names=LPAR2
LPAR2,5,1

Part 4: (Optional) Test vNIC Failover (VIOS failure)

Important

Extremely important!
This part of the lab exercise must be done as a team because you are stopping a shared VIOS
partition.
There should be two students per team. As you did earlier, you must coordinate with your other
team members before doing this part of the exercise. Only one team member can execute the
steps in this lab.
If you are uncertain who your team members are, ask the instructor.

__ 1. Did you read the Important message above? If not, read it now.
__ 2. Are you aware of your team members, and are you working as a team? If not, contact the
instructor for directions. Do not proceed in this exercise without coordination with your team
members.
__ 3. Choose one of the team’s LPAR2 sessions to work in.
__ 4. Ensure that you have a login session to VIOS2.
__ 5. In the LPAR2 session, start a ping to the HMC.

© Copyright IBM Corp. 2009, 2019 7-22

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
__ 6. In the HMC CLI session, run a more informative loop to see the status of the vNIC backing
devices as they change when you shutdown VIOS2. The lshwres command should be
typed as a single command all the way through the --header flag. Substitute the following
values in the example below:
T# = your assigned team number
T#_VIOS1 = your team’s VIOS1 LPAR name
T#_VIOS2 = your team’s VIOS2 LPAR name
while [ TRUE ]
do
lshwres -r virtualio -m <sys-name> --rsubtype vnicbkdev -F lpar_name status
is_active --filter "\"lpar_names=T#_VIOS1,T#_VIOS2"\" --header
echo " "
sleep 3
done
The output of the above command should look like this:
lpar_name logical_port_id status is_active
VIOS1 27004011 Operational 0
VIOS2 27004012 Operational 1

lpar_name logical_port_id status is_active

VIOS1 27004011 Operational 0
VIOS2 27004012 Operational 1
... ... ...
__ 7. In the VIOS2 session, type: shutdown -force
__ 8. Immediately begin monitoring the ping in the LPAR2 window, and the loop in the HMC
session. When the ping pauses in the VIOS2 window, pay close attention to the HMC
window, as the status of VIOS2 will change.
You should see a break in the pings in the LPAR2 window:
64 bytes from 10.40.226.240: icmp_seq=322 ttl=64 time=0 ms
64 bytes from 10.40.226.240: icmp_seq=323 ttl=64 time=0 ms
64 bytes from 10.40.226.240: icmp_seq=330 ttl=64 time=0 ms
64 bytes from 10.40.226.240: icmp_seq=331 ttl=64 time=0 ms
64 bytes from 10.40.226.240: icmp_seq=332 ttl=64 time=0 ms

© Copyright IBM Corp. 2009, 2019 7-23

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty
Along with an indication in the HMC CLI session of a status change for VIOS2 from
Operational to “Powered Off” and VIOS1 becoming the active port:
lpar_name status is_active
VIOS1 Operational 0
VIOS2 Operational 1 <- VIOS2 is active

lpar_name status is_active

VIOS1 Operational 1 <- VIOS1 is now active
VIOS2 "Powered Off" 0

lpar_name status is_active

VIOS1 Operational 1
VIOS2 "Powered Off" 0
__ 9. In the LPAR2 session, restart the ping to the HMC if you stopped it, and in the HMC CLI
session, restart the loop, if you stopped it.
__ 10. In the HMC GUI, reactivate VIOS2 to current configuration.
__ 11. Immediately begin monitoring the ping in the LPAR2 window, and the loop in the HMC
session. When the ping pauses in the LPAR2 window, pay close attention to the HMC
window, as the status of VIOS2 will change.
You should see a break in the pings in the LPAR2 window:
64 bytes from 10.40.226.240: icmp_seq=217 ttl=64 time=0 ms
64 bytes from 10.40.226.240: icmp_seq=218 ttl=64 time=0 ms
64 bytes from 10.40.226.240: icmp_seq=222 ttl=64 time=0 ms
64 bytes from 10.40.226.240: icmp_seq=223 ttl=64 time=0 ms

But this time, what happened to the active port in the HMC CLI?
As before, there is an indication of a status change for VIOS2 from “Powered Off” to
Operational as expected, but VIOS2 becomes the active port:
lpar_name status is_active
VIOS1 Operational 1 <- VIOS1 is active
VIOS2 "Powered Off" 0

lpar_name status is_active

VIOS1 Operational 0
VIOS2 Operational 1 <- VIOS2 is now active
__ 12. In the monitor loop, which VIOS is now active? Surprised? VIOS2 is now active. It is not all
that surprising, because earlier in the exercise you changed the priorities of the VIOS
backing devices (and enabled automatic fallback) which will cause automatic fallback to the
highest priority VIOS.
So, make note...this is the second time a change to the state of the vNIC could cause an
interruption in traffic on the vNIC. The first time was when you changed the priorities of the
backing devices. You must be careful and do the proper planning before making changes.

© Copyright IBM Corp. 2009, 2019 7-24

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

Questions

There was an outage for the vNIC on the client when VIOS2 came back online. Is that acceptable?

__ 13. That is the end of this lab exercise. Notify your instructor that you are finished.

End of exercise

© Copyright IBM Corp. 2009, 2019 7-25

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

© Copyright IBM Corp. 2009, 2019 7-26

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

© Copyright IBM Corp. 2009, 2019 7-27

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

© Copyright IBM Corp. 2009, 2019 7-28

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

© Copyright IBM Corp. 2009, 2019 7-29

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 7. Virtual Network Interface Controllers (vNICs) and vNIC Failover

EXempty

© Copyright IBM Corp. 2009, 2019 7-30

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty

Exercise 8. PowerVM system

maintenance
(with hints)

Estimated time
01:00

Overview
This exercise provides you with an opportunity to perform a couple of the most common PowerVM
maintenance tasks.

Objectives
At the end of this exercise, you should be able to:
• Update the Virtual I/O Server software to a newer fix level
• Backup the device configuration of the Virtual I/O Server using the viosbr command. (The
viosbr command includes device attributes, as well as logical and virtual device mapping
information.)
• Use cron to schedule the Virtual I/O Server backup

Introduction
The Virtual I/O Server software continues to change. Some of the updates fix problems and others
add new functionality. The latest available level of software is always considered the recommended
level. This software can be acquired by anonymous FTP, HTTP (web browser), and ISO images.
This exercise uses files which were previously downloaded using FTP.
The last part of this exercise demonstrates how cron can be used to schedule a backup of the
Virtual I/O Server.

Requirements
• This workbook.
• A student workstation equipped with a web browser and a utility for running SSH.
• Access to a POWER8 processor-based managed system and a Hardware Management
Console running HMC V8 configured for this course.

© Copyright IBM Corp. 2009, 2019 8-1

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty
Exercise instructions with hints

Preface
• All procedures in this exercise depend on the availability of specific equipment in your
classroom.
• All hints are marked by a » sign.

Part 1: Update the Virtual I/O Server software

At the time of writing, the latest fix pack for VIOS 2.2.6 was 2.2.6.41. This fix is available from an
NFS mount in your VIOS. Your instructor might direct you to install a newer fix in a different location.
Your system has been configured with a combination of availability options including MPIO for disk
redundancy and shared Ethernet adapter failover for redundancy in bridging virtual Ethernet traffic.
You will update each Virtual I/O Server; one at a time to avoid impacting your client LPARs.

Note

You need to plan the VIO server updates as a team of two students. Refer to the table below for
your team number. In each team, the student with the odd number ID should complete part 1 of the
exercise first. The student with the even number ID should wait until the student with the odd
number ID indicates they have completed the update of their VIO server. In this way, the AIX client
partitions will have access to the network and their VSCSI disks at all times.

Table 10. Team numbers

Team number 1 1 2 2
Student number 1 2 3 4
Assigned Virtual I/O Server name T1_VIOS1 T1_VIOS2 T2_VIOS1 T2_VIOS2
Client LPAR name lpar1 lpar1 lpar1 lpar1

__ 1. Log in to your assigned Virtual I/O Server and perform the following steps:
__ a. Determine the name of the SEA configured in the partition using the following command:
$ lsdev -virtual | grep Shared
__ b. Determine if the SEA is currently acting as the primary SEA in a failover configuration
using the following command, replacing SEAdev with the device name obtained in the
previous step:
$ entstat -all SEAdev | grep "^ *State"

© Copyright IBM Corp. 2009, 2019 8-2

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty
__ c. If the SEA is currently acting as the primary device, the previous command output will be
PRIMARY. If the SEA is primary, change the ha_mode attribute to standby using the
command shown below, replacing SEAdev with the device name obtained in a previous
step. If the SEA is currently acting as the backup device, then no action is required.
$ chdev -dev SEAdev -attr ha_mode=standby
__ d. Check your current Virtual I/O Server level. The output should say 2.2.6.21.
$ ioslevel
__ e. All interim fixes currently applied must be manually removed before applying the fix
pack. Use the emgr command to list any installed interim fixes.
$ oem_setup_env
# emgr -l
__ f. If the output from the emgr command indicates there are interim fixes installed on your
system (it should not), remove each one individually using the command shown below.
Do not reboot the VIOS partition, even if the command output indicates this should be
done.
# emgr -r -L <fix label name>
__ g. When all interim fixes have been removed, return to the VIO CLI environment.
# exit
__ h. All uncommitted updates must be committed prior to installing new updates. Execute the
following command:
$ updateios -commit
The fixes for Virtual I/O Server 2.2.6.41 have been downloaded and are available in an
NFS directory. Mount the products NFS directory and verify that the fixes are available in
/mnt/VIOS/VIOS_2.2.6.41.
$ mount 10.8.252.1:/export/products /mnt
$ more /mnt/VIOS/VIOS_2.2.6.41/VIOS_SP_2.2.6.41.dd.xml
Search for the string “Description”. You will see that this is the service pack for VIOS
2.2.6.41.
__ i. To verify the VIOS update files before installation, follow the steps below. Your instructor
may tell you to use a different path name for the directory that contains the fixes you are
using.
$ cp /mnt/VIOS/VIOS_2.2.6.41/ck_sum.bff /home/padmin
$ chmod 775 ck_sum.bff
$ ck_sum.bff /mnt/VIOS/VIOS_2.2.6.41
__ j. Execute the following command to install the updated Virtual I/O Server software:
$ updateios -dev /mnt/VIOS/VIOS_2.2.6.41 -install -accept
Once the command examines the software contained in the specified directory, you will
be prompted to confirm that the installation should proceed. Type a y to proceed.
Continue bos.rte.install installation [y|n]?

© Copyright IBM Corp. 2009, 2019 8-3

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty
y

The initial confirmation will result in the installation of the update to the bos.rte.install
fileset. When this has been completed, the command will once again examine the
contents of the directory and prompt you to confirm that the installation should proceed.
Again, type a y to proceed. The software installation could take 15 minutes or more to
complete.
__ k. If your student ID is an odd number (1, 3, and so on), then reboot the Virtual I/O Server
to implement the changes. If your student ID is an even number (2, 4, and so on), then
do not reboot the Virtual I/O Server until your teammate has confirmed that they have
finished the update of their Virtual I/O Server. In this way, the AIX client partitions will still
have access to disk and network services. If both Virtual I/O Server partitions are
rebooted at the same time, then the AIX client partitions will experience severe issues,
since they will no longer have disk or network access.
» $ shutdown -force -restart
__ l. When the partition has rebooted, log on as padmin and run ioslevel to list the new
software level.
» Example command and output after installing VIOS 2.2.6.41 is shown below:
$ ioslevel
2.2.6.41
__ m. If you changed the ha_mode attribute of the SEA device to standby in a previous step,
then change it back to auto using the following command, replacing SEAdev with the
device name:
$ chdev -dev SEAdev -attr ha_mode=auto
__ 2. If your student ID is an odd number, inform your teammate that you have completed the
update of your VIO partition. Your teammate can now perform the update of their VIO
partition.

Part 2: Using viosbr to backup the virtual and logical configuration

This part of the exercise will use the viosbr command to backup the virtual and logical
configuration of the Virtual I/O Server. This viosbr command backs up all the relevant data to
recover a Virtual I/O Server after a new installation.
__ 3. Use the viosbr command to backup all the device attributes, as well as logical and virtual
device mappings on your Virtual I/O Server to a backup file called myvios-backup.
» Example command:
$ viosbr -backup -file myvios-backup
__ 4. Use the viosbr command to list the available backup files.
» Example command and expected output:
$ viosbr -view -list
myvios-backup.tar.gz
» Note that the backup file is compressed.

© Copyright IBM Corp. 2009, 2019 8-4

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty
__ 5. Use the viosbr command to display the contents of the backup file.

© Copyright IBM Corp. 2009, 2019 8-5

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty
» Example command and output:
$ viosbr -view -file myvios-backup.tar.gz
Details in: myvios-backup
===============================================================

Controllers:
============
Name Phys Loc
---- --------
iscsi0
pkcs11
fcs0 U78CB.001.WZS01VH-P1-C6-T1
fcs1 U78CB.001.WZS01VH-P1-C6-T2
sfwcomm0 U78CB.001.WZS01VH-P1-C6-T1-W0-L0
sfwcomm1 U78CB.001.WZS01VH-P1-C6-T2-W0-L0
fscsi0 U78CB.001.WZS01VH-P1-C6-T1
ent0 U78CB.001.WZS01VH-P1-C11-T1
fscsi1 U78CB.001.WZS01VH-P1-C6-T2
ent1 U78CB.001.WZS01VH-P1-C11-T2
ent2 U8284.22A.214834V-V1-C2-T1
sfw0
fcnet0 U78CB.001.WZS01VH-P1-C6-T1
fcnet1 U78CB.001.WZS01VH-P1-C6-T2

Physical Volumes:
=================
Name Phys Loc
---- --------
hdisk0 U78CB.001.WZS01VH-P1-C6-T1-W500507680130B855-L0
hdisk1 U78CB.001.WZS01VH-P1-C6-T1-W500507680130B855-L1000000000000
hdisk2 U78CB.001.WZS01VH-P1-C6-T1-W500507680130B855-L2000000000000
hdisk3 U78CB.001.WZS01VH-P1-C6-T1-W500507680130B855-L3000000000000
hdisk4 U78CB.001.WZS01VH-P1-C6-T1-W500507680130B855-L4000000000000

Optical Devices:
================
Tape Devices:
=============
Ethernet Interfaces:
====================
Name
----
en0
en1
en2
en3
IP Address Hostname Auto
---------- -------- ----
6.109.201 sys8981_vios1 -

Storage Pools:
==============
SP Name PV Name
------- -------
rootvg hdisk0

© Copyright IBM Corp. 2009, 2019 8-6

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty
File Backed Storage Pools:
==========================
Optical Repository:
===================
Ether Channel Adapters:
=======================
Shared Ethernet Adapters:
=========================
Name Physical Adapter Default Adapter Virtual Adapters
---- ---------------- --------------- ----------------
ent3 ent0 ent2 ent2

Virtual Server Adapters:

========================
SVSA Phys Loc VTD
---- -------- ---
vhost1 U8284.22A.214834V-V1-C3 lpar1_MPIO

vhost2 U8284.22A.214834V-V1-C4 lpar2_MPIO

Virtual Suspended Adapters:

===========================
Virtual Suspended Devices:
========================
Suspend Device Phys Loc
-------------- --------
suspended0

LPM Adapters:
=============
LPM Adapter Phys Loc
----------- --------
vioslpm0

USB Mass Storage Devices:

=========================
SVFC Adapters:
==============
Name FC Adapter Phys Loc
---- ---------- --------
vfchost0 fcs0 U8284.22A.214834V-V1-C21

VBSD Pools:
===========
VRM(AMS) Pages:
===============
RSCT:
=====
ct_node_id
-----------
2700171055b643f4

Cluster:
========

© Copyright IBM Corp. 2009, 2019 8-7

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3
Exercise 8. PowerVM system maintenance

EXempty
Cluster State
------- -----

Virtual Log Repositories:

=========================
Virtual Log Repository State
---------------------- -----
vlogrepo0 AVAILABLE

Part 3: Using cron to schedule tasks

In this part of the exercise you will use cron to schedule tasks. In the following steps, you use cron
to schedule a backup of the Virtual I/O Server.
__ 6. Use the crontab command to schedule the creation of a mksysb backup of your Virtual I/O
Server into a file called vios_backup located in the /home/padmin directory at 11am.
» Use the crontab -e command to edit the crontab file for the padmin user. Add the following on
one line to the file, and then save the file and quit the editor:
0 11 * * * /usr/ios/cli/ioscli backupios -file ~/vios_backup -mksysb
__ 7. Use the crontab command to view the crontab file of the padmin user.
» Use the crontab -l command to view the crontab file for the padmin user. The output should
reflect the entry you created in the previous step.
__ 8. Let your instructor know when you have completed the exercise.

End of exercise

© Copyright IBM Corp. 2009, 2019 8-8

Course materials may not be reproduced in whole or in part without the prior written permission of IBM.
V11.3

backpg

© Copyright International Business Machines Corporation 2009, 2019.