ibm wave redbook

Draft Document for Review February 26, 2014 8:59 am SG24-8192-00

ibm.com/redbooks

Front cover

IBM Wave for z/VM - Installation, Implementation and Exploitation

Lydia ParzialeSusan Greenlee

Ernest HornGene Ong

Richard LewisSean McBride

Steve McGarril

Simplify Linux on System z and z/VM administration tasks

Install,customize and use IBM Wave with detailed step-by-step instructions.

Undertand the business benefits of IBM Wave for z/VM

http://www.redbooks.ibm.com/

International Technical Support Organization

IBM Wave - Installation, Implementation and Exploitation

September 2013

Draft Document for Review February 26, 2014 8:58 am 8192edno.fm

SG24-8192-00

© Copyright International Business Machines Corporation 2013. All rights reserved.Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP ScheduleContract with IBM Corp.

8192edno.fm Draft Document for Review February 26, 2014 8:58 am

First Edition (September 2013)

This edition applies to IBM Wave for z/VM v1.1.0 (PID number 5648-AE1 1.1), Version 6, Release 3 of z/VM (product number 5741-A07), SUSE Enterprise Linux Server 11 sp2 and Red Hat Enterprise Linux release 6.4.

This document was created or updated on February 24, 2014.

Note: Before using this information and the product it supports, read the information in “Notices” on page 3.

Contents 1

Draft Document for Review February 26, 2014 8:58 am SG248192TOC.fm

Contents

Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Authors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5Now you can become a published author, too! . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6Stay connected to IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Part 1. Architecture and installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

Chapter 1. Introduction to IBM Wave for z/VM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.1 What is IBM Wave for z/VM? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121.2 Benefits of using IBM Wave for z/VM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

Chapter 2. IBM Wave for z/VM architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.1 The IBM Wave for z/VM basic architecture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

Chapter 3. Planning and design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.1 Planning for installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263.2 Blank worksheets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Chapter 4. Installation and configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314.1 Pre-installation tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 324.2 IBM Wave for z/VM Installation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Part 2. Using IBM Wave for z/VM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71

Chapter 5. How to manage and operate the user interface environment . . . . . . . . . . 735.1 The IBM Wave for z/VM User Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 745.2 The Tool Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805.3 System Viewers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 815.4 The IBM Wave for z/VM Viewers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 925.5 The Status Bar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 935.6 Group Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 945.7 Project Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 975.8 Creating Reports . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1105.9 User mangement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1175.10 Backup option for IBM Wave. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

Chapter 6. How to manage z/VM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1436.1 Managing z/VM User IDs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1446.2 Managing Device Pools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1626.3 Managing DASD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1706.4 Managing the network. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

Chapter 7. How to manage the Linux virtual machines. . . . . . . . . . . . . . . . . . . . . . . . 1837.1 Preparing to install Linux in the virtual machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1847.2 Create a z/VM Guest. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1907.3 Initializing guests using IBM Wave for z/VM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

SG248192TOC.fm Draft Document for Review February 26, 2014 8:58 am

2 IBM Wave - Installation, Implementation and Exploitation

7.4 Create a golden master. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2167.5 Cloning virtual machines from a golden master . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222

Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227Other publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

© Copyright IBM Corp. 2013. All rights reserved. 3

Draft Document for Review February 26, 2014 8:58 am 8192spec.fm

Notices

This information was developed for products and services offered in the U.S.A.

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, Armonk, NY 10504-1785 U.S.A.

The following paragraph does not apply to the United Kingdom or any other country where such provisions are inconsistent with local law: 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 states 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 supply in any way it believes appropriate without incurring any obligation to you.

Any performance data contained herein was determined in a controlled environment. Therefore, the results obtained in other operating environments may vary significantly. Some measurements may have been made on development-level systems and there is no guarantee that these measurements will be the same on generally available systems. Furthermore, some measurements may have been estimated through extrapolation. Actual results may vary. Users of this document should verify the applicable data for their specific environment.

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 the names and addresses used by an actual business enterprise is entirely coincidental.

COPYRIGHT LICENSE:

This information contains sample application programs in source language, which illustrate programming techniques on various operating platforms. You may copy, modify, and distribute these sample programs in any form without payment to IBM, for the purposes of developing, using, marketing or distributing application programs conforming to the application programming interface for the operating platform for which the sample programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore, cannot guarantee or imply reliability, serviceability, or function of these programs.

8192spec.fm Draft Document for Review February 26, 2014 8:58 am


Trademarks

IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. These and other IBM trademarked terms are marked on their first occurrence in this information with the appropriate symbol (® or ™), indicating US registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml

The following terms are trademarks of the International Business Machines Corporation in the United States, other countries, or both:

AIX®DB2®DirMaint™HiperSockets™IBM®MVS™

Power Systems™PR/SM™Redbooks®Redbooks (logo) ®System z10®System z®

WebSphere®z/OS®z/VM®z10™

The following terms are trademarks of other companies:

Linux is a trademark of Linus Torvalds in the United States, other countries, or both.

Windows, and the Windows logo are trademarks of Microsoft Corporation 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 company, product, or service names may be trademarks or service marks of others.

http://www.ibm.com/legal/copytrade.shtml


Draft Document for Review February 26, 2014 8:58 am 8192pref.fm

Preface

IBM® Wave for z/VM® is a virtualization management solution for IBM z/VM and Linux on System z®. This virtualization management software provides a simplified and cost effective way for companies to harness the consolidation capabilities of the IBM System z platform and its ability to host the workloads of tens of thousands of commodity servers. IBM Wave for z/VM is a complete management solution for System z-based virtual server farms.

This IBM Redbooks® publication provides a guide to understanding IBM Wave for z/VM by providing information on the IBM Wave for z/VM architecture and how it fits into the Cloud. Also provided will be a planning and design guide based on common scenarios. This IBM Redbooks publication will provide installation and configuration task information and how to manage and operate the environment.

The intended audience for this IBM Redbooks publication is IT Architects responsible for planning their IBM Wave for z/VM environments and IT Specialists responsible for implementation of them.

Authors

This book was produced by a team of specialists from around the world working at the International Technical Support Organization, Poughkeepsie Center.

Lydia Parziale is a Project Leader for the ITSO team in Poughkeepsie, New York, with domestic and international experience in technology management including software development, project leadership, and strategic planning. Her areas of expertise include Linux on System z and database management technologies. Lydia is a Certified IT Specialist with an MBA in Technology Management and has been employed by IBM over 25 years in various technology areas.

Susan Greenlee is the System z WW Technical Enablement team lead for Cloud and for Virtualization and Linux. Susan has over 20 years experience with IBM including Software Engineer for z/OS® Development, Technical Support Management for Virtualization and Linux, and for the past 7 years was the Project Manager for the Virtualization and Linux on System z Project Development Team. This is Susan's second IBM Redbooks Publication.

Ernest Horn works for IBM US on the ATS IBM Wave Tiger Team.

Gene Ong is a Consulting IT Specialist working for IBM since 1987. He spent 11 years in large system software design and development which included products such as TSO/E, APPC/MVS™, APPC/RRS and System Logger. He joined IBM’s System z Client Center in 1998. He was involved in the original Redbook for Linux on System z in 2001. He is the Linux on System z Technical Lead for the Client Center working on benchmarks, proofs of concepts, and demos involving Linux on System z and associated middleware such as WebSphere® Products, DB2® UDB, and Oracle. He holds a Masters degree in Computer Science from the Polytechnic University of New York.

Note: The terms virtual machine,virtual server, guest, and z/VM user all refer to a virtual machine. We use these terms interchangeably throughout this book.

8192pref.fm Draft Document for Review February 26, 2014 8:58 am


Richard Lewis is currently the team leader for the z/VM and Linux on System z team in ATS at the IBM Washington System Center. Richard has been with IBM for 30 years supporting z/VM as a System Programmer on internal IBM systems and then as a member of the Washington System Center. Richard has also worked with AIX® on Power Systems™ and with Linux on System z since it was originally ported to the System z platform.

Sean McBride is a Linux and z/VM specialist with the Advanced Technical Skills (ATS) organization. He holds a Bachelors degree in German and Computer Science from the United States Military Academy at West Point and a Master of Information Technology & Management degree from the Illinois Institute of Technology. He is also the founder of the Millennial Mainframer blog, which provides “a fresh and modern look at all things mainframe” from college students, young mainframe professionals, and mainframe thought leaders.

Steve McGarril iis a Systems Programmer in the System z Client Center in Poughkeepsie, New York, working with business partners, Independent Software Vendors and clients on benchmarks, Proof of Concepts, Proof of Technologies and customer demos. He began his IBM career working in MVS Operations and moved into a mainframe systems support capacity and his areas of expertise include both the z/OS and z/VM operating systems. Steve is a Certified IT System Specialist and has been employed by IBM for over 31 years working with System z in its various forms.

Thanks to the following people for their contributions to this project:

Roy P. CostabInternational Technical Support Organization, Poughkeepsie Center

Erin Farr and the IBM Wave for z/VM development teamIBM

Sharon Chen, Amir Glaser, Erin Farr, Sevilay Kurt, Sean P. McBride, Fred Bader, Li Yong Li and Klaus EgelerIBM

Now you can become a published author, too!

Here’s an opportunity to spotlight your skills, grow your career, and become a published author—all at the same time! Join an ITSO residency project and help write a book in your area of expertise, while honing your experience using leading-edge technologies. Your efforts will help to increase product acceptance and customer satisfaction, as you expand your network of technical contacts and relationships. Residencies run from two to six weeks in length, and you can participate either in person or as a remote resident working from your home base.

Find out more about the residency program, browse the residency index, and apply online at:

ibm.com/redbooks/residencies.html

Comments welcome

Your comments are important to us!

We want our books to be as helpful as possible. Send us your comments about this book or other IBM Redbooks publications in one of the following ways:

http://www.redbooks.ibm.com/residencies.html

http://www.redbooks.ibm.com/residencies.html

Preface 7

Draft Document for Review February 26, 2014 8:58 am 8192pref.fm

! Use the online Contact us review Redbooks form found at:

ibm.com/redbooks

! Send your comments in an email to:

[email protected]

! Mail your comments to:

IBM Corporation, International Technical Support OrganizationDept. HYTD Mail Station P0992455 South RoadPoughkeepsie, NY 12601-5400

Stay connected to IBM Redbooks

! Find us on Facebook:

http://www.facebook.com/IBMRedbooks

! Follow us on Twitter:

http://twitter.com/ibmredbooks

! Look for us on LinkedIn:

http://www.linkedin.com/groups?home=&gid=2130806

! Explore new Redbooks publications, residencies, and workshops with the IBM Redbooks weekly newsletter:

https://www.redbooks.ibm.com/Redbooks.nsf/subscribe?OpenForm

! Stay current on recent Redbooks publications with RSS Feeds:

http://www.redbooks.ibm.com/rss.html



http://www.redbooks.ibm.com/contacts.html

http://www.facebook.com/IBMRedbooks

http://www.linkedin.com/groups?home=&gid=2130806

http://www.redbooks.ibm.com/rss.html

8192pref.fm Draft Document for Review February 26, 2014 8:58 am



Draft Document for Review February 26, 2014 8:58 am 8192p01.fm

Part 1 Architecture and installation

In this part we describe IBM Wave for z/VM and its business value. We also describe the basic architecture of the IBM Wave for z/VM environment as well as our own lab environment used in the creation of this IBM Redbooks publication. Additionally, we discuss planning and design decisions that must be made prior to installation as well as the basic installation tasks.

Part 1

8192p01.fm Draft Document for Review February 26, 2014 8:58 am



Draft Document for Review February 26, 2014 8:58 am 8192ch01.fm

Chapter 1. Introduction to IBM Wave for z/VM

In this chapter, the following topics are discussed:

! What is IBM Wave for z/VM

! Benefits of using IBM Wave for z/VM

1

8192ch01.fm Draft Document for Review February 26, 2014 8:58 am


1.1 What is IBM Wave for z/VM?

IBM Wave for z/VM is a z/VM hypervisor manager product that was acquired with the IBM acquisition of CSL International. This product provides a graphical interface tool that simplifies the managment and administration of z/VM and Linux environments on IBM System z. The scope of IBM Wave for z/VM can span multiple LPARs, SSI clusters, and servers.

IBM Wave for z/VM provides a visualization that represents virtual resources and their relationships. It provides a customizable graphical or tabular view to help adminstrators understand the status of the entire system or a subset such as system, network or storage topology in a single view. The status of servers, networks and network devices, and storage is automatically detected by IBM Wave for z/VM. Changes in resources and their relationships are monitored and reflected in displays. Many of the IBM Wave for z/VM displays have advanced filters, tagging, layout and layer selection to provide customized views.

This simplified user interface can be used to manage virtual servers and resources that would otherwise need to be performed with z/VM system commands by a skilled z/VM systems programmer. With IBM Wave for z/VM the user can point and click to:

! Perform basic z/VM guest actions such as activate, deactivate, recycle, pause and resume

! Provisions virtual resources such as z/VM guests, network and storage

! Capture and clone virtual servers across LPARs

! Create and configure vSwitch and guest LANs

! Connect virtual servers to virtual networks

! Install Linux on a virtual guest

! Relocate virtual guests with live guest relocation

! Display and monitor page and SPOOL and add and remove disks

! Provision and track used or free OSA nd Hipersocket devices using Device Pools

! Manage storage using the Manage Storage Wizard that enables management at the z/VM level such as dedicating devices, adding minidisks, as well enabling management on the Linux level such as the creation and expansion of LVM volume groups, regular partitions, and logical volumes and adding FCP attached storage to z/VM guests

1.2 Benefits of using IBM Wave for z/VM

IBM Wave for z/VM has quite a few benefits for today’s z/VM environment. It simplifies management and administration of a z/VM environment by abstracting z/VM resources and providing a GUI to automate z/VM functions. With a simplified GUI and its point and click functionality, there is much less need to understand extensive z/VM command syntax. This is particularly true for Linux administrators that are assigned Projects.

A unique capability of IBM Wave for z/VM is its ability to create Projects, a set of virtual resources consisting of any combination of servers, networks, and storage across the enterprise. The definition of Projects can enable different constituencies, such as operations, systems programming, application development, project management or end users to exercise their authority to manage appropriate aspects of the z/VM environment.

The scope and permissions a Linux administrator (also called a regular IBM Wave for z/VM user) has within a Project can be uniquely assigned. Scope refers to what a Linux administator can view or see. Permissions refer to what actions a Linux administrator can take

Chapter 1. Introduction to IBM Wave for z/VM 13


within the assigned scope. A Linux administrator may have different scopes and permissions for different projects. Likewise, different Linux administrators may have different scopes and permissions for the same project.

The delegation of administrative capability can also be assignede based on defined roles - project administration, network administration, storage administration, operator. The flexibility of setting up Linux administrators and assigning unique projects, scope, and permissions, and the use of defined roles makes it feasible to distribute administration responsbility to specific individuals across organizations facilitating project self administration. IBM Wave for z/VM provides an environment that allows a level of autonomy and self-service without extensive z/VM skills.

IBM Wave for z/VM not only improves the ease of use of z/VM but provides additional functionality and flexibility to manage a virtualized environment:

! Most actions can be carried out against multiple objects whether they are on a single or multiple LPARs.

! Intelligent icons help the user understand the status of the system and its resources. For example an icon can indicate whether a virtual guest is running, whether it is running a Linux distribution from Suse or Red Hat, or whether it is a CMS guest, a system service machine or a IBM Wave for z/VM internal virtual server.

! Definition and use of functionality types ensures ordered activation and deactivation of z/VM guests. For example, a database server can be brought up followed by a transaction server and a web server with deactivation occurring in the reverse order.

! Custom attributes and their associated values can be defined and associated with virtual guests. Multiple attributes can be assigned to guests and then used for grouping and filtering in any combination to take action on a set of guests. For example, an attribute called Server Use could be defined with possible values Production, QA, Development. Another attribute called Server Type could be defined with possible values Database, Transaction, Web.

! Ability to broadcast a message to a set of Linux users. This can be finely filtered with the use of custom attributes.

! Intelligent Active Notes (IANs) allow free form text to be associated with an object and displayed during a mouse hover. Initiating an action on the object causes the IAN to pop up.

! At a glance views of overall system storage utilization with the ability to drill down to the volume level for site level administrators and for Linux administrators views of storage groups to which they are uniquely permitted

! Simplified mechanism of storage assignment and storage management allowing the IBM Wave for z/VM user to interact with DASD Volumes, DASD Groups and DIRMAINT regions.

! Performance view of all managed systems includes current CPU utilization, page and spool space and virtual to real storage ratios. Drill down to the Linux guest level is also supported

! Generate concise reports for selected objects with the ability to use complex filters and customization



Chapter 2. IBM Wave for z/VM architecture

This chapter provides an overview of the basic IBM Wave for z/VM architecture and how it fits into your enterprise, whether small or large. We also describe the foundational components of IBM Wave for z/VM within each tier of the architecture.

2



2.1 The IBM Wave for z/VM basic architecture

IBM Wave for z/VM implements a three tier architecture shown in Figure 2-1:

! A graphical user interface (GUI) client! The WAVESRV Linux guest server! The Target Virtualization Platform (TVP) application programming interface (API)

Figure 2-1 IBM Wave for z/VM three tier architecture

In this section we discuss each and how they communicate with each other. A high level overview of this communication is shown in Figure 2-2

Figure 2-2 Communication between tiers of IBM Wave for z/VM

2.1.1 Tier 1: The GUI client

The GUI client controls the execution, attributes and behavior of the Linux virtual servers and the rest of the managed z/VM resources as per the scope and permissions of the IBM Wave user. It provides a graphical interpretation of the knowledge base and allows the user to

© 2013 IBM Corporation34

z/VM LPAR API Server

WAVESRV Server

Background TaskScheduler (BTS)

IBM Wave for z/VM

database

IBM Wave for z/VMGUI client

IBM Wave for z/VM

web server

Chapter 2. IBM Wave for z/VM architecture 17


interact with the TVP using point-and-cllick and drag-and-drop operations. It’s responsible for the facilitation of all interaction and viewing of the objects managed by IBM Wave for z/VM. The GUI engine also facilitates all of the interaction with the z/VM LPAR.

The GUI client is a Java WebStart application and is supported on MS Windows 7. The Apache web server on the WAVESRV Linux guest server contains a link which downloads and installs the IBM Wave GUI client on the workstation. For MS Windows users, it also creates a link on the desktop and in the Start menu.

Once the application is installed, starting the application can be done by simply double clicking on the icon or shortcut. The Java Web Start facility includes an automatic search for updates. This means that after applying IBM Wave maintenance, which updates the GUI Client, it is not necessary to re-install IBM Wave. The application will be automatically updated the next time it is launched.

The IBM Wave GUI engine is in constant contact with the Background Task Scheduler (BTS) on the WAVESRV Linux guest server to ensure the GUI display represents the most current state of the environment. All communications between the GUI, the BTS and the IBM Wave service machines are encrypted using a proprietary protocol.

Additionally, a user has the ability to turn on the Trace GUI option in the User Tasks menu as shown in Figure 2-3.

Figure 2-3 Trace GUI

If the Trace GUI option is selected, the user can set the GUI Debug Level by navigating from Adminstrative → IBM Wave Parameters. A pop-up will open similar to that found in Figure 2-4. Select the GUI tab, choose the trace level you desire and set the GUI Entry Limit, which is the maximum number of log entries that appear in the Log Viewer in the General Status Viewer.



Figure 2-4 Setting the Debug Level for Trace GUI

The GUI Debug Log resides on the client workstation, under a temporary directory. Generally, you can find it by navigating to C:\Users\<<username>>\AppData\Local\Temp and the format for the name of the log is WAVEGuiLog-<<dd-mm-yy>>.txt .

2.1.2 Tier 2: The WAVESRV Linux guest server

The WAVESRV Linux guest server runs within a z/VM LPAR or as a native Linux for System z partition on PR/SM™. It can run on Red Hat Enterprise Linux version 6 or SUSE Linux Enterprise Server 11.

The WAVESRV Linux guest server is a virtual server and contains the foundational components for IBM Wave for z/VM - the application knowledge database (or knowledgebase), a web server and the Background Task Scheduler (BTS) as shown in Figure 2-2.

The application knowledgebaseThe application knowledge database currently used with IBM Wave for z/VM is the MySQL data base. It resides on the WAVESRV Linux guest server and contains information on all z/VM elements managed by IBM Wave as well as the IBM Wave Regular Log, the Wave Users and more:



! Objects managed by IBM Wave for z/VM: CPCs, z/VMs, z/Linux virtual servers and their z/VM Guests, Virtual Networks and Virtual Networks connections' definitions, Prototypes, IBM Wave logs, IBM Wave users, Projects, Storage Groups, Disks and their attributes

! For each unique directory IBM Wave detects, a Directory record will be saved in the knowledgebase, and any z/VM System which uses that directory will be associated with the directory record automatically. IBM Wave monitors the directory on a periodic basis, and therefore if, for some reason, the unique identification of the directory changes, these changes will be reflected in the knowledgebase and the GUI clients.

! When a z/VM System is added to IBM Wave management, the auto-detection process starts automatically. The “auto-detect mechanism” accesses the z/VM LPAR and queries various aspects of the z/VM System. The results of these queries populate the IBM Wave knowledgebase.

! IBM Wave automatically detects z/VM Accounts used in managed z/VM Systems and stores information about them in the IBM Wave knowledgebase.

The IBM Wave for z/VM knowledgebase resides on the /var filesystem. It is highly recommended to configure this filesystem as a logical volume managed by LVM.

Background Task Scheduler (BTS)The BTS updates the knowledgebase with live data from all managed and active z/VM LPARs. The BTS, along with the z/VM API server, provides information from the knowledgebase to the GUI client. So even if changes are made outside of IBM Wave, say by a z/VM Systems Programmer using a 3270 emulator, they will be reflected in the GUI client.

The BTS handles requests using worker threads via a proprietary communication protocol based on TCP/IP. There are two types of worker threads:

1. User worker threads: Responsible for running user-generated BTS Workunits and Requests. This is a configurable option. To determine the number of User Worker Threads, use the following formula:<Number of concurrent IBM Wave Users>*2 + <Number of concurrent cloneoperations>*<maximum of minidisks in a source guest for cloning>

2. Internal worker threads: Responsible for running the internal IBM Wave Workunits and Requests such as the periodic tasks. This is not customizable and is determined by the number of z/VM systems managed by the BTS.

Every request running in the BTS is run under a BTS Workunit. BTS Work Units can be viewed from the BTS Workunit viewer in the GUI Client as shown in Figure 2-5.

Figure 2-5 BTS Work Unit viewer

A BTS Work Unit is made up of one or more BTS requests. All output generated by the BTS and all BTS requests are stored in the IBM Wave common output repository (COR). The COR is made up of the log COR entry which is used to log all activity of a specific BTS request.



Each BTS request is allocated at least one COR entry (a Log COR Entry) upon initiation, which logs all the activity of the BTS request. The Log COR Entry for a specific request can be viewed by accessing the BTS Work Unit to which this BTS request belongs, and selecting the specific request. The log COR entry also has System COR entries that can be viewed from the System COR entry viewer in the GUI as shown in Figure 2-6.

Figure 2-6 System COR entry viewer

Other Log COR Entries can be viewed from the BTS Work Units Details window, accessed by double clicking a specific BTS Workunit, and selecting a specific BTS request. You can manually delete COR entries from the BTS Work Unit Details tab. When a BTS Work Unit is deleted from IBM Wave, COR entries owned by the BTS request are also automatically deleted.

The BTS also contains a scheduling mechanism which automatically runs at specific intervals. Some of these intervals (shown in Figure 2-7) can be changed using the Manage Parameters option of the Administrative menu. You can also change the level of debug from this option.



Figure 2-7 BTS parameter intervals

The scheduling requests:

! Monitor the status of internal IBM Wave components such as the IBM Wave service machines and the WAVESRV file system. They also clean the BTS work units that have expired

! Update the IBM Wave knowledgebase with information from the managed z/VM systems on a periodic basis

Some of these requests are global and will be added only once, while others are CPC or z/VM LPAR-specific and will be automatically added to the scheduler when a z/VM LPAR or CPC is added to IBM Wave management. You can control these scheduling entries from the BTS Manager option accessible from the Administrative menu then right clicking on an event under the Scheduling tab as shown in Figure 2-8.



Figure 2-8 Updating the BTS Scheduler

For more information on the BTS Scheduler, see IBM Wave for z/VM: Administration and Customization, SC27-6118.

The web serverThe WAVESRV Linux guest server also runs a minimal Apache web server which is installed to enable the download of the GUI client using the Java Web Start application.

2.1.3 Tier 3: The Target Virtualization Platform (TVP)

The TVP is a mediation layer that provides the interface into the managed environment. Currently, only z/VM versions 5.4 and above are supported as the TVP. IBM Wave for z/VM uses the system management application program interface, or SMAPI, in order to mediate requests from the BTS and the GUI Client.

The TVP represents the hypervisor which hosts the virtual guests. The BTS and the GUI clients use the TVP API to query and perform changes to the TVP and hosted virtual guests.

Specific functions executed on the z/VM System use the following service machines:

WAVEWRKS The Short Service Machine executes various compiled REXX EXECs to interact with the z/VM environment

WAVEWRKL The Long Service Machine provides an additional thread of execution to run longer scripts or executes some directory manager commands

WAVEWRKC This Cross-System Cloning Service Machine is used to stream minidisks from a source z/VM system to a target z/VM system during cloning actions.

The names of these service machines used here are default names and can be configured with different names within IBM Wave for z/VM.

As part of the auto-detection process, when adding a new z/VM System to IBM Wave management, these 3 service machines are created and started on the z/VM System automatically.



IBM Wave performs various queries and executes commands on the TVP using these service machines and SMAPI. These intereactions are done through the BTS. The communication between the service machines and the BTS can optionally be encrypted using SSL.

The service machines run in any z/VM LPAR that is managed by IBM Wave and must be kept up and running. IBM Wave monitors these service machines and will issue alerts if there are any errors.



Chapter 3. Planning and design

In this chapter we provide guidance in planning for your IBM Wave for z/VM installation. We also provide some design guidelines. We provide worksheets with examples of each that will assist in installation of IBM Wave for z/VM in your own enterprise.

3



3.1 Planning for installation

Some of the things that you need to consider before installation are the sizing both for memory and your file systems, including the filesystems used for logging. Additionall you will need to consider the location of the WAVESRV Linux server.

The minimum requirements are:

! 1 GB RAM! / - 2GB (approximately 3000 cylinders)! /boot - 100MB (approximately 100 cylinders)! / var - 3 GB (approximately 4500 cylinders)

By default, the logs are stored in /var so you should configure this filesystem as a logical volume under LVM so it can be extended when needed. This is due to the fact that the IBM Wave knowledgebase resides on this file system and it might need to be expanded later on, depending on a variety of factos such as the amount of historical data set to be kept, the number of z/VM Systems added to IBM Wave management, etc.

The WAVESRV Linux server can be implemented as a virtual server within a z/VM LPAR.

Example 3-1 provides a sample directory entry for the WAVESRV Linux server.

Example 3-1 Sample directory entry with comments for the WAVESRV Linux server

*****************************************************************USER WAVESRV <Password Here> 1G 2G GCCPU 00IPL CMSMACHINE ESA 4OPTION QUICKDSPCONSOLE 0009 3215NICDEF 0800 TYPE QDIO LAN SYSTEM <GLAN/VSWITCH Name here>SPOOL 000C 3505 ASPOOL 000D 3525 ASPOOL 000E 1403 ALINK MAINT 0190 0190 RRLINK MAINT 019D 019D RRLINK MAINT 019E 019E RRMDISK 0191 3390 1 3 <DASD Volume Name here> - This is a CMS minidiskMDISK 0150 3390 1 3000 <DASD Volume Name here> - This minidisk will be used for “/”MDISK 0151 3390 1 4500 <DASD Volume Name here> - This minidisk will be used for “/var”MDISK 0152 3390 1 200 <DASD Volume Name here> - This minidisk will be used for swap*****************************************************************

3.2 Blank worksheets

Throughout the process of installing and configuring IBM Wave, you will require numerous pieces of information related to your z/VM environment and how you want to configure IBM Wave. These worksheets are provided to help you find this information before starting the installation process to allow for easy reference later.

Chapter 3. Planning and design 27


Linux Guest for Use as IBM Wave ServerA major prerequisite for installing Wave is to create the Linux guest that will serve as the IBM Wave Server. Fill in the following information to allow you to SSH and install the IBM Wave RPM.

IBM Wave Administrator IDWhen you launch the IBM Wave webstart application for the first time, you are immediately prompted with a panel to create the IBM Wave Administrator. Fill in the following information

Central Processing Complex DetailsAfter you install IBM Wave, you define your Central Processing Complex. You will provide a name to uniquely identify your CPC, the model of your CPC (e.g. 2827 / zEC12), and the CPU ID of your system. This CPU ID is used to validate that your system is licensed to run IBM Wave. To obtain this information, issue the CP command QUERY CPUID. This returns a response in the following format:

CPUID = aassssssccccdddd

The ssssss string (third through eighth hexadecimal digits) is the value that you need to enter in IBM Wave when defining a CPC.

z/VM System DetailsAfter defining your CPC, you can add z/VM Systems. This process includes using the z/VM system name to create an External Entity and then using all other values to complete the “Create New z/VM System” panel.

The Name value is used to uniquely identify your z/VM system in IBM Wave. A good choice for this name would be your z/VM system id as defined in the z/VM SYSTEM CONFIG file.

The z/VM Version and z/VM Service Level values identify the level of z/VM running on this z/VM system. These values can be obtained by running the CP command QUERY CPLEVEL.

The IPv4 Address, IPv6 Addres, and Hostname fields define how the Wave server can establish a TCP/IP connection to the z/VM system.

The 3270 Port # and TLS/SSH values reflect how your z/VM environment is configured to allow tn3270 terminal sessions. In most environments, the 3270 Port # is 23 or 3270.

The Performance Toolkit and TCP/IP service machine names are the names of the z/VM users running these service machine functions in your z/VM environment. Typically, the

Distribution IP Address Root Password

User Name

Password

Security Question

Security Question Answer

Name Model CPU ID



Performance Toolkit service machine is named PERFSVM and the TCP/IP service machine is named TCPIP. You can check this by running the CP command QUERY NAMES.

The TCP/IP service machine owns a minidisk used to store the TCPIP PROFILE file. By default, this is stored on a minidisk at the 592 address. You can confirm this by accessing TCPIP’s 592 minidisk and looking for the PROFILE file.

Directory Manager DetailsIn order to use IBM Wave, your z/VM environment must run a directory manager, such as DirMaint™.

In order to interact with the directory manager in your z/VM environment, IBM Wave requires the user name and password of a z/VM user authorized to issue SMAPI commands. In the case of DirMaint, this user requires specific privilege classes and the ability to execute commands without entering a password. This user name and password can be any z/VM user, including the Wave server itself.

In z/VM environments running DirMaint, IBM Wave creates a Dummy Region of one cylinder. This is used as a placeholder because DirMaint is unable to create empty storage groups. The Name and VOLID values are used to create this Dummy Region during the initial Auto-Detection process for adding a new z/VM system. Although this region has to be defined on a real DASD volume, this Dummy Region is never actually allocate. Enter a valid VOLID and a name for this Dummy Region, such as DUMMY.

Name

z/VM Version

z/VM Service Level

IPv4 Address

IPv6 Address

Hostname

3270 Port #

3270 via TLS/SSH?

Performance Toolkit Service Machine Name

TCPIP Service Machine Name

Address of TCPIP Minidisk containing PROFILE

Type of Directory Manager

User Name

Password

VOLID for Wave Dummy Region

Name of Wave Dummy Region

Chapter 3. Planning and design 29


Wave Service Machine DetailsAlthough IBM Wave uses the Systems Management API (SMAPI), it also uses three CMS service machines. the Short Service Machine executes various compiled REXX EXECs to interact with the z/VM environment on behalf of the z/VM GUI. The Long Service Machine performs the same functionality, but it provides an additional thread of execution to run longer scripts in a way that does not prevent the execution of the numerous shorts IBM Wave scripts constantly running against your z/VM environment. The Cross-System Cloning Service Machine is used to stream minidisks from a source z/VM system to a target z/VM system during cloning actions. Typically, these service machines are named WAVEWRKS, WAVEWRKL, and WAVEWRKC respectively. In addition to providing the names of these service machines, you need to provide a storage group to allow IBM Wave to allocate various minidisks for these service machines.

VSwitch DetailsAfter IBM Wave completes the auto-detection process for you z/VM system, it will create metadata objects for each of your VSwitches. However, IBM Wave will flag each of your VSwitches as “Objects that require attention” in the System Status tab of your z/VM system’s Current System View. This is because the metadata objects of your VSwitches have not yet had Default NIC values defined. When IBM Wave is used to connect z/VM guests to VSwitches, IBM Wave adds a virtual NIC to each of these guests with an address equal to the Default NIC value of the VSwitch to which the guest is connecting. Because a virtual NIC is composed of three OSA addresses, the Default NIC value is actually only the first of three cosecutive address. For example, if a VSwitch has a Default NIC value of AC00, then this NIC uses AC00, AC01, and AC02. When planning the Default NIC values for your VSwitches, it is important to ensure that the Default NIC values do not overlap. So if one VSwitch uses AC00, then another VSwitch cannot use AC01 or AC02, as this would overlap. The second VSwitch would have to be AC03 to prevent an overlap.

Note: Ensure that your Default NIC values start with an alphabetic character. IBM Wave has problems with Default NICs that begin with numeric values like 0AC0. We have found that using sequences of AC00, AC03, AC06, and so function properly most consistently.

Short Service Machine Name

Long Service Machine Name

CSC Service Machine Name

Storage Group for Wave Service Machines’ minidisks

VSwitch Name Wave Default NIC



Virtual Network Segment DetailsIBM Wave automatically generates Virtual Network Segments based on the assumption that all segments are IPv4 Class C blocks. This assumption is likely not valid, requiring you to manually define your Virtual Network Segments correctly.

Name Network IP Gateway IP Netmask Broadcast IP VLAN ID


Draft Document for Review February 26, 2014 8:58 am 8192ch04 - 3_3_0.fm

Chapter 4. Installation and configuration

This chapter provides step-by-step instructions to guide you in the installation of IBM Wave for z/VM. We start with pre-installation tasks for both your z/VM environment as well as your Linux on System z guests. Next, we guide you in the actual installation of IBM Wave for z/VM in your environment. Finally, we provide instructions in post-configuration tasks required before you can start detecting your environment.

4

8192ch04 - 3_3_0.fm Draft Document for Review February 26, 2014 8:58 am


4.1 Pre-installation tasks

In order to successfully install IBM Wave for z/VM it is necessary to make sure you have all of the prerequisite software installed and configured on z/VM and the Linux system that will be used as the WAVESRV Linux server. The following sections list pre-requisite software and walk through pre-installation configuration tasks.

4.1.1 z/VM pre-requisites

The pre-requisites for the z/VM system where IBM Wave for z/VM will be installed are:

! z/VM Release must be V5.4, V6.2 or higher.! A Directory Manager such as DIRMAINT must be installed and active.! If Performance monitoring for z/VM Systems is to use the z/VM Performance Toolkit, this

has to be installed and configured properly (optional but suggested) See z/VM: Getting Started with Linux on System z, SC24-6194 for more information on how to do this.

! IBM System z10® or later

4.1.2 Workstation pre-requisites

For every workstation running the IBM Wave for z/VM graphical user interface (GUI) client:

! JAVA Runtime 1.7 with Web Start support must be installed! Ensure access to the IP addresses of all managed z/VM systems and to the WAVESRV

server.! Must have MS Windows 7 ! Firefox or Internet Explorer! PuTTY or equivalent telnet/SSH Client

4.1.3 WAVESRV pre-requisites

The pre-requisites for the WAVESRV Linux guest server are:

! The Linux system where the IBM Wave for z/VM rpm will be installed must be either SUSE Linux Enterprise Server 11, or Red Hat Enterprise Linux 6.– The system can be a z/VM Guest or LPAR– The Linux system must also have Java SE Runtime 1.7 installed– For Linux on System z, the virtual machine must:– Have at least 1GB of memory.– Have a file system that is at least 6GB.– MySQL v12.22 or higher– Apache

! The Linux system must also have Java SE Runtime 1.7 installed

4.1.4 Managed Linux System pre-requisites

! For every Linux on System z virtual machine managed by IBM Wave for z/VM:– SSH, VMCP and CMSFS (if using SUSE Linux Enterprise Server) packages must be

installed

Chapter 4. Installation and configuration 33


4.1.5 Pre-installation configuration tasks for z/VM

Before installing the IBM-WAVE for z/VM rpm into the WAVESRV system, it is best to complete the following z/VM configuration tasks.

z/VM System Management API (SMAPI)IBM-WAVE uses the SMAPI to accomplish various tasks on the z/VM system. It is therefore necessary to make sure that the SMAPI is configured and running prior to the installation of IBM-WAVE. The exact configuration steps for SMAPI differ based up on the version and release level of z/VM. Since our environment is running z/VM 6.3, we document here the steps used to configure the SMAPI for use with IBM-WAVE

1. Modify the DMSSISVR NAMES file to uncomment the directory manager definition, and the dump handler definition. Since DMSSISVR NAMES is a replacement maintained part, it is recommended to use the VMSES/E LOCALMOD command to make changes to this file. The LOCALMOD command is defined in the VMSES/E z/VM Service Guide, GC24-6232.a. From the MAINT630 userid enter the command as shown in Figure 4-1:

Figure 4-1 VMSES/E Update DMSSISVR Names

b. Enter 1 when prompted to continue. The system will respond by placing you in an XEDIT session on the file DMSSISVR NAMES

c. Scroll to the lines shown in Figure 4-2:

Figure 4-2 Default DIRMAINT Section

d. Modify the lines as shown in Figure 4-3:

Figure 4-3 Uncommented DIRMAINT Section

e. Scroll to the lines shown in Figure 4-4:

Figure 4-4 Default Dump Handler Section

f. Modify the lines as shown in Figure 4-5:

Figure 4-5 Uncommented Dump Handler Section

LOCALMOD CMS DMSSISVR NAMES

*:server.DIRMAINT *:type.DMGR

:server.DIRMAINT :type.DMGR

* Dump Handler *:server.OPERATNS*:type.WORKER *:short.DMPH

* Dump Handler :server.OPERATNS :type.WORKER :short.DMPH



g. File these changes to save them . The system will create a DMSSISVR NAML0002 file on the CMS LOCALMOD minidisk (MAINT630 3C4).

h. Enter the command shown in Figure 4-6 to rebuild the file including the new modification:

Figure 4-6 Rebuild modified DMSSISVR NAMES File

i. Enter the command shown in Figure 4-7 to move the new DMSSISVR NAMES file from the alternate build disk to the runtime disk (MAINT 193)

Figure 4-7 Put Updated DMSSISVR NAMES Into Production

2. Since many of the SMAPI API calls are carried out by the directory manager virtual machine (DIRMAINT) it is necessary to authorize the SMAPI worker virtual machines to use DIRMAINT. This configuration work can be done from the MAINT630 userid.a. When you configured DIRMAINT for use you would have created a configuration

override file to make changes to the IBM supplied CONFIG DATADVH file. On our system we named that file CONFIGAA DATADVH.

b. Enter the command shown in Figure 4-8 for DIRMAINT to send you the config file

Figure 4-8 Get CONFIG File

c. Receive the spool file sent by DIRMAINT using the CMS RECEIVE command. This will place the configuration file on the MAINT630 191 minidisk.

d. Use XEDIT to modify the file by adding the lines shown in Figure 4-9:

Figure 4-9 Updates to CONFIGAA DATADVH

e. File the changes to save them.f. Use the commands shown in Figure 4-10 to send the modified file back to DIRMAINT

and to have DIRMAINT reprocess the contents of the file

Figure 4-10 Send And Activate Config File

g. Use the command shown in Figure 4-11 to have DIRMAINT send you the AUTHFOR CONTROL file

SERVICE CMS BUILD

PUT2PROD

DIRM SEND CONFIGAA DATADVH

ALLOW_ASUSER_NOPASS_FROM= VSMGUARD *ALLOW_ASUSER_NOPASS_FROM= VSMWORK1 *ALLOW_ASUSER_NOPASS_FROM= VSMWORK2 *ALLOW_ASUSER_NOPASS_FROM= VSMWORK3 *ASYNCHRONOUS_UPDATE_NOTIFICATION_EXIT.TCP= DVHXNE EXEC ASYNCHRONOUS_UPDATE_NOTIFICATION_EXIT.UDP= DVHXNE EXEC

DIRM FILE CONFIGAA DATADVH ADIRM RLDDATADIRM RLDCODE



Figure 4-11 Get AUTHFOR file

h. Receive the spool file sent by DIRMAINT using the CMS RECEIVE command. This will place the config file on the MAINT630 191 minidisk.

i. Use XEDIT to modify the file by adding the lines shown in Figure 4-12:

Figure 4-12 Updates to AUTHFOR CONTROL

j. File the changes to save them .k. Use the command shown in Figure 4-13 to send the modified AUTHFOR CONTROL

file back to dirmaint, and to cause DIRMAINT to reprocess the content of the changed file:

Figure 4-13 Send And Activate AUTHFOR Changes

When IBM Wave for z/VM uses SMAPI it must authenticate using an existing z/VM virtual machine name and password. The virtual machine used does not need any special privilege classes, and will never be logged on. It simply needs to be defined in the z/VM USER DIRECTORY for authentication purposes.

The SMAPI authorized users file is named VSMWORK1 AUTHLIST, and it is located in the VMSYS: filepool under the root directory for VSMWORK1 (VMSYS:VSMWORK1.). Care needs to be taken when updating this file. The default file that comes with z/VM 6.3 has the data shown in Figure 4-14.

Figure 4-14 Default AUTHLIST File

We used the MAINT userid for authorization within IBM Wave for our environment. If you would like to add a new userid to the file it is recommended that you copy an existing line, and then only overtype the userid text on the duplicated line.

Once the IBM Wave rpm is installed on the WAVESRV virtual machine and running, it is possible to access the z/VM API Java testing application from the IBM Wave web page. An example of this application is shown in Figure 4-15.

DIRM SEND AUTHFOR CONTROL

ALL VSMGUARD * 140A ADGHMOPSALL VSMGUARD * 150A ADGHMOPSALL VSMWORK1 * 140A ADGHMOPSALL VSMWORK1 * 150A ADGHMOPSALL VSMWORK2 * 140A ADGHMOPSALL VSMWORK2 * 150A ADGHMOPSALL VSMWORK3 * 140A ADGHMOPSALL VSMWORK3 * 150A ADGHMOPS

DIRM SEND AUTHFOR CONTROL ADIRM RLDDATADIRM RLDCODE

DO.NOT.REMOVE DO.NOT.REMOVE MAINT ALL ZHCP ALL

Sean McBride



Figure 4-15 Test API Begin

The IP address you enter is the IP address of your z/VM system. The API User Name and password corresponds to a userid listed in the VSMWORK1 AUTHLIST file. When you press the Test button you should see a result similar to what is shown in Figure 4-16.



Figure 4-16 Test API Results

DIRMAINTThe IBM Wave for z/VM product also has worker virtual machines that need to be authorized to use DIRMAINT facilities directly (i.e. not through SMAPI). You will once again need to update the configuration override file and AUTHFOR CONTROL file.

1. As mentioned in number 2a above, the name of our configuration override file is CONFIGAA DATADVH. The command shown in Figure 4-17 issued from the MAINT630 virtual machine will retrieve that file from the DIRMAINT service machine.

Figure 4-17 Get Config File

2. The system will respond by sending the file as a spool file to the virtual reader of MAINT630. Use the CMS RECEIVE command to place the file on MAINT630’s 191 minidisk. The lines shown in Figure 4-18 need to be added to the CONFIGAA DATADVH file.

Figure 4-18 Updates To CONFIGAA DATADVH

DIRM SEND CONFIGAA DATADVH

ALLOW_ASUSER_NOPASS_FROM= WAVEWRKS * ALLOW_ASUSER_NOPASS_FROM= WAVEWRKC * ALLOW_ASUSER_NOPASS_FROM= WAVEWRKL *



3. If you did not activate DATAMOVE when you originally configured DIRMAINT for use, you should do so now as IBM Wave for z/VM requires that the DATAMOVE function be available. Add the statements shown in Figure 4-19 to file CONFIGAA DATADVH:

Figure 4-19 Activate DATAMOVE If Not Already Done

4. File the changes to save them . The commands shown in Figure 4-20will send the file back to DIRMAINT and so that DIRMAINT can reprocess the file.

Figure 4-20 Send And Activate Config Changes

5. The command shown in Figure 4-21 will cause DIRMAINT to send the AUTHFOR CONTROL file as a spool file to MAINT630’s virtual reader..

Figure 4-21 Get Authfor File

6. Use the CMS RECEIVE command to write the file on MAINT630’s 191 minidisk. 7. Add the lines shown in Figure 4-22 to this file using the XEDIT editor.

Figure 4-22 Updates To AUTHFOR CONTROL

8. File the changes. 9. The commands shown in Figure 4-23 will send the updated file back to DIRMAINT and

cause DIRMAINT to reprocess the contents of the file.

Figure 4-23 Send And Activate AUTHFOR Changes

10.You can make sure that your AUTHFOR updates completed properly with the DIRMAINT command shown in Figure 4-24

Figure 4-24 Check AUTHFOR Changes

DISK_CLEANUP= YES DATAMOVE_MACHINE= DATAMOVE * *

DIRM SEND CONFIGAA DATADVH ADIRM RLDDATADIRM RLDCODE

DIRM SEND AUTHFOR CONTROL

ALL WAVEWRKS * 140A ADGHMOPSZALL WAVEWRKS * 150A ADGHMOPSZALL WAVEWRKL * 140A ADGHMOPSZALL WAVEWRKL * 150A ADGHMOPSZALL WAVEWRKC * 140A ADGHMOPSZALL WAVEWRKC * 150A ADGHMOPSZ

DIRM SEND AUTHFOR CONTROL ADIRM RLDDATADIRM RLDCODE

DIRM FOR ALL AUTHFOR ?



The command response should show that VSMGUARD, VSMWORK1 - 3 and WAVEWRKS, WAVEWRKL, and WAVEWRKC are all authorized.

11.To support IBM Wave for z/VM, the DIRMAINT virtual machine must run with the C and E PRIVCLASSes added to the default of BD and G. The command shown in Figure 4-25 will add privclasses C and E to the DIRMAINT virtual machine.

Figure 4-25 Add PrivClasses To DIRMAINT

The new privilege classes will not take effect until the DIRMAINT virtual machine is logged off and logged back on. You can accomplish this with the DIRM SHUTDOWN command and then the XAUTOLOG DIRMAINT command once the DIRMAINT machine has fully logged off the system.

Since IBM Wave for z/VM will be allocating DASD space it is necessary to configure the DIRMAINT EXTENT CONTROL file before installing IBM Wave for z/VM. At a minimum you should have one region defined. When IBM Wave for z/VM is installed, a dummy region of one cylinder will be created as a placeholder for “empty” storage groups. While this region is defined on a real volume it is never physically allocated. If you have already updated EXTENT CONTROL to add regions and groups, you may use one of your already allocated regions.

In our environment, we created regions with the region name equal to the volume serial number of the DASD, and then created a group named VMLNX2 for use by IBM Wave for z/VM when creating new virtual machines.

To avoid problems during the IBM Wave install you will want to make sure that EXCLUDE section is properly configured to avoid any problems due to full pack overlap allocations for IBM virtual machines such as SYSDUMP or MAINT.

To double check the region name and volume name you will use during IBM Wave for z/VM install you may want to execute the commands shown in Figure 4-26 and ensure that a return code of zero is returned. In our example, we used region name and volume name LX9220. You will need to substitute a value for LX9220 that matches information in your EXTENT CONTROL file.

Figure 4-26 Check Dummy Region

If you do not receive a return code of 0 from the commands above, you will need to fix any problems in your EXTENT CONTROL file before proceeding with the install of IBM Wave for z/VM.

TCP/IP for z/VMSince IBM Wave for z/VM has the ability to establish 3270 sessions from within the WAVE GUI it is necessary to update the PROFILE TCPIP file for the z/VM TCP/IP stack to authorize two of the WAVE worker machines to use the obeyfile command. You will also want to make sure that the internal client is active, and the FTP server is active.

The PROFILE TCPIP file is by default created on the TCPMAINT virtual machine’s 198 minidisk. Add the virtual machines to the OBEY section of that file as shown in Figure 4-27.

DIRM FOR DIRMAINT PRIV +CE

DIRM DASD QUERY REGION LX9220DIRM DASD QUERY VOLUME LX9220



Figure 4-27 Updates to OBEY Section

In Figure 4-27, the first line following the keyword OBEY is part of the default definitions when z/VM 6.3 is installed and the IPWIZARD utility is used to configure an initial TCP/IP connection. We added the WAVEWRKS and WAVEWRKL virtual machines to this list.

Ensure that the PORT section of your PROFILE TCPIP contains the uncommented lines shown in Figure 4-28.

Figure 4-28 Review PORT Section

Finally, make sure that the FTPSERVE virtual machine is automatically started by TCPIP. You should see an AUTOLOG section similar to that shown in Figure 4-29.

Figure 4-29 Review AUTOLOG Section

After making changes to the PROFILE TCPIP file, you will need to restart your TCPIP virtual machine to pick up these changes.

Performance ToolkitIf the Performance Toolkit is available, then IBM Wave for z/VM will use this feature to display performance information for individual virtual machines. More information regarding setup and configuration of Performance Toolkit can be found in z/VM Performance Tookit Reference (SC24-6210).

In order for IBM Wave for z/VM to be able to use Performance Toolkit, the short duration worker machine (WAVEWRKS) needs to be authorized to gather data from Performance Toolkit. This is accomplished by adding WAVEWRKS to the FCONRMT AUTHORIZ file maintained on the 191 minidisk of the PERFSVM virtual machine. In order to update this file you can either logoff the PERFSVM virtual machine, link to it’s 191 minidisk and modify the file (this means performance monitoring is temporarily suspended), or logon to the PERFSVM virtual machine and update the file while it is still running. If you choose to update the file while PERFSVM is still running you need to be sure to run the FCONTROL RELOAD AUTHORIZ command following changes to the file.

Update the FCONRMT AUTHORIZ file with the entry shown in Figure 4-30:

OBEY OPERATOR TCPMAINT MAINT MPROUTE REXECD SNMPD SNMPQE LDAPSRV MAINT630 WAVEWRKS WAVEWRKL ENDOBEY

PORT 20 TCP FTPSERVE NOAUTOLOG ; FTP Server 21 TCP FTPSERVE ; FTP Server 23 TCP INTCLIEN ; TELNET Server

AUTOLOG FTPSERVE X ENDAUTOLOG



Figure 4-30 Update FCONRMT AUTHORIZ File

IBM Wave for z/VM general requirements for Performance Toolkit include making sure that the VMCF interface is turned on. This is accomplished by uncommenting the FC MONCOLL VMCF ON statement in the FCONRMT PROFILE. It is also necessary for the CP Monitor to be configured for collection of sampe records from the processor, storage and user (all) domains. This is accomplished by making certain that the PROFILE EXEC for PERFSVM has the commands shown in Figure 4-31 uncommented:

Figure 4-31 CP Monitor Sample Domains

These configuration steps are commonly done when setting Performance Toolkit up for general system use. It is good however to review the FCONRMT PROFILE and PERFSVM PROFILE EXEC to make certain the necessary configuration steps have been completed.

4.2 IBM Wave for z/VM Installation

Once all of the prerequisite work for z/VM is completed, and a Linux system matching the requirements listed earlier has been created, the installation of IBM Wave for z/VM is quite easy. It simply involves transferring the IBM Wave for z/VM rpm file to the target Linux system and installing the rpm.

In our example, we created a SUSE SLES 11 SP2 Linux virtual machine and installed the IBM Wave 3.3 rpm. We transferred the rpm file in binary mode to the WAVESRV virtual machine, and then invoked the command shown in Figure 4-32.

Figure 4-32 Install IBM Wave 3.30

When the command finishes, you should see a message indicating successful installation similar to Figure 4-33.

WAVEWRKS * DATA CMD

'CP MONITOR SAMPLE ENABLE PROCESSOR' 'CP MONITOR SAMPLE ENABLE STORAGE' 'CP MONITOR SAMPLE ENABLE USER ALL'

rpm -ivh IBM-Wave-3.30-1.s390x.rpm



Figure 4-33 RPM Install Complete

At this point all of the WAVE components are running and waiting for a connection via browser to setup the environment.

4.2.1 First Login

The next step in the installation process is to start a browser and open a connection to the IP address of the WAVESRV system similar to what is seen in Figure 4-34.



Figure 4-34 Splash screen

The Lauch z/VM API Testing Application button was mentioned earlier in this chapter. To continue with the installation, click the button labeled Launch IBM Wave v3.3.0 to start the IBM Wave graphical user environment. You will be prompted to define an administrator userid as shown in Figure 4-35 on page 44.



Figure 4-35 Define Administrator Userid

We entered WAVEADM as the Userid along with a password and challenge question. Figure 4-36 on page 44 displays the information that we provided.

Figure 4-36 Define Administrator Userid Completed

After completing the panel press the Go button. The system will respond by displaying the IBM Wave main display as in Figure 4-37 on page 45.



Figure 4-37 IBM Wave Main Display

4.2.2 Add New CPC

Now that IBM Wave for z/VM is installed, the next step is to define a Central Processing Complex (CPC) that hosts a z/VM system to be managed. On the main display right click inside the hardware viewer panel. You will see an option to Add New CPC similar to Figure 4-38 on page 46.



Figure 4-38 Add New CPC

In response to clicking on Add New CPC a panel will be opened to collect information about the new CPC similar to Figure 4-39 on page 47.



Figure 4-39 Create New CPC

The first text box on the panel is where you type in a name for this CPC. This can be any name you choose. Select the CPC Model from the CPC Model drop down. Paste in the CPU serial number for the CPC in the CPC CPUID text box. You might also want to type in a description of the system in the Description text box. After entering all of the information press the Create button. The system will respond with a success message similar to Figure 4-40 if no errors are encountered.

Figure 4-40 CPC Successfully Created

After pressing OK you will be returned to the main display. A graphic representing the CPC just added will be visible in the hardware viewer section of the main display, similar to Figure 4-41 on page 48.



Figure 4-41 Hardware Viewer CPC Graphic

4.2.3 Define New System

In order to define a z/VM system on the new CPC for IBM Wave for z/VM to manage it is necessary to first create an external managed entity for this system. From the main display click on the Administration category at the top of the display, then select Site Management, and External Entities Manager similar to Figure 4-42 on page 49.



Figure 4-42 External Entities Manager Navigation

The system will respond by displaying a panel listing any external entities already defined. Since we have not yet defined any, the panel will be blank similar to Figure 4-43 on page 49.

Figure 4-43 External Entities Manager Display

Click the Create IBM Wave External Entity button to add a new entity to IBM Wave. The system will respond with a panel to gather the name and type of entity being added similar to Figure 4-44 on page 50.



Figure 4-44 External Entity VMLINUX2

Enter the name of the LPAR that you want IBM Wave for z/VM to manage. In this example, we use VMLINUX2 as the name of the entity, and then select z/VM as the type from the drop down box. Add a description to the Description text box (in this example, ITSO System) . Once all of the information is added press the Create button. The system will respond by submitting the add request to the background task scheduler (BTS) and displaying a pop-up box with that information. Press OK to clear the pop-up box. When the add request is complete a new external entity will show up in the External Entitities Manager window, similar to Figure 4-45 on page 50.

Figure 4-45 External Entity Add Complete

Press the Close button to remove the Enternal Entities Manager window and return to the main display. Now that a type z/VM external entity has been created it is possible to add a new z/VM system for IBM Wave for z/VM to manage. This is done by selecting the CPC in the hardware viewer and right clicking on it. Select More Actions and then Add New System similar to Figure 4-46 on page 51.



Figure 4-46 Add New System Navigation

The system will respond by displaying a panel to collect information about the new z/VM system to be managed similar to Figure 4-47 on page 52.



Figure 4-47 Add New System Panel

The Add New System Panel is displayed with some of the information already filled in. The System Name and CPC Name were taken from External Entity definition and CPC Object. The IBM Wave Service Machine Information contains typical defaults. Figure 4-48 on page 53 displays the panel with information added from our lab environment.



Figure 4-48 Completed Create System Panel

The IP Address entered is the IP address associated with the z/VM TCP/IP stack (TCPIP virtual machine). This same IP address is entered as the Service Machine IP since the IBM Wave for z/VM worker machines listen on ports hosted by the z/VM TCP/IP stack. The API User Name entered is a name from the VSMWORK1 AUTHUSER file discussed earlier in this chapter. In our lab environment, we chose to use MAINT since it was already defined in the default file. We then entered the name of a DIRMAINT Region and the VolSer for that device. We kept the Region Names and VolSer the same for each device defined in the DIRMAINT EXTENT CONTROL file. All of this information should have been previously defined in the worksheets from Chapter 3, “Planning and design” on page 25 for easy reference. Since we licensed Performance Toolkit, we need to enter PERFSVM as the Performance Machine. After entering all of the information, press the Create button. The system will respond by creating the new system object and displaying a panel indicating the auto detect process will begin, similar to Figure 4-49 on page 54.



Figure 4-49 Auto Detect Begin

4.2.4 Auto Detect New System

The Auto Detect process is where IBM Wave for z/VM creates the worker machines and then discovers as much information as possible from the z/VM system. Information discovered consists of all of the defined virtual machines, DASD information, networking information, and so on. To kick off the Auto Detect process press the Next button. The system will display a panel prompting you to enter the user name and password of the Authorized API User. This is the user name defined in the VSMWORK1 AUTHLIST file discussed earlier. Figure 4-50 on page 54 shows this panel.

Figure 4-50 Authorized API Credentials



We used MAINT in our lab environment as the Authorized API User since MAINT was already defined in the VSMWORK1 AUTHLIST file. When you press the Next button, the system will respond by displaying a panel with information and actions relating to the IBM Wave Service Machines, similar to Figure 4-51.

Figure 4-51 Service Machines

Since the IBM Wave service machines do not already exist on the system, the default action is to Create, and then Populate with code. The WAVEWRKS virtual machine contains all of the executable code. WAVEWRKL and WAVEWRKC link to the minidisk owned by WAVEWRKS to access the code. These service virtual machines augment the capability provided by the z/VM SMAPI. The WAVEWRKS virtual machine is used for short duration requests. The WAVEWRKL virtual machine is used for long duration requests (such as processing the User DIRECT file). The WAVEWRKC is used for cross system cloning operations.

You will typically leave the IBM Wave Service Machine information as is. In the box next to Use DASD Group enter the name of the Group defined in EXTENT CONTROL to be used for the definition of the service machine minidisks.

When you press the Next button the system will respond with a panel displaying device pool information similar to Figure 4-52.



Figure 4-52 Device Pools

Since information for this system is being acquired, the default action will be to create new device pools for all of the various pool types (DASD, OSA, HiperSockets™, and FCP). You can accept these defaults and press the Next button. The system will respond by displaying a panel to collect additional parameter information similar to Figure 4-53 on page 56.

Figure 4-53 Additional Parameters

The Additional Parameters panel displays information that is typical for an unmodified z/VM system. If you have changed the name of the TCP/IP service machine from the IBM default of TCP/IP to something else, you would enter the name you used on this panel. Similarly if you moved the TCP/IP client executables from the default location of TCPMAINT 592, then you should specify the new location on this panel also. The only non-default piece of information is the specification of a Region name (from the EXTENT CONTROL file) that IBM Wave can



use to create a Dummy Region. The Dummy Region is a one cylinder region that is never physically allocated. It exists to handle the limitation of a Group definition in EXTENT CONTROL cannot be empty. In our lab environment, we specified the name LX106B which is the name of a Region in EXTENT CONTROL, and also the Volume Serial number of the device the region is defined on. When you press the Next button, the system will respond with a summary panel similar to Figure 4-54 on page 57.

Figure 4-54 Summary

If all of the information looks correct in the summary, press the Go button. The system will respond with a confirmation pop-up box similar to Figure 4-55.

Figure 4-55 DIRMAINT Warning

Since IBM Wave will use the long duration IBM Wave service machine, WAVEWRKL, to create the Dummy Region, this warning is reminding you that the IBM Wave service machines need to be authorized with DIRMAINT to execute commands such as this. If you followed the steps in this chapter for updating and activating the changes to AUTHFOR CONTROL and the CONFIGxx DATADVH file you have completed the steps this warning is referring to. Press Yes to continue with the auto detection process.

As the auto detect process runs, a BTS Work Unit will show a progress bar in the IBM Wave viewers section of the main display, similar to Figure 4-56 on page 58.



Figure 4-56 Auto Detect Running

Details associated with this work unit can be viewed by double clicking on the row in the table under the BTS Work Unit tab. This will open a Work Unit Details panel similar to Figure 4-57 on page 58.

Figure 4-57 Auto Detect Work Units

When the auto detect work unit has completed, the system might still be processing the data gathered. A progress message will be updated in the system viewer portion of the main display similar to Figure 4-58 on page 59.



Figure 4-58 System Auto Detect Status

When all of the queued update events have been processed, the z/VM groups will be displayed for the current system (VMLINUX2) similar to Figure 4-59 on page 59.

Figure 4-59 Auto Detect Complete Site Defined Groups

4.2.5 Correct Alerts From Auto Detect

The System Status tab may show red in your system. In our system, it showed red for VMLINUX2, indicating there are some items that need to be addressed. Clicking on the System Status tab will change to that display in the system viewer similar to Figure 4-60 on page 60.



Figure 4-60 System Status Tab

The red alert lines are displayed because IBM Wave for z/VM does not know what default virtual NIC address should be used when connecting virtual machines to the three VSWITCH objects discovered on the system. We can easily correct that problem from the Network tab in the system viewer. Section , “Fix Alerts For VSWITCH” on page 60 describes how.

Fix Alerts For VSWITCHFrom the System Status tab, double clicking on the red alert for the first VSWITCH (DTCSMAPI) will cause the Network display to appear with focus centered on the DTCSMAPI VSWITCH icon similar to Figure 4-61 on page 61.



Figure 4-61 Add Default NIC

In order to update the information for this VSWITCH, right click on the icon and select Update Information similar to Figure 4-62.

Figure 4-62 Add Default NIC Navigation

When Update Information is clicked, the system will respond by displaying an information panel similar to Figure 4-63.



Figure 4-63 Add Default NIC Data

To remove the alert, enter a virtual address in the Default NIC text box.We entered F000 as the default NIC for this VSWITCH. If a Guest is connected to a Virtual Network through this VNS using IBM Wave, the NICDEF statement in the guest's User Directory will specify F000 as the virtual NIC address, and all Linux configuration files will refer to the F000 virtual address.

Press the Update button to complete the process. The system will respond by updating the properties for the VSWITCH and returning to the Network panel of the main display. We will proceed to do the same steps on the reminaing two VSWITCHes in order to remove the alert condition from them as well. When updates have been completed on all of the VSWITCHes the icons on the network display will no longer show the attention alert condition.

Update Virtual To Real Ratio ThresholdSince memory is constrained in our lab environment, the virtual to real status gauge showed red as well. Since we intend to run this system with a fairly high level of memory over commit we can update the threshold so that IBM Wave does not display this as an alert condition.

The threshold values are part of the global parameters maintained by IBM Wave. To modify these parameters click Administration on the menu bar then select Manage Parameters as shown in Figure 4-64 on page 63.



Figure 4-64 Manage Parameters

Clicking on Manage Parameters will cause the panel shown in Figure 4-65 .

Figure 4-65 Virtual To Real Threshold Defaults

We modified the defaults for our environment as shown in Figure 4-66 .



Figure 4-66 Updated Virtual To Real Thresholds

After entering new values for the thresholds, press the Update button to save and activate these new values.

The System Status tab also showed yellow alerts for the new device pools discovered during the auto detect process. The alerts indicate that the new device pools do not have default devices specified. We can eliminate these alerts by adding default devices to the device pool objects.

Update Device PoolsIn order to update the device pool objects click on Administration in the Menu Bar, and select Manage Device Pools as shown in Figure 4-67 on page 65.



Figure 4-67 Manage Device Pools

When you click on Manage Device Pools the system will display the Device Pool Manager panel. Each line in the panel represents a new device pool discovered during the auto detect process. Click on one of the lines to select that line, then right click to open a context menu and select Update Details as shown in Figure 4-68.

Figure 4-68 Update Details

When you click Update Details the system will display an Update Device Pool panel. Enter a value for Default Virtual Device and click on the Update button as shown in Figure 4-69 on page 66.



Figure 4-69 Add Default Virtual Device

When the Update button is pressed the system will add the default virtual device to the particular device pool and return to the Device Manager panel. Continue entering default devices for all of the new device pools.

After updating all of these items, the System Status Tab and the Attention Required area of the IBM Wave Viewer should not longer show attention items similar to Figure 4-70.

Figure 4-70 Updated Status



Mark WAVESRV As IBM Wave Service MachineOne last task to complete after installation is to mark the WAVESRV virtual machine (the Linux virtual machine running the IBM Wave background services) as an IBM Wave service machine.

IBM Wave for z/VM is aware that the worker virtual machines are part of the product since those virtual machines were created during the auto detect process. However IBM Wave for z/VM is not yet aware of which virtual machine the IBM Wave for z/VM background processes are running in. In order to mark the WAVESRV virtual machine in this manner go the the z/VM User Groups display in the Current System View and expand the group USER-LOCAL. Right click on the icon for WAVESRV and select More Actions and Mark As IBM Wave Service Machine from the context menu as shown in Figure 4-71.

Figure 4-71 Mark as IBM Wave Service Machine

When you click on Mark As IBM Wave Service Machine, a confirmation panel will be displayed. Press the Go button as shown in Figure 4-72 on page 68.



Figure 4-72 Mark Service Machine Action

Pressing the Go button will cause the system to lock the virtual machine from further changes and move it to the group of virtual machines designated as WAVE-INTERNAL. The action panel will reflect the outcome of the action similar to Figure 4-73 on page 68.

Figure 4-73 Mark Service Machine Complete

Press the Close button to return to the main display. Expanding the WAVE-INTERNAL group icon will show that the WAVESRV virtual machine is now part of that group similar to Figure 4-74 on page 69.



Figure 4-74 Updated WAVE-INTERNAL Group.

At this point, the alerts resulting from the Auto Detect process have been addressed. IBM Wave for z/VM has been installed and a z/VM system is now ready to be managed.


Draft Document for Review February 26, 2014 8:58 am 8192p02.fm

Part 2 Using IBM Wave for z/VM

In this part we describe actually using IBM Wave for z/VM to manage your environment. We discuss such tasks as:

! How to manage and operate the IBM Wave for z/VM graphical user interface. ! How to use IBM Wave for z/VM to manage z/VM! How to manage the Linux virtual machines

We discuss such topics as creating a z/VM guest, installing Linux in a virtual machine, creating a golden master and cloning virtual machines from a golden master. All done with the use iof IBM Wave for z/VM.

Part 2

8192p02.fm Draft Document for Review February 26, 2014 8:58 am



Draft Document for Review February 26, 2014 8:58 am 8192HowtoManageOperateUI.fm

Chapter 5. How to manage and operate the user interface environment

This chapter provides an introduction to the graphical user interface (GUI) for IBM Wave for z/VM, and then covers a number of tasks that exploit the GUI. We will go through the process of creating groups, projects and custom attributes and see how these objects help manage a z/VM system. We will then go through the process of creating IBM Wave for z/VM users and assigning scopes, permissions, group associations, and project associations. Lastly we will cover creating reports and backup functions available through the GUI.

5

8192HowtoManageOperateUI.fm Draft Document for Review February 26, 2014 8:58 am


5.1 The IBM Wave for z/VM User Interface

When you login to IBM Wave for z/VM, the first display you will see is the main IBM Wave screen. Figure 5-1shows that screen with annotations to various sections that are described following the figure.

Figure 5-1 IBM Wave Main Screen

5.1.1 The Title Bar

Figure 5-2 shows the main display title bar.

Figure 5-2 IBM Wave Title Bar

The Title Bar contains basic information about the IBM Wave for z/VM server and buttons to manipulate the primary IBM Wave window. From left to right the Title Bar contains:

! The IBM Wave icon! The version of IBM Wave! The hostname of the server running IBM Wave! The IP address of the server running IBM Wave! The minimize, maximize, and close window buttons

5.1.2 The Menu Bar

The Menu Bar contains a variety of drop-down submenus that provide a variety of functions. While the large number of submenus may seem intimidating many of these menus contain

z/VM environment viewers

IBM Wave viewer

Status Bar

Tool Bar

Title BarMenu Bar

Chapter 5. How to manage and operate the user interface environment 75


only one function. Additionally, most of these functions can be accessed through other means, such as the Tool Bar or context menus in various viewers.

The File MenuFigure 5-3 shows the File Menu.

Figure 5-3 The File Menu

The File Menu contains a single action to Exit IBM Wave for z/VM.

The Auto Detect MenuThe Auto Detect Menu contains all of the functionality for autodetecting new z/VM systems and refreshing Viewers for IBM Wave managed z/VM systems. Because all of these functions operate at the level of a specific z/VM system or member, all selections are greyed out and inaccessible unless a particular z/VM system is selected in the Hardware Viewer. Figure 5-4 shows the Auto Detect Menu.

Figure 5-4 Auto Detect Menu

The 'Run Autodetect Wizard' action is used to perform the initial configuration needed to allow IBM Wave to manage a z/VM system. The process discovers all resources associated with a z/VM System and builds the IBM Wave metadata repository. This action is only accessible if a z/VM System is selected in the Hardware Viewer that has not yet run the Autodetect Wizard.

The Refresh Submenu provides the capability to refresh specific metadata for a selected z/VM system. Since IBM Wave updates its metadata repository on a periodic basis, these selections provide a manual method of ensuring that the metadata repository is fresh.

User-Group, Network Management, Prototype, and Storage Management Menus

Figure 5-5 shows the User-Group, Network Management, Prototype, and Storage Management Menus.



Figure 5-5 User-Group, Network, Prototyp, and Storage Management

These menus contain various options to select which Viewer to display in the Current System View. All of the actions in these menus are only accessible when a specific z/VM system is selected in the z/VM Hardware Viewer.

The Administrative MenuThe Administrative Menu contains a variety of actions related to the IBM Wave Administrator role. This menu is only visible for IBM Wave users with the IBM Wave administrator role. Figure 5-6 shows the Administrative Menu.

Figure 5-6 The Administrative Menu

! The Site Management Submenu contains a variety of actions related to managing physical resources, such as mainframe CPCs, z/VM systems, routers, z/OS system, Linux repositories, device pools, FCP devices, and network segments. It also contains actions for configuring custom attributes for managed guests and managing z/VM account codes.– The External Entitities Manager action is used to define metadata for major entities that

exist external to IBM Wave, including z/OS or z/VM systems, storage controllers, and routers.

– The Manage Device Pools action is used to create and manage pools of DASD, OSA, HiperSockets, and FCP devices.

– The Manage IBM Wave Virtual Network Segment action is used to the virtual network topology of the z/VM environment to IBM Wave.

– The Custom Attribute Manager action is used to define possible custom attributes that IBM Wave users can assign to guest virtual machines.

– The z/VM Directory Manager action is used to view and update the z/VM directories managed by IBM Wave.



– The z/VM Account Manager action is used to create, update, and delete z/VM accounts associated with virtual machines.

– The AGC Manager action is used to configure IBM Wave's Automatic Guest Configuration Engine, which is used to create rules that either allow auto-discovery to automatically assign IBM Wave metadata or allow changes to IBM Wave metadata to make changes to the z/VM Directory. At this time the only object that is changed in the z/VM Directory is the ACCOUNT statement.

– The Manage FCP Information action is used to assign WWPNs to FCP storage controllers in order to allow IBM Wave to manage FCP disk.

– The IBM Wave Linux Repository Manager action is used to create, modify, and delete Linux Repositories for use by the IBM Wave Bare Metal Installation function.

– The Add New CPC action is used to add a new mainframe CPC to IBM Wave.– The Delete CPC submenu is used to fully delete an existing CPC from IBM Wave. The

submenu is populated with each of the CPCs managed by IBM Wave.– The Tools menu contains the Import Guest Information action, which allows an IBM

Wave user to import guest metadata from a CSV file formatted according to the criteria in the IBM Wave User Guide.

! The Toggle Single User Mode action puts IBM Wave in single user mode, which is a mode where the IBM WAVE administrator is the only IBM Wave user currently logged into IBM Wave. This mode is required for certain IBM Wave actions, such as shutting down a z/VM system.

! The Backup IBM Wave Database task is used to export the IBM Wave metadata repository to unencrypted flat files.

! The Manage IBM Wave Users action is used to create or delete IBM Wave users and assign users roles, privileges, and scope.

! The Manage IBM Wave User Profiles action is used to create User Profiles to associate with LDAP groups when using LDAP for access control.

! The Project Manager action is used to create or delete IBM Wave projects.! The View Logged In Users action is used to see which users are logged into IBM Wave

and optionally force the user to log off if the user is hung because the end-user did not gracefully log out of IBM Wave.

! The View WRS Elements action views all of the active IBM Wave Resource Serialization (WRS) Elements, which are the locks on various IBM Wave elements. WRS elements are used to make sure that certain panels are locked when a IBM Wave user has it open to prevent two users from submitting conflicting requests to the IBM Wave server at the same time. However, it is possible that a WRS element can get hung, preventing any user from opening a particular panel. This action allow the IBM Wave administrator to forceably terminate these WRS elements in case they get hung.

! The BTS Manager action allows an administrator to display and manipulate a variety of information about IBM Wave's Background Task Scheduler, including connected users, the status of each IBM Wave worker thread, the schedule which determine how often IBM Wave periodic updates occur, historical statistics for all workunits submitted to the BTS, and a log of all BTS activity.

! The Send Message and Broadcast Messages actions allow a IBM Wave administrator to directly message a specific IBM Wave user or broadcast a message to all IBM Wave users. This is useful when notifying about systems coming down or when requesting that users log off IBM Wave to allow the administrator to enter single user mode for system-wide actions.

! The Recycle Service Machines and Recycle API Server actions are used to recycle the WAVEWRK* service machines and the z/VM Service Management API respectively.

! The Manage Parameters action is used to configure a wide variety of system-wide IBM Wave settings.



The User Tasks MenuThe User Tasks Menu contains actions that regular IBM Wave users are able to perform. Figure 5-7 shows the User Tasks Menu.

Figure 5-7 User Tasks Menu

! The View Log action allows a IBM Wave user to diplay all of the logs produced by the Wave server.

! The Change IBM Wave User Password action allows a user to change his/her password.! The Change IBM Wave User Preferences action configures user settings for the current

IBM Wave user. This includes display settings, configuration settings for an external SSH terminal, and SSH and 3270 connection settings.

! The Display Effective Scope/Permissions for Session action displays the IBM Wave's scope and permission within IBM Wave, which determines which actions the user is able to perform.

! The IBM Wave User Scripts Manager allows a user to create or delete Linux Shell Scripts that the Wave server can execute across the managed z/VM environment.

! The Display IBM Wave Managed Entitities, Display IBM Wave Virtual Network Segments, Display IBM Wave Linux Repositories, and Display IBM Wave Messages actions give a normal IBM Wave user read-only access to a subset of the elements that the IBM Wave administrator configured using the actions in the Administrative menu.

! The Init Users for IBM Wave User action configures a z/VM Linux guest to be eligible for management by the IBM Wave interface. This action must be completed to be able to manipulate the guest using IBM Wave interfaces.

! The Show Changes Log action shows the changes that the IBM Wave user has executed during the current session.

! The Trace GUI action toggles on or off if the client-side IBM Wave GUI is outputting status and error messages to the local Java console.

! The Gather Debug Information action is used to generate a large report that summarizes the current state of the installed IBM Wave environment to allow IBM to troubleshoot IBM Wave errors.



The Reports MenuThe Reports Menu contains the Report Manager action, which allows an IBM Wave user to generate printable reports showing the status of selected aspects of the z/VM environment. Figure 5-8 shows the Reports Menu.

Figure 5-8 The Reports Menu

More will be written about reports later in this chapter.

The Window MenuThe Window Menu contains actions for controlling how viewers are displayed. Figure 5-9 shows the Window Menu.

Figure 5-9 The Window Menu

! The Overview action provides a tool to more easily control the zoom level and positioning of the Current System View

! The Bottom Panel Viewer action toggles whether the IBM Wave Viewers are displayed or minimized

! The Left Panel Viewer action toggles whether the tabbed element containing the Hardware Viewer, Enterprise Viewer, and Dashboard Viewer is displayed or minimized.

The Help MenuThe Help Menu contains information about the versions of IBM Wave and your JVM and a few links to IBM Wave documentation. Figure 5-10 shows the Help Menu as well as the About Wave panel.

Figure 5-10 The Help Menu

! The About action shows version information about IBM Wave.! The About Java action shows the version of the Java Virtual Machine running on your

workstation.



! The View User Guide opens the IBM Wave User Guide in PDF form.! The View Administration Guide opens the IBM Wave Administration Guide in PDF form.

5.2 The Tool Bar

The IBM Wave Tool Bar is composed of small icons that execute a variety of actions. Some of the actions are only visible to IBM Wave Administrators. Figure 5-11 on page 80 shows the Tool Bar.

Figure 5-11 Tool Bar

From left to right, the Tool Bar functions are as follows:

! "The Icon of a Door executes the Exit action to logoff your IBM Wave user and exit the IBM Wave program. This functionality is also accessible by File ' Exit in the Menu Bar or by the close button on the Title Bar.

! "The Icon of a Book executes the View Log action to to diplay all of the logs produced by the Wave server. This functionality is also accessible by User Tasks ' View Log in the Menu Bar or double-clicking on the IBM Wave Log tab in the IBM Wave Environment Viewer.

! "The Icon of a Person with Pencil executes the Change IBM Wave User Preferences action, which configures user settings including display settings, configuration settings for an external SSH terminal, and SSH and 3270 connection settings. This functionality is also accessible by User Tasks ' Change IBM Wave User Preferences in the Menu Bar.

! "The Icon of Two People with a Pencil executes the Manage IBM Wave Users action, which is used to create or delete IBM Wave users and assign users roles, privileges, and scope. This action is only available to Administrative users. This functionality is also accessible by Administrative ' Manage IBM Wave Users in the Menu Bar.

! "The Icon of Bullet Points and Lines executes the Manage IBM Wave User Profiles action, which is used to create User Profiles to associate with LDAP groups when using LDAP for access control. This action is only available to Administrative users. This functionality is also accessible by Administrative ' Manage IBM Wave User Profiles in the Menu Bar.

! "The Icon of a Flow Chart and a Triangle executes the Project Manager action, which is used to create or delete IBM Wave projects. This action is only available to Administrative users. This functionality is also accessible by Administrative ' Manage Projects in the Menu Bar or the Context Menu when the z/VM User Groups tab of the Current System Viewer is grouped by project.

! "The Icon of a Pencil and a Spreadsheet executes the Manage Parameters action, which is used to configure a wide variety of system-wide IBM Wave settings. This action is only available to Administrative users. This functionality is also accessible by Administrative ' Manage Parameters in the Menu Bar.

! "The Icon of a Cylinder and a Spreadsheet executes the View WRS Elements action, which is used to view and clear the locks on various IBM Wave elements. This action is only available to Administrative users. This functionality is also accessible by Administrative ' View WRS Elements in the Menu Bar.

! "The button labeled Stop Updates is used to halt and queue all of IBM Wave's periodic update process to ensure that the IBM Wave metadata repository reflects the current state of the z/VM environment. Once stopped, this button is relabeled as 'Process Updates' and executes the queue of IBM Wave updates that had been halted. This action is only available to Administrative users, and is the only functionality on the Tool Bar that cannot be executed through other means in the IBM Wave GUI.



5.3 System Viewers

The System Viewers are an important piece of the IBM Wave interface, as they are the location where systems management tasks are executed. Figure 5-12 on page 81 shows the System Viewers.

Figure 5-12 System Viewers

There are three graphical elements related to System Viewers:

! "The Viewer Navigator is in the top left. It is subdivided into tabs for the Hardware Viewer, the Enterprise Viewer, and the Dashboard Viewer. These three viewers are the high level organization for this portion of the GUI and are used to select and filter what IBM Wave displays in the View Area.

! "The View Area is on the right. It displays the view associated with what is selected in the Viewer Navigator. The Current System View corresponds to the z/VM system selected in the Hardware Viewer. The Enterprise View corresponds to the filter criteria set in the Enterprise Viewer. The Dashboard View corresponds to the filter criterial set in the Dashboard Viewer.

! "The Property Viewer displays tabs of information related to items selected in both the Viewer Navigator and the View Area.

5.3.1 The Hardware Viewer

The Hardware Viewer visualizes mainframe Central Processor Complexes (CPCs), instances of the z/VM hypervisor, and Single System Image (SSI) clusters. Figure 5-13 shows the context menus available from the Hardware Viewer.



Figure 5-13 Hardware Viewer

Right clicking on white-space in this area opens a context menu that allows a IBM Wave administrator to add a new mainframe server through the Add New CPC task.

Right clicking on a mainframe CPC opens a context menu that allows a IBM Wave administrator to delete the CPC, update CPC information (such as model and CPUID), and add a z/VM system.

Right clicking on a z/VM system opens a context menu that allows a IBM Wave administrator to delete the system from the IBM Wave metadata repository, update the z/VM system's metadata, purge the z/VM system's spool, add page or spool space to the system, and shutdown (and optionally restart / re-IPL) the entire z/VM system.

Left clicking on any element in this view opens a property tab for that element in the Property Viewer. Additionally, left clicking on z/VM systems opens a Current System View of that z/VM system in the View Area.

5.3.2 The Current System View

The Current System View shows the system currently selected in the Hardware Viewer.

z/VM User Groups TabThe z/VM User Groups tab of the Current System View displays the guest virtual machines running on the currently selected z/VM system. Through the Group By Drop-down menu, the guest virtual machines can be organized by Site Defined Groups, Projects, Operating System



Distribution, or z/VM Account. Figure 5-14 on page 83 shows the z/VM Groups Tab, and the context menus available at the group level.

Figure 5-14 z/VM Groups Tab

Right clicking on white space in this view opens a context menu that allows the IBM Wave user to arrange items in order to clean up the diplay, collape or expand all groups, or add a new instance of the group selected in the Group By menu.

Right clicking on a group opens a context menu that allows the IBM Wave user to diplay information on the group or collapse/expand the group. Collapsing and expanding the group is also accomplished by double clicking on the group.

Right clicking on a virtual machine within a group opens a context menu that allows many actions on the virtual machine as shown in Figure 5-15 on page 84.



Figure 5-15 Virtual Machine Context Menus

Right clicking on a guest virtual machine opens a context menu that allows for executing just about any possible IBM Wave function on this guest. Here is a brief breakdown of functionality in this context menu and submenus:

! Display Information on all aspects of the guest from the hypervisor to guest operating system to IBM Wave metadata.

! Update Information related to the guest's z/VM directory entry (virtual memory, processors, guest password)

! Assign and Remove Metadata, including OS Distibution, Projects, Site Defined Groups, Custom Attributes, accounts, and functionality

! Definine the Default z/VM System that you would like this guest to IPL (boot) on when starting a z/VM system.

! Access the guest via read-only or full-featured 3270 terminals.! Access a Linux shell via SSH or through a feature called CLC that screen-scrapes the

guest's z/VM console (intended for emergency access when TCP/IP network connectivitiy in unavailable).

! Create a new guest by cloning a guest's z/VM directory entry and disk volumes.! Convert a guest into an "Associated Prototype" for use as a Golden Master.! Install Linux on this guest via the Bare Metal Installation feature.! Lock or unlock the guest to prevent/allow changes via IBM Wave! Add, update, or delete a sticky note to the guest to communicate something to all IBM

Wave users.! Update metadata! Create or expand LVMs in the guest's Linux filesystem.! Execute a REXX script on the guest! Delete the guest.



Network TabThe Network tab of the Current System View displays the end-to-end virtual network topology of the selected z/VM system, including virtual LANs, network device pools, virtual switches, virtual network segments, and guests with TCP/IP stacks and IP addresses. This view can be customized by the panel on the left-hand, which allows the IBM Wave user to select predefined views, manually create a view, and pick a particular layout. Figure 5-16 shows the Network Tab.

Figure 5-16 Network Tab

Double-clicking any of the depicted elements brings up a detailed view of that element. Right-clicking on any of the elements allow a IBM Wave user to lock or unlock the selected element. The behavior of guests in this view is identical to their behavior in the z/VM User Groups panel.

Right-clicking on white-space in the panel brings up a detailed context menu that allows a IBM Wave user to define a new guest LAN or virtual switch and configure the display of the network view as shown in Figure 5-17 on page 86.



Figure 5-17 Network Tab Menus

Prototypes TabThe Prototypes tab of the Current System View shows the Prototypes defined in your Directory Manager and the Associated Prototypes defined in IBM Wave. By default when DIRMAINT is installed, two prototype definitions are supplied, one for Linux and one for CMS. These will show up in the Prototypes Tab as shown in Figure 5-18 on page 86.

Figure 5-18 Prototypes Tab

Right clicking on a prototype brings up a context menu that allows a IBM Wave user to create new guests by cloning this prototype. It also brings up options for converting the Prototype



back into a normal z/VM guest to allow the master image to be updated and options for deleting the prototype as shown in Figure 5-19 on page 87.

Figure 5-19 Prototype Menus

Storage TabThe Storage tab of the Current System View contains three different views into the storage system at various levels of granularity. These views can be selected by tabs on the lefthand side of the panel.

The first storage view is Distribution, which presents a pie chart that depicts the portion of the storage distribution in each possible status. Each "pie slice" can be double-clicked to drill down into a view of the DASD volumes with that status. Figure 5-20 on page 88 shows the distribution view in the Storage Tab.



Figure 5-20 Storage Tab Distribution

The second storage view is Groups, which shows each storage group alongside graphical and textual depictions of the amount of storage space that is free for allocation as minidisks to guest virtual machines.

Right-clicking on white space presents a context menu that allows for the creation of a new DASD Group. Right-clicking on an existing DASD Group presents options to rename the group as seen in Figure 5-21 on page 88.

Figure 5-21 Storage Tab Groups

The second storage view is Volumes, which shows actual DASD volumes. Because a typical z/VM environment has far too many DASD volumes to be effectively visualized on a single view, this view uses a filter located at the bottom of the view as seen in Figure 5-22.



Figure 5-22 Storage Tab Volumes

Right-clicking on a specific DASD volume brings up a context menu that allows a IBM Wave user to execute a wide variety of tasks, including:

! Assigning or Unassigning the volume to or from a DASD Group.! Defining or Undefining a Region on the volume! Varying the DASD volume on or offline! Marking or unmarking the DASD as page or SPOOL space for z/VM! Adding the DASD volume to z/VM as a page or SPOOL volume.

System Status TabThe System Status Tab of the Current System View displays performance and storage information related to the guest and any administrative actions that IBM Wave believes that a IBM Wave administrator needs to complete. All of the panels can be double-clicked to bring up a panel with more detailed information. The "Objects that require attention" can be right-clicked to bring up a context menu to clear or ignore the item. Figure 5-23 shows the System Status Tab.



Figure 5-23 System Status Tab

Session Tasks TabThe Session Tasks Tab diplays activities that are executing on multiple z/VM systems in parallel. When the environment consists of a single z/VM system, this tab always displays the string "No Session Tasks Exist” as shown in Figure 5-24 on page 90.

Figure 5-24 Session Tasks Tab

5.3.3 The Enterprise Viewer

The Enterprise Viewer and View differs from the Hardware Viewer and Current System View in that it displays information across all instances of the z/VM hypervisor. This enables an enterprise-wide view of z/VM LPARs as shown in Figure 5-25.



Figure 5-25 Enterprise Viewer

The left-side of the Enterprise Viewer area consists of an area that allows for one or more filter criteria to be applied to the view in the Guests tab on the right. This is an ideal way to visualize virtual resources across z/VM systems by project, account, or other such attribute.

The functionality in the Guests tab of the Enterprise View is identical to the User Group View of the Hardware View. The only exception is that the Enterprise View is the only location that can be used for IBM Wave's drag-and-drop implementation of Live Guest Relocation between z/VM systems.

5.3.4 The Dashboard Viewer

The Dashboard Viewer and View displays a single screen that shows performance data for all of the z/VM systems managed by IBM Wave as shown in Figure 5-26.

Figure 5-26 Dashboard Viewer and View

Double-clicking on the z/VM System Name column or selecting Expand All or Collapse All alternates the view between the analog dials shown above and the following compressed view, which is able to show more data on a single screen as seen in Figure 5-27.



Figure 5-27 Dashboard View Compressed

Double clicking on the dials or % fields brings up more detailed performance information and a textual view of the z/VM Performance Toolkit data used to build these views.

When the Dashboard View tab is open, the Viewer area on the left becomes a filter that can be used to limit the report to particular z/VM system or mainframe hardware.

5.4 The IBM Wave for z/VM Viewers

The IBM Wave Environment Viewers are focused on IBM Wave concepts, such as messages, Backgroup Task Scheduler (BTS) data related to workunits submitted by IBM Wave actions, and lists of IBM Wave objects that require manual intervention by a IBM Wave administrator.

The IBM Wave Log tab displays messages received from the Wave server or users with the IBM Wave administrator role during. The full message can be displayed by double clicking on the row of the desired message. By double-clicking on the IBM Wave Log tab, a IBM Wave user executes the View Log action, which allows users to search for messages that were received during previous sessions. Figure 5-28 shows the IBM Wave Log tab.

Figure 5-28 BTS Log Tab

The Background Task Scheduler (BTS) Work Units tab displays a list of the workunits submitted to the BTS by all IBM Wave users. Work units can be double-clicked for more information, including a step-by-step list what of the workunit did to the system. A work unit can be deleted from the repository by right clicking on a work unit and selecting the Delete BTS Workunit action from the context menu. Figure 5-29 shows the BTS Work Unit Tab.

Figure 5-29 BTS Work Units

The Background Task Scheduler (BTS) System Common Output Repository (COR) tab displays a list of output messages from the BTS logged to the current COR log. The rows in this tab cannot be double-clicked or right-clicked. Double-clicking on the BTS System COR tab itself opens a panel that allows a IBM Wave user to browse the contents of older COR log files. Figure 5-30 shows the BTS Common Output Repository Tab.



Figure 5-30 BTS COR

The Background Task Scheduler (BTS) Log tab contains a human-readible log of the processing done by the BTS. By right clicking on the text and selecting the Scroll to Bottom when Messages Arrive action, this log will function like a console and automatically scroll as messages arrive. Double-clicking on the BTS Log tab itself opens the BTS Manager, which configured the settings for the BTS and stores all BTS logs. Figure 5-31 on page 93 shows the BTS Log Tab.

Figure 5-31 BTS Log Tab

The Attention Required tab displays a list of all actions that IBM Wave believes requires manual IBM Wave administrator intervention. Right-clicking on a row opens a context menu that allows a IBM Wave administrator to ignore the issue or change the numeric severity level. Figure 5-32 shows the Attention Required Tab.

Figure 5-32 Attention Required Tab

5.5 The Status Bar

The Status Bar displays a few items shown in Figure 5-33.



Figure 5-33 Status Bar

From left to right, the Status Bar has:

! A Message Area that displays messages from the IBM Wave client - in this case it indicates that SSL has been enabled

! A cell that shows the current IBM Wave user.

5.6 Group Management

IBM Wave for z/VM is designed to help System Programmers and System Administrators manage even the most complex z/VM environments. A large z/VM system may consist of hundreds or thousands of virtual machines. Keeping track of all of these virtual machines, and managing them requires some mechanism to summarize what is in the directory. IBM Wave for z/VM accomplishes this by assigning virtual machines to groups during the auto detect process, and then providing the capability to create site defined groups for better granularity.

The grouping capability only exists within the context of IBM Wave, and does not involve any modifications to a managed z/VM system. The default groups for a small z/VM 6.3 system are shown in Figure 5-34.

Figure 5-34 Default Site Defined Groups

IBM Wave for z/VM places virtual machines that are part of the z/VM product installation in particular groups so that related virtual machines are kept together. For example the IBM-SMAPI Group contains all of the z/VM System Management API requestor and worker



virtual machines. The USER-LOCAL group contains all of the virtual machines that IBM Wave was not able to place in a specific group during the auto detect process. On a large z/VM system this group might contain many virtual machines. It is typically the content of this group that you will move to new groups that you create to better categorize the virtual machines running on a managed z/VM system.

In order to create a new group, simply right click on any portion of the white background of the z/VM User Groups tab. This will open a context menu with a selection to create a new group as seen in Figure 5-35.

Figure 5-35 New Site Defined Group

Clicking on the Add New Site Defined Group option will cause a panel to be displayed to collect information about the new group as shown in Figure 5-36.

Figure 5-36 Create New Site Defined Group



In the panel displayed you must give the new group a name. Providing a Full Name, or Description is optional. When the Create button is clicked a new group will be created and displayed in the z/VM User Groups tab as seen in Figure 5-37.

Figure 5-37 New Group Created

Once a new group is created, new virtuaul machines can be created it in, and existing virtual machines can be moved from an existing group to the new group. To move an existing virtual machine to the new group, right click on the virtual machine and choose Update, Assign Site Defined Group, and select the group from the list, as shown in Figure 5-38 on page 96.

Figure 5-38 Assign Site Defined Group



It is also possible to expand the existing and new groups, select the virtual machine from the existing group, hold down the Ctrl key and drag the virtual machine icon to the new group as shown in Figure 5-39.

Figure 5-39 Drag To New Group

As is evident from the figures above, IBM Wave for z/VM groups are local to a particular z/VM system. The next section on Project Management will cover organizing virtual machines across z/VM systems and CPCs.

5.7 Project Management

Since Site Defined Groups only exist within the context of a single z/VM system, Site Defined Groups are of limited use for organizing virtual machine in clustered environments or environments that span multiple LPARs or CPCs. For these sorts of complex environments, IBM Wave for z/VM provides the concept of projects, which allow categorization and logical groupings of virtual machines and other virtual resources across multiple z/VM systems, and CPCs.

Projects are also IBM Wave’s primary means of limiting the scope of IBM Wave users. This will be discussed in greater detail later.

Projects are created using the Project Manager from the Administrative pull down in the Menu Bar as shown in Figure 5-40.

Note: a virtual machine may only belong to one group.



Figure 5-40 Project Manager Navigation

Clicking on Project Manager will display the Project Manager panel listing all of the projects currently defined as shown in Figure 5-41 on page 98.

Figure 5-41 Project Manager Panel

Click on Add New Project to create a new project as shown in Figure 5-42 on page 99.



Figure 5-42 Add New Project

On the Add New Project panel enter a name for the project, description, contact name, phone, and e-mail. Enter also the Project Team and Project Department. Click the Add button and the system will create the new project and return to the project manager panel as shown in Figure 5-43 on page 99.

Figure 5-43 Project Added

Once a project has been created it is possible to assign virtual machines to the project, and give IBM Wave for z/VM users permission to the project. In order to assign a virtual machine to a project, right click on the icon for a particular virtual machine. Click Update, then click Assign Project, and select a project from the displayed list as shown in Figure 5-44 on page 100.



Figure 5-44 Assign Virtual Machine To Project

When a project is selected, a BTS transaction is started to associate the virtual machine with the selected project. The benefit of assigning virtual machines to projects is seen when using the Enterprise viewer with the “Group by” filter set to Project as shown in Figure 5-45.

Figure 5-45 Enterprise Viewer Projects

The project named “No Project Assigned” contains all of the virtual machines throughout the enterprise that have not yet been assigned to projects. To see the virtual machines assigned to the project created earlier (ITSOPROJ1) expand the project icon as shown in Figure 5-46 on page 101.



Figure 5-46 Expand Project Icon

The virtual machines in the example project above are each from different z/VM LPARs.

5.7.1 Automatic guest classification

When you implement projects to help organize and manage virtual machines across your enterprise, one of the challenges becomes maintaining the association between virtual machines and projects. This is where the Automatic Guest Classification (AGC) capability of IBM Wave can assist. With AGC you can setup an association between z/VM account codes assigned to virtual machines with the ACCOUNT directory statement and projects defined within IBM Wave using the Project Manager.

AGC will then automatically change the project to which a virtual machine is associated if a change occurs to the z/VM account code associated with the virtual machine, and if a change to the project associated with a particular account code changes, then IBM Wave will make sure that the directory information for virtual machines associated with that project is updated to reflect the new value on the ACCOUNT directory statement.

By default when IBM Wave for z/VM is installed, the AGC functionality is not enabled. Before you can make use of this capability, you will need to enable the AGC functionality. This is done by clicking Administration from the Menu Bar, and then selecting Manage Parameters. Click the Functionality tab at the top of the Manage Parameters tab, and then near the bottom of the panel click the check box for Automatic Guest Classification as shown in Figure 5-47.



Figure 5-47 Manage parameters using AGC

Press the Update button to save the changes, and activate the AGC capability.

Before defining AGC entries you need to make sure that the projects to be associated with z/VM ACCOUNT record values have been defined. In the previous section we discussed the creation of projects. We will now use the AGC Manager to create the association between an ACCOUNT value and a particular project.

The AGC Manager is started by Clicking the Administration entry in the Menu Bar, then selecting Site Management and AGC Manager as shown in Figure 5-48 on page 103.



Figure 5-48 Navigate To AGC Manager

When you click on AGC Manager, a new panel will be opened similar to Figure 5-49.

Figure 5-49 AGC Manager Panel

Click Create New AGC Entry to define a new criteria for automatic guest classification. The system will respond with a panel to enter the AGC information similar to Figure 5-50 on page 104.



Figure 5-50 Create New AGC Entry

Give the new entry a Name, then select an AGC Value from the selection box. The values are all of the various values discovered on ACCOUNT statements when the USER Directory was processed during the auto detection of the particular z/VM system. In our example we want to associate the ACCOUNT value of SERVICE with a Project named Service. Entering a description of the AGC entry is optional. You will then right click in the area at the bottom of the panel to bring up the selection of Create new AGC Property association. Clicking on this selection will open a panel where you can select the project to associate with the AGC value selected, similar to Figure 5-51 on page 104.

Figure 5-51 AGC Entry Data Properties

In the Property selection box you can choose from Project, OS Distribution, Default z/VM System, Lock, Site Defined Group, or Functionality. Functionality is an attribute assigned to virtual machines which is itself defined in the global IBM Wave parameters. In our example we are choosing the Project Value of Service to associate with the AGC entry. Press the Add button to return to the Create New AGC panel as shown in Figure 5-52 on page 105.



Figure 5-52 Create AGC Entry Go

When you press the Go button, the new AGC entry will be created and displayed on the left side of the AGC Manager panel. Click on the entry to see the details displayed on the right side of the AGC Manager panel as shown in Figure 5-53 on page 105.

Figure 5-53 Run AGC

Once the AGC entry has been created you can press the Run AGC button to submit a BTS entry to process all virtual machines and apply the automatic classification criteria just defined. When you press the button a confirmation panel will be displayed simlar to Figure 5-54.



Figure 5-54 Run AGC Confirmation

Press the Yes button to proceed with running AGC against all z/VM virtual machines. It is also possible to run AGC against just one or a small group of virtual machines by selecting them from the enterprise viewer, or from a group in the Site Defined Groups display. Right click on the guests and select More Actions followed by Run AGC.

To see the results of running AGC it is easiest to switch the Group By filter on the Current System display from Site Defined Groups to Project. This will allow you to easily see what virtual machines were added to the Project associated with the AGC Entry. In our example when we display the Service Project we see a single virtual machine named TESTSUSE was added. When we display the information for that virtual machine we see that it is now associated with this project as shown in Figure 5-55 on page 106.

Figure 5-55 View Project Service

Since we setup our AGC Entry with bidirectional synchronization, if we update the account information on another virtual machine to be SERVICE, it will automatically appear in this Project folder as well. Figure on page 107 shows the result of updating the WAVE320 virtual machine ACCOUNT to be SERVICE.



Figure 5-56 Change Virtual Maching ACCOUNT

With this AGC entry, anytime the primary ACCOUNT value for a virtual machine is changed to the value SERVICE, it will automatically be associated with Project Service. If at a later time we change the AGC Value in the AGC Entry to a different ACCOUNT value then IBM Wave will change the primary ACCOUNT value in the z/VM Directory to the new value for all of the virtual machines in Project Service. If this behavior is not what is desired, it is possible to change the AGC Entry to use a synchornization method different from the default of bi-directional.

5.7.2 Defining Custom Attributes

Custom attributes apply to all virtual servers. If a defined attribute is not assigned to a virtual server, the value will be the default value “Not Assigned”.



To specify Custom Attributes and their values, choose the Custom Attribute Manager as shown in Figure 5-57.

Figure 5-57 Choose Custom Atribute Manager

Click on Add Custom Attribute and define an attribute, for example Company, and its possible values as shown in Figure 5-58.

Figure 5-58 Set values for custom attribute

To change a custom attribute or its values, from the Custom Attribute Manager, right click on the custom attribute to be changed, delete the attribute value by right clicking on the value as shown in Figure 5-59 on page 109 and add the correct value.



Figure 5-59 Update custom attributes

To assign a virtual server a custom attribute, right click on the virtual server in the z/VM guest view and choose Update, and Assign Custom Attribute as shown in Figure 5-60

Figure 5-60 Assign custom attribute

On the Assign Custom Attributes to z/VM guests panel, select the desired attribute and value and click Go as shown in Figure 5-61 on page 110. Each attribute assignment sends a request to the BTS server.



Figure 5-61 Assign custom attributes to z/VM guests

To display with filtering by custom attributes, use the Enterprise Viewer to display guests across all LPARs based on the attributes selected. In Figure 5-62, the Enterprise Viewer is used to display guests associated with Company = Colab Consultants, with Server Type = Production Server type. In the Guests tab the grouped by z/VM System is selected and a filter of Show Empty Groups = No was selected, so that LPARs with no entries would not be displayed.

Figure 5-62 Company Production Servers by z/VM Systems

5.8 Creating Reports

IBM Wave for z/VM has a facility to generate reports which provide information about managed z/VM systems. The IBM Wave Report Manager provides the capability to generate two types of reports, global reports, and private reports. User scope and permissions determine the objects included in your reports.



You must have site level administration authority to save defined global reports. Once saved, global reports can be used by all IBM Wave users. Private reports can be generated by an IBM Wave user and only the creator of the report can view, edit and delete the report. If you have site level admin authority it is possible to convert global reports to private reports, or convert private reports to global reports.

Creating and generating new reports, or generating saved reports is done using the Report Manager. The Report Manager is started by clicking on the Reports item in the Menu Bar, and then selecting Report Manager as shown in Figure 5-63.

Figure 5-63 Report Manager Navigation

Clicking on Report Manager will open the Report Manager panel as shown in Figure 5-64.

Figure 5-64 Report Manager Panel

The Report Manager panel shows the various reports that can be created on the left side of the panel, and then lists any saved reports in the body of the panel. Currently the following report types are available.

! z/VM CPC Report - shows information about all CPCs under IBM Wave management.



! z/VM System Report - shows information about all z/VM LPARs defined to IBM Wave! z/VM User Report - shows information about each virtual machine across all IBM Wave

managed z/VM systems, or a particular z/VM system.! z/VM GLAN/VSWITCH report - shows virtual switch and guest lan definition information.! z/VM GLAN/VSWITCH connection report - shows network related information for virtual

switches and guest lans.! z/VM Prototype report - shows information related to prototypes such as a virtual machine

associated with the prototype, and other descriptive information.! z/VM DASD Group report - shows information about all DASD groups defined across all

managed z/VM systems.! z/VM DASD Volume report - shows information about all DASD volumes available in all

managed z/VM systems.! Flagged z/VM Object report - shows a list of all objects that require attention in all

managed z/VM systems.! IBM Wave User report - lists all of the users defined to IBM Wave along with the attributes

associated with each user.! IBM Wave User Permission report - lists access level, permissions and scopes for all IBM

Wave users, or lists IBM Wave users with access to a particular project or other managed object.

To create a new report Click File and then select the report from the context menu as shown in Figure 5-65.

Figure 5-65 Create z/VM Users Report

Clicking the report name will cause the system to display a panel to customize the content of the report. We selected the z/VM User Report type as shown in Figure 5-66 on page 113.



Figure 5-66 z/VM User Report Panel

This panel allows you to restrict the report output to particular systems, projects, distributions, or groups. In addition you can include both Active and Inactive virtual machines, or just Linux only virtual machines. You can also customize the column data by selecting or deslecting various items in the Table Fields portion of the panel. To see all Active and Inactive virtual machines across all systems we put an asterisk in the User Name field and pressed the Generate button as shown in Figure 5-67 on page 114.



Figure 5-67 z/VM User Report Output

The report output is placed in the center of the panel with two lines showing how much of the information will be printed with standard letter size paper or A4 paper in landscape mode before output is folded. At this point, you might want to deselect some columns to make output better fit a printed report, or simply leave the report content as is.

Clicking on File in the upper left corner displays choices to print the report output or save the report criteria for easy generation at another time. To save the report click File, then Save As. This will result in the display of a small panel to assign the report a name and provide a description of the report as shown in Figure 5-68 on page 114.

Figure 5-68 Save z/VM Users Report



You must have site level authority to save the report as a Global Report. By saving the report as a Global Report it is available for all IBM Wave users to generate. When confirmation is received that the report was saved it will show up in the Report Manager window as seen in Figure 5-69.

Figure 5-69 New Saved Report

To generate the saved report simply double click on the line for the report. This will result in the report generation panel seen earlier, with all of the selections made prior to the report being saved. When the generate button is clicked, the report content is displayed in the center of the panel as before. Figure 5-70 shows the generated report.



Figure 5-70 Generate Saved z/VM Users Report

To print the report, click File, then Print. This will cause a print preview panel to display as shown in Figure 5-71 on page 117.



Figure 5-71 Print Preview

Clicking Report on the Print Preview panel provides a choice to send the previewed report to a printer.

5.9 User mangement

In a traditional z/VM environment, z/VM systems programmers often carry the burden of administering both hypervisor and guests. This is because traditional 3270-based CMS interfaces often prove too foreign and complex for Linux systems administrators working in large multi-hypervisor environments. While other hypervisors have offered methods of easily configuring guests through GUI-based management consoles, z/VM has lacked this sort of systems management interface. This led Linux administrators and middleware owners to rely of z/VM systems programmers for tasks that typically are handled though self-service systems management GUIs on other hypervisors. Need more processors, memory, or disk? Ask the systems programmer. Need to reset the z/VM user password? E-mail the systems programmer. For environments with strong z/VM systems programming staff, this wasn’t an issue. However, for new mainframe customers consolidating Linux workloads onto System z, the complexity of traditional z/VM management tools was a stumbling block.

The introduction of IBM Wave for z/VM changes this dynamic by providing a graphical management tool that provides functionality suitable for different classes of users. Middleware owners can activate, deactivate, and reset the guests running the software that they are responsible for. Linux administrators can provision guests from golden images, resize LVMs, and SSH into guests to install packages. z/VM systems programmers can configure VSwitches and DASD storage pools, write advanced REXX EXECs, launch 3270 sessions for traditional CMS-based tools, and manage the catalog of golden images.



Because IBM Wave offers features that appeal to many classes of users, the functionality to create different classes of users and determine permissions and scope is especially critical. After all, do you want to receive an e-mail from your database administrators informing you that he or she accidentally ran a function called “re-IPL this z/VM system” and then all connectivity to Linux guest was lost?

To prevent these sorts of issues, it is critical to consider the following questions:

→ What sorts of different roles (either IT or business unit personnel) use Linux guests running on z/VM on a regular basis?

→ What sort of functionality does your company’s Linux systems administrators regularly use in a self-service fashion with other hypervisors?

→ How skilled is the IT department at z/VM systems programming?

→ What sorts of methods do you currently have in place to segment guests between projects or business units? Is segmentation provided manually by human administration or by technical means?

The answers to these questions should help you determine the classes of users that should exist in your IBM Wave environment, the subset of functionality that each class should possess, and the preferred means of segmenting access in multi-tenant environments.

5.9.1 Overview of the IBM Wave User Manager tool

IBM Wave administrators manage the creation and modification of users (deletion of users is impossible) and the granting and revocation of scope and privileges through the IBM Wave User Manager. This tool is available from the Tool Bar by clicking the icon that resembles a pencil hovering over two people as shown in Figure 5-72.

Figure 5-72 The User Manager icon in the IBM Wave Toolbar

It is also available from the Menu Bar by selecting the Manage IBM Wave Users task from the Administrative menu as shown in Figure 5-73.



Figure 5-73 The Manage IBM Wave Users task in the Administrative Menu of the Menu Bar

Either of these methods opens the IBM Wave User Manager.

This tool offers a number of key functions:

! List all IBM Wave users through the a sortable and filterable table on the main IBM Wave User Manager window.

! Create new IBM Wave users from scratch using the Create New IBM Wave User task available from the Add New IBM Wave User button.

! Create new IBM Wave users by cloning the password, details, user type, scope, and permissions of an existing users through the Clone IBM User tool available from the context menu via the Clone the IBM Wave User action.

! Grant or revoke roles, permissions, and scope to or from a single IBM Wave user via the Update IBM Wave User panel available from the context menu via the Update IBM Wave User action.

! Grant or revoke roles, permissions, and scope by making a user’s roles, permissions, and scope identical to another user via the Copy Selected Scopes and Permissions tool. This tool is available from the context menu of the “destination” user via the “Copy scope and permissions to selected user” action.

! Audit and revoke permissions across all IBM Wave users via the IBM Wave User Permissions Cleaner tool available from the Review Scope and Permissions button.

! Modify a single user’s password or security question or force them to change their password upon their next login using the Update IBM Wave User panel available from the context menu opened by right clicking on a user in the table of existing users.

! Suspend or resume a single user via the Update IBM Wave User panel available from the context menu via the Update IBM Wave User action.

These features provide two workflows for administering users. The first workflow is creating IBM Wave users one at a time and manually setting the type, scope, and permissions of each user. While this is the more fine-tuned approach, it is time consume and does not scale well to



larger IBM Wave environments. The second workflow is creating multiple IBM Wave users in batch operations by cloning known users or adjusting user scope and permissions for many

Given these options, it is a best practice is to create “golden masters” users for each particular class of user (z/VM Systems Programmers, Linux Systems Administrators, DB2 DBAs, etc.) and use these “golden master” users to clone individual users of each class.

In order to demonstrate how to use IBM Wave to manage users and scope, this section will walk through the example of creating multiple user IDs for a WebSphere Application Server (WAS) middleware team (through the use of a project). These users will have permissions limited to the basic sorts of IBM Wave tasks that Linux administrators care about, including:

! Activating, deactivating, and resetting the Linux guests running WAS.

! Cloning from a Linux golden master preinstalled with W

! Creating and expanding Logical Volumes (LVMs)

! Accessing Linux guests via SSH.

Additionally, they will have a scope limited to viewing and interacting with guest virtual machines in their departmental project.

5.9.2 Creating a “Golden Master” User

In order to ease the task of user management, we will now create a “golden master” user that possess the roles, permissions, and scope common to all administrators on the WAS middleware team. No user will actually log into IBM Wave with this ID, but rather we will clone all instances of WAS administrator users from this “golden master.”

To create this new WAS Administrator user, click the ‘Add New IBM WAVE User’ button from the main IBM Wave User Manager window as shown in Figure 5-74 on page 121.

Note: IBM Wave also allows use of LDAP for single-signon and the use of Active Directory groups to predefine the permissions and scope associated with classes of users. For more information, see the IBM Wave Administration Guide.



Figure 5-74 Click the Add New IBM Wave User button

This opens the ‘Create New IBM WAVE User’ panel on the ‘General Details’ tab. We will now walkthrough completing this panel as shown in Figure 5-75 on page 122.

To signify that this is a “Golden Master” user, we have given this user the user name “WebSphere Admin Master.”

We assigned this user the password “password.” Although this is not a secure password, assigning a simple and memorable password to “Golden Master” users is critical because all users cloned from this “Golden Master” inherit the same password. However, by keeping this user in the status “A-Suspend” except for the brief periods when using this guest for cloning actions, this guest can remain unusable. Additionally, by checking the “Change Password On Next Login” field, users can be forced to change from the initial “password” string to something more secure.

The Security Question and Answer values have not been completed as normal. This is because we expect to clone numerous users from this “Golden Master” user. If we were to clone multiple users with the same security question, then a malicious user could access other user’s logins by using the security question and answer to reset the password. To prevent this, we have set the Security Question to “Contact IBM Wave Admin” and the password to a secret value known only by the IBM Wave admin team. The “Contact IBM Wave Admin” string informs users to contact an IBM Wave admin for help resetting their IBM Wave password. If a user updates their preferences at a later time, they can change their security question and answer to something personalized in order to be able to use the Security Question feature as intended.

The Description field is a 255 character area for entering human-readable information to make it easier to organize and identify IBM Wave users. Considering that cloned users have the same description as source users, this field is a good way to associate classes of users. This allows all instances of a class of user to be viewed together by sorting on the Description field in the IBM Wave User Manager window. We have entered the string “WebSphere Admin” to describe this class of user.



Figure 5-75 General Details Tab of Create New IBM Wave User Wizard

Click on the User Type tab to select the type of user. Within this context, the Administrator user type refers to IBM Wave administrators. Linux or Middleware administrators are considered Regular Users. The Network Admin and Site Level Admin checkboxes are associated with particular permissions.

In the case of our “WebSphere Admin Master” user, we ensure that Regular User is selected as in Figure 5-76.



Figure 5-76 User Type Tab of Create New IBM Wave User Wizard

Note that the Scope and Permissions tab is greyed and and inaccessible. In order to be able to set Scope and Permissions for our user, we must first create our user and then access this tab through the “Update IBM Wave User” action from the context menu.

So we click Create to instruct IBM Wave to use the values that we entered to create a new user. We receive a confirmation panel like Figure 5-77when this is complete.

Figure 5-77 Create New IBM Wave User Confirmation Panel

Once we return back to the main IBM Wave User Manger panel, we now see the “WebSphere Admin Master” user listed within the table of existing users. Right click on this user and select the Update IBM Wave User action from the context menu as shown in Figure 5-78.



Figure 5-78 Select Update IBM Wave User from the WebSphere Admin Master user’s context menu

The opens the Update IBM Wave User panel as in Figure 5-79 on page 125. This is nearly identical to the Create IBM Wave panel that we used earlier to create the WebSphere Admin Master user. However, this time the User Name field is greyed out and inaccessible (IBM Wave user names can never be changed after initial creation) and the Scope and Permissions tab is now accessible.



Figure 5-79 Update IBM Wave User Panel

Click on the Scope and Permissions tab and then click the Add Permission button. We are now going to give our user limited permissions to visualize z/VM systems and display the current status of the system (active, inactive, etc.). All other functionality related to z/VM systems will be restricted. The completed window should look like Figure 5-80.

Figure 5-80 Adding Permissions for z/VM Systems



Click the Add Permission button to return to the Scope and Permissions tab of the Update IBM Wave User panel. Notice that the permission and scope that we set is listed in the table. Click the Add Permission button to add another permission.

We are now going to give the user permission to perform varoius actions within the scope of the a project called “WebSphere Middleware Team” as in Figure 5-81 on page 126. This project contains the Associated Prototypes and Linux guests used by the WebSphere middleware team. Because this permission is limited to a particular project, this user cannot see any guests outside of this project. Once the panel is complete, click the Add Permission Button.

Figure 5-81 Adding Permissions for IBM Wave Projects

Once back at the Scope and Permissions tab of the Update IBM Wave User panel, click the Add Permission button. We are now going to add permissions for DASD volumes as in Figure 5-82. Specifically, we are going to give the user the permission to add disk from a DASD storage group to a LInux system. If this permission is not given, then it is functionally impossible to create new Linux guests or use DASD to expand Linux logical volumes.



Figure 5-82 Adding Permissions for z/VM DASD Storage Groups

Once you complete this permission, click the Add Permission button. Once back at the Scope and Permissions tab of the Update IBM Wave User panel, close the Update IBM Wave User panel to return to the main IBM Wave User Manager window.

You have now created a “Golden Master” user for WebSphere administrators.

5.9.3 Clone Individual Users from a “Golden Master” User.

Now that we have created a “Golden Master” user, we can easily clone new instances of this type of user.

Recall that we put our WebSphere Admin Master user in an suspended state to prevent users from using it to log into IBM Wave because it had the string ‘password’ set as the user’s password. If were were to clone new users from this “Golden Master” user at this point, the clones would also be in a suspended state and have to be activated individually. This can be inefficient for large batches of clones. It is more efficient to update the “Golden Master” users’ status to Active prior to a cloning action. Then when the cloning action is complete, the “Golden Master” user can be set back to a suspended state.

Right-click on the WebSphere Admin Master user and select Update IBM Wave User from the context menu as shown in Figure 5-83



Figure 5-83 Select Update IBM Wave User from WebSphere Admin Master’s context menu

In the Update IBM Wave User panel, click on the Active selector and click the Update button as in Figure 5-84.

Figure 5-84 Set WebSphere Admin Master to Active User Status

After a confirmation message, you will be returned to the main IBM Wave User Manager window. Right click on the WebSphere Admin Master user and select the Clone this IBM Wave User action from the context menu as in Figure 5-85.



Figure 5-85 Select Clone this IBM Wave User from WebSphere Admin User’s context menu

We are going to clone users for three new WebSphere admins, so enter 3 in the Number of users field. Because we do not want to prefix each of these usernames with the same string, leave the Base Name field blank. Your panel should resemble Figure 5-86. Finally click Update.

Figure 5-86 Set the Number of User Instances to clone form WebSphere Admin Master

This populates the table in the middle of the window with three rows. At this point, each of the rows have the value <Insert user name here> in the User Name column like in Figure 5-87, signifying that we need to enter a unique user name in each of these fields. Double click on each field, type a user name, and then hit the enter key to record your entry in the cell.



Figure 5-87 Enter User Names for each cloned IBM Wave User

Once you have entered a name for each user like Figure 5-88, click the Go button.

Figure 5-88 Click Go to begin the IBM Wave User Cloning Process

The status cell of each row will show the progress of the cloning process. Close the Clone IBM Wave User window when all rows show the status ‘Done.’

Your three new WebSphere admin IDs have been created and are ready for their initial login. However, we need to put the WebSphere Admin Master user back in a suspended state to prevent security loopholes. Right click onthe WebSphere Admin Master and select the Update IBM Wave User task from the context menu as in Figure 5-89 on page 131.



Figure 5-89 Select the Update IBM Wave User action from WebSphere Admin Master’s context menu

Click on the “A-Suspend” selector and click the Update button as in Figure 5-90.

Figure 5-90 Set WebSphere Admin Master to Suspended User Status

You have now created a “Golden Master” user used explicitly for cloning and used this user to rapidly clone three user IDs.



Revoke Permissions from an IBM Wave User

When an employee changes roles or leaves your company, it is important to revoke permissions from the employee’s IBM Wave user. This is performed from the IBM Wave User Manager window by right-clicking on a user and selecting the Update IBM Wave User action from the context menu as shown in Figure 5-91.

Figure 5-91 Update IBM Wave User

In the Update IBM Wave User window, click on the Scope and Permissions tab. This brings you to a table of the user’s current permissions. Highlight a permission that you would like to delete and click the Delete Permission button as shown in Figure 5-92.

Note: To change the permissions of an IBM Wave user, the user ID must not currently be logged into an active session.



Figure 5-92 Select and Delete IBM Wave User Permissions

This can be repeated for all of the permissions availabe to the user. Once complete, click the Close button to return to the main IBM Wave User Manager window.

Change the Class of IBM Wave User by Cloning Another Golden MasterIn previous examples, we demonstrated how to create a “Golden Master” template user preconfigured with the scope and permission needed by a particular class of users. We named this user “WebSphere Admin Master” to signify its status as a “Golden Master” template. We then cloned numerous instaces of users with the class of WebSphere admin by cloning from WebSphere Admin Master.

However, perhaps one of these users changes roles in the company and moves from the role of WebSphere Admin to Database Admin.

Just as we used “Golden Master” templates to create new instances of a particular class of user, we can use these templetes to modify a particular user ID from one class of user to another by selectively cloning the scopes and permission of another user template.

To grant an existing IBM Wave user with all of the scope and permissions of another class of users, right click on a user and select the “Copy scopes and permission to selected user” action from the context menu as shown in Figure 5-93.

Note: Assuming that a user is changing jobs, all existing permissions must be deleted as described in the section Revoke Permissions from an IBM Wave User prior to cloning the scopes and permissions of the new user template. This is because the clone action appends additional permissions rather than replace existing permissions. This allows a user to possibly have the permissions of multiple classes of users.



Figure 5-93 Select Copy Scopes and Permissions to Selected Users

In the Copy Selected Scopes and Permissions window, ensure the target user is selected. From the “Select IBM Wave User to copy scopes and permissions from” drop down box, select the source user. In Figure 5-94 on page 134, we are going to copy the scopes and permissions from the WebSphere Admin Master to user lydia.

Figure 5-94 Select IBM Wave User to Copy Scopes and Permissions From



Your user now has been granted all of the scopes and permissions associated with this new class. However, unlike when using the cloning action to create new IBM Wave users, this cloning action does not impact the other attributes of the user, including password, challenge question, and description. Because our previous examples show the use of the Description field to organizes IBM Wave users by class, it is now important to modify this field to match the class that we used to clone the scopes and permissions from. To do this, right click on the user that you just added a new class of permission to and select the Update IBM Wave User action from the context menu. In the Update IBM Wave User window, change the Description field to match the name of the newly-assigned class of users and click Update as shown in Figure 5-95.

Figure 5-95 Update the User’s Description field to match the new Scopes and Permissions

Notice that the main IBM Wave User Manager window now correctly reflects the newly-assigned class of user as shown in Figure 5-96. You can now click on the Decription column header to resort the IBM Wave users by class and see your user listed alongside the template and other users of that class.

Figure 5-96 Notice the new value in the Description field

Suspend an IBM Wave UserOccasionally, your organization will have employees choose to leave or move to positions that do not require access to a z/VM systems management tool. In these cases, it is important to revoke the employee’s IBM Wave user. Due to auditing requirements in certain industries, IBM Wave is designed to make it impossible for an IBM Wave user to be entirely deleted. Rather, a user is administratively suspended indefinately when a particular ID is no longer needed. Prior to this, we suggest several steps:



1. Delete all Permissions to ensure that the IBM Wave user cannot be used if it is somehow reactivated.

2. Change the Description field of the user to “Suspended” to help sort and differentiate indefinitely suspended users from all other IBM Wave users.

3. Once this is done, administratively suspend the user by right clicking on a user listed in the IBM Wave User Manger and selecting the Update IBM Wave User action from the context menu. In the Update IBM Wave User window, change the User Status to A-Suspend as shown in Figure 5-97 and click the Update button.

Figure 5-97 Change User Status to A-Suspend

The IBM Wave user is now suspended.

5.10 Backup option for IBM Wave

Because IBM Wave runs on a Linux guest (typically running on z/VM), this guest can be backed up and restored using the same processes as any other Linux guest. To learn more about general backup techniques for z/VM and Linux guests, refer to Set up Linux on System z for Production, SG24-8137 andUsing z/VM v 6.2 Single System Image (SSI) and Live Guest Relocation (LGR), SG24-8039. These IBM Redbook publications are available from the ITSO Redbooks site at http://www.redbooks.ibm.com/.

In addition to traditional backup techniques, IBM Wave also allows IBM Wave administrators to backup the IBM Wave database using SQL scripts. The IBM Wave backup action is useful for capturing snapshots of the IBM Wave metadata in order to reverse changes to pure-metadata objects, such as projects, site defined groups, IBM Wave users, etc. The rest of this section will deal with this level of backup.



5.10.1 Backing up the IBM Wave database

To back up the IBM Wave for z/VM database, select the Backup IBM Wave for z/VM Database action from the Administrative menu as shown in Figure 5-98.

Figure 5-98 Selecting the backup option

As soon as you select the Backup IBM Wave for z/VM Database action, the Wave server examines its database and determines the size of the SQL script that would be generated from this action. It then checks the capacity available in the volume holding the /usr/wave/DBBackup directory. This is critical because this script may be larger than one gigabyte.

If the filesystem does not have enough available space, IBM Wave presents an error message like Figure 5-99 and terminates the backup action. When this occurs, an IBM Wave administrator must free sufficient space by resizing the volume holding the /usr/wave/DBBackup directory or deleting older backup files via an SSH session and then re-run the Backup IBM Wave for z/VM Database action..

Figure 5-99 Not Enough Free Space Error Message

If IBM Wave determines that the appropriate Wave server volume contains sufficient space, it generates an SQL script in /usr/wav/DBBackup named according to the following format:

backupSQLDump-<Day>-<month>-<Day of Month>-<Hour>-<minute>-<Second>-<Year>.sql



These files can be accessed by SSH or FTP. Example 5-1 shows an example of the /usr/wav/DBBackup directory with a backup file.

Example 5-1 IBM Wave Backup

wavesrv:/usr/wave/DBBackup/ # ls -latotal 115700drwxr-xr-x 2 root root 4096 Jan 15 14:53 .drwxr-xr-x 9 waveuser root 4096 Jan 15 14:53 ..-rw-r--r-- 1 root root 118342493 Jan 30 14:54 backupSQLDump-Thu-Jan-30-10-01-40-EST-2014.sql

5.10.2 Enter Single User Mode

In order to be able to backup the IBM Wave database, your IBM Wave session needs to enter Single User Mode. Under Single User Mode, IBM Wave only allows a single user session to be active at any one time. If other IBM Wave users are active when you attempt to enter Single User Mode, you will be given a series of windows allowing you to send messages to the other active IBM Wave users to tell them to log off. You are then able to request a log off option on their behalf, which opens the logoff confirmation panel on their screens.

To enter Single User Mode, select the Toggle Single User Mode action from the Administrative menu of the menu bar as shown in Figure 5-100.

Figure 5-100 Toggle Single User Mode

If no other users are active, you will immediately enter single user mode. This will prevent other IBM Wave users from being able to login. However, if other IBM Wave users are in active sessions, then you will be presented with the Request Users Logoff panel as in Figure 5-101. Click on the Send Message button.

Note: If the IBM Wave for z/VM database is backed up more than once in a period of a day, IBM Wave for z/VM will prompt the user if the current day’s backup should be overwritten.



Figure 5-101 Request Users Logoff Panel

Clicking the Send Message button opens the Broadcast Message panel to allow you to send a message to all other active users. Type your message into the text area and then click Go as shown on Figure 5-102 on page 139

Figure 5-102 Broadcast Messages Panel

The other IBM Wave users then receive a pop-up notification for this message as in Figure 5-103.



Figure 5-103 Other IBM Wave users receive message that the Admin requesting a Logoff

Having read this message, the other IBM Wave users learn that they need to finish their work and begin the process of logging out. As the users log out, the users disappear from the Current Logged In Users table in the Request User Logoff window. However, it’s possible that one or more users are not actually at the workstation and have left their IBM Wave session unattended. Alternatively, it is possible that the user has a hung session if connectivity between the user’s workstation and the Wave server was interrupted prior to a graceful logoff. IBM Wave anticipates this scenarios by automatically replacing the Send Message button with a Request Logoff button after you send a message to the other IBM Wave users. For example, in Figure 5-104, the IBM Wave user gene did not logoff as requested by the message from the IBM Wave administrator because he was not at his desk. By examining the workunits tab, the IBM Wave administrator determines that gene is not currently active on IBM Wave. In this situation, an IBM Wave administrator can click the Request Logoff button.

Figure 5-104 Force an IBM Wave User off via the Request Logoff button



This action instructs the IBM Wave server to attempt to contact the selected user’s IBM Wave client. If successful, it instructs the Wave client to initiative a logoff action. At this point, the user’s IBM Wave session has been remotely terminated and displays the panel shown in Figure 5-105. Once the user clicks OK, the IBM Wave client closes.

Figure 5-105 You have been logged off by an IBM Wave administrator dialogue

In the case of a hung session where the IBM Wave server is unable to reach to the IBM Wave client, the Request Logoff action will not terminate the user session. This is because it is possible that session tasks might have been in progress when the session hung, and ending a session in such a state might lead to unpredictable results. Nevertheless, if the IBM Wave administrator still needs to terminate the session, the Request Logoff button is replaced by a Force logoff button if the Request Logoff action failed. Initiating the Force Logoff action instructs the IBM Wave server to unilaterally terminate a selected IBM Wave session. Because this might potentially be a destructive action, IBM Wave presents a warning message to the IBM Wave administrator as shown in Figure 5-106:

Figure 5-106 Force Logoff Warning

Click yes to continue.

Between all of these actions, the IBM Wave administrator should now have successfully terminated all other user sessions as in Figure 5-107.



Figure 5-107 All Users are Logged Off

Now that all other users are logged off, the IBM Wave administrato can close the Request Users logoff window. This automatically places IBM Wave into Single User Mode and presents a dialogue like Figure 5-108 to the IBM Wave administrator.

Figure 5-108 Single User Mode is now enabled


Draft Document for Review February 26, 2014 8:58 am 8192HowtoManagezVMMaster.fm

Chapter 6. How to manage z/VM

This chapter describes how to use the IBM Wave for z/VM interface to provision, deprovision, and manage virtual hardware, including systems, disk, and networking equipment. Typically, these tasks are the responsibility of a z/VM systems programmer.

By the end of this chapter, the reader will learn to:

! Provision z/VM User IDs for use as CMS guests or Linux guests

! Login to guests as a 3270 console or a ASCII terminal

! Deprovision z/VM User IDs and return disks to DIRMAINT

! Organize disk by group

! Set disks as Page or SPOOL space

! View the Network Topology of your environment

! Provision virtual network switches and segments

6

8192HowtoManagezVMMaster.fm Draft Document for Review February 26, 2014 8:58 am


6.1 Managing z/VM User IDs

In this section, we discuss and describe the following:

! Define a z/VM User ID! View details of the z/VM User ID! Activate a new z/VM User ID! Log into CMS as a new z/VM User ID! Remove a z/VM User ID

6.1.1 Define a z/VM userID

To define a new z/VM User ID from scratch, please complete the following steps:

1. On the top left panel, select “Hardware Viewer” to display your physical systems and logical partitions, and select the z/VM logical partition in which you would like to provision a new z/VM User ID as shown in Figure 6-1.

Figure 6-1 Select a z/VM LPAR from the Hardware Viewer

2. Ensure that the Group By Drop-Down Menu is set to Site Defined Group in order to display icons representing each of the Site Defined Groups as shown in Figure 6-2 on page 145

Note: Due to cultural differences and the historical role of z/VM as a timesharing operating system, the terminology used to describe guest operating systems running on the z/VM hypervisor is commonly different from other hypervisors. For the purposes of this IBM Redbooks publication, please consider “z/VM User ID” to be synonymous with the terms “guest” and “virtual machine.”

Chapter 6. How to manage z/VM 145


Figure 6-2 Select “Site Defined Groups” from the Group By drop-down menu.

3. Double-click on USER-LOCAL or a custom Site Defined Group that you’ve created to open that group. The group is now depicted by a box containing icons representing each of the z/VM User IDs contained in this group. To create a new z/VM User ID within this Site Defined Group, right click within this box and select z/VM User Actions → Create New z/VM User from the context menu as shown in Figure 6-3 on page 146.

Note: A Site Defined Group is a IBM Wave for z/VM concept that allows site administrators to define logical grouping of related z/VM guests. During auto-discovery, IBM Wave for z/VM creates and locks numerous Site Defined Groups containing IBM and ISV service machines. For example, all of the service machines associated with DIRMAINT are stored in IBM-DIRM. All other z/VM User IDs found during auto-discovery are placed in a Site Defined Group called USER-LOCAL.



Figure 6-3 RIght-Click within the Site Defined Group Box and Navigate to “Create New z/VM User”

Starting the “Create New z/VM User” task will open up a pop-up window with a variety of fields needed to provision a new z/VM user. The fields are as follows:

User Name The z/VM User ID of your new guest. This is case insensitive and limited to up to 8 characters.

Password The password for logging into z/VM as the new z/VM User ID.Storage The text field and drop-down menu identify the size and units of the

storage that you would like to define and allocate to your new guest. The drop-down to the right of the label “from” is used to identify the storage group from which you intend to allocate your storage.

Memory Min / Max (MB) These fields identify the minimum and maximum amount of memory

that your system can be allocated by the z/VM Control Program.

Note: The way that IBM Wave for z/VM allocates and attaches minidisks may look unusual to an experienced z/VM systems programmer. After submitting the “Create New z/VM User” panel, IBM Wave for z/VM determines the size of DASD volumes available in the selected storage group, and allocates one or more minidisks with a capacity totaling the amount that you selected. As an end-user of IBM Wave for z/VM, there is no way to know the size and number of individual minidisks that your guest will be allocated. In order to have greater control over types of DASD (e.g. Mod-9 versus Mod-27), create and label storage groups that seperate out the available DASD volumes by type (e.g. MOD9GRP and MOD27GRP). Additionally, IBM Wave for z/VM attaches these new minidisks at addresses starting at 192 and incrementing by one. For example, if IBM Wave for z/VM attaches three minidisks to your z/VM User, they are attached at 192, 193, and 194. At the completion of the “Create New z/VM User” task, the minidisks are in an unformatted state.



Depending on how your guest is configured, this might be useful because your guest might be able to dynamically resize this value as needed. However, because the CP command DEFINE STORAGE responsible for this value results in a system restart, dynamic resizing of memory cannot typically be done with Linux guests. Furthermore, there is no functionality within IBM Wave for z/VM to resize this memory value. Therefore size the Memory Min (MB) value for the memory requirements of your workload. IBM Wave for z/VM will not allow you to perform a Bare Metal Installation of Linux if your Memory Min (MB) is smaller than 1000 MB.

Virtual CPUs This field set the number of virtual processors available to the guest.Project This drop-down menu allows the guest to be assigned to a pre-defined

project. The detailed information associated with each project can be viewed by selecting a project and clicking on the Details button to the right of the drop-down. If your site has not yet defined any projects, then you may be unable to select any projects. If this is the case, then projects can be defined and assigned at a later time.

Functionality: This drop-down is used to express the interdependencies between Linux guests. For example, if an application server depends on a database server, then the database server would have a higher activation level to ensure that it is available when the application server starts. If your site is not using this functionality, then you will only have an option called “N/A (Application Level 1),” which means that all Linux guests have the same priority for startup.

Description This is a field to insert a textual description of the function performed by the guest. This is useful because of the z/VM limitation that a User Name can be no longer than 8 characters prevents the user name from effectively conveying the purpose of the guest. This field is stored as metadata in IBM Wave for z/VM’s database.

Default z/VM System This field defines the default member of an SSI cluster that the guest

would IPL on if the entire cluster is restarted from scratch.Network Information This area displays a list of available virtual network segments for your

guest. The number of available segments can be anywhere from zero to many depending on the network configuration of your z/VM environment. If you have defined a virtual switch, then you likely will see at least one segment automatically generated to connect your guest to your virtual switch. Our values are shown in Figure 6-4.



Figure 6-4 Complete the Create New z/VM Guest Panel.

Note: IBM Wave for z/VM has a variety of parameters that set the default and maximum values for Memory Min (MB), Memory Max (MB), CPUs, and Disk Space. To change these values, select the Administrative drop-down menu and select Manage Parameters as shown in Figure 6-5. All of the fields are under the first Threshold tab and are shown in Figure 6-5 on page 149.



Figure 6-5 IBM Wave for z/VM Parameters Threshold Tab

4. When you’ve completed the panel, click the Create button. If you have an invalid entry in any of the required fields, then you will receive an error message that forces you back to the panel to correct the field. If you have a missing optional field (such as in the Description field), you will receive a pop-up asking you to confirm that you would like to create the guest without the optional field. Once all fields are accepted, a progress bar will appear on the bottom of the panel alongside a text area that explains the z/VM tasks that IBM Wave for z/VM is automatically performing (See Figure 6-6).

Note: To the left of the “Create” button is a “Close” button. This completely cancels out of the panel without a confirmation screen, resulting in data loss from all fields. Only click “Close” when you intend to cancel the “Create New z/VM User” task.



Figure 6-6 Bottom of “Create New z/VM Guest” Panel with Progress Bar and Status Text

5. Once the “Create New z/VM User” task is complete, you will see a popup stating “New z/VM User ****** created successfully.” as shown in Figure 6-7.

Figure 6-7 Confirmation Pop-up for completed “Create New z/VM Guest.”

Your new z/VM User ID now appears in the selected Site Defined Group as an inactive guest as shown in Figure 6-8.



Figure 6-8 New z/VM user now visible in Site Defined Group

6.1.2 View details of z/VM user id

Double click on the icon of the new z/VM user id to display its details. The most useful information is found in the z/VM view, which displays the User ID’s complete directory entry, as shown in Figure 6-9.



Figure 6-9 Directory Entry in z/VM View of Display Guest Details Panel.

The new user’s directory entry has an IPL CMS statement. This means that the guest will automatically boot into the CMS operating sytem when activated, even if the IBM Wave for z/VM end-user installs Linux on this guest. In order to boot the Linux kernel, the end-user would have to manualy log into the z/VM User ID to enter the CMS environment and then enter the command IPL 0192 (assuming that the Linux kernel is installed to the 0192 minidisk). You can change this by logging into a 3270 terminal emulator session and issuing the command shown in Example 6-1.

Example 6-1 Changing the boot disk

Dirm for <<userid>> IPL 192

where <<userid>> is the new z/VM user id that you just created.

The new user links to MAINT’s 0190, 019D, and 019E minidisks in read-only mode. This allows the new guest to IPL CMS off MAINT’s 0190 minidisk without having a local CMS filesystem. These statements also make the new user ineligible for live-guest relocation to another member of an SSI environment.

In order to have your guest automatically boot into Linux and be eligible for live-guest relocation, the z/VM User ID’s directory entry must be manually edited through traditional 3270 methods.

Example 6-2 Example of changing the IPL CMS statement using DIRMAINT

dirm for testfive ipl 192 DMSWSF006E No read/write filemode accessed for TESTFIVE NETLOG DVHCMD1186W Filemode A is not a R/W disk or directory. Your DIRMAINT DVHCMD1186W NEWMAIL date/time stamp will not persist across an IPL. DVHXMT1181R Enter the current logon password of TESTFIVE at ITSOZVM1 for



DVHXMT1181R authentication. It will not be displayed on the DVHXMT1181R terminal. To exit without processing the command, just DVHXMT1181R press ENTER.Author note: the password was typed in and enter was pressed DVHXMT1185W Filemode A is not a R/W disk or directory. Your request DVHXMT1185W counter will not persist across an IPL. DVHXMT1191I Your IPL request has been sent for processing to DIRMAINT at DVHXMT1191I ITSOZVM1. Ready; T=0.01/0.01 15:03:18 DVHREQ2288I Your IPL request for TESTFIVE at * has been accepted. DVHBIU3450I The source for directory entry TESTFIVE has been updated. DVHBIU3423I The next ONLINE will take place via Diagnose 84. DVHRLA3891I Your DSATCTL request has been relayed for processing. DVHRLA3891I Your DSATCTL request has been relayed for processing. DVHRLA3891I Your DSATCTL request has been relayed for processing. DVHBIU3428I Changes made to directory entry TESTFIVE have been placed DVHBIU3428I online. DVHREQ2289I Your IPL request for TESTFIVE at * has completed; with RC DVHREQ2289I = 0

Open the Network View to see a graphical representation of the network segments connecting your UserID to a VSWITCH. Our network view is shown in Figure 6-10.

Figure 6-10 New z/VM Guest in Network View



6.1.3 Activate new z/VM UserID

After creating a new z/VM user, the user must be activated in order to be used. This process is analagous to powering on a physical server. This is done by right-clicking on the icon that represents your z/VM User and selecting “Activate” (see Figure 6-11). Multiple z/VM users can be activated simultaneously by highlighting more than one icon prior to right-clicking and selecting “Activate.”

Figure 6-11 Selecting Activate on a z/VM User

This action will open up a dialog bog called z/VM User Activation (Figure 6-12). Verify that the one or more z/VM users that you want to activate are all checked and then click the “Go” button.



Figure 6-12 z/VM User Activation Dialog

Clicking Go will cause a message to pop-up that the “Activate z/VM User” task has been dispatched to the Background Task Scheduler. Click okay and close the “z/VM User Activation Panel” to return to the Current System View in the main IBM Wave for z/VM window. Once the z/VM user is running, the icon representing the new z/VM user will have be partially covered by a green circle containing a running man as shown in Figure 6-13

Figure 6-13 User activated, up and running.

6.1.4 Log into CMS as new z/VM UserID

Once a z/VM user is active, a IBM Wave for z/VM user is able to log into CMS as this user. This allows a IBM Wave for z/VM user to perform z/VM systems management and programming tasks using the traditional 3270-based CMS interface

To enter this environment, right-click on the icon of your new running z/VM user and select “3270” to open an integrated 3270 terminal as shown in Figure 6-14 on page 156



Figure 6-14 Selecting the 3270 Console

This will open a window that prompts for the password associated with this z/VM user. Enter the password into the field to the right of the “Enter Password:” field and press the Login button as shown in Figure 6-15.

Figure 6-15 Password Prompt for 3270 Connection

This will open a 3270 terminal session via an integrated Java-based 3270 terminal emulator and log you in as the z/VM user you selected.

Once logged on, you can access the traditional CMS environment by issuing the IPL CMS command. However, because IBM Wave for z/VM is designed with the assumption that z/VM users created via the GUI will be used for Linux guests, your new z/VM user lacks a 191 minidisk at filemode A, meaning that there is no local R/W filesystem formatted for CMS. Because of this, numerous CMS utilities will issue an error similar to that shown in Example 6-3.



Example 6-3 Error issued when missing a 191 minidisk

DMSACP113S A(191) not attached or invalid device address

By issuing a QUERY DASD command as shown in Example 6-4, you can see that your user has a large DASD at 0192. This is equal to the disk size that you requested on the “Create New z/VM Guest” panel.

Example 6-4 Results from the QUERY DASD command

query dasd DASD 0190 3390 ZVM4RS R/O 214 CYL ON DASD 991B SUBCHANNEL = 0007 DASD 0192 3390 ZVM4U1 R/W 2884 CYL ON DASD DB53 SUBCHANNEL = 000A DASD 019D 3390 ZVM4RS R/O 292 CYL ON DASD 991B SUBCHANNEL = 0008 DASD 019E 3390 ZVM4RS R/O 500 CYL ON DASD 991B SUBCHANNEL = 0009

Notice that the 192 disk is unformatted because IBM Wave for z/VM assumes that this will be formatted as a Linux root filesystem via the IBM Wave for z/VM Linux installation wizard. This means that it is impossible to access the 0192 disk as a CMS filemode. Example 6-5 demonstrates the type of error that you will receive if you attempt to do this.

Example 6-5

acc 0192 b DMSACP112S B(192) device error

Despite these limitations, the CMS environment can still be used to manually edit configuration files via XEDIT. Utilities such as the TCP/IP utilities on TCPMAINT’s 592 disk, are also potentially accessable and useable (see Figure 6-16 on page 157).

Figure 6-16 Logged into CMS as a z/VM User.



6.1.5 Deactive z/VM User

To deactivate a z/VM user, right click on the user’s icon, and select the command “Deactivate” as shown in Figure 6-17. You can simultaneously deactive multiple z/VM users by selecting multiple icons, right-clicking, and selecting “Deactivate.”

Figure 6-17 Deactivating a z/VM user

This will display the “Deactive z/VM Users” window as shown in Figure 6-18. Ensure that the users that you want to deactive are checked. Additionally, if this is a Linux guest, you have the option of either issuing a shutdown -h command to the Linux guest’s shell or forcing the guest off via a z/VM user logoff. A z/VM user logoff is the equivalent of powering off a machine, which is destructive to a Linux guest that has not been shut down. This means that if you want to deactivate a z/VM user running Linux, you need to run the ‘Deactive z/VM User' task twice. It must first be run with the Issue shutdown -h with timeout option selected, and then a second time with the “Force z/VM User Logoff” option selected.



Figure 6-18 Deactivate z/VM Users Window

Once you have selected “Force z/VM User Logoff” and clicked Go, you are presented with a Confirmation Pop-Up titled “Force confirmation” shown in Figure 6-19.

Figure 6-19 Force confirmation popup

6.1.6 Remove a z/VM userid

There will be times that you will need to remove a z/VM userid. To do this, perform the following steps:

1. Right-click one or more userids. 2. Select More Actions 3. Select Delete. (See Figure 6-20 on page 160)



Figure 6-20 Right click and select Delete

4. Confirm that the checkbox next to each of userids that you’d like to delete is checked as shown in Figure 6-21 on page 161



Figure 6-21 Panel to delete a z/VM user

5. Select if you’d like to Erase Disk Contents. We are unsure if this zeros out the disks only or if it returns the disk to the storage group for re-allocation.

6. Click the Go button7. You will see a confirmation screen as shown in Figure 6-22 on page 161

Figure 6-22 Confirm delete action

8. Click Yes if you are sure that you would like to proceed. You will receive a message showing that your request has been submitted to the BTS.

If you would like to see the progress of the work unit, open the BTW Work Units tab, and double-click on your “Delete z/VM Guests” task as shown in Figure 6-23 on page 162



Figure 6-23 Workunit details for delete z/VM guests task

When the task is complete, the z/VM user icon will disappear from its Site Defined Group

6.2 Managing Device Pools

BM Wave provides functionality to create, modify and update Device Pools. It is also possible to transfer Real Devices from one Device Pool to another.

To get to the Manage Device Pools screen, click on Administrative then Site Management as shown in Figure 6-24



Figure 6-24 Manage Device Pools menu option

This will display all the Device Pools accessible to the current IBM Wave User (Figure 6-25). Only Device Pools which are in the IBM Wave User’s scope will be shown here. The window displayed is a display-only version of the Device Pool Manager, accessible from the “Administrative” menu.

Figure 6-25 Device Pool manager

The Device Pools Manager employs three device-class pools:

! DASD (FCP and ECKD)! HiperSockets™! OSAs

It can manage multiple pools for each device class (Test/PROD/SYS etc…).



Drilling into the Device Pool Manager, we get to the actual Device Pools, which lists information about each device in the pool (Figure 6-26)

Figure 6-26 down into the Device Pool manager

The fields at the top of this window are:

! Name - The name of the Device Pool. ! Type - The type of the Device Pool.! Created By/Last Modified By - The IBM Wave user who created/last modified this Device

Pool coupled with timestamp

Under the Descritive tab, you would see:

! Default Virtual Device - an optional device address (4 hexidecimal digist) which specifies the default virtual device to use when dedicating real devices from this device pool to z/VM guests

! A Description - an optional description for the Device Pool

Moving on to the Devices tab you would see all the devices owned by the Device Pool.

! System Name - This table displays all z/VM System associated with the Device Pool. When a z/VM System is selected, the Real Devices table will populate with Real Devices from the selected z/VM System owned by the Device Pool.

! Real Devices - This table lists the Real Devices and their System-Addresses which belong to the selected z/VM System and are owned by the Device Pool. Each line in the table provides information about one Real Device – it’s address, online status, dedicated status etc.

The next tab is the Connected IBM Wave Resources . This tab displays a table which lists all the WAVE Managed Entities connected to the Device Pool.

6.2.1 Create a Device Pool

To create a new Device Pool:

1. Open the Device Pools Manager by clicking on the Administrative menu and selecting the “Manage Device Pools” from the “Site Management” submenu



2. Click Create New Device Pool (Figure 6-27)

Figure 6-27 Create new Device Pool

3. In the “Create New Device Pool Dialog”, Enter a Device Pool Name and select the TYPE of Device Pool, (e.g. OSA). A description is optional. You can also optionally specify a Default Virtual Device and Click the “Create” button (Figure 6-28).

Figure 6-28 Create new Device Pool

4. After clicking “Create,” you will see a pop-up that states that the workunit has been dispatched to the Background Task Scheduler (BTS). You can then close the “Create New Device Pool” panel .Figure 6-29 shows, clockwise, a composite of the Workunit being submitted to the BTS, the workunit details and the Device Pool manager with the new device pool.



Figure 6-29 Device Pool creation process

5. Now that you have an empty Device Pool created, you need to associate it with a z/VM system before you an transfer devices from one pool to another. A Device Pool can be associated with several z/VM system as the same time. To do this, right click on the newly created pool entry, and select “Update Details” (Figure 6-30).

Figure 6-30 Device Pool creation process

6. A dialog will open for you to update the selected device pool. Select the Devices tab, and click Associate System.



Figure 6-31 Assocate Device Pool to z/VM system

7. In the popup window, select which z/VM system that you want to associate with, then click “Go” button.

Figure 6-32 Picking a z/VM system to associate

8. After clicking “Go,” you will see a pop-up that states that the workunit has been dispatched to the Background Task Scheduler (BTS). And you will see the Associated z/VM Systems ( )Notice that, No devices found for the selected z/VM System yet.



Figure 6-33 Associated z/VM system

6.2.2 Transfer Real Devices from one Pool to another

In order to transfer Real Devices from one Pool to another, the source pool must have some free devices. Of course, these free devices must be available on the z/VM system that the target Device Pool is associated with.

1. To transfer Real Devices from one OSA pool to the newly created pool, in the “Device Pool Manager” dialog box, right click on the source pool entry and select “Update Details” (Figure 6-34)

Figure 6-34 Transfering Real Devices from one pool to another

2. Select the “Devices” tab, and select the Real Devices that you want to transfer (in this cases, 2009-200B) by clicking on the ones you want, holding the CTL key and clicking on additional devices. Right Click and select Transfer To and then the Pool you want to transfer to from the submenu as shown in Figure 6-35.

.



Figure 6-35 Transfering Real Devices from one pool to another

3. After clicking on it, you will see a pop-up that states that the workunit has been dispatched to the Background Task Scheduler (BTS). Close that pop-up dialog, and you will notice that the real devices (2009-200B) are gone from the source pool. If you open the target Pool, you will notice the devices have been successfully transferred there.

6.2.3 Remove a Device Pool

To remove a Device Pool, transfer all the devices it has to another device pool.By following the steps in section 6.2.2, “Transfer Real Devices from one Pool to another” on page 168 to transfer devices back until there are no devices found for the selected z/VM System

If there is a 0 in the Devices column it means that there is no real device that belongs to this pool, and you can delete it by right-clicking, and select “Remove Device Pool” from the “More Actions” submenu as shown in Figure 6-36



Figure 6-36 Removing a Device Pool

After clicking Remove Device Pool, you will receive a popup asking you to confirm that you want to remove the specific Device Pool. Click Yes, your request will be submitted to the BTS and your Device Pool will be removed.

Some notes about IBM Wave for z/Vm and Device Pools:

! IBM Wave uses internal z/VM data to retrieve a unique ID for each Real Device.– This unique ID is made up of the Logical Control Unit ID with other elements – This unique ID is a unique identification for each device.

! Attention Required status – If any real device was dedicated to z/VM guest on two or more z/VM systems, IBM

Wave would raise an Attention Required status for the real devices.– If this situation is acceptable, the Attention Required Entry can be ignored.

! Transfer Real Devices– Since Device Pools are associated with Real Devices by their Unique ID, all Real

Devices with same Unique ID are transferred to the target Device Pool. – Would happens in PAV/HIPER PAV environment.

! It is possible to link a Managed Entity to a Device Pool. – It is pure meta-data. – For example, it is possible to link an OSA Device Pool to a Router Managed Entity. The

Network Viewer will display a link between the Device Pool and the Router.

6.3 Managing DASD

In this section, we describe how to manage DASD using IBM Wave for z/VM

6.3.1 Attaching DASD Volumes to System

In order to attach DASD values to a system, open the Hardware Viewer and click on the Storage Tab as shown in Figure 6-37 on page 171. On the leftmost column of the Current System View, there are three buttons entitlted Distribution, Groups, and Volumes. Click on the Volumes Tab.



Figure 6-37 Storage View by Volumes

Depending on the number of DASD volumes accesible by your system, you may need to apply a filter to be able to find the icons represting the volumes that you would like to add. In this case, it is likely easiest to sort the volumes by providing a range of real addresses, so we click on Real Address as shown in Figure 6-38.

Figure 6-38 Selecting Filter Criteria

Once the filter criteria is applied, the number of volumes displayed will decrease and the icons will resize to fill the display as shown in Figure 6-39 on page 172



Figure 6-39 Storage Volumes filtered by Real Address

Once you have displayed the specific volumes that you would like to attach to the system, select one or more volumes and then right-click and select “Attach to System” as shown in Figure 6-40

Figure 6-40 Highlighting Storage Volumes and Selecting the Attach to System Task

Starting the “Attach to System” task will open a window called “Attach DASD Volumes to z/VM System.” The one or more volumes that were highlighted when you right clicked and selected the action are listed in this window. By default, each of these volumes are highlighted. Click the “Go” button to start the task. See Figure 6-41 on page 173 for our example.



Figure 6-41

After clicking “Go,” you will see a pop-up that states that the workunit has been dispatched to the Background Task Scheduler (BTS). You can then close the “Attach DASD Volumes to z/VM System” panel.

Now that the DASD volumes have been attached to the system, you must define the volume in DIRMAINT extent control as a whole-volume region. This is done in IBM Wave for z/VM by selecting one or more volumes that have been attached to the system, right-clicking, and selecting the task “Define Region.” See Figure 6-42

Figure 6-42 Selecting Define Region from multiple attached Storage Volumes.



Clicking on “Define Region” will open the “Define Regions to DASD Volumes” window with one or more attached volumes. By detault, all listed volumes are checked. To define all listed volumes to DIRMAINT extent control as regions, click “Go” as shown in Figure 6-43

Figure 6-43 Define Regions to DASD Volumes” Window.

After clicking “Go,” you will see a pop-up that states that the workunit has been dispatched to the Background Task Scheduler (BTS). You can then close the “Define Regions to DASD Volumes” window

Once the attached volumes have been defined as whole-volume regions to DIRMAINT extent control, they not must be added to a DIRMAINT group. This can be done by selecting one or more volumes, hovering over the Assign Volume to Group” menu option and then clicking on one of the groups that appears in the sub-menu. See Figure 6-44



Figure 6-44 Assigning the volumes to a group

Clicking on a Group name in the “Assign Volumes to Group” sub-menu will open the “Assign DASD Volumes to DASD Group” window with one or more attached volumes. By detault, all listed volumes are checked. To assign all listed volumes to the selected DIRMAINT group, click “Go.” See Figure 6-45 on page 175

Figure 6-45 Assigning DASD volumes to a DASD group

After clicking “Go,” you will see a pop-up that states that the workunit has been dispatched to the Background Task Scheduler (BTS). You can then close the “Assign DASD Volumes to DASD Group ******” window.



Each volume is then fully configured and ready for use by IBM Wave for z/VM (see Figure 6-46 on page 176).

Figure 6-46 New volumes fully configured.

6.4 Managing the network

In this section, we discuss how to create a new VSwitch and then, once that is complete, how to create a virtual network segment.

To see the network configuration, go to the Hardware Viewer display on the left side of the IBM Wave for z/VM display and highlight the z/VM LPAR that you want to work with. On the right side select the network tab to say the layout of device pools, vswitches, virtual network segments and connected users.

6.4.1 Creating a new VSwitch

To create a new VSwitch, perform the following steps:

1. On the network topology screen, right click on any white(unused) space.



Figure 6-47 Creating a new vswitch

2. Click the Add New VSwitch option, a pop-up will appear similar to the one shown in Figure 6-48:

Figure 6-48 New Virtual Network



The two required fields to fill in are the Name: and Default NIC:. You will notice that you can input up to 8 characters in the Name: field but it will only display six characters. You can optionally check the VLAN Aware box and input a Default VLAN ID:. The Layer type: input field defaults to the IP address but you also have the option to choose MAC. Once you select the Create button, a request will be sent to the BTS and the new VSwitch will be added to the network topology, in this case VSWITCH3, as shown in Figure 6-49.

Figure 6-49 Creating a new VSwitch

6.4.2 Creating a Virtual Network Segment

Now that the VSwitch is created you will need to ‘fill in the cloud’ associated with it by adding a virtual network segment. To do this:

1. From the Administrative pulldown at the top of the screen, select Site Management → Manage IBM Wave for z/VM Virtual Network Segment.

2. When the pop-up box appears, select the Create New Virtual Network Segment option and you will see information similar to that shown in Figure 6-50 on page 179:



Figure 6-50 Create a virtual network segment

3. Enter the name of the VNS in the Name: field and the IP Information section4. Select the Create button. 5. Once it is created, go to the Auto Detect pulldown at the top of the screen and select

Refresh → Schedule z/VM Network Update. This will send a request to the BTS. 6. When this has been created you will connect it to a virtual network. Go to

Administrative → Site Management → Manage IBM Wave for z/VM Virtual Network Segment , highlight and right click the newly created network segment and select Update Information.

7. Select the Connected Virtual Networks tab and then click on Connect to Virtual Network, as shown in Figure 6-51 on page 180



Figure 6-51 Connect the virtual network segment to the virtual network

When you select the Go button, a WorkUnit will be sent to BTS and then you will see that the network IP address that you supplied in the Network field in the Network Segment Manager box will now appear in the VNS cloud, as shown in Figure 6-52 on page 181.



Figure 6-52 VNS Designation


Draft Document for Review February 26, 2014 8:58 am 8192HowtoManageLinuxMaster.fm

Chapter 7. How to manage the Linux virtual machines

A golden master is a template for a virtual machine (VM) It can also be referred to as a golden image, a master image or a base image. Usually, a golden image is manually installed and clones are created with minimal modifications. Using IBM Wave for z/VM, it is possible to create a prototype that will automate the cloning of a golden image.

This chapter describes how to use the IBM Wave for z/VM interface to install Linux guests from scratch using installation media mounted on a file server, convert Linux guests into golden master prototypes, and clone guests from a prototype. This chapter builds on some of the core concepts covered in Chapter 6, “How to manage z/VM” on page 143. Traditionally, these tasks have been performed by a z/VM systems programmer. However, the Bare Metal Installation and Cloning features simplify z/VM administration to enable Linux administrators to independently provision and deprovision Linux guests on z/VM.

In this chapter are instructions on how to:

! Install Linux onto a z/VM user through the Bare Metal Installation feature.! Convert a Linux guest into a IBM Wave for z/VM Associated Prototype! Clone guests from an IBM Wave for z/VM Associated Prototype

7

8192HowtoManageLinuxMaster.fm Draft Document for Review February 26, 2014 8:58 am


7.1 Preparing to install Linux in the virtual machine

IBM Wave has two features that make it possible to use the graphical interface for the manual creation of Linux guests from scratch. This functionality offers the possibility of building Linux golden images without the use of 3270 interfaces.

There will be times when the system programmer utilizes traditional 3270 methods to build a golden image and then use IBM Wave for z/VM to turn this golden image into Associated Prototypes. This is perfectly acceptable!

There will also be times when a Bare Metal Installation is necessary. A Bare Metal Installation can serve as a suitable means for building golden images when a z/VM systems programmer is not available. In this section, we discuss a bare metal install to build a golden image.

7.1.1 Install an X Server on your Workstation

VNC is not currently supported by IBM Wave during a Bare Metal Installation, so you will have to install an X-server on your workstation in order to tunnel X-windows through an SSH connection. This will enable the use of the fully graphical Linux installation wizard. There are many options available including XMing, Cygwin with XFree86/X.org, and Hummingbird Exceed.

If you are unable to install X11, you will have to install the pseudo-graphical Ncurses from YaST as shown in Figure 7-1.

Figure 7-1 Ncurses pseudo-graphical interface

After accepting all defaults during the installation wizard, start your Xserver. With most X11 applications, you will know that the Xserver has started when either the X icon in your system tray changes color from red to green or an X icon appears in your system tray.

Note: At this time, Bare Metal Installation of Red Hat Enterprise Linux is not available.

Chapter 7. How to manage the Linux virtual machines 185


While your Xserver is running, other applications such as PuTTY will have access to the X window services.

7.1.2 Log in as the IBM Wave administrator

Log in as the IBM Wave for z/VM administrator. If you are already logged into IBM Wave, log out and log back in as the administrator.

7.1.3 Configure an external SSH Program to launch sessions with X services

PuTTY is a free implementation of Telnet and SSH for Windows and Unix platforms, along with an xterm terminal emulator. You can use any SSH terminal, but in this section, we describe how to configure your SSH terminal using PuTTY in IBM Wave for z/VM.

To configure this in IBM Wave, open User Preferences by clicking the small task bar icon in the upper left corner of your IBM Wave GUI that shows a person with a pencil as shown in Figure 7-2.

Figure 7-2 User preferences

This will open a Change Preferences panel that will allow you to set preferences specific to your IBM Wave userid. This panel is shown in Figure 7-3.

Note: If you are a Macintosh user, you can use XQuartz, which is available from http://xquartz.macosforge.org/landing/. Run the *.dmg and copy the application icon to your Application folder.

If you are a Linux user, then you most likely already have an X server preinstalled on your system

http://xquartz.macosforge.org/landing/



Figure 7-3 Change preferences panel

Click the Browse... button next to the External SSH Program Location box. This will open a popup window. Navigate to the location of your putty.exe file and press Select as shown in Figure 7-4.

Figure 7-4 Navigating to the External SSH Program Location - in this case, PuTTY

This will place the path of your external SSH program into the External SSH Program Location field, as shown in Figure 7-5.



Figure 7-5 External SSH Program Location field filled in

IBM Wave needs to understand the proper format for passing in the IP address, username and password to the SSH program as arguments. In the case of PuTTY, this is done by right clicking in the white space of the External SSH Program Location field. When you do that, a submenu will pop up similar to that shown in Figure 7-6. This will format the field with the symbolic notation that will allow IBM Wave to dynamically insert arguments at runtime.

Figure 7-6 How to insert default PuTTY Syntax



Click the Update button to complete this task. The External SSH Program Location field should now look similar to what is shown in Figure 7-7.

Figure 7-7 External Program Location with Default PuTTY syntax

In order to launch your SSH program sessions with X services, you will need to modify the statement in the External SSH Program Location field to include the -X parmeter. This parameter indicates to the SSH program that an X server is locally available from your workstation for this SSH session. An X server makes it possible to launch graphical windows such as yast2 from your SSH session.

Add the -X parameter to the External SSH Program Location as shown in Example 7-1.

Example 7-1 Including the -X parameter

C:\putty.exe -X <LOGIN_USER>@<IP_ADDRESS> -pw <LOGIN_PASSWORD>

Click the Update button to save your settings. Your SSH sessions will now use X windows when possible.

7.1.4 Launch an SSH program from IBM Wave for z/VM

Now that your clones are active and your external SSH program is configured, you can launch SSH sessions and access any of your clones running the Linux operating system. To do this, right click on the active Linux system that you would like to access. In Figure 7-8, we right clicked on TESTSUSE. Select Access... → SSH Access

Note: If you use the same IBM Wave userid on multiple computers, then you will have to ensure that the path to putty.exe is identical on each computer to avoid having to reconfigure this setting.



Figure 7-8 Access an external SSH program

Enter the user name and password for the system in the popup (Figure 7-9).

Figure 7-9 SSH credentials

This will connect you to your Linux guest using the external SSH application, in our case, PuTTY and we used XMing as our X windows provider.



7.2 Create a z/VM Guest

Once the user preference has been updated, a new z/VM guest needs to be created. We will now use IBM Wave’s ties to z/VM to use the graphical environment to define a new guest virtual machine. During this process, IBM Wave presents a series of windows to allow you to set the characteristics of the guest that you would like to define. In the background, IBM Wave then works through a directory manager to define a new directory entry for your guest and dynamically places this directory entry into production. Finally, IBM Wave refreshes it’s view of the underlying z/VM environment to graphically indicate this new guest.

The author’s view is currently set at “Site Defined Groups”. Ensure that you have the correct group selected when creating your z/VM user. We will be using the Group by: Project group to create our user as shown in Figure 7-10 on page 190.

Figure 7-10 Group by: Project

We are now going to define a new guest in the IBM Wave project named ITSO Redbook. Right-click anywhere in the white space within the context of the project to bring up a context menu and select z/VM User Actions Æ�Create New z/VM User as shown in Figure 7-11 on page 191.



Figure 7-11 Create new z/VM user

This brings up a pop-up window called “Create New z/VM Guest” with the fields required to create a new z/VM user. An example of how we filled in these fields is shown in Figure 7-12



Figure 7-12 Parameters to create new z/VM guest

The fields are as follows:User Name The z/VM User ID of your new guest. This is limited to up to 8

characters. The authors used the value bareinst1.

Password The password for logging into z/VM as your new guest.

Storage The number and drop-down identify the size of the storage that you would like to define and allocate to your new guest. The drop-down to the right of the label “from” is used to identify the storage group from which you intend to allocate your storage and the author’s is $3390$. Keep the default of 4096 MB next to the add: on this panel drop down.

Memory

Min / Max (MB) These fields identify the minimum and maximum amount of memory that your system can be allocated by the z/VM Control Program. IBM-Wave does resize these memory values. Size the memory values for the memory requirements of your workload. IBM-Wave will not



allow you to perform a Bare Metal Installation if your memory minimum is smaller than 1000 MB.

Virtual CPUs This field sets the number of virtual processors available to the guest. The default value is 1.

Project This drop-down menu allows the guest to be assigned to a pre-defined project. When this task is executed from within the context of an IBM Wave project, the name of the project is shown here and the drop-down menu is inaccessible and greyed out. If you have executed this task outside of the context of a project, you would have to select a project here. You can view the detailed information of your selected project by clicking the details button.

Functionality This drop-down is used to express the interdependencies between Linux guests. For example, if an application server depends on a database server, then the database server would have a higher activation level to ensure that it is available when the application server starts. Since we are not exploiting this functionality in this example, “N/A (Application Level 1)” is the only option, which means that all Linux guests have the same priority for startup.

Description This is a field to insert a text description of the function performed by the guest. This is useful because of the z/VM limitation that a user name can be no longer than 8 characters, which prevents the user name from effectively conveying the purpose of the guest. This field is stored as metadata in the IBM Wave’s database. Enter a string of your choosing in this field.

Default z/VM System This field defines the default member of an SSI cluster that the guest would IPL on if the entire cluster is restarted. You can also select a stand alone z/VM system here.

Network Information This area displays a list of available virtual network segments for your guest. The number of available segments can be anywhere from zero to many depending on the network configuration of your z/VM environment. If you have defined a virtual switch, then you likely will see at least one segment automatically generated to connect your guest to your virtual switch. Place a check in the check-box for the VSWITCH or NIC you want your z/VM user to use.

When you’ve completed the panel, click the Create button. If you have an invalid entry in any of the required fields, you will receive an error message that forces you back to the panel to correct the field. If you have a missing optional field (such as in the Description field), you will receive a pop-up asking you to confirm that you would like to create the guest without the optional field. Once all fields are accepted, a progress bar will appear on the bottom of the panel alongside a text area that explains the z/VM tasks that IBM-Wave is automatically performing. This will be followed by a panel confirming the creation of your new z/VM userid as shown in Figure 7-13 on page 193.

Figure 7-13 Confirmation of z/VM userid creation.



To see your new z/VM userid, double click the project group where it was created, in this case the ITSO Redbook project. You will see BAREINST listed and greyed out signifying that it is inactive as shown in Figure 7-14.

Figure 7-14 New z/VM guest, BAREINST, in the ITSO Redbook project group

Double click on the new userid to see general information about your userid. The first tab, Data which is shown in Figure 7-15, will show which member of your cluster the new userid is eligible to run on (if you are using an SSI cluster). Keep in mind that if the cluster is recycled, your userid will IPL on the z/VM system that it was created in.



Figure 7-15 New userid general information - Data tab

You can examine your z/VM directory information by clicking on the z/VM view tab. Here you will see that your userid has been created with the name BAREINST, with one virtual CPU, a NIC connected to a VSWITCH named VSWITCH1 and a mini disk 4GB in size.



Figure 7-16 New userid general information - z/VM tab

USER BAREINST is userid name,XXXXXXX represents the password and 1024M and 2048M are the min and max storage values. The last parameter, G, is the privilege class, in this case General

CPU In this case 00 represents a single virtual process reserved for this guest

NICDEF Network Interface Card at virtual address C200 that connects this userid to vswitch VSWITCH1

MDISK 0192 is the IBM Wave default minidisk address that is always used for guests created by the ‘Create New z/VM Guest’ panel. The size of this minidisk is 5769 cylinders on volume LXD124

By opening the Network View you will see a graphical representation of the new userid, in our case BAREINST. Figure 7-17 illustrates its connection to ‘Unknown IP(VSWITCH1)’. There is not an IP address assigned to the BAREINST user because it is still an empty shell without an operating system associated with it.



Figure 7-17 Network view of new guest machine BAREINST

7.2.1 Activation, login, and deactivation of a new guest virtual machine

Your new guest is currently an empty shell that lacks a bootable disk or operating system.Nonetheless, all guest virtual machines running under z/VM have access to the traditionalConversational Monitoring System (CMS) operating system stored in a common read-onlydisk controlled by the z/VM Control Program. This enables guests that lack their ownoperating systems to perform z/VM systems management and programming tasks using the traditional 3270-base CMS interface

Prior to logging into the new guest machine, BAREINST, must be activated. Return to the z/VM User Groups tab and go to the Group By: Project selection and right click on the BAREINST icon and click on Activate as seen in Figure 7-18



Figure 7-18 Activating the new BAREINST guest ID

A confirmation panel will pop up( Figure 7-19). Ensure that the guest you want activated is checked and click Go.



Figure 7-19 Selecting a single guest ID for activation

A request will be sent to the BTS. Upon completion, the BAREINST icon in the Project Group will now have a green running man symbol indicating that this guest is now active as shown in Figure 7-20.

Figure 7-20 Activation completion for new guest ID

To access the BAREINST 3270 console, right click the BAREINST icon and select Access... and then CSL-Wave 3270 Console as seen in



Figure 7-21 Accessing the 3270 console

You will be prompted for the BAREINST password as shown in Figure 7-22.

Figure 7-22 Password prompt

This will open the BAREINST console using the IBM Wave integrated 3270 terminal emulator as shown in Figure 7-23.



Figure 7-23 3270 console display

Issue the command Q DASD to see the DASD that has been assigned to you by IBM Wave and the directory manager (Figure 7-24).



Figure 7-24 z/VM console commands

Notice the 192 disk which has 5769 cylinders on volume LXD124 that is shown in the output of the command. Type b on the command line and click the Go button to enter the CMS environment. This will allow you to submit additional CMS commands. Enter the command access 192 b to assign this disk to filemode B. Enter the command q disk to view the various allocated and linked disks for this user.

Click the Red X in the upper right corner to close BAREINST’s console and return to IBM Wave.

In order to proceed with the installation of Linux onto the new BAREINST guest, it would need to be deactivated. Right click on the BAREINST icon and select the Deactivate option as shown in Figure 7-25.



Figure 7-25 Deactivating a guest ID

When the deactivation completes, the BAREINST icon will now be grayed out. At this point you are ready for Linux installation.

7.2.2 Install Linux on your Guest Virtual Machine

You will now initiate the SUSE Linux Enterprise Server (SLES) Installation Wizard through the IBM Wave Bare Metal Installation feature. This feature automates the process of building bootable Linux media capable of logging into a Linux Repository via TCP/IP, transferring this media to BAREINST, and booting BAREINST using this install media. This eliminates the need to manually manipulate SPOOL devices to get to a Linux Installation Wizard.

To initiate a Bare Metal Installation, right click on BAREINST and select Install -> Launch Linux Installation from the context menu and you will see a pop-up similar to that shown in Figure 7-26:



Figure 7-26 BMI Wizard Installation Wizard welcome page

Click Next which will bring up another panel, shown in Figure 7-27. Ensure the Hostname and Networking Interface reflect your installation. Change the Temporary Installation Password to the permanent value that you want.

Figure 7-27 Customizing the installation

There is also a Show Password button that can be selected to see, in clear text, the value you have entered (Figure 7-28).



Figure 7-28 Password display

Click Next and you will see a flashing screen while ‘Verifying Repository’ and after a few seconds you will see a ‘Repository Verified’ message (Figure 7-29).

Figure 7-29 Linux repository verification

Click Next and you will be brought to another panel in the Repository Verification process and on the Select Storage Group, select your storage group, in our case it was $3390$ (Figure 7-30).



Figure 7-30 Repository Verification Storage Group selection

After the repository has been verified click Next to view the Summary (Figure 7-31).

Figure 7-31 Repository Information Summary

You should verify in the summary, the following:

! You are installing the Linux system you expect (SLES 11 SP 2 in our example)! Virtual NIC addresses are assigned to the VSwitch (c200-c202 a in our example)! TCP/IP information regarding the IP address for the Host, Gateway, Netmask and

Broadcast! The install repository location



Enter Go and this submits the Launch Linux Installation workunit which submits a request to the BTS.

Figure 7-32 Workunit progress display

Double click on the BTS Work Units tab to view more detailed information (Figure 7-33).



Figure 7-33 BTS Workunit Details

Upon completion, another panel will appear that indicates the Interactive Installation is ready (Figure 7-34).

Figure 7-34 Interactive Installation message

From here you will have to continue manually by clicking Ok and NOT Start SSH Session . For information regarding a manual Linux install see The Virtualization Cookbook for IBM z/VM 6.3, RHEL 6.4, and SLES 11 SP3, SG24-8147.

After installation completes, you will notice that a penguin has been added to the BAREINST icon, as shown in Figure 7-35.



Figure 7-35 Linux installation completion

The process of initializing guests for IBM Wave management has a number of prerequisites. For Suse Linux Enterprise Server installs, a package called cmsfs is required, which allows Linux to be able to read from CMS minidisks. In order to be able to initialize BAREINST for IBM Wave management, you will need to use SSH to access BAREINST and manually install this package. To do this, right mouse click on your guest, select Access.. and then SSH Access as shown in Figure 7-36.



Figure 7-36 SSH Access

Once you select SSL Access you will be prompted for a userid and password as shown in Figure 7-37.

Figure 7-37 SSH password prompt

Once you enter Login an SSH session will pop-up (Figure 7-38).



Figure 7-38 SSH session

Ensure that you have an X11 terminal emulator installed. In our case we are using MobaXterm. If you don’t have an emulator initialized you will get a ‘cannot open display’ error.

Enter the command yast2 to open the YaST2 Control Center in an X Window. Scroll down to the Software section and click on Software Management (Figure 7-39).



Figure 7-39 YaST2 Control Center

When the next panel appears enter cmsfs in the box next to the Search button, then click on Search. When cmsfs is found, place a check mark in the box next to the package you want installed and click Accept, as shown in Figure 7-40.



Figure 7-40 cmsfs package search

Once the install of cmsfs is complete, you will be returned to the YaST2 Control Center. Click on the red X in the upper right corner to close it. At this point you will be back in your SSH session where you can verify that cmsfs has been installed by entering the command:

rpm -qa | grep ‘cmsfs’

If the cmsfs package is displayed, then you have successfully installed the package. Execute the command ‘exit’ and close your SSH program (Figure 7-41).

Figure 7-41 cmsfs install confirmation

You are now able to successfully initialize BAREINST for IBM Wave management. Right click on the BAREINST icon and select More Actions -> Init User for IBM Wave Use from the context menu.



Figure 7-42 Init User for IBM Wave use

In the ‘Initialize User for IBM Wave Use’ panel, ensure that BAREINST is checked, enter the password and click Go. This submits a workunit to the IBM Wave Background Task Scheduler.



Figure 7-43 Selection of BAREINST with password prompt for use by IBM Wave

After a few moments, the caution symbol on the icon and the *Unknown* text beneath the userid field disappear and BAREINST has reappeared, as shown in Figure 7-44.

Figure 7-44 Completed BMI guest icon

You have now successfully created a SLES guest from scratch using IBM Wave’s Add New z/VM User and Bare Metal Installation functionality.

7.3 Initializing guests using IBM Wave for z/VM

After installing or importing a new Linux guest, the guest must be initialized by IBM Wave for z/VM to be able to be managed in the GUI environment.

Running the “Init User for IBM Wave for z/VM use” task on a Linux guest does the following:

! Opens an SSH tunnel into the Linux guest! Verifies that a version of IBM Wave is not installed on the guest! Verifies root privileges



! Verifies Linux guest prerequisites:– The CMSFS package containing the driver and utility required to allow Linux to access

CMS files and minidisks.– The VMCP package containing the driver and utility needed to allow Linux to issue

commands to the z/VM Control Program.! Verifies the guest has the LOGOFF ON SHUTDOWN flag set in the zipl.conf! Verifies that both the zipl.conf and fstab files adhere to the supported formats of IBM Wave! Links the WAVEWRKS 399 minidisk to the z/VM guest running Linux in the directory entry

with read-only access.! Adds a series of scripts to /usr/wave! Creates a Linux user id named waveuser and adds it to the /etc/sudoers file and sets its

password

7.4 Create a golden master

A prototype is a complete z/VM User Directory entry that is used by the directory manager to dynamically create and add a new guest virtual machine to the z/VM user directory. This allows a z/VM systems programmer to add a new guest with the same directory attributes as the prototype using the LIKE parameter.

A golden master is a template for a virtual machine (VM) It can also be referred to as a golden image, a master image or a base image. Usually, a golden image is manually installed and clones are created with minimal modifications.

Using IBM Wave for z/VM, it is possible to create a prototype that will automate the cloning of a golden image. In order to do this, the following should be in place before a clone can be created:

! Sufficient storage space (DASD)! Sufficient memory (RAM)! A VSWITCH! A pool of IP addresses connected to the VSWITCH! A fully installed Linux server that has been deactivated

In this section, we will change one of our Linux guest servers to become an IBM Wave for z/VM prototype You can call this a “golden master” or template of future clones. After we create a prototype, we will describe the steps necessary to clone by using this prototype

In IBM Wave for z/VM, the concept of a golden master is implemented using a feature called an “Associated Prototype.” An Associated Prototype is made up of both a traditional prototype directory entry and an asscociated z/VM guest with assigned minidisks and an installed guest operating system. This is synonymous with the term “golden master.” The associated guest is deactivated and has a password set to NOLOG to prevent it from being actively used. Within the IBM Wave for z/VM GUI, the guest is filtered from the z/VM User Groups tab and only visible in the Prototypes tab.

Using associated prototypes, the IBM Wave for z/VM administrator can clone a commonly-used pattern. When combined with the functionality of the Script Manager, the IBM Wave for z/VM Administrator can automate much of the process of dynamically provisioning guest Linux virtual machines.

Important: Once you change a Linux server to become an IBM Wave for z/VM prototype, the Linux server you choose will no longer be able to be logged into and therefore will no longer be bootable.



Associated prototypes allow for physical cloning, which is the process of creating a new z/VM guest definition according to the definition of the prototype, or full cloning, which is the process of creating a new z/VM guest definition and copying the data from the assigned user to the newly created user.

A disassociated prototype does not have a z/VM guest associated with it. When cloning from a disassociated prototype, the user can only choose physical cloning. The disassociated prototype is only visible to the IBM Wave user that has the Site Level administrator role.

In order to create and be able to convert a z/VM guest into an associated prototype, the guest must be inactive. For instructions on how to deactivate a z/VM guest, refer to Chapter 6, “How to manage z/VM” on page 143.

If you are running in a non-SSI environment, then you can tell at a glance if a member is inactive, as the z/VM guest icon is grayed out. If you are running in an SSI environment, then it is impossible to tell at a glance if a z/VM guest is inactive on all members of the cluster, as the views are generated on a per-member basis. For this reason, you must open up the detailed information panel to determine if this guest is active on any members in the SSI. This panel is accessible by either double-clicking on the guest icon or right clicking on the guest icon in any view and selecting “Display Information.” The middle of the data tab displays a table showing the state of the guest on each member of the SSI cluster. In Figure 7-45, guest LNXSL1 is Inactive in all members of the cluster.

Figure 7-45 Status of guest in members of a cluster

Once the z/VM guest is inactive, right click on the icon and select Cloning --> Convert VM User to Prototype, as shown in Figure 7-46



Figure 7-46 Select “Convert VM User to Prototype” from an Inactive z/VM Guest

This opens a “Create Prototype From z/VM User” panel as shown in Figure 7-47.

Figure 7-47 Create Prototype from z/VM User



The panel has the following fields:

z/VM Guest Name: This field actually has nothing to do with a z/VM guest. This value sets the name of the new prototype directory entry that IBM Wave for z/VM will instruct your directory manager to create. This name is restricted to eight characters.

z/VM Full Name: Because the name passed to the directory manager is limited to eight characters, IBM Wave for z/VM provides an additional field to allow for a more verbose or descriptive name. This field is stored in the IBM Wave for z/VM knowledgebase as metadata.

Description: This field allows for additional informaiton about the prototype. This could be used to describe software packages and versions associated with this prototype. Given that clones from a IBM Wave for z/VM prototype duplicate the Linux usernames and passwords of the original prototype, this could also list the usernames and passwords on the machine.

Project: This field allows for the prototype to be associated with a default project. While this is an optional field, IBM Wave for z/VM will present an error message every time a user attempts to clone from a prototype that lacks a default project.

DASD Group: This field selects the storage group to be used for creating new minidisks during cloning operations from this prototype.

When the fields are complete, click the create button.A workunit is dispatched to the BTS, as shown in Figure 7-48.



Figure 7-48 The Create Prototype from z/VM User Panel

Once the prototype is created, the z/VM guest used to create the prototype disappears from the z/VM User Groups tab and the associated prototype appears in the Prototypes tab (Figure 7-49).

Figure 7-49 The new IBM Wave for z/VM Associated Prototype



Double clicking on the prototype will display information about the prototype on the Data tab (Figure 7-50).

Figure 7-50 The VIew Prototype Panel of the New IBM Wave for z/VM Associated Prototype

This is nearly identical to the directory entry of the orignal guest. The only differences are as follows:

! The directory entry has been renamed from the name of the original guest to the 1-8 character string we specified in the z/VM field from the “Create Prototype From z/VM User” panel.

! The MDISK statements have been changed into directory manager commands to Add Minidisk (AMD) with AUTOG keywords requesting that the directory manager dynamically create 3390 DASD from the storage group specified on the “Create Prototype From z/VM User” panel.

The z/VM View tab of the View Prototype panel displays the prototype directory entry associated with our IBM Wave for z/VM prototype (Figure 7-51).



Figure 7-51 The Prototype Directory Entry of the new IBM Wave for z/VM Associated Prototype

7.5 Cloning virtual machines from a golden master

Right click and select “Clone From This Prototype (Figure 7-52).

Figure 7-52 Clone from this prototype

if your prototype does not have a Default Project assigned, then you will receive an error message similar to that in Figure 7-53.



Figure 7-53 Error message

This will open the “Cloning from z/VM Prototype” panel (Figure 7-54)

Figure 7-54 Cloning from z/VM Prototype” panel

1. Select the Target z/VM System Name.2. Enter the number of clones desired.



3. Enter the base name for the clones. If multiple clones are to be created, this will be the base name of the clone and each clone will have the base plus a number.

4. Enter a password and verify the password. 5. Select the Update button. This will populate the “Clone the following users” table in the

middle of the panel, as shown in Figure 7-55. An IP address is sggested and if you have multiple clones, the Virtual Network Segment (VNS) column will be populated with suggested IP addresses on that VNS’s IP segment. The IP address can be changed by double clicking on it. A pop-up will appear as shown in Figure 7-55. Double click on the IP address, edit the field with your change and click the IP button above it to change it in the table.

6. Click the Go button to submit the clone request to the BTS.

Figure 7-55 Clone from z/VM prototype

The clones appear in the z/VM User Group Panel under USER-LOCAL as shown in Figure 7-56. Note that the icons have a red circle with a white slash over them, indicating that the WAVESRV Linux server does not have IP connectivity to the z/VM Guest. (Figure 7-56)



Figure 7-56 No connectivity indicator on new Linux guest

Connectivity is actually live (Figure 7-57) and you are able to SSH via an external ASCII terminal emulator

Figure 7-57 Verifying network connectivity

This indicates that the IBM Wave for z/VM metadata needs to be refreshed - go to Auto Detect → Refresh →Schedule z/VM Network Update as shown in Figure 7-58.

Figure 7-58 Update the z?VM network metadata



Once the metadata has been refreshed, you can see the red circle with white slash has gone away (Figure 7-59) indicating the network is now all connectable.

Figure 7-59 Network connectiable


Draft Document for Review February 26, 2014 8:58 am 8192bibl.fm

Related publications

The publications listed in this section are considered particularly suitable for a more detailed discussion of the topics covered in this book.

IBM Redbooks

The following IBM Redbooks publications provide additional information about the topic in this document. Note that some publications referenced in this list might be available in softcopy only.

! Set up Linux on IBM System z for Production, SG24-8137

! Using z/VM v 6.2 Single System Image (SSI) and Live Guest Relocation (LGR), SG24-8039

You can search for, view, download or order these documents and other Redbooks, Redpapers, Web Docs, draft and additional materials, at the following website:

ibm.com/redbooks

Other publications

These publications are also relevant as further information sources:

! IBM Wave for z/VM: Administration and Customization, SC27-6118

! IBM Wave for z/VM: User Guide and Reference, SC27-6119

Online resources

These websites are also relevant as further information sources:

! IBM Wave for z/VM product page

http://www-03.ibm.com/systems/z/solutions/cloud/wave/index.htmlDescription2

! IBM Offering Information page (announcement letters and sales manuals):

http://www.ibm.com/common/ssi/index.wss?request_locale=en

Help from IBM

IBM Support and downloads

ibm.com/support

IBM Global Services

ibm.com/services



http://www.ibm.com/support/

http://www.ibm.com/support/

http://www.ibm.com/services/

http://www.ibm.com/services/

http://www-03.ibm.com/systems/z/solutions/cloud/wave/index.htmlDescription2

http://www.ibm.com/common/ssi/index.wss?request_locale=en

8192bibl.fm Draft Document for Review February 26, 2014 8:58 am



Draft Document for Review February 26, 2014 8:58 am 8192IX.fm

Index

RRedbooks website 227

Contact us 7

8192IX.fm Draft Document for Review February 26, 2014 8:58 am


To determine the spine w

idth of a book, you divide the paper PP

I into the number of pages in the book. A

n example is a 250 page book using P

lainfield opaque 50# smooth w

hich has a PP

I of 526. Divided

250 by 526 which equals a spine w

idth of .4752". In this case, you would use the .5” spine. N

ow select the S

pine width for the book and hide the others: S

pecial>Conditional

Text>Show

/Hide>S

pineSize(-->H

ide:)>Set . M

ove the changed Conditional text settings to all files in your book by opening the book file w

ith the spine.fm still open and File>Im

port>Formats the

Conditional Text S

ettings (ON

LY!) to the book files.

Draft D

ocument for R

eview February 26, 2014 8:58 am

8192spine.fm231

(0.1”spine)0.1”<->0.169”

53<->89 pages

(0.2”spine)0.17”<->0.473”

90<->249 pages

(1.5” spine)1.5”<-> 1.998”

789 <->1051 pages

(1.0” spine)0.875”<->1.498”

460 <-> 788 pages

(0.5” spine)0.475”<->0.873”

250 <-> 459 pages

IBM W

ave - Installation, Implem

entation and Exploitation

IBM W

ave - Installation, Im

plementation and Exploitation

IBM W



IBM W

ave - Installation, Implem

entation and Exploitation

(2.0” spine)2.0” <-> 2.498”

1052 <-> 1314 pages

(2.5” spine) 2.5”<->nnn.n”

1315<-> nnnn pages

To determine the spine w

idth of a book, you divide the paper PP

I into the number of pages in the book. A

n example is a 250 page book using P

lainfield opaque 50# smooth w

hich has a PP

I of 526. Divided

250 by 526 which equals a spine w

idth of .4752". In this case, you would use the .5” spine. N

ow select the S

pine width for the book and hide the others: S

pecial>Conditional

Text>Show

/Hide>S

pineSize(-->H

ide:)>Set . M

ove the changed Conditional text settings to all files in your book by opening the book file w

ith the spine.fm still open and File>Im

port>Formats the

Conditional Text S

ettings (ON

LY!) to the book files.

Draft D

ocument for R

eview February 26, 2014 8:58 am

8192spine.fm232

IBM W



IBM W



®

SG24-8192-00 ISBN

Draft Document for Review February 26, 2014 9:00 am

INTERNATIONAL TECHNICALSUPPORTORGANIZATION

BUILDING TECHNICAL INFORMATION BASED ON PRACTICAL EXPERIENCE

IBM Redbooks are developed by the IBM International Technical Support Organization. Experts from IBM, Customers and Partners from around the world create timely technical information based on realistic scenarios. Specific recommendations are provided to help you implement IT solutions more effectively in your environment.

For more information:ibm.com/redbooks

®

IBM Wave for z/VM - Installation, Implementation and

tbd

Back cover

ibm wave redbook

Documents