Implementing tws extended agent for tivoli storage manager sg246030

Implementing TWS
Extended agent for
Tivoli Storage Manager
Insider’s guide to writing a TWS
Extended agent

Ready-to-use solution for TSM
and TWS integration

TSM Extended agent
code included

Vasfi Gucer
Henry Daboub
Warren Gill
Denise Kikumoto
Tina Lamacchia

ibm.com/redbooks

SG24-6030-00

International Technical Support Organization

Implementing TWS
Extended agent for Tivoli Storage Manager

March 2001

Take Note!
Before using this information and the product it supports, be sure to read the general information in
Appendix C, “Special notices” on page 89.

First Edition (March 2001)

This edition applies to Tivoli Workload Scheduler 7.0 and Tivoli Storage Manager 4.1

Comments may be addressed to:
IBM Corporation, International Technical Support Organization
Dept. JN9B Building 003 Internal Zip 2834
11400 Burnet Road
Austin, Texas 78758-3493

When you send information to IBM, you grant IBM a non-exclusive right to use or distribute the
information in any way it believes appropriate without incurring any obligation to you.

© Copyright International Business Machines Corporation 2001. All rights reserved.
Note to U.S Government Users – Documentation related to restricted rights – Use, duplication or disclosure is subject
to restrictions set forth in GSA ADP Schedule Contract with IBM Corp.

Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Tables. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .ix
The team that wrote this redbook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Comments welcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x

Chapter 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
1.1 TWS overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
1.1.1 Software configurations used for this redbook . . . . . . . . . . . . . . .2
1.1.2 TWS concepts and terminology . . . . . . . . . . . . . . . . . . . . . . . . . .2
1.1.3 TWS architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
1.1.4 Extended agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4
1.2 TSM overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7
1.2.1 What are the benefits of a TSM Extended agent for TWS? . . . . . .9
1.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Chapter 2. Extended agent functions . . . .. . . . . .. . . . . . . . . . . . . . . . 13
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 13
2.2 Workstation definition . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 13
2.3 Method options file . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 14
2.4 Access method interface . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 15
2.4.1 Method command line syntax . . . . .. . . . . .. . . . . . . . . . . . . . . . 15
2.4.2 Task options . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 16
2.4.3 Example . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 18
2.5 Method response messages . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 20
2.6 Execution and troubleshooting . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 21
2.6.1 Executing the method . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 21
2.6.2 Killing a job. . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 21
2.6.3 Method troubleshooting . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 21
2.7 Summary . . . . . . . . . . . . . . . . . . . . . . . .. . . . . .. . . . . . . . . . . . . . . . 22

Chapter 3. Case study - TSM Extended agent . . . . . . . . . . . . . . . . . . . 23
3.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

Chapter 4. Sample scenarios . . . . . . . . . . .. . . . . .. . . . .. . . . . .. . . . . 43
4.1 Description of the scenarios . . . . . . . . . .. . . . . .. . . . .. . . . . .. . . . . 43
4.1.1 Database backup . . . . . . . . . . . . . .. . . . . .. . . . .. . . . . .. . . . . 43
4.1.2 Backup device configuration . . . . . .. . . . . .. . . . .. . . . . .. . . . . 63
4.1.3 Backup volume history . . . . . . . . . .. . . . . .. . . . .. . . . . .. . . . . 64
4.1.4 Clean volume history . . . . . . . . . . .. . . . . .. . . . .. . . . . .. . . . . 65
4.1.5 Expiration process . . . . . . . . . . . . .. . . . . .. . . . .. . . . . .. . . . . 66
4.1.6 Reclamation process . . . . . . . . . . .. . . . . .. . . . .. . . . . .. . . . . 67

iii

4.1.7 Migration process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
4.1.8 Restore . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
4.2 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

Appendix A. TSW Extended agent for TSM source code. . . . . . . . . . . . 71

Appendix B. Using the additional material . . . . . . . . ....... ...... .. 87
B.1 Locating the additional material on the Internet . . . . . ....... ...... .. 87
B.2 Using the Web material. . . . . . . . . . . . . . . . . . . . . . . . ....... ...... .. 87
B.2.1 How to use the Web material . . . . . . . . . . . . . . . ....... ...... .. 87

Appendix C. Special notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89

Appendix D. Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
D.1 IBM Redbooks collections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
D.2 Other resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
D.3 Referenced Web sites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

How to get IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
IBM Redbooks fax order form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

Abbreviations and acronyms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

IBM Redbooks review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103

iv Implementing TWS Extended agent for Tivoli Storage Manager

Figures

1. Extended agent network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2. Extended agent processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3. Extended agent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4. Requesting a backup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5. Verifying that a backup was unsuccessful . . . . . . . . . . . . . . . . . . . . . . . . . 25
6. Backing up a storage pool . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
7. Reclamation in the tape device . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
8. Migration process. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
9. Device configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
10. Maestro (TWS) Main Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
11. Maestro (TWS) Composer window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
12. List of jobs scheduled on your server . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
13. Changing job definitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
14. Maestro (TWS) Console Manager . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
15. Jobs scheduled on TWS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
16. Right-click to see the status of the job . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
17. Status of the job . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
18. Maestro (TWS) Main Window . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
19. Select Jobs to create a new job . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
20. Select the CPU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
21. New Job window for database backup . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
22. Job definition window for Database Backup . . . . . . . . . . . . . . . . . . . . . . . 47
23. List of jobs, including BACKUPDB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
24. Selecting Schedules to add a new schedule . . . . . . . . . . . . . . . . . . . . . . . 48
25. Creating a new schedule for database backup . . . . . . . . . . . . . . . . . . . . . 49
26. Defining a new schedule for database backup . . . . . . . . . . . . . . . . . . . . . 50
27. New schedule - backup db. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
28. On/Except window to definite the new schedule . . . . . . . . . . . . . . . . . . . . 51
29. Options to define a new schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
30. Follows Sched/Job window for database backup . . . . . . . . . . . . . . . . . . . 53
31. Opens Files window for database backup . . . . . . . . . . . . . . . . . . . . . . . . . 54
32. Needs resources window for database backup . . . . . . . . . . . . . . . . . . . . . 55
33. Jobs window for database backup. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
34. Adding jobs to the Current Jobs list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
35. New schedule on the List of Schedules . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
36. Selecting Conman from the Maestro (TWS) Main Window . . . . . . . . . . . . 58
37. BACKUPDB schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
38. Standard output for database backup . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
39. Checking the status of a job. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
40. Stdlist for BACKUPDB . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

v

41. Job Definition window for storage pool backup . . . . . . . . . . . . . . . . . . . . . 62
42. Job Definition window for backup device configuration . . . . . . . . . . . . . . . 64
43. Job Definition window for backup volume history . . . . . . . . . . . . . . . . . . . 65
44. Job Definition window for clean volume history . . . . . . . . . . . . . . . . . . . . . 66
45. Job Definition window for the expiration process. . . . . . . . . . . . . . . . . . . . 67
46. Job Definition window for the reclamation process . . . . . . . . . . . . . . . . . . 68
47. Job Definition window for the migration process . . . . . . . . . . . . . . . . . . . . 69
48. Job Definition window for restore. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

vi Implementing TWS Extended agent for Tivoli Storage Manager

Tables

1. Task types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2. Task options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

© Copyright IBM Corp. 2001 vii

viii Implementing TWS Extended agent for Tivoli Storage Manager

Preface

Tivoli Workload Scheduler (TWS) is Tivoli’s strategic, multiplatform
distributed scheduling product, providing high-volume, complex scheduling
capability. Although TWS provides native support for many platforms and
applications, it is possible to extend its robust scheduling capabilities to cover
additional platforms and applications. Writing an Extended agent
accomplishes this.

This redbook will show you how to write a TWS Extended agent. The
Extended agent allows you to schedule on platforms and applications for
which TWS has no native agent. Using the Extended agent, you can integrate
Tivoli Storage Manager (TSM) with TWS. TWS’s scheduling facility allows
you to assign dependencies among TSM-scheduled tasks or to assign limits
or priorities. By extending TWS to schedule these TSM tasks, you can take
advantage of its advanced scheduling capabilities.

This redbook will be essential for those who will write a TWS Extended agent
or who will use TWS to schedule TSM functions.

The team that wrote this redbook
This redbook was produced by a team of specialists from around the world
working at the International Technical Support Organization (ITSO), Austin
Center.

Vasfi Gucer is a Project Leader at the ITSO, Austin Center. He worked with IBM
Turkey for 10 years and has been with the ITSO since January 1999. He has
more than eight years of experience in systems management, networking
hardware, and distributed platform software. He has worked on various Tivoli
customer projects as a systems architect in Turkey and the U.S. Vasfi is also a
Certified Tivoli Consultant.

Henry Daboub is currently a member of the Tivoli Customer Support Global
Response Team (GRT). The GRT provides on-site support to Tivoli
customers with an emphasis on large environments. Henry has more than 12
years of experience in software development and support, and has supported
IBM workload management products since 1997.

Warren Gill is the Product Marketing Manager for TWS and Tivoli Operations
Planning and Control. He has been a courseware designer and trainer for
TWS and its predecessor Maestro, as well as a customer support engineer.

© Copyright IBM Corp. 2001 ix

Warren has been with Tivoli Systems since December 1997, when Tivoli
acquired Unison Software. He has been using TWS since 1990.

Denise Kikumoto is an IT specialist working for IBM Brazil since May 1997.
She is responsible for AIX server and Lotus Notes administration. She also
has extensive experience with TSM. She holds a degree in systems analysis.

Tina Lamacchia has been in the IT realm for 18 years. Her career started
with programming, and then administrating hundreds of UNIX systems. She
has been with Tivoli Systems for more than three years. She has worked with
TWS for two years and managed a TWS support team. Prior to joining Tivoli
Systems, she was a TWS customer for three years and deployed the product
set (including SAP Extended agent) in an Enterprise Systems Management
(ESM) environment. Currently she is working in the GRT to assist ESM
customers with various deployments in TWS and Tivoli Management
Environment (TME).

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

Tivoli Systems, International Technical Support Organization, Austin
Center
Sergio Juri, Richard Harrison, Cesare Pagano

IBM, International Technical Support Organization, Austin Center
Leah Nelson

Comments welcome
Your comments are important to us!

We want our redbooks to be as helpful as possible. Please send us your
comments about this or other redbooks in one of the following ways:
• Fax the evaluation form found in “IBM Redbooks review” on page 103 to
the fax number shown on the form.
• Use the online evaluation form found at ibm.com/redbooks
• Send your comments in an Internet note to redbook@us.ibm.com

x Implementing TWS Extended agent for Tivoli Storage Manager

Chapter 1. Introduction

This redbook is designed to illustrate the implementation of a Tivoli Workload
Scheduler (TWS) Extended agent. The implementation the redbook team
designed facilitates a communication between TWS V7.0and Tivoli Storage
Manager (TSM) V4.1 to initiate regularly scheduled tasks on the TSM server.
TSM’s server-prompted scheduling facility allows us to initiate client backups
on machines defined as TSM clients in server-prompted scheduling mode.

1.1 TWS overview
TWS is a multiplatform distributed scheduling system that provides
high-volume, complex scheduling capability. A powerful scheduling language
allows for precise coding of job streams with dependencies on other jobs,
files, virtual resources, operator prompts, and jobs in other scheduling
environments.

A Java-based graphical user interface (GUI) known as the Job Scheduling
Console (JSC) provides the user with the option of a dialog-based graphical
interface. JSC allows a user: to create/modify scheduling objects, to create
an execution plan for a batch workload, and to monitor/manage the plan
when it is executing. The more experienced user can use a command line
interface (CLI) for monitoring, scheduling, or troubleshooting.

An application programming interface (API) allows TWS to be integrated with
any application with a scheduling facility of its own. This redbook explores
that interface. Chapter 5 of the Tivoli Workload Scheduler 7.0 Reference
Guide, GC32-0424 provides an overview of the TWS Extended agent API.

There are two basic aspects to job scheduling in TWS: the database and the
plan. The database contains all the definitions for scheduling objects (for
example, jobs, job streams, resources, and workstations). It also holds
statistics about job and job stream execution, as well as information on the
user ID that created an object and when an object was last modified. The plan
contains all job scheduling activity planned for a one-day period. In TWS the
plan is created every 24 hours and consists of all the jobs, job streams,
dependencies, and other scheduling objects referenced in the upcoming day.
All job streams for which you have created a run cycle are automatically
scheduled and included in the plan. At the end of the day, the jobs and job
streams not successfully executed can be rolled over into the next day’s plan.

© Copyright IBM Corp. 2001 1

1.1.1 Software configurations used for this redbook
An IBM RS/6000 43P workstation running AIX V4.3.3.0 was used in the
development of this redbook. The following software was loaded onto this
system:
• Tivoli Workload Scheduler V7.0
• Tivoli Management Framework V3.7
• Tivoli Job Scheduling Services V1.1
• Tivoli TWS Connector V7.0
• Tivoli Storage Manager V4.1 for AIX

1.1.2 TWS concepts and terminology
TWS uses these important concepts:
• Job streams and calendars
The job streams created using the JSC are central to TWS’s ability to
manage batch job execution. Each job stream is scheduled to run on a
specific set of dates and times, and consists of a list of jobs that execute
as a unit (such as the weekly backup application), along with times,
priorities, and other dependencies that determine the exact order of
execution.
Job streams are dated using actual dates, days of the week, or calendars.
A calendar is a set of specific dates. You can create as many calendars as
required to meet your scheduling needs. For example, you can define a
calendar named PAYDAYS containing a list of pay dates, a calendar
named MONTHEND containing a list of each last business day of the
month for the current year, and a calendar named HOLIDAYS containing a
list of your company’s holidays. At the start of each processing day, TWS
automatically selects all the job streams that run on that day, and carries
forward uncompleted job streams from the previous day.
• Workstations
A workstation is usually an individual computer on which jobs and job
streams are executed. A workstation definition is required for every
computer that executes jobs in the TWS network. Workstation definitions
primarily refer to physical workstations.
However, in the case of Extended agents and network agents, the
workstations are logical definitions that must be hosted by a physical TWS
workstation.

2 Implementing TWS Extended agent for Tivoli Storage Manager

There are several types of workstations in a TWS network:
• Master Domain Manager
The Domain Manager in the topmost domain of a TWS network. It
contains the centralized database files used to document scheduling
objects. It creates the production plan at the start of each day, and
performs all logging and reporting for the network.
• Domain Manager
A Domain Manager acts as a repeater, which allows TWS to scale to large
numbers of machines. The Domain Manager forwards messages between
the Master Domain Manager and the agents.
• Backup Master
A Fault-tolerant Agent (FTA) capable of temporarily assuming the
responsibilities of its Domain Manager in a failover situation. This is
enabled by marking the workstation as Full Status and Resolve
Dependencies.
• Fault-tolerant Agent
A workstation capable of resolving local dependencies and launching its
jobs in the absence of a Domain Manager. The FTA is initialized at the
beginning of the production day with all of the scheduling objects needed
to perform the assigned workload.
• Standard agent
A workstation that launches jobs only under the direction of its Domain
Manager. Each job launch on the standard agent is directed in real time by
the Domain Manager or Master.
• Extended agent
A logical workstation definition that enables you to launch and control jobs
on other systems and applications, such as PeopleSoft, Oracle
Applications, SAP R/3, and MVS JES2 and JES3. Users can also write
Extended agents, as presented in this redbook.
• Network agent
A logical workstation definition for creating dependencies between jobs
and job streams in separate TWS networks.

1.1.3 TWS architecture
The Master Domain Manager contains the centralized database files used to
store scheduling objects. It creates the production plan at the start of each
day, distributes the plan to the FTAs and Domain Managers in the Master
Domain, and logs all transactions on the TWS network.

Chapter 1. Introduction 3

All communications to agents are routed through the Domain Manager, which
is the management hub in a domain. The network can be managed by a mix
of agents. FTAs are capable of resolving local dependencies and launching
jobs should a network interruption cause them to lose communication with
their Domain Managers. They can do this because each one is given a set of
scheduling instructions at the beginning of every processing day.

Version 7.0 introduced a new Java GUI: the Job Scheduling Console (JSC).
The JSC provides a common interface to both TWS and Operations, Planning
and Control (OPC).

1.1.4 Extended agents
TWS Extended agents are programs that allow TWS to manipulate and to get
information about objects in other scheduling environments. Extended agents
use open scheduling APIs and protocols, and they provide an effective
mechanism for extending TWS’s scheduling capabilities to foreign platforms
and applications. Extended agents also allow segregation of applications at
the workstation level by providing an alternative method other than jobmanrc
process. This allows some applications to be treated differently using an
alternative job scheduling method.

The connection between TWS and an Extended agent is shown in Figure 1.
The Extended agent is installed on one of the TWS hosts. TWS accepts
information from the Extended agent. The interface between TWS and an
Extended agent is called the access method. The access method is a shell
script or program which resides in ~TWSHOME/methods directory.

.

Figure 1. Extended agent network


Extended agent processing is explained in Figure 2. The sequence of
operations is as follows:

Note
This explanation does not consider network communication issues. It is
accepted that the Extended agent resides on a TWS server. If the
Extended agent resides on an FTA, the sequence of operations for
communication with the Extended agent will be the same but there will be
additional processes for communication between the TWS Master and the
FTA.

1. Batchman process on the FTA talks to the jobman (jobman.exe on an
NT system), which is owned by the root user ID.
2. Jobman invokes JOBMAN (jobmon.exe on an NT system) process in
the context of the TWS user (maestro, for example).
3. JOBMAN talks to the access method.
4. The access method invokes a Remote Function Call (RFC).
5. The access method consults with the <method.opts> file.
6. The access method talks to the system or the application’s job.
7. The access method and the job communicate with each other.
8. The method passes the information back to the TWS host through
JOBMAN.

Figure 2. Extended agent processing


The JOBMAN process launches the access method script to perform one of
these tasks:
• Launch a job
• Manage a job
• Check for the existence of a file to satisfy an OPENS dependency
• Get the status of an external job

The syntax of the method execution is as follows:

methodname -t task options -- taskstring

where:
task: Is one of the following:
• Launch a job (LJ)
• Manage a previously launched job (MJ)
• Check availability of a file – OPENS dependency (CF)
• Get the status of an external job (GS)
options: Are the list of job properties.
taskstring: Is the string to execute from the job definition.

The following are the options (related with the job’s properties) that should be
passed to the access method:
• Workstation/Host/Master
• Workstation’s node definition
• Workstation’s port definition
• The current production run number
• The job’s run number
• The job’s schedule name
• The date of the schedule (in two formats)
• The user ID (logon) for the job
• The path to the output (stdlist) file for the job
• The job ID
• The job number
• The string from the SCRIPTNAME or DOCOMMAND entry in the job definition


An example method invocation is as follows, where job TEST is executed on
the Extended agent workstation itso6 using the user ID itso and method
name wgetmethod:

wgetmethod -t LJ
-c ITSO6,ITSO7,ITSO7
-n ITSO6
-p 31111 -r 143,143 -s MACDSH91
-d 20000410,955381986 -l itso
-o /opt/maestro/stdlist/2000.04.10/O13676.1053
-j TEST,13676 -- /home/itso//batchjobs/killer

The current Extended agents TWS provides are:
• UNIX Remote Shell
• UNIX Local
• MVS (JES,OPC, CA7)
• SAP R/3 Batch
• PeopleSoft
• Oracle Applications

Note
In addition to TWS-provided Extended agents, you can always write your
own Extended agents for platforms or applications thatare not TWS
supported using the open API that is documented in the Tivoli Workload
Scheduler 7.0 Reference Guide, GC32-0424.

1.2 TSM overview
TSM is an end-to-end, scalable storage management solution spanning palm
tops to mainframes on more than 35 platforms. Its features include:
• Centralized storage management
• Storage Area Network (SAN) features, such as LAN-free backup and tape
sharing
• Automated network incremental and subfile backup, archive, and retrieval
• Fast recovery time
• Space management file migration


• High speed policy-based disaster recovery
• Data protection offered for most popular groupware, e-mail, databases,
and applications

TSM is the core product of the Tivoli Storage Management product set. It
provides a solution for distributed data and storage management in an
enterprise network environment. It is the next generation of the product
originally released by IBM as ADSTAR Distributed Storage Manager (ADSM).

The base functions TSM and its complementary products provide are:
• Data protection, including:
- Operational backup and restore of data: The backup process
creates a copy of the data protect against the operational loss or
destruction of file or application data. The customer defines how often
to back up (frequency) and how many numbers of copies (versions) to
hold.
The restore process places the backup copy of the data into a
customer-designated system or workstation.
- Disaster recovery: All activities to organize, manage, and automate
the recovery process from a major loss of IT infrastructure and data
across the enterprise. This includes processes to move data offsite into
a secure vault location, to rebuilt IT infrastructure, and to reload data
successfully in an acceptable time frame.
• Storage resource management, including:
- Vital record retention, archive, and retrieval: The archive process
creates a copy of a file or a set of files representing an endpoint of a
process for long-term storage. Files can remain on the local storage
media or can be deleted. The customer controls how long (retention
period) an archive copy is to be retained.
The retrieval process locates the copies within the archival storage and
places them into a costumer designated system or workstation.
- Hierarchical space management: This process provides the
automatic and transparent movement of operational data from the user
system disk space to a central storage repository. If the user accesses
this data, it is dynamically and transparently restored to the client
storage.


1.2.1 What are the benefits of a TSM Extended agent for TWS?
TSM administrators must perform several types of operations regularly each
day. TSM contains a built-in scheduling facility, which provides a simple
mechanism to automate routine tasks. This scheduling facility does not
provide the ability to assign dependencies among scheduled tasks or to
assign limits or priorities. By extending TWS to schedule these tasks, you can
take advantage of its advanced scheduling capabilities.

The following common TSM tasks were implemented for this example:
• Database backup (BACKUP DB)
• Volume history backup (BACKUP VOLHISTORY)
• Device configuration backup ( BACKUP DEVCONFIG)
• Delete volume history ( DELETE VOLHISTORY)
• Inventory expiration ( EXPIRE INVENTORY)
• Client backup

All TSM tasks executed by the TSM Extended agent ( tsmxagent), use the TSM
Administrative Client Interface ( dsmadmc). The user ID used to log in to the
dsmadmc interface is fetched from the tsmAdmin variable in the
tsmxagent.opts file in the TWS methods directory. The password is retrieved
from the TWS parameter object named TSMPASS.

Client backups are performed as follows:
• A schedule is defined to run a backup immediately (DEFINE SCHEDULE).
• The schedule is associated with the client specified (DEFINE ASSOCIATION).
• The schedule is monitored until completion or timeout ( QUERY EVENT).
• The schedule is deleted ( DELETE SCHEDULE).
• The status is reported to TWS.

As shown in Figure 3 on page 10, Extended agents are used to extend the
job scheduling functions of TWS to other systems and applications.


Extended agent External
Host system or
application

TWS network

Figure 3. Extended agent

An Extended agent is defined as a workstation that has a host and an access
method. The host is any other workstation, except another Extended agent.
The access method is a Tivoli-supplied or user-supplied script or program
that the host executes whenever the Extended agent is referenced in the
production plan. For example, to launch a job on an Extended agent, the host
executes the access method, passing it job details as command line options.
The access method communicates with the external system or application to
launch the job and return the status of the job.

Each Extended agent must have a logical workstation definition. This logical
workstation must be hosted by a TWS physical workstation, either a Master,
Domain Manager, or FTA workstation. The Extended agent workstation
definition references the name of the access method and the host
workstation. When jobs are launched on the Extended agent workstation, the
access method is called and passes the job information to the external
system. For an example of defining an Extended agent workstation, see the
Tivoli Workload Scheduler 7.0 User’s Guide, GC32-0423.

1.3 Summary
TWS is a multiplatform distributed scheduling program that provides
high-volume, complex scheduling capability for many systems and
applications. Master Domain Manager, Domain Manager, Backup Master,
Fault-tolerant Agent and standard agent are various types of workstations in
a TWS network.

You can extend the number of platforms and applications TWS supports by
writing TWS Extended agents which allow TWS to manipulate and to get
information about objects in other scheduling environments.


TSM is a scalable storage management solution spanning palm tops to
mainframes on more than 35 platforms. Centralized storage management,
LAN-free backup and tape sharing, and automated network incremental and
subfile backup, archive, and retrieval are some of the important functions that
TSM provides.


Chapter 2. Extended agent functions

This chapter describes the functions of the Tivoli Workload Scheduler (TWS)
Extended agent method, the access method interface and the
communications protocols between jobs running within an access method.

2.1 Introduction
The Extended agent is defined as a workstation that has a host and an
access method. The host is any other TWS workstation, except another
Extended agent. The access method is a Tivoli-supplied or user-supplied
script or program that the host executes whenever the Extended agent is
referenced in the production plan.

For example, to launch a job on an Extended agent, the host executes the
access method, passing it job details as command line options. The access
method communicates with the external system or application to launch the
job and return the status of the job.

2.2 Workstation definition
Each Extended agent must have a logical workstation definition. This logical
workstation must be hosted by a TWS physical workstation, either a Master,
Domain Manager, or FTA workstation. The Extended agent workstation
definition references the name of the access method and the host
workstation. Workstation definitions are created using the command line
interface, Composer, or Job Scheduling Console (JSC). When jobs are
launched on the Extended agent workstation, the access method is called
and passes the job information to the external system.

Suppose, for the purpose of illustration, that you have a TWS network
comprised of three workstations: a Master workstation (called Copernicus), a
Fault-tolerant Agent (FTA) workstation (Kepler), and an Extended agent
workstation (Galileo). The following is an example of the Extended agent
workstation definition:
CPUNAME galileo
OS OTHER
NODE focus
TCPADDR 1609
FOR MAESTRO
HOST kepler
ACCESS nuncius


END

In this case, Galileo is the name of the Extended agent workstation, focus is
the node name, 1609 is the TCP address, Kepler is the host, and nuncius is
the name of the access method. An executable file (script or program) named
nuncius must reside in the /methods subdirectory of the TWS installation. For
Extended agents only, the node and tcpaddr definitions are left to the method
developer to define. In other words, depending on the nature of the access
method, node might have different uses.

2.3 Method options file
By default, when executing an Extended agent job, TWS tries to execute the
method file as the logon defined for each job to be run by the method. When
executing the method script to check for an OPENS dependency (using the
check file [CF] task) TWS executes the method as root by default. You can
change this behavior by using a method options file to specify special login
information and other options for your access method. For example, you may
wish to use the logon field of your agent’s job definitions to represent a logon
to a foreign system that does not exist on the Extended agent’s host.

TWS reads the file, if it exists, before executing a method. If the file is
modified after TWS is started, the changes take effect when it is stopped and
restarted. The file can contain TWS options and any other method
information you wish to include. The options recognized by TWS are:
LJuser=username
CFuser=username

where:
LJuser=username: Specifies the login to use for the launch job (LJ) and
manage job (MJ) tasks. The default is the login from the
job definition.
CFuser=username: Specifies the login to use for the CF task. The default is
root for UNIX, and for Windows NT it is the user name of
the account in which TWS was installed.


Note
If the Extended agent’s host is a Windows NT computer, these users must
be defined as TWS user objects. The options file must have the same path
name as its access method, with an .opts file extension. For example, the
Windows NT path name of an options file for a method named nuncius is
WShomemethodsnuncius.opts.

2.4 Access method interface
The interface between TWS and an access method consists of information
passed to the method on the command line, and messages returned to TWS
in stdout.

2.4.1 Method command line syntax
The TWS host runs an access method using the following command line
syntax:
methodname -t task options -- taskstring

where:

methodname: Specifies the file name of the access method. In our example,
this is nuncius. All access methods must be stored in the
WShome/methods directory.

-t task: Specifies the task to be performed, where task is one of the
options listed in Table 1 on page 16:

Chapter 2. Extended agent functions 15

Table 1. Task types

Task Description

LJ Launches a job. The access method is called with this task option when
TWS launches a job. TWS will expect the access method to continue
running until the job it executes is complete.

MJ Manages a previously launched job. Use this option to synchronize job(s)
if a prior LJ task terminated unexpectedly. The MJ method is called for
under only a few circumstances. For example, TWS or its batchman
process terminates (an operator issues a Conman STOP command to the
agent’s host) while an Extended agent job is running. When the TWS
processing is restarted, it will execute the Extended agent’s access method
for each job that was previously running, passing it the MJ task and the
same arguments initially sent to the job. The access method should use the
MJ task to determine the state of the previously running job and report that
state back to TWS.

CF Checks the availability of a file. Use this option to check file OPENS
dependencies. Using the Galileo workstation example, if a schedule
contained the dependency (OPENS GALILEO#/this/is/my/file) TWS
would call the access method for Galileo and pass it the CF task and the
path (/this/is/my/file).

options: Specifies the options associated with the task. See
Section 2.4.2, “Task options” on page 16 for more information.

taskstring: A string of up to 255 characters associated with the task. See
Section 2.4.2, “Task options” on page 16 for more information.

2.4.2 Task options
The task options are listed in Table 2. An “x” means the option is valid for the
task.
Table 2. Task options

Task Options Task string

-t -c -n -p -r -s -d -l -o -j -q

LJ x x x x x x x x x ljstring

MJ x x x x x x x x x mjstring

CF x x x x cfstring


-c xagent,host,master: Specifies the TWS names of the Extended
agent, the host, and the Master Domain
Manager, separated by commas.
-n nodename: Specifies the node name of the computer
associated with the Extended agent, if any.
This is defined in the Extended agent’s
workstation definition node field.
-p portnumber: Specifies the TCP port number associated
with the Extended agent, if any. This is
defined in the Extended agent’s workstation
definition TCP Address field.
-r currentrun,specificrun: Specifies the current run number of TWS
and the specific run number associated with
the job, separated by a comma. The current
and specific run numbers might be different
if the job is carried forward from an earlier
run.
-s jstream: Specifies the name of the job’s job stream.
-d scheddate,epoch: Specifies the schedule date (yymmdd) and
the epoch equivalent, separated by a
comma.
-l user: Specifies the job’s user name. This is
defined in the job definition Logon field.
-o stdlist: Specifies the full path name of the job’s
standard list file. Any output from the job
must be written to this file.
-j jobname,id: Specifies the job’s name and the unique
identifier assigned by TWS, separated by a
comma. The name is defined in the job
definition Job Name field.
-q qualifier: Specifies the qualifier to be used in a test
command issued by the method against the
file.
-- ljstring: Used with the LJ task. The string from the
script file or command field of the job
definition.


-- mjstring: Used with the MJ task. The information
provided to TWS by the method in a %CJ
response to an LJ task. Usually, this
identifies the job that was launched. For
example, a UNIX method can provide the
process identification (PID) of the job it
launched, which TWS sends as part of an
MJ task.
-- cfstring: Used with the CF task. For a file OPENS
dependency, the string from the Opens Files
field of the job stream definition.

2.4.3 Example
Using our previous workstation definition and the options previously listed,
here is an example of how your access method would be called from TWS. In
this example, a job defined as tsmjob1 runs the command ADMIN VHBACKUP
/opt/tsm/dat/fa20001020 in the job stream (schedule) DAILYBAK:
/opt/maestro/methods/nuncius -t LJ -c GALILEO,KEPLER,COPERNICUS -n focus -p
1609 -r 109,109 -s DAILYBAK -d 20001207,976147200 -l root -o
/opt/maestro/stdlist/2000.12.07/O9041.1414 -j TSMJOB1,9041-- ADMIN VHBACKUP
/opt/tsm/dat/fa20001020

Let us examine the command structure from the example.

/opt/maestro/methods/nuncius

This is the directory to the access method script.

-t LJ

The task option LJ was sent, indicating that a job is being run.

-c GALILEO,KEPLER,COPERNICUS

This option shows that Galileo is the Extended agent’s name, Kepler is the
agent’s host’s name, and Copernicus is the TWS Master’s name.

-n focus

Focus is the node name of the Extended agent. This name comes directly
from the workstation definition for Galileo.


-p 1609

The number 1609 is the tcpaddr port definition from the workstation definition
for Galileo.

-r 109,109

The current run number (TWS production day) is 109, as is the job’s run
number. If, for example this option was “-r 110,109”, it would signify that the
job stream (schedule) from which this job is running has been carried forward
from a previous day.

-s DAILYBAK

The job stream (schedule) name from which this job is running is DAILYBAK.

-d 20001207,976147200

The current scheduling production run date is December 7, 2000. Remember
that the scheduling production run date may be different than the actual
calendar run date.

-l root

The job run by this method should log on to the system as root. This
information comes from the job definition.

-o /opt/maestro/stdlist/2000.12.07/O9041.1414

All output (stdout and stderr) is being redirected to the file
/opt/maestro/stdlist/2000.12.07/O9041.1414 on the host workstation.

-j TSMJOB1,9041

The name of the job being run is TSMJOB1, and its job number (as seen in the
Conman SHOWJOBS command) is 9041. On UNIX systems only, 9041 is also
the parent process ID for this job.

-- ADMIN VHBACKUP /opt/tsm/dat/fa20001020

The rest of the arguments after a double-hyphen are directly from the
SCRIPTNAME or DOCOMMAND entry in the TWS job definition. In this case,
the application understands the command ADMIN VHBACKUP and processes the
command appropriately. Typically the access method ignores any options
after the double hyphen and passes them directly to the target application.
This is, of course, up to the method’s author to determine.


Upon execution of this job, its status would be shown in the Console Manager
as:

Est) (Est)
CPU Schedule Job State Pr Start Elapse Dependencies

GALILEO #DAILYBAK ******** SUCC 10 14:14 01:10
TSMJOB1 SUCC 10 14:14 01:10 #J9041
%

2.5 Method response messages
Methods returns information to TWS in messages written to stdout. Each line
starting with a percent sign (%) and ending with a new line is interpreted as a
message. The messages have the following format:
%CJ state [mjstring]
%JS [cputime]
%UT [errormessage]

where:
CJ: Changes the job state.
state: The state to which the job is changed. All TWS job
states are valid except hold and ready.
mjstring: A string of up to 255 characters that TWS will include in
any MJ task associated with the job.
JS [cputime]: Indicates successful completion of a job and provides its
elapsed run time in seconds. This message must be
presented (usually with an echo statement) before the
method ends in order for the job to be considered
successful.
UT [errormessage]: Indicates that the requested task is not supported by the
method. Displays a string of up to 255 characters that
TWS will include in its error message (stored in the TWS
standard list file; typically this file is
~maestro/stdlist/yyyy.mm.dd/MAESTRO).

You can use these messages within your access method to relay information
to the TWS Console Manager. By changing the status of your job, operators
can pay special attention to your jobs by filtering their console windows by job
status. For example, you might have jobs toggle between wait and exec


states by issuing an echo %CJ WAIT or echo %CJ EXEC while the job connects to
the application and executes its function.

2.6 Execution and troubleshooting
The Extended agent’s access method is executed by TWS’s jobman process,
much as a standard job is launched through jobmanrc.

2.6.1 Executing the method
Before the method is launched, TWS establishes an execution environment
for the method:
• The LJuser parameter is read from the method options file to determine
the user account with which to run the method. If the parameter is not
present or the options file does not exist, the user account specified in the
Logon field of the job’s definition is used. In addition, the following
environment variables are set:
- HOME: User’s home directory
- LOGNAME: Login user’s name (from the job definition)
- PATH: For UNIX, /bin:/usr/bin; for Windows, %SYSTEM%SYSTEM32
- TZ: Time zone

2.6.2 Killing a job
While an access method is running an MJ or LJ task, it should trap a
SIGTERM signal (signal 15). This is the signal sent to the job when an
operator issues a kill command from the user interface. When a kill signal is
caught, the method should attempt to stop or kill the job and exit without
writing a %JS message.

2.6.3 Method troubleshooting
If the method cannot be executed, its state is set to fail. All output messages
from an access method, except those that start with a percent sign (%), are
written to the job’s standard list (stdlist) file. For GS and CF tasks that are not
associated with TWS jobs, messages are written to TWS’s standard list file.
For information regarding a problem of any kind, check these files.

For Extended agents, error, warning, and information messages are written to
TWS’s stdlist file.


• A successful job launch generates the following message:
Launched job jobname for wkstation ,#J jobid for user username .
• Failure to launch a job generates the following message:
Error launching jobname for wkstation :errortext
• Failure of a CF task generates the following message:
Error invoking methodname for wkstation :errortext
• Failure of an MJ task generates the following message:
Error managing jobname for wkstation using methodname :errortext
• When a method sends a message to Jobman that is not recognized, the
following message is generated:
Error:message invalid for jobname ,#j jobnumber for wkstation using
methodname ."first 64 characters of the offending message "

2.7 Summary
In this chapter we covered the functions of the TWS Extended agent method,
the Access Method Interface and the communications protocols between jobs
running within an access method.

The main functions used in an Extended agent method are:
• LJ: Launch job
• MJ: Manage job
• CF: Check file dependency

The Extended agent’s access method is executed by the jobman process,
much as a standard job is launched through jobmanrc.


Chapter 3. Case study - TSM Extended agent

This chapter describes scripts that will be useful to Tivoli Workload Scheduler
(TWS) and Tivoli Storage Manager (TSM) administrators.

These scripts are about TSM processes such as backup, storage pools,
migration, reclamation, volume history, and device configuration. Because, in
practice, TSM administrators generally schedule these process, it makes
sense to write a TSM Extended agent for TWS that incorporates these
functions. Using the TSM Extended agent, administrators can schedule these
TSM operations through TWS.

First, here is some background information about these operations. The
following sections should be especially useful if you are not familiar with TSM
scheduling operations.
• Backup database: Use this command to back up a TSM database to
sequential access volumes (Figure 4 on page 24). To determine how much
additional storage space a backup will require, use the QUERY DB command.
This command displays the database pages (in megabytes) that have
changed since the last backup.
The syntax of the command is:

backup db devclass=dbbackup type=full sctatch=yes wait=yes

where:
backupdb: Is the name of the devclass defined on the TSM server.
type=full: Specifies that you can run the full backup on the TSM
database. This parameter is optional. Possible options
are: incremental, full and dbsnapshot. The default is
incremental.
scratch=yes: To take the device that is scratch. The default value is yes.
This parameter is optional. If you put scratch=no, then it
will specify that scratch volume cannot be used.
wait: Specifies whether to wait for the server to complete
processing this command in the foreground. The default is
no. Possible options are: yes (the server processes the
command in foreground), and no (the server processes the
command in background).
You can find more details about the syntax of the commands in the Tivoli
Storage Manager for AIX Administrator’s Reference, GC35-0404.


Backup process

Requests a
backup

TSM server Library
User

Figure 4. Requesting a backup

If a backup completes unsuccessfully, you can verify and restart it. For
example, in Figure 5 on page 25 the backup was unsuccessful for some files
and restarted.


Check .NSF files

Adminstration
Server

vasfi.nsf vasfi.nsf

desinha.nsf desinha.nsf vasfi.nsf

danes.nsf Backup danes.nsf mcelo.nsf

mcelo.nsf mcelo.nsf

luisv.nsf luisv.nsf

Library

Figure 5. Verifying that a backup was unsuccessful

• Backup storage pool: Use this command to back up primary storage pool
files to a copy storage pool. The operation is shown in Figure 6 on
page 26.
If a file already exists in the copy storage pool, the file is not backed up
unless the copy is marked as damaged. However, a new copy is not
created if the file in the primary storage pool is also marked as damaged.
In a random-access storage pool, neither cached copies of migrated files
nor damaged files are backed up.
If migration for a storage pool starts during a storage pool backup, some
files may be migrated before they are backed up. You may want to back up
storage pools that are higher in the migration hierarchy before backing up
storage pools that are lower. For example, when performing a storage
pool backup to copy the contents of a storage pool offsite, if the process is
not done according to the existing storage pool hierarchy, some files may

Chapter 3. Case study - TSM Extended agent 25

not be copied to the copy storage pool. This could be critical for disaster
recovery purposes. When performing a storage pool backup on multiple
storage pools, the primary storage pool should be completed before the
backup process on the next storage pool.
The syntax of the command is as follows:

backup stgpool spacemgtpool copypool

where:
spacemgtpool: Is the primary backup that you want to back up.
copypool: Is the copy storage pool defined on TSM server.

Figure 6. Backing up a storage pool

• Reclamation: Specifies when the server reclaims a volume, based on the
percentage of reclaimable space on a volume (Figure 7 on page 27).
Reclamation makes the fragmented space on volumes usable again by
moving any remaining active files from one volume to another volume,
thus making the original volume available for reuse. You can specify an
integer from one to 100. The value 100 means that reclamation is not
performed.
If you change the value from the default of 100, specify a value of 50
percent or greater so that files stored on two volumes can be combined
into a single output volume.
To move data from the reclaim storage pool back to the original storage
pool, use the storage pool hierarchy. Specify the original storage pool as
the next storage pool for the reclaim storage pool.

update stgpool copypool reclaim=50


where:
reclaim: Is the percentage of the reclamation in copypool.

Figure 7. Reclamation in the tape device

• Migration: You can use a primary storage pool as the destination for
backup files, archive files, or files migrated from client nodes.
The following command is used for this purpose:

define stgpool somepool dbbackup pooltype=primary access=readwrite maxsize=nolimt
highmigh=90 lowmig=70 migprocess=1 collocate=no migcontinue=yes reusedelay=0
migdelay=0

where:
somepool: Is the pool name defined on the TSM server.
dbbackup: Is the device class name defined on the TSM server.
pooltype: Specifies that you want to define a primary storage pool.
The default value is primary.


access: Specifies how client nodes and server processes (such as
migration and reclamation) can access files in the storage
pool. The default value is readwrite, but you can also
define as readonly and unavailable.
maxsize: Specifies the maximum size for a physical file size that the
server can store in the storage pool during a session with
a client. The default value is nolimit.
highmig: Specifies that the server starts migration for this storage
pool when the amount of data in the pool reaches this
percentage of the pool’s estimated capacity. You can
specify an integer from zero to 100. The default value is
90. When the storage pool exceeds the high migration
threshold, the server can start migration of files by node,
to the next storage pool, as defined with the
NEXTSTGPOOL parameter. You can specify highmig=100
to prevent migration for this storage pool.
lowmig: Specifies that the server stops migration for this storage
pool when the amount of data in the pool reaches this
percentage of the pool’s estimated capacity. You can
specify an integer from zero to 99. The default value is 70.
When the storage pool reaches the low migration
threshold, the server does not start migration of another
node’s files. Because all file spaces that belong to a node
are migrated together, the occupancy of the storage pool
can fall below the value you specified for this parameter.
You can set lowmig=0 to permit migration to empty the
storage pool.
migprocess: Specifies the number of processes the server uses for
migrating files from this storage pool. You can specify an
integer from one to 999. The default value is one. During
the migration the server runs this number of processes in
parallel to provide the potential for improved migration
rates.
migdelay: Specifies the minimum number of days a file must remain
in a storage pool before the file becomes eligible for
migration from the storage pool. The server counts the
number of days from the day the file was stored in the
storage pool or retrieved by a client, whichever is more
recent. The default is zero, which means you do not want
to delay migration.


migcontinue: Specifies whether you allow the server to migrate files that
do not satisfy the migration delay time. The default value
is yes. Because you can require that files remain in the
storage pool for a minimum number of days, the server
may migrate all eligible files to the next storage pool yet
not meet the low migration threshold. This parameter
allows you to specify whether the server is allowed to
continue the migration process by migrating the files that
do not satisfy the migration delay time.
Figure 8 shows the migration process.

Migration process

Client Client
TSM server

Disk Disk

Backup

Library

Figure 8. Migration process

• Backup device configuration: Use this command to back up the
following information in one or more files:
- Device class definitions
- Library definitions


- Drive definitions
To restore the TSM database, the device configurations must be available.
You can use the devconfig server option to specify one or more files in
which to store device configuration information. TSM updates the files
whenever a device class, library, or drive is defined, updated, or deleted
(Figure 9).

backup devconfig filenames=device

where:
device: Is the directory that it will back up the information about
devconfig.

Device class definitions

Drive definitions

Devices

Library definitions

Figure 9. Device configuration

• Backup volume history: Use this command to back up sequential
volume history information to one or more files. You can use volume
history information when you reload the database and audit affected
storage pool volumes. If you cannot start the server, you can use the
volume history file to query the database about these volumes. The
volume history includes information about the following types of volumes:
- Database backup volumes
- Database dump volumes
- Export volumes


- The following sequential access storage pool volumes:
• Volumes added to storage pools
• Volumes reused through reclamation or move-data operations
• Volumes removed by using the delete volume command or during
reclamation of scratch volumes
The syntax of the command is:

backup volhistory filenames=volhist

where:
volhist: Is the name of the filename that will contain information about
the volhistory backup.
• Delete volume history: Use this command to delete volume history file
records that are no longer needed (for example, records for obsolete
database backup volumes).
When you delete records for volumes that are not in storage pools (for
example, database backup or export volumes), the volumes return to
Source Manager, which acquires them as scratch volumes. Scratch
volumes of device-type files are deleted. When you delete the records for
storage pools volumes remain in the TSM database. When you delete
records for recovery-plan file objects from a source server, the objects on
the target server are marked for deletion.

delete volhistory type=rpfile todate=11/31/00

where:
type: Specifies the type of records, which also meet the date and
their criteria, to delete from the volume history file. Rpfile
specifies to delete only records that contain information about
full and incremental database backup volumes and recovery
plan file volumes.
todate: Specifies the date to use to select sequential volume history
information to be deleted. TSM deletes only those records
with a date on or before the date you specify.
For more details, you can see the Tivoli Storage Manager for AIX
Administrator’s Reference , GC35-0404.


• Expire inventory : Use this command to manually start inventory
expiration processing. This inventory expiration process removes the
client backup and archive file copies from server storage based on the
policy specified in the backup and archive copy groups of the
management classes to which the files are bound. The inventory
expiration process that runs during server initialization does not remove
these virtual volumes.
Only one expiration process is allowed at any time. If an expiration
process is currently running you can not start another process.

expire inventory quiet=no wait=no skipdirs=no duration=40

where:
quiet: Specifies whether the server suppresses detailed messages
about the policy changes during the expiration processing.
This parameter is optional. The default value is no, which
specifies that the server sends detailed informational
messages. You can also put the value yes, which specifies
that the server sends only summary messages.
wait: Specifies whether to wait for the server to complete
processing this command in the foreground. This parameter is
optional. Possible values are no, which specifies that the
server processes this command in the background, and yes,
which specifies that the server processes this command in the
foreground.
skipdirs: Specifies whether the server skips directory type during the
expiration processing. Possible values are: no, which specifies
that the server will expire files and directories based upon the
appropriate policy criteria, and yes, which specifies that the
server will skip directory type objects during expiration
processing even if the directories are eligible for expiration.
duration: Specifies the maximum number of minutes for the expiration
process to run. The process stops when the specified number
of minutes have passed or when all eligible expired objects
are deleted, whichever comes first. You can specify a number
from one to 999,999 (optional).
• Restore database : To restore a database or volume to its most current
state, the log mode must have been set to roll-forward continuously from
the time that the last backup series was created.


If the log mode is set to roll-forward after a point-in-time database
restoration, a database backup starts when the server is brought up for the
first time. This can cause loss of data: A tape can have current data on it,
but because of the point-in-time restoration, it can be marked as scratch.
When the server starts for the first time, TSM may use this tape to write
the database backup, thus destroying the original data on this tape.
You can restore a database if the following are true:
- The log mode was set to roll-forward continuously from the time the last
backup series was created.
- An intact recovery log is available.
- An intact volume history file is available.
TSM requests volume mounts to load the most recent backup series and
then uses the recovery log to update the databases to its most current
state.
If the volume history file is unavailable, you can use one or more dsmc
restore db commands to restore the database to a specific point in time.
For example, to load a full backup and one or more incremental backups,
issue a dsmc restore db command for the full backup and an additional
dsmc restore db command for each incremental backup. When you use
multiple dsmc restore db commands, specify commit=no for each command
except the last one. For the last command, specify commit=yes. The
database remains in an inconsistent and unusable state until you issue a
dsmc restore db command with a commit=yes.

dsmc restore db devclass=device_class_name volumename=volume_name commit=no

where:
devclass: Specifies the name of the sequential access device class to
use. The device class must be defined in a device
configuration file.
volumename: Specifies the backup volume to use to restore the database.
Possible values are:
volume_name: Specifies the names of the volumes. To
specify multiple volumes, separate the names
with commas without intervening spaces.
file:file_name: Specifies the name of a file that contains a list
of the volumes.


commit: Specifies whether this is the last restore command needed to
restore the database. This parameter is optional. The default
value is no (specifies that you will issue one or more additional
dsmc restore db commands), but you can also define the value
yes (specifies that this is the last restore command to restore
the database).

After you analyze all the scripts, you can also test the environment and
commands in the TWS structure.

Initially install the TWS software on your system. Decide on which system
your Master will be (this is the system on which the TWS database resides,
as well as the system on which all updates are received). Please follow the
installation steps in the Tivoli Workload Scheduler 7.0 Planning and
Installation Guide, GC32-0422. Once the software is installed correctly, make
sure your display and path are setup as well.

To access the TWS database, execute the gmaestro command (which should
be located in the TWShome/bin directory). Make sure you are logged on with
the user maestro (default TWS user), or the name of the maestro user you
have chosen. Gmaestro contains the Composer screen and the Conman
screen. Further details on their functionality is explained in the document.

Figure 10 shows the Maestro Main Window the gmaestro command launched.

Figure 10. Maestro (TWS) Main Window


You can choose the Database Manager (Composer), which will allow you to
enter the CPU, CPU classes, domains, calendars, jobs, and users, or
Console Manager (Conman), which displays information about schedules,
jobs, CPUs, and parameters.

Composer can be accessed from the Master only or wherever the mozart
directory is accessible. We do not recommend that the mozart directory is
mounted or mapped as the TWS database should not be accessible to all
users. Composer reads the mozart directory and the corresponding file
pertaining to your choice. At this point you can add, modify, copy, and delete
data. Make sure you have the proper security access when performing such
operations.

Conman displays job status, allows for ad hoc submission, and allows you to
display previous production days and their status. Again, all features may be
performed if the proper access is given.

When you click the Composer button, a window (Figure 11) will appear:

Figure 11. Maestro (TWS) Composer window

You can display, add, and modify information about CPU, CPU classes,
domains, and users by clicking the corresponding button.

If you choose the Jobs button, you may use the filter feature to list all jobs
according to the chosen CPU or you can display all jobs in the TWS


database. Clicking the Jobs button, you will see a list of jobs, as seen in
Figure 12 on page 36.

Figure 12. List of jobs scheduled on your server

You can execute modifications, additions, deletions, and/or replication (Save
As choice) on an existing job. To do that, right-click the desired job, and
select Modify. The following window (Figure 13 on page 37) will appear.
Follow the instructions to make your changes or just display the job.


Figure 13. Changing job definitions

As illustrated, adding a new job should be relatively easy. However, a few
explanations are needed in order for you to take full advantage of TWS.

With TSM and TWS, you can take advantage of several options to ensure
your TSM jobs are successful. The Recovery Options, for instance can be
used to ensure that a backup is successfully executed every time. By
choosing the Continue option or Rerun option, the Recovery Job and/or
Recovery Prompt can be used to ensure backups are always successful. You
can key in the recovery job, which may be, for example, another set of tapes
to back up to if the primary tapes are not available or a different
directory/system if a file system full occurs. Whichever you chose, a backup
should always be successful. Also, you can use the Recovery Prompt as a
messaging vehicle to the operations staff on the proper execution of the
recovery job.

The Interactive option is for NT systems within your TWS environment. This
option is used if the scheduled job requires manual intervention.

In Conman (Console Manager), you can display various information as well
as submit ad hoc jobs and schedules. The information is retrieved from the
Symphony file located in the maestrohome directory.

When you click this button, a window (Figure 14 on page 38) will appear:


Figure 14. Maestro (TWS) Console Manager

By double-clicking the buttons, the corresponding information is displayed, as
summarized here:
• CPUs: Status on each TWS system is displayed. The information on its
connectivity with the Master, the run number, and the node name are
displayed. The exact information can be accessed from the command line
interface through the command conman -v.
• Domains: Displays the Master Domain, the associated Domain Manager,
and its parent.
• Schedules: Displays all schedules along with their status, priority, start
times, and dependencies
• Jobs: Displays all jobs along with their status, priority, start times, and
dependencies.
• Resources: Displays resources and their associated CPUs. Resources
can be used when two or more jobs require the use of the same resource,
for example a set of tapes (the tape is the resource). Once one job is
finished utilizing the one set of tapes or a resource, TWS releases the
resource and then next job commences.
• Prompts: Displays the prompt’s description, and its associated CPU,
schedule, and job.


• Files: Displays the files with their associated CPUs, schedules, and jobs.

All of these features can be used to compliment any Tivoli product, especially
TSM. Each feature will be discussed in more detail in Chapter 4, “Sample
scenarios” on page 43.

Figure 15 shows you the scheduled jobs.

Figure 15. Jobs scheduled on TWS

You can see the status of the scheduled job by right-clicking. This shows you
whether the job was successful, or abended or failed. Also the screen can
show you if the job and/or schedule is ready and when it will start, according
to start time or dependency columns, which can include resources, files, or
other jobs and/or schedules.

If a job fails or abends, you can see the reason why it was unsuccessful by
right-clicking on the job (Figure 16 on page 40).


Figure 16. Right-click to see the status of the job

The Stdlist option allows you to read the log (located in the
maestrohome/stdlist/MM.DD.YYYY directory) associated with that job. For
further explanation on various other options, please refer to the Tivoli
Workload Scheduler 7.0 User's Guide, GC32-0423. Figure 17 shows a
sample stdlist output.

Figure 17. Status of the job


The stdlist header contains the job name, and associated schedule and CPU
names, as well as the login name and the script location. As also shown, it
displays the process ID, and the date and time of the job execution.

The body of the stdlist contains the steps executed for the job to became
successful or unsuccessful. The information contained depends solely on its
script and the information to standard out (stdout). You can also print this
output.

3.1 Summary
In this chapter we described the following TSM functions that can be
scheduled with the TWS:
• Backup database: To backup a TSM database to sequential access
volumes.
• Backup storage pool: To backup primary storage pool files to a copy
storage pool.
• Reclamation: To reclaim a volume, based on the percentage of
reclaimable space on a volume.
• Migration: To copy files to a library when a threshold is reached on the
disk.
• Backup device configuration: To backup the device class, library and
drive definitions.
• Backup volume history: To backup sequential volume history information
to one or more files.
• Expire inventory: To manually start inventory expiration processing.
• Restore database: To restore a database or volume to its most current
state.

The information in this chapter will be especially useful if you are not familiar
with TSM scheduling operations.


Chapter 4. Sample scenarios

This chapter will address several scenarios when utilizing Tivoli Workload
Scheduler (TWS) with Tivoli Storage Manager (TSM) product sets.

This chapter will demonstrate how to schedule TSM functions within TWS.
Primarily we will create jobs and schedules in TWS to handle such functions.
We will address dependency options and display jobs/schedules output.

Note
In these scenarios we used the legacy TWS user interface (Composer and
Conman) instead of the new Job Scheduling Console (JSC), which was
available with the TWS V7.0. The TWS Extended agent for the TSM code
we introduce in this redbook is valid for TWS V7.0, and for TWS versions
prior to 7.0, in which JSC is not supported.

4.1 Description of the scenarios
The scenarios described involve information about backup databases,
backup storage pools, migration, and various other typical scenarios.

4.1.1 Database backup
Our first scenario is database backup. You can specify the type of backup that
your company requires. This depends on the necessary data’s level of
availability for recovery purposes, as well as addressing service level
agreements (SLAs) between your company and third party vendors.

We will start with scheduling a TSM function in TWS:
1. Execute the TWS interface and choose Composer from the main menu
(Figure 18 on page 44).


.

Figure 18. Maestro (TWS) Main Window

2. From the Scheduling Objects selections, choose Jobs (Figure 19 on
page 45).


.

Figure 19. Select Jobs to create a new job

3. From the Actions menu, choose New Job.
4. Select the CPU definition in which the new job will reside (Figure 20). In
TWS, a CPU must be identified with the job or where the backup script
resides.

Figure 20. Select the CPU

Chapter 4. Sample scenarios 45

5. Once you have chosen the associated CPU, type the job name and click
OK (Figure 21). The job name should be your identifier for easy ad hoc
scheduling as well as monitoring. A naming convention is highly
recommended when planning your TWS environment.

Figure 21. New Job window for database backup

6. Complete the necessary fields of the new Job Definition window
(Figure 22 on page 47). Logon refers the user ID that will execute the
job (script or command). Enter the script name or the command, making
sure the path is included (or the .jobmanrc is present in the user’s
directory. See Tivoli Workload Scheduler 7.0 Planning and Installation
Guide, GC32-0422). The path can also be a parameter if necessary
(see Tivoli Workload Scheduler 7.0 User’s Guide,GC32-0423 for more
details on the Parameter object). The Interactive option is for Windows
NT jobs to run interactively on the Windows NT desktop. Recovery
Options is an important field to consider. When a job does not does not
complete successfully (or the state is abended) the entered Recovery
Job is executed.


.

Figure 22. Job definition window for Database Backup

7. Choose File->Save and then File->Close.
8. As jobs are entered in the database, utilizing the View Filter option can
help you search for a specific job. The list of jobs is seen in Figure 23 on
page 48.


Figure 23. List of jobs, including BACKUPDB.

Once a job is defined, it must be entered in an existing or new schedule. The
following steps will assist you with schedule definitions:
1. From Composer, choose Schedules (Figure 24).

Figure 24. Selecting Schedules to add a new schedule


2. From the Actions menu, select New, to add a schedule (Figure 25). If
you want to change the parameters of an existing schedule, choose
Modify.

Figure 25. Creating a new schedule for database backup

3. If you choose the New schedule option, select the CPU associated with
the schedule and enter the schedule name (Figure 26 on page 50).
Naming convention methodology is highly recommended for easy
monitoring and ad hoc scheduling.


Figure 26. Defining a new schedule for database backup

4. Once the correct information is entered, click OK (Figure 27).

Figure 27. New schedule - backup db

5. The Schedule Definition window (Figure 28 on page 51) requires
various pieces of information.


Figure 28. On/Except window to definite the new schedule

6. Choose the date on which the schedule will execute from the
On/Except tab (Figure 28). In our scenario, the schedule will run
Everyday. Choose this option. If this was a weekly backup, you would
only wish to choose Saturday or Sunday. You can also create a
calendar with scheduled dates.


Figure 29. Options to define a new schedule

7. Under the Options tab, enter the schedule’s run time (Figure 29).
Indicate the schedule’s priority, as well as the limit of jobs running per
schedule. Please note that the offset cannot be used at the job and
schedule level.


Figure 30. Follows Sched/Job window for database backup

8. The Follows Sched/Job option (Figure 30) is one of the many TWS
dependency options. For instance, if this schedule consisted of a job
that starts the backup schedule, you may have this schedule and/or job
dependent on an existing job that exports the database. This means the
backup schedule would only start if the exporting of the database was
completed successfully or the job state was SUCC.
9. In the Opens Files tab (Figure 31 on page 54) it is not necessary to
enter a file as a dependency for our scenario. Leave this blank.


Figure 31. Opens Files window for database backup

10. Clicking the Needs Resources tab (Figure 32 on page 55), brings up
the Needs Resources window. Resources represent physical or logical
scheduling resources that can be used as dependencies for jobs and
job streams. The resource may be an available device (a tape or tape
device). If other schedules/jobs are dependent on the same device, you
may use the resource option to run schedules/jobs in a particular
sequence.


Figure 32. Needs resources window for database backup

11. The Jobs tab allows the existing job(s) to be defined for the schedule.
You must initially select the CPU and the desired job (Figure 33 on
page 56).


.

Figure 33. Jobs window for database backup

12. After selecting the desired job, click the Add button. The job should
appear in the Current Jobs column (Figure 34). Repeat this step until all
desired jobs are displayed in this column.

Figure 34. Adding jobs to the Current Jobs list


13. When you finish adding jobs, select File->Save, and then File->Close.
14. Your schedule (Figure 35) is now complete and ready to be executed
after a Jnextday run, as it will be part of your production Symphony file.
If you wish to submit the schedule immediately, you can do this from the
Conman window.

Figure 35. New schedule on the List of Schedules

Choose Conman from the main TWS window (Figure 36 on page 58) to
execute the new schedule and/or job. The report displays job
information and you can use it for various monthly reports.


Figure 36. Selecting Conman from the Maestro (TWS) Main Window

Important

You will not be able to display the newly created schedule and/or job until a
Jnextday run. The Jnextday script processes the following steps:
a. Runs schedulr to select the appropriate schedules for the new day.
b. Runs compiler to create an interim Production Control file.
c. Runs reptr to print the pre-production reports.
d. Stops TWS processing.
e. Runs stageman to carry forward uncompleted schedules, to log the
old Production Control file, and to install the new file.
f. Starts TWS processing for the new day.
g. Runs reptr to print the post-production reports from the most recent
log file.
h. Runs logman to log job statistics from the most recent log file.

15. If you wish to submit the schedule and/or job immediately, choose the
Submit option from the header tab. In the Submit window, you can
choose Schedules, Jobs, File, or Command . In this case choose Job
or Schedule. Complete the appropriate fields and your job/schedule is
submitted.


16. When a Jnextday is completed, choose the appropriate object and
display the production environment. Depending on security, this may not
be possible. Ask your TWS administrator for your security options.
Choose the Schedules object to display the schedule name, state of
the schedule, its priority, the schedule’s duration, the number of jobs in
that schedule, and the schedule’s limit and dependencies.
In our example, the BACKUPDB schedule contains its state (SUCC),
priority, start time, and other pertinent information (Figure 37).

Figure 37. BACKUPDB schedule

17. By clicking the right button of the mouse, you can view the standard
output (Figure 38).

Figure 38. Standard output for database backup


Figure 39. Checking the status of a job

18. The standard output (better known in TWS as the stdlist) displays the
header information: logon user, the job ID (or process ID), and the start
date and time (Figure 39 and Figure 40). The body of the stdlist
contains the standard output of the job and its exit code, which will
determine the status of the job.

Figure 40. Stdlist for BACKUPDB

Follow the same steps when scheduling any job and/or schedule. The
dependencies are the only differences.


Let’s summarize the steps needed to schedule a job in TWS to back up a
database. To back up a database, you will need to define a copy of a storage
pool on your server. Eliminating this copy would result in errors when
scheduling in TWS. To define a copy of a storage pool on your server, please
refer to the relevant user guide for your server platform.
1. From the Maestro (TWS) Main Window, click on Composer (Database
Manager).
2. Choose Jobs from the Scheduling Objects’ selections.
3. From the Actions menu, choose New Job.
4. After you select the corresponding CPU and enter the job name
(following your naming conventions).
5. Enter the logon ID. This is the user ID that will execute the job.
6. Enter the script name (including path).
7. Enter the command or script (including path). The path can be entered
in a parameter, in the .jobmanrc file, or in the command or script field.
8. If the job resides on an NT system and it requires a manual intervention,
Interactive must be selected.
9. Enter the Recovery Option. If the job does not complete successfully,
you must choose an action. The Stop option will wait for your manual
next step. The Continue option would not be used in this scenario.The
Rerun option allows the job to be executed again without manual
intervention. Figure 41 on page 62 displays the Job Definition window.


.

Figure 41. Job Definition window for storage pool backup

10. Select File->Save, and then File->Close .
To define a new schedule for this job, choose Composer, and then
select Schedules.
Our objective is to create a new schedule.
11. On Actions menu, click New.
12. Choose the associated CPU and enter the Schedule Name.
13. Select the date(s) or pre-defined calendar to execute the schedule as
well as its dependencies (if any).
14. Select File->Save, and then File->Close .
You now have a job and schedule in the TWS database to backup a
storage pool. There are several ways to execute your schedule.
a. If you have added this schedule and job in a production
environment, you can only submit this schedule in an ad hoc
fashion. Submit the job or schedule from the Conman window,
choosing Submit Schedule or Submit Job, answering the
necessary prompt.
b. If the schedule and job were created in a test environment, you can
rerun the Jnextday script from the command line. This will allow the
schedule and job to run in the Symphony file (or TWS production


day). The schedule will execute immediately (if no time was
entered) or it will execute on the given time.

Note
A new run number is generated when a Jnextday process is executed and
all executing jobs will appear as CF (carryforward schedules.)

To view the state of the job and schedule:
15. Go to Maestro (TWS) Main Window and click on Conman (Console
Manager).
16. Choose Schedules or Jobs to view the state of the job.
17. To display the standard output or result of the job, right-click the mouse
button and choose Stdlist.

4.1.2 Backup device configuration
You can use this command to backup information about device class
definitions, library definitions, and drive definitions. If you do not have a
library installed on your server, you can also back up information about your
disk.

In order to schedule backup device configuration using TWS, you must define
a new schedule and job. Follow the same steps described in Section 4.1.1,
“Database backup” on page 43 to prepare your job and schedule.

The Job Definition window for backup device configuration is seen in
Figure 42 on page 64.


Figure 42. Job Definition window for backup device configuration

Once you create the new schedule and job to back up a device configuration,
test your newly created schedule and job as in Section 4.1.1, “Database
backup” on page 43.

4.1.3 Backup volume history
You can use this command to back up sequential volumes to one or more
files.

To define a new job and a new schedule for this event, follow the steps
described in Section 4.1.1, “Database backup” on page 43.

The Job Definition window for backup volume history is seen in Figure 43 on
page 65.


Figure 43. Job Definition window for backup volume history

4.1.4 Clean volume history
Use this command to delete volume history file records that are no longer
needed (that is, obsolete databases).


The Job Definition window for clean volume history is seen in Figure 44 on
page 66.


Figure 44. Job Definition window for clean volume history

4.1.5 Expiration process
This inventory expiration process removes client backup and archive file
copies from server storage based on policy specified in the backup and
archive copy groups of the management classes to which the files are bound.


The Job Definition window for the expiration process is seen in Figure 45 on
page 67.


Figure 45. Job Definition window for the expiration process

4.1.6 Reclamation process
Reclamation makes the fragmented space on volumes usable again by
moving any remaining active files from one volume to another volume, thus
making the original volume available for reuse.


The Job Definition window for the reclamation process is seen in Figure 46 on
page 68.


Figure 46. Job Definition window for the reclamation process

4.1.7 Migration process
You can use a primary storage pool as the destination for backup files,
archive files, or files migrated from client nodes.


The Job Definition window for the migration process is seen in Figure 47 on
page 69.


Figure 47. Job Definition window for the migration process

4.1.8 Restore
You can restore files, a single database, databases, or volume history files.
For more information, see the Tivoli Storage Manager for AIX Administrator’s
Reference, GC35-0404.


The Job Definition window for restore is seen in Figure 48 on page 70.


Figure 48. Job Definition window for restore

4.2 Summary
In this chapter we investigated the following TSM functions that can be
scheduled by using TWS Extended agent for TSM.
• Backup device configuration
• Backup volume history
• Clean volume history
• Database backup
• Expiration process
• Migration process
• Reclamation process
• Restore

Instead of using the native TSM scheduler, we used our TWS custom
Extended agent for TSM to schedule these tasks through TWS. This way we
were able to use TWS’s extended scheduling capabilities.


Appendix A. TSW Extended agent for TSM source code

#!/bin/ksh

set +u

#-------------------------------------------------------------------------
---

# This method script executes commands on the Tivoli Storage Manager server

# for administrative and client backup purposes

#

# In order to use it, install it in the ~maestro/methods directory,

# and create an x-agent workstation definition that uses it. For

# example:

# CPUNAME TSMADMIN

# OS OTHER

# NODE dontcare

# TCPADDR 9000

# FOR MAESTRO

# HOST TWSFTA

# ACCESS tsmxagent

# END

#

# Create jobs for this x-agent with the following commands in the

# "script name" field on the job definition:

#

# ADMIN DBBACKUP type devclass

#

# where type is either full or incremental

# and where devclass is a devclass name


# defined in TSM.

# This will run a backup database command.

#

#

# ADMIN VHBACKUP filename

#

# where filename is a valid unix file for output

# This will run a backup volhistory command.

#

#

# ADMIN DCBACKUP filename

#

# where filename is a valid unix file for output.

# This will run a backup devconfig command.

#

# ADMIN DELVOLHIST type todate

#

# where type is a string for the type parameter

# and where todate is a date of form MM/DD/YYYY

# This will run a delete volhistory command.

# See the TSM Admin Reference for type parameter.

#

# ADMIN EXPIRE INVENTORY duration

#

# where duration is a numeric value

# This will run an expire inventory command.

#


#

# CLIENT client

#

# where client is the name of a server-prompted

# TSM client ready for scheduling

# This will run a client backup schedule for the

# client specified.

#-------------------------------------------------------------------------
---

PATH=$PATH:/bin:/usr/bin:/sbin:/usr/sbin:/etc/:/posix:/usr/ucb:/usr/bsd

export PATH

# Initialize some of the variables.

SCHEDULE_NAME=""

FORMATTED_SCHED_DATE=""

UNIVERSAL_DATE=""

NODE_NAME=""

JOB_NAME=""

JOB_ID=""

STREAM_LOGON=""

MAESTRO_CPU=""

MAESTRO_HOST=""

MAESTRO_MASTER=""

PORT_NUMBER=""

TEST_OPTIONS=""

CURRENT_RUN_NUMBER=""

REQUIRED_RUN_NUMBER=""

STDLIST=""

TASK=""

Appendix A. TSW Extended agent for TSM source code 73

PARENT="NO"

XARG1=""

XARG2=""

XARG3=""

TMPJOBNAME=""

JOBSTATUS=""

DEVCLASS=""

BACKUPTYPE=""

FILENAMES=""

TODATE=""

DURATION=""

RET9=""

DSMADMC="/usr/bin/dsmadmc"

# The number of seconds to pause between client status checks

CHECKFREQ=10

# The number of times to check client status

JOBTIMEOUT=120

if [ -z "$UNISON_EXEC_PATH" ]

then

PROG_NAME=`basename $0`

EXEC_PATH=`dirname $0`

else

PROG_NAME=`basename $UNISON_EXEC_PATH`

EXEC_PATH=`dirname $UNISON_EXEC_PATH`

fi

BANNER="TWS Extended agent for TSM"


MAE_HOME=`dirname $EXEC_PATH`

# Add maestro home to PATH.

PATH=$MAE_HOME/bin:$MAE_HOME:$PATH

export PATH

echoerr()

# Function to display error messages.

{

TIME=`date +%H:%M`

echo "${PROG_NAME}:$TIME/" "$@" >&2

}

# The name of the TSM adminstrator logon is in the tsmagent.opts file

if [ -f ${PROG_NAME}.opts ]

then

USERID=`awk -F= '$1 ~ /tsmAdmin/ { print $2 }' ${PROG_NAME}.opts`

else

echoerr "Using default admin logon"

USERID="admin"

fi

USAGE="Invalid Usage. See Tivoli Workload Scheduler Users Guide for correct
usage ."

print_version()

{

echo "$BANNER"

}

command_exist()

{

if type "$@" 2>&1 | grep -i "not *found" > /dev/null 2>&1

then


echoerr "Command $@ not found."

return 1

else

return 0

fi

}

Check_File()

# Check for existance of a file on the extended agent machine.

{

[ -z "$TEST_OPTIONS" ] && TEST_OPTIONS="-f"

test $TEST_OPTIONS "$FILE_NAME"

exit $?

}

fail_job()

{

send_Message "CJ FAIL"

exit 1

}

abend_job()

{

send_Message "CJ ABEND"

exit 1

}

succeed_job()

{

send_Message "JS"

exit 0


}

tsm_admin_command()

# Run dsmserv admin command

{

case $XARG2 in

DBBACKUP|dbbackup)

send_Message "CJ EXEC"

BACKUPTYPE=$XARG3

DEVCLASS=$XARG4

$DSMADMC -id=$USERID -password=$PASSWORD backup db

devclass=$DEVCLASS type=$BACKUPTYPE scratch=yes wait=yes

[ ! $? -eq 0 ] && fail_job

succeed_job

;;

VHBACKUP|vhbackup)


FILENAMES=$XARG3

$DSMADMC -id=$USERID -password=$PASSWORD backup volhistory

filenames=$FILENAMES

[ ! $? -eq 0 ] && fail_job

succeed_job

;;

DCBACKUP|dcbackup)


FILENAMES=$XARG3

$DSMADMC -id=$USERID -password=$PASSWORD backup devconfig

filenames=$FILENAMES wait=yes


[ ! $? -eq 0 ] && fail_job

succeed_job

;;

DELVOLHIST|delvolhist)


BACKUPTYPE=$XARG3

TODATE=$XARG4

$DSMADMC -id=$USERID -password=$PASSWORD delete volhistory

type=$BACKUPTYPE todate=$TODATE

[ ! $? -eq 0 ] && fail_job

succeed_job

;;

EXPIRE|expire)


DURATION=$XARG3

$DSMADMC -id=$USERID -password=$PASSWORD expire inventory

duration=$DURATION wait=yes

[ ! $? -eq 0 ] && fail_job

succeed_job

;;

esac

}

tsm_client_command()

# run tsm client backup

{

TEMPJOBNAME="TWS"$JOB_ID



$DSMADMC -id=$USERID -password=$PASSWORD define schedule

standard $TEMPJOBNAME

action=incremental startdate=today starttime=now dayofweek=any

expiration=TODAY

[ ! $? -eq 0 ] && fail_job

$DSMADMC -id=$USERID -password=$PASSWORD define association standard

$TEMPJOBNAME $XARG2

[ ! $? -eq 0 ] && fail_job

COUNTER=JOBTIMEOUT

while [ 1 ]

do

sleep $CHECKFREQ

COUNTER=`expr $COUNTER - 1`

if [ $COUNTER -eq 0 ]

then

$DSMADMC -id=$USERID -password=$PASSWORD delete schedule


fail_job

else

JOBSTATUS=`$DSMADMC -id=$USERID -password=$PASSWORD query event

standard * | grep $TEMPJOBNAME | grep Completed`

if [ ! -z "$JOBSTATUS" ]

then



succeed_job


fi

fi

done



[ ! $? -eq 0 ] && fail_job

}

Launch_Job()

# Launch the job

{

if [ $# -lt 2 ]; then

echoerr "Invalid job definition. Too few arguments!"

fail_job

fi

[ -z "$SCRIPT_NAME" ] && exit 1

touch "$STDLIST"

TMP_PATH=/tmp

XARG1=`echo "$SCRIPT_NAME" | cut -d' ' -f1`




case $1 in

ADMIN|admin)

tsm_admin_command $@

;;

CLIENT|client)

tsm_client_command $@


;;

*)

echoerr "Invalid job definition. Must be ADMIN or CLIENT!"

fail_job

esac

exit 0

}

Check_Connection()

#Check connection

{

exit 0

}

Manage_Job()

{

exit 1

}

send_Message()

{

echo "%" "$@"

return 0

}

echo $BANNER

if command_exist parms

then

PASSWORD=`parms TSMPASS`

else

echoerr "Could not find the TWS parms utility."


exit 3

fi

#Parse the arguments

while [ $# -gt 0 ]

do

OPT="$1"

[ "ècho %$OPT | cut -c2`" != "-" ] && break

shift

case $OPT in

-c)

OPT_ARG="$1"

shift

MAESTRO_CPU=ècho "$OPT_ARG" | cut -d',' -f1`

MAESTRO_HOST=ècho "$OPT_ARG" | cut -d',' -f2`

MAESTRO_MASTER=ècho "$OPT_ARG" | cut -d',' -f3`

;;

-d)

OPT_ARG="$1"

shift

FORMATTED_SCHED_DATE=ècho "$OPT_ARG" | cut -d',' -f1`

UNIVERSAL_DATE=ècho "$OPT_ARG" | cut -d',' -f2`

;;

-j)

OPT_ARG="$1"

shift

JOB_NAME=ècho "$OPT_ARG" | cut -d',' -f1`

JOB_ID=ècho "$OPT_ARG" | cut -d',' -f2`


;;

-l)

OPT_ARG="$1"

shift

STREAM_LOGON="$OPT_ARG"

;;

-n)

OPT_ARG="$1"

shift

NODE_NAME="$OPT_ARG"

;;

-o)

OPT_ARG="$1"

shift

STDLIST="$OPT_ARG"

;;

-p)

OPT_ARG="$1"

shift

PORT_NUMBER="$OPT_ARG"

;;

-q)

OPT_ARG="$1"

shift

TEST_OPTIONS="$OPT_ARG"

;;

-r)


OPT_ARG="$1"

shift

CURRENT_RUN_NUMBER=`echo "$OPT_ARG" | cut -d',' -f1`

REQUIRED_RUN_NUMBER=`echo "$OPT_ARG" | cut -d',' -f2`

;;

-s)

OPT_ARG="$1"

shift

SCHEDULE_NAME="$OPT_ARG"

;;

-t)

OPT_ARG="$1"

shift

TASK="$OPT_ARG"

;;

-V)

print_version

exit 0

;;

--)

break

;;

*)

echoerr "$USAGE"

exit 1

;;

esac


done

REMAINING_ARGS="$*"

if [ ! "$TASK" = "CC" ]

then

if [ "$#" = "0" ] || [ -z "$TASK" ]

then

echoerr "$USAGE"

exit 1

fi

fi

case $TASK in

CF)

FILE_NAME="$REMAINING_ARGS"

Check_File

;;

MJ)

JOB_PID=`echo "$REMAINING_ARGS" | cut -d" " -f1`

STDLIST=`echo "$REMAINING_ARGS" | cut -d" " -f2`

Manage_Job

;;

LJ)

SCRIPT_NAME="$REMAINING_ARGS"

send_Message "CJ WAIT"

Launch_Job $SCRIPT_NAME

;;

CC)

Check_Connection


;;

*)

echoerr $USAGE

send_Message "UT $USAGE"

exit 2

;;

esac

exit 3


Appendix B. Using the additional material

This redbook also contains Web material. See the appropriate section below
for instructions on using and downloading Web material.

B.1 Locating the additional material on the Internet
The Web material associated with this redbook is available in softcopy on the
Internet from the IBM Redbooks Web server. Point your Web browser to:

ftp://www.redbooks.ibm.com/redbooks/SG246030

Alternatively, you can go to the IBM Redbooks Web site at:

ibm.com/redbooks

Select Additional materials and open the directory that corresponds with the
redbook form number.

B.2 Using the Web material
The additional Web material that accompanies this redbook includes the
following:
File name Description
TSMX-agent.zip TWS Extended agent for Tivoli Storage Manager
code (zipped)
X-agent.zip Warren Gill’s presentation about how to write a
TWS Extended agent (zipped and PowerPoint
Viewer included)

B.2.1 How to use the Web material
Create a subdirectory (folder) on your workstation and copy the contents of
the Web material into this folder.


88 Implementing TWS Extended agent for Tivoil Storage Manager

Appendix C. Special notices

This publication is intended to help Tivoli professionals who want write a
Tivoli Workload Scheduler Extended agent or to integrate Tivoli Workload
Scheduler and Tivoli Storage Manager. The information in this publication is
not intended as the specification of any programming interfaces that are
provided by Tivoli Workload Scheduler or Tivoli Storage Manager. See the
PUBLICATIONS section of the IBM Programming Announcement for Tivoli
Workload Scheduler and Tivoli Storage Manager for more information about
what publications are considered to be product documentation.

References in this publication to IBM products, programs or services do not
imply that IBM intends to make these available in all countries in which IBM
operates. Any reference to an IBM product, program, or service is not
intended to state or imply that only IBM's product, program, or service may be
used. Any functionally equivalent program that does not infringe any of IBM's
intellectual property rights may be used instead of the IBM product, program
or service.

Information in this book was developed in conjunction with use of the
equipment specified, and is limited in application to those specific hardware
and software products and levels.

IBM may have patents or pending patent applications covering subject matter
in this document. The furnishing of this document does not give you any
license to these patents. You can send license inquiries, in writing, to the IBM
Director of Licensing, IBM Corporation, North Castle Drive, Armonk, NY
10504-1785.

Licensees of this program who wish to have information about it for the
purpose of enabling: (i) the exchange of information between independently
created programs and other programs (including this one) and (ii) the mutual
use of the information which has been exchanged, should contact IBM
Corporation, Dept. 600A, Mail Drop 1329, Somers, NY 10589 USA.

Such information may be available, subject to appropriate terms and
conditions, including in some cases, payment of a fee.

The information contained in this document has not been submitted to any
formal IBM test and is distributed AS IS. The use of this information or the
implementation of any of these techniques is a customer responsibility and
depends on the customer's ability to evaluate and integrate them into the
customer's operational environment. While each item may have been
reviewed by IBM for accuracy in a specific situation, there is no guarantee


that the same or similar results will be obtained elsewhere. Customers
attempting to adapt these techniques to their own environments do so at their
own risk.

Any pointers in this publication to external Web sites are provided for
convenience only and do not in any manner serve as an endorsement of
these Web sites.

The following terms are trademarks of the International Business Machines
Corporation in the United States and/or other countries:
IBM ® AS/400
CT Current
e (logo)® Lotus
Lotus Notes Notes
Netfinity Redbooks
Redbooks Logo RS/6000
System/390 XT

The following terms are trademarks of other companies:

Tivoli, Manage. Anything. Anywhere.,The Power To Manage., Anything.
Anywhere.,TME, NetView, Cross-Site, Tivoli Ready, Tivoli Certified, Planet
Tivoli, and Tivoli Enterprise are trademarks or registered trademarks of Tivoli
Systems Inc., an IBM company, in the United States, other countries, or
both. In Denmark, Tivoli is a trademark licensed from Kjøbenhavns Sommer
- Tivoli A/S.

C-bus is a trademark of Corollary, Inc. in the United States and/or other
countries.

Java and all Java-based trademarks and logos are trademarks or registered
trademarks of Sun Microsystems, Inc. in the United States and/or other
countries.

Microsoft, Windows, Windows NT, and the Windows logo are trademarks of
Microsoft Corporation in the United States and/or other countries.

PC Direct is a trademark of Ziff Communications Company in the United
States and/or other countries and is used by IBM Corporation under license.

ActionMedia, LANDesk, MMX, Pentium and ProShare are trademarks of Intel
Corporation in the United States and/or other countries.

UNIX is a registered trademark in the United States and other countries


licensed exclusively through The Open Group.

SET, SET Secure Electronic Transaction, and the SET Logo are trademarks
owned by SET Secure Electronic Transaction LLC.

Other company, product, and service names may be trademarks or service
marks of others.

Appendix C. Special notices 91

Appendix D. Related publications

The publications listed in this section are considered particularly suitable for a
more detailed discussion of the topics covered in this redbook.

D.1 IBM Redbooks collections
Redbooks are also available on the following CD-ROMs. Click the CD-ROMs
button at ibm.com/redbooks for information about all the CD-ROMs offered,
updates and formats.
CD-ROM Title Collection Kit
Number
IBM System/390 Redbooks Collection SK2T-2177
IBM Networking Redbooks Collection SK2T-6022
IBM Transaction Processing and Data Management Redbooks Collection SK2T-8038
IBM Lotus Redbooks Collection SK2T-8039
Tivoli Redbooks Collection SK2T-8044
IBM AS/400 Redbooks Collection SK2T-2849
IBM Netfinity Hardware and Software Redbooks Collection SK2T-8046
IBM RS/6000 Redbooks Collection SK2T-8043
IBM Application Development Redbooks Collection SK2T-8037
IBM Enterprise Storage and Systems Management Solutions SK3T-3694

D.2 Other resources
These publications are also relevant as further information sources:
• Tivoli Workload Scheduler 7.0 Reference Guide, GC32-0424
• Tivoli Workload Scheduler 7.0 Planning and Installation Guide,
GC32-0422
• Tivoli Workload Scheduler 7.0 User’s Guide,GC32-0423
• Tivoli Storage Manager for AIX Administrator’s Reference, GC35-0404

D.3 Referenced Web sites
These Web sites are also relevant as further information sources:
• http://www.tivoli.com/ Tivoli’s external Web site


How to get IBM Redbooks

This section explains how both customers and IBM employees can find out about IBM Redbooks,
redpieces, and CD-ROMs. A form for ordering books and CD-ROMs by fax or e-mail is also provided.
• Redbooks Web Site ibm.com/redbooks
Search for, view, download, or order hardcopy/CD-ROM Redbooks from the redbooks Web site.
Also read redpieces and download additional materials (code samples or diskette/CD-ROM images)
from this redbooks site.
Redpieces are redbooks in progress; not all redbooks become redpieces and sometimes just a few
chapters will be published this way. The intent is to get the information out much quicker than the
formal publishing process allows.
• E-mail Orders
Send orders by e-mail including information from the IBM Redbooks fax order form to:
e-mail address
In United States or Canada pubscan@us.ibm.com
Outside North America Contact information is in the “How to Order” section at this site:
http://www.elink.ibmlink.ibm.com/pbl/pbl
• Telephone Orders
United States (toll free) 1-800-879-2755
Canada (toll free) 1-800-IBM-4YOU
Outside North America Country coordinator phone number is in the “How to Order”
section at this site:
• Fax Orders
United States (toll free) 1-800-445-9269
Canada 1-403-267-4455
Outside North America Fax phone number is in the “How to Order” section at this site:

This information was current at the time of publication, but is continually subject to change. The latest
information may be found at the Redbooks Web site.

IBM Intranet for Employees
IBM employees may register for information on workshops, residencies, and Redbooks by accessing
the IBM Intranet Web site at http://w3.itso.ibm.com/ and clicking the ITSO Mailing List button.
Look in the Materials repository for workshops, presentations, papers, and Web pages developed
and written by the ITSO technical professionals; click the Additional Materials button. Employees may
access MyNews at http://w3.ibm.com/ for redbook, residency, and workshop announcements.


IBM Redbooks fax order form

Please send me the following:
Title Order Number Quantity

First name Last name

Company

Address

City Postal code Country

Telephone number Telefax number VAT number

Invoice to customer number

Credit card number

Credit card expiration date Card issued to Signature

We accept American Express, Diners, Eurocard, Master Card, and Visa. Payment by credit card not
available in all countries. Signature mandatory for credit card payment.


Abbreviations and acronyms

ADSM ADSTAR Distributed TMR Tivoli Management
Storage Manager Region

AIX Advanced Interactive TSM Tivoli Storage Manager
Executive TWS Tivoli Workload
API application Scheduler
programming interface
CF check file
CLI command line interface
CPU central processing unit
ESM Enterprise Systems
Management
FTA Fault-tolerant Agent
GRT Global Response Team
GS get status of job
GUI graphical user interface
IBM International Business
Machines Corporation
ITSO International Technical
Support Organization
JSC Job Scheduling
Console
LJ launch job
MJ manage job
MVS Multiple Virtual Storage
OPC Operations, Planning
and Control
PID process identification
RFC Remote Function Call
SAN Storage Area Networks
SLA service level agreement
TCP Transmission Control
Protocol
TME Tivoli Management
Environment


Symbols export volumes 30
.jobmanrc 46 Extended agent 4, 9, 10, 13
access method 10
access method interface 15
A API 1
abended 46
architecture 5
access method 4, 5
example 18
ADSM
executing the method 21
See ADSTAR Distributed Storage Manager
host 10
ADSTAR Distributed Storage Manager 8
interface 4
application programming interface 1
interface between TWS 4
archival storage 8
jobname 17
archive 7, 8
killing a job 21
logical workstation definition 10
B messages 20
Backup Master 3 method command line syntax 15
backup volumes 30 method options file 14
batch job execution 2 method troubleshooting 21
Batchman process 5 nodename 17
password 9
portnumber 17
C processing 5
CA7 7
qualifier 17
centralized database 3
referenced 10
CF task 14
sample options file 14
client backups 1
sample scenarios 43
command line interface 1, 13
segregation of applications 4
commit 34
special login information 14
common interface 4
stdlist 17
compiler 58
troubleshooting 21
user 17
D Windows NT 15
data protection 8 workstation definition 13
databases 8 external job 6
DEFINE ASSOCIATION 9 external system 13
DEFINE SCHEDULE 9
DELETE SCHEDULE 9
dependencies 1 F
fast recovery 7
device class definitions 29
Fault-tolerant Agent
disaster recovery 8
See FTA
DOCOMMAND 6
Follows Sched/Job option 53
Domain Manager 3, 4
foreign platforms 4
drive definitions 30
FTA 5
dump volumes 30
Full Status 3
duration 32

E G
groupware 8
e-mail 8

99

H NEXTSTGPOOL 28
hierarchical space management 8 node 14

I O
Interactive option 46 offset 52
OPC 4
open API 7
J OPENS dependency 6
Java 1
Opens Files 53
JES2 3
operator prompts 1
JES3 3
Oracle Applications 3, 7
job execution 1
job ID 6
job number 6 P
Job Scheduling Console 1, 2, 13 PeopleSoft 3, 7
job stream execution 1 physical workstation 13
jobman.exe 5 PID 18
jobmanrc 4 port definition 2, 6
jobmon.exe 5 production run number 6
JSC
See Job Scheduling Console
Q
QUERY EVENT 9
L
LAN-free backup 7
launch a job 6
R
Recovery Options 46
library definitions 29 remote console 3
line 13 Remote Function Call 5
LJ task 14 repeater 3
local dependencies 4 reptr 58
logical workstation 13 Resolve Dependencies 3
retrieval 7
M roll over 1
Maestro 34 root user ID 5
manage a job 6 run cycle 1
management hub 4
Master Domain Manager 3
method.opts 5
S
SAN 7
methods directory 4 SAP R/3 3
migration 7 Batch 7
MJ task 14 schedule name 6
MVS JES2 3 scheduling API 4
MVS JES3 3 scheduling protocol 4
schedulr 58
N SCRIPTNAME 6
Needs Resources 54 service level agreement 43
Network agent 3 shell script 4
network communication 5 skipdirs 32
SLA

100 Implementing TWS Ex tended agent for Tivoli Storage Manager

See service level agreement volume history 23
space management 7 TSM scenarios
stageman 58 backup device configuration 63
Standard agent 3 backup volume history 64
stdout 41 clean volume history 65
Storage Area Networks database backup 43
See SAN expiration process 66
storage resource management 8 migration process 68
subfile backup 7 reclamation process 67
Submit option 58 restore 69
Symphony file 62 TSMPASS 9
synchronize job 16 tsmxagent 9
tsmxagent.opts 9
TWS 2, 4
T architecture 3
tape sharing 7
Backup Master 3
tcpaddr 14
calendars 2
timeout 9
concepts 2
Tivoli Job Scheduling Services 2
Conman 57
Tivoli Management Framework 2
Connector 2
Tivoli Storage Management 8
CPU 38
Tivoli Storage Manager 11
current run number 17
See TSM
database 1
Tivoli TWS Connector 2
database files 3
troubleshooting 1
default user 34
TSM 7, 11, 23
Domain Manager 3
Administrative Client Interface 9
Extended agent 3
archive process 8
Fault-tolerant Agent 3
backup 23
Files 39
backup storage pool 25
job streams 2
backup volume history 41
Jobs 38
client backups 9
JSC Client 3
Composer 34
legacy user interface 43
data protection 8
Master 5
database 23, 31
Master Domain Manager 3
disaster recovery 8
methods directory 9
Extended agent benefits 9
network 2
incremental backup 7
plan 1
Main Window 34
production day 62
migration 23, 27
production plan 13
overview 7
Prompts 38
processes 23
Recovery Options 37
reclamation 23, 26
Resources 38
Restore Database 41
scaling 3
restore database 32
Schedules 38
retrieval process 8
scheduling API 4
scheduling facility 1
separate networks 3
scratch volumes 31
server 5
task 9
Standard agent 3

101

standard output 60
task options 16
terminology 2
types of workstations 3
versions prior to 7.0 43
workstations 2

U
UNIX Local 7
UNIX Remote Shell 7

V
View Filter option 47
virtual resources 1
vital record retention 8

W
wait 32
wgetmethod 7

102 Implementing TWS Ex tended agent for Tivoli Storage Manager

IBM Redbooks review
Your feedback is valued by the redbook authors. In particular we are interested in situations where a
redbook "made the difference" in a task or problem you encountered. Using one of the following
methods, please review the redbook, addressing value, subject matter, structure, depth and
quality as appropriate.
• Use the online Contact us review redbook form found at ibm.com/redbooks
• Fax this form to: USA International Access Code + 1 845 432 8264
• Send your comments in an Internet note to redbook@us.ibm.com

Document Number SG24-6030-00
Redbook Title Implementing TWS Extended agent for Tivoli Storage Manager

Review

What other subjects would you
like to see IBM Redbooks
address?

Please rate your overall O Very Good O Good O Average O Poor
satisfaction:

Please identify yourself as O Customer O Business Partner O Solution Developer
belonging to one of the O IBM, Lotus or Tivoli Employee
following groups: O None of the above

Your e-mail address:
The data you provide here may
be used to provide you with O Please do not use the information collected here for future
information from IBM or our marketing or promotional contacts or other communications beyond
business partners about our the scope of this transaction.
products, services or activities.

Questions about IBM’s privacy The following link explains how we protect your personal information.
policy? ibm.com/privacy/yourprivacy/


90<->249 pages
0.17”<->0.473”
(0.2”spine)
Implementing TWS Extended agent for Tivoli Storage Manager

®

Implementing TWS
Extended agent for Tivoli
Storage Manager
Insider’s guide to Tivoli Workload Scheduler (TWS) is Tivoli’s strategic,
writing a TWS multiplatform distributed scheduling product that provides
INTERNATIONAL
Extended agent high-volume, complex scheduling capability. Although TWS TECHNICAL
has native support for many platforms and applications, it is SUPPORT
Ready-to-use possible to extend the TWS’s robust scheduling capabilities to ORGANIZATION
cover platforms and applications beyond those TWS supports.
solution to integrate
The way to do this is to write an Extended agent. This redbook
TSM and TWS
shows you how to write a TWS Extended agent that allows you
to schedule on platforms and applications for which TWS has BUILDING TECHNICAL
TSM Extended agent no native agent. Tivoli Storage Manager (TSM) is used as a INFORMATION BASED ON
code included PRACTICAL EXPERIENCE
sample application to integrate with TWS using the Extended
agent. Using TWS’s scheduling facility allows you to assign
dependencies among TSM scheduled tasks or to assign limits IBM Redbooks are developed by
or priorities. By extending TWS to schedule these TSM tasks, the IBM International Technical
you can take advantage of its advanced scheduling Support Organization. Experts
from IBM, Customers and
capabilities.
Partners from around the world
create timely technical
This redbook will be essential for those who will write a TWS information based on realistic
Extended agent or will use TWS to schedule TSM functions. scenarios. Specific
recommendations are provided
to help you implement IT
solutions more effectively in
your environment.

For more information:
ibm.com/redbooks

SG24-6030-00 ISBN 0738419753

Implementing tws extended agent for tivoli storage manager sg246030

More Related Content

What's hot (17)

Similar to Implementing tws extended agent for tivoli storage manager sg246030 (20)

More from Banking at Ho Chi Minh city (20)

Recently uploaded (20)

Implementing tws extended agent for tivoli storage manager sg246030